The Nanoscience and Technology of Renewable Biomaterials

Document Sample
The Nanoscience and Technology of Renewable Biomaterials Powered By Docstoc
					The Nanoscience and
   Technology of
     Renewable
    Biomaterials

                       Edited by

 LUCIAN A. LUCIA AND ORLANDO J. ROJAS
      Department of Forest Biomaterials,
     North Carolina State University, USA




          A John Wiley and Sons, Ltd., Publication
The Nanoscience and Technology of
     Renewable Biomaterials
The Nanoscience and
   Technology of
     Renewable
    Biomaterials

                       Edited by

 LUCIAN A. LUCIA AND ORLANDO J. ROJAS
      Department of Forest Biomaterials,
     North Carolina State University, USA




          A John Wiley and Sons, Ltd., Publication
This edition first published 2009
c 2009 Blackwell Publishing Ltd
Registered office
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom
For details of our global editorial offices, for customer services and for information about how to apply for permission to
reuse the copyright material in this book please see our website at www.wiley.com.
The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs
and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form
or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright,
Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in
electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product
names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The
publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate
and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is
not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a
competent professional should be sought.
The publisher and the author make no representations or warranties with respect to the accuracy or completeness of the
contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness
for a particular purpose. This work is sold with the understanding that the publisher is not engaged in rendering professional
services. The advice and strategies contained herein may not be suitable for every situation. In view of ongoing research,
equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of
experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the
package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes
in the instructions or indication of usage and for added warnings and precautions. The fact that an organization or Website
is referred to in this work as a citation and/or a potential source of further information does not mean that the author or
the publisher endorses the information the organization or Website may provide or recommendations it may make. Further,
readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work
was written and when it is read. No warranty may be created or extended by any promotional statements for this work.
Neither the publisher nor the author shall be liable for any damages arising herefrom.
Copyright Acknowledgments
A number of articles in The Nanoscience and Technology of Renewable Biomaterials have been written by government
employees in the United States of America. Please contact the publisher for information on the copyright status of such
works, if required.
Works written by US government employees and classified as US Government Works are in the public domain in the United
States of America.
The contributions by Dr Goodrich and Dr Bhattacharya appearing in Chapter 8 on “Cellulose and Chitin as Nanoscopic
Biomaterials” are copyright of Eastman Chemical Company, all rights reserved.




Library of Congress Cataloging-in-Publication Data:
The nanoscience and technology of renewable biomaterials / edited by Lucian A. Lucia and Orlando J. Rojas.
    p. cm.
  Includes bibliographical references and index.
  ISBN 978-1-4051-6786-4 (cloth)
 1. Nanostructured materials. 2. Natural products. 3. Renewable natural resources. I. Lucia, Lucian A.                II. Rojas,
Orlando J.
  TA418.9.N35N3457 2007
  620.1 1—dc22
                                          2009009713
A catalogue record for this book is available from the British Library.
ISBN: 978-1-4051-6786-4
Typeset in 10/12pt Times-Roman by Laserwords Private Limited, Chennai, India.
Printed and bound in Singapore by Fabulous Printers Private Ltd
                              Contents


Preface                                                                      xi
Acknowledgements                                                            xv
Contributors                                                               xvii

 1 A Fundamental Review of the Relationships between
   Nanotechnology and Lignocellulosic Biomass                                1
   Theodore H. Wegner and E. Philip Jones
    1.1    Introduction                                                      1
    1.2    Use of Lignocellulosic-based Materials                            3
    1.3    Green Chemistry and Green Engineering                             4
    1.4    Nanotechnology                                                    6
    1.5    Nanotechnology-enabled Product Possibilities                      8
    1.6    Wood Nanodimensional Structure and Composition                   10
    1.7    Nanomanufacturing                                                11
    1.8    Nanotechnology Health and Safety Issues                          15
    1.9    Instrumentation, Metrology, and Standards for Nanotechnology     16
    1.10   A Nanotechnology Agenda for the Forest Products Industry         17
    1.11   Forest Products Industry Technology Priorities                   21
    1.12   Nanotechnology Priority Areas to Meet the Needs of the Forest
           Products Industry                                                23
           1.12.1 Achieving Lighter Weight, Higher Strength Materials       23
           1.12.2 Production of Nanocrystalline Cellulose and Nanofibrils
                    from Wood                                               25
           1.12.3 Controlling Water/Moisture Interactions with Cellulose    26
           1.12.4 Producing Hyperperformance Nanocomposites from
                    Nanocrystalline Cellulose                               29
           1.12.5 Capturing the Photonic and Piezoelectric Properties of
                    Lignocelluloses                                         30
           1.12.6 Reducing Energy Usage and Reducing Capital Costs in
                    Processing Wood to Products                             33
    1.13   Summary                                                          37
           References                                                       38

 2 Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine           43
   Candace H. Haigler and Alison W. Roberts
vi       Contents

          2.1   Introduction                                                     43
          2.2   Background                                                       44
          2.3   CesA Protein is a Major Component of the Plant CSC               45
          2.4   The Functional Operation of the CSC                              47
                2.4.1    Assemble with Genetically Determined Morphology         48
                2.4.2    Stabilize in Operational Form in the Plasma Membrane    50
                2.4.3    Acquire UDP-Glucose Substrate                           50
                2.4.4    Polymerize Glucose with β-1,4-Linkage                   51
                2.4.5    Operate so that Fibrils Emerge Outside the Plasma
                         Membrane                                                51
                2.4.6    Control Cellulose Chain Length                          51
                2.4.7    Control Cellulose Nanofibril Diameter                    52
                2.4.8    Control Crystallization?                                53
                2.4.9    Move in the Plasma Membrane as it Spins out Cellulose
                         Nanofibrils                                              53
          2.5   Phylogenetic Analysis                                            53
                2.5.1    Possible Functional Diversification of CS Proteins       53
          2.6   Conclusion                                                       55
                References                                                       55

     3    Tools for the Characterization of Biomass at the Nanometer Scale       61
          James F. Beecher, Christopher G. Hunt and J.Y. Zhu
          3.1   Introduction                                                     61
          3.2   Water in Biomass                                                 61
          3.3   Measurement of Specific Biomass Properties                        62
                3.3.1    Pore Structure and Accessibility                        62
                3.3.2    Cellulose Crystallinity                                 66
          3.4   Microscopy and Spectroscopy                                      68
                3.4.1    Specimen Preparation                                    68
                3.4.2    Scanning Probe Microscopies                             71
                3.4.3    Focused Beam Microscopies                               75
                3.4.4    Transmission Electron Microscopy                        78
          3.5   Summary                                                          80
                References                                                       80

     4    Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin
          Films: Applications in the Area of Adsorption and Friction             91
          Junlong Song, Yan Li, Juan P. Hinestroza and Orlando J. Rojas
          4.1   Introduction                                                     91
          4.2   Polyampholytes Applications in Fiber Modification                 92
          4.3   Cellulose Thin Films                                             95
          4.4   Friction Phenomena in Cellulose Systems                          97
          4.5   Lubrication                                                      98
          4.6   Boundary Layer Lubrication                                       99
                                                                      Contents    vii

          4.6.1    Thin Films: Property Changes and Transitions                   99
          4.6.2    Orientation of Lubricant Films                                101
   4.7    Techniques to Study Adsorption and Friction Phenomena                  102
   4.8    Surface Plasmon Resonance (SPR)                                        103
   4.9    Quartz Crystal Microbalance with Dissipation (QCM)                     105
   4.10   Application of SPR and QCM to Probe Adsorbed Films                     107
          4.10.1 Monitoring Adsorption and Desorption of
                   Macromolecules                                                107
          4.10.2 Conformation of Adsorbate Layers Revealed by the
                   QCM-D                                                         108
          4.10.3 Coupling QCM and SPR Data                                       109
   4.11   Lateral Force Microscopy                                               112
   4.12   Summary                                                                115
          Acknowledgements                                                       116
          References                                                             116

5 Polyelectrolyte Multilayers for Fibre Engineering                              123
                o                             a
  Rikard Lingstr¨ m, Erik Johansson and Lars W˚ gberg
   5.1    Background                                                             123
   5.2    The Formation of PEM on Wood Fibres                                    125
   5.3    Formation of PEM with Different Polyelectrolytes and the
          Properties of the Layers Formed                                        129
   5.4    Formation of PEM on Fibres                                             132
   5.5    Influence of PEM on Properties of Fibre Networks                        139
   5.6    Influence of PEM on Adhesion between Surfaces                           141
   5.7    Concluding Remarks                                                     144
          Acknowledgements                                                       145
          References                                                             145

6 Hemicelluloses at Interfaces: Some Aspects of the Interactions                 149
                                            ¨
  Tekla Tammelin, Arja Paananen and Monika Osterberg
   6.1    Overview                                                               149
   6.2    Introduction                                                           150
   6.3    Theoretical Basis for Interpreting QCM-D and AFM Data                  152
          6.3.1    QCM-D Data                                                    152
          6.3.2    Measuring Interaction Forces with AFM                         153
   6.4    Experimental                                                           154
          6.4.1    Materials                                                     154
          6.4.2    Methods                                                       155
   6.5    Results                                                                158
          6.5.1    Adsorption of Hemicelluloses on Cellulose                     158
          6.5.2    Viscoelastic Properties of the Hemicellulose Layers           160
          6.5.3    Effect of Xylan Adsorption on the Interaction between
                   Cellulose Beads                                               163
viii   Contents

             6.5.4   Effect of Electrolyte on the Interaction between
                     Xylan-coated Cellulose Surfaces                         164
       6.6   Discussion                                                      164
             6.6.1   Adsorption of Dissolved Hemicelluloses on Cellulose     164
             6.6.2   Adsorption Behavior and Interaction Forces between
                     Xylan and Cellulose                                     166
       6.7   Conclusions                                                     168
             Acknowledgements                                                168
             References                                                      168

  7    Lignin: Functional Biomaterial with Potential in Surface
       Chemistry and Nanoscience                                             173
       Shannon M. Notley and Magnus Norgren
       7.1   Introduction                                                    173
       7.2   Lignin Synthesis and Structural Aspects                         174
       7.3   Isolation of Lignin from Wood, Pulp and Pulping Liquors         177
             7.3.1    Isolation of Lignin from Wood and Pulp Fibres          178
             7.3.2    Isolation of Lignin from Spent Pulping Liquors         180
       7.4   Solution Properties of Kraft Lignin                             181
       7.5   Surface Chemistry of Solid State Lignin                         187
             7.5.1    Preparation and Properties of Lignin Thin Films        188
             7.5.2    Use of Lignin Thin Films for the Investigation of
                      Surface Chemical Properties                            191
       7.6   Lignin: Current and Future Uses                                 196
       7.7   Concluding Remarks                                              198
             References                                                      198

  8    Cellulose and Chitin as Nanoscopic Biomaterials                       207
       Jacob D. Goodrich, Deepanjan Bhattacharya and William T. Winter
       8.1   Overview                                                        207
       8.2   Introduction                                                    207
       8.3   Preparation and Microscopic Characterization of Cellulose and
             Chitin Nanoparticles                                            210
       8.4   NMR Characterization of Cellulose and Chitin Nanoparticles      214
       8.5   Chemical Modification of Cellulose and Chitin Nanoparticles      220
       8.6   Nanocomposite Properties                                        225
       8.7   Conclusions                                                     227
             Acknowledgements                                                228
             References                                                      228

  9    Bacterial Cellulose and Its Polymeric Nanocomposites                  231
       Marie-Pierre G. Laborie
       9.1   Introduction                                                    231
       9.2   Bacterial Cellulose: Biosynthesis and Basic Physical and
             Mechanical Properties                                           232
                                                                      Contents    ix

            9.2.1   Synthesis and Properties of BC                               232
            9.2.2   Performance of BC Mats                                       232
     9.3    BC Nanocomposites by in situ Polymerization                          234
            9.3.1   BC Nanocomposites with Thermosetting Phenolic and
                    Epoxy Resins                                                 234
            9.3.2   BC Nanocomposites with Acrylic Resins                        235
     9.4    BC Nanocomposites by Polymer Impregnation and Solution
            Casting                                                              242
            9.4.1   BC/Biopolymer Nanocomposites                                 243
            9.4.2   BC/Synthetic Polymer Nanocomposites                          247
     9.5    BC Nanocomposites via Biomimetic Approaches                          248
            9.5.1   BC/Xyloglucan Nanocomposites                                 250
            9.5.2   BC/Mannan Nanocomposites                                     255
            9.5.3   BC/Pectin Nanocomposites                                     257
            9.5.4   BC/Xyoglucan/Pectin Nanocomposites                           258
            9.5.5   BC/Lignin Nanocomposites                                     259
            9.5.6   BC/Synthetic Polymer Nanocomposites                          261
     9.6    BC/Polymer Nanocomposites Based on Bacterial Cellulose
            Nanocrystals                                                         263
     9.7    Conclusions and Prospects                                            266
            References                                                           267

10   Cellulose Nanocrystals in Polymer Matrices                                  273
     John Simonsen and Youssef Habibi
     10.1   Introduction                                                         273
     10.2   Background on CNXL Material Science                                  273
     10.3   Polymer Nanocomposite Systems                                        277
     10.4   Thermal Properties                                                   278
     10.5   Mechanical Properties                                                279
     10.6   Transport Properties                                                 283
            References                                                           287

11   Development and Application of Naturally Renewable Scaffold
     Materials for Bone Tissue Engineering                                       293
     Seth D. McCullen, Ariel D. Hanson, Lucian A. Lucia
     and Elizabeth G. Loboa
     11.1   Introduction                                                         293
     11.2   Natural Renewable Materials for Bone Tissue Engineering
            (BTE)                                                                295
     11.3   Bone Background                                                      296
            11.3.1 Progenitor Cells for Tissue Engineering Bone                  297
            11.3.2 Natural Renewable Materials Used for Bone Tissue
                     Engineering                                                 298
            11.3.3 Naturally Occurring Polysaccharide Materials in BTE           298
            11.3.4 Naturally Occurring Fibrous Protein Materials in BTE          301
x    Contents

                11.3.5  Naturally Occurring Inorganic Matrices in Bone Tissue
                        Engineering                                             304
      11.4      Conclusions and Future Directions                               306
                References                                                      306

12    Template Synthesis of Nanostructured Metals Using Cellulose
      Nanocrystal                                                               315
      Yongsoon Shin and Gregory J. Exarhos
      12.1      Overview                                                        315
      12.2      Introduction                                                    316
      12.3      Metal Oxide and Metal Carbides                                  317
                12.3.1 Porous Anatase                                           317
                12.3.2 SiC Nanorods                                             320
      12.4      Metal Nanoparticles on CNXL                                     321
                12.4.1 Transition Metal Nanoparticles                           321
                12.4.2 Precious Metal Nanoparticles: Ag, Au, Pd, Pt             324
                12.4.3 Nanocrystalline Se                                       326
      12.5      Conclusion                                                      330
                Acknowledgements                                                331
                References                                                      331

Index                                                                           337
                                    Preface


This book is a compilation of contributions in the area of nanoscience as it applies to
renewable biomaterials. It elegantly paints a broad picture of some of the nanotechnolog-
ical ramifications of such materials that have to some extent been previously overlooked,
despite the vast opportunities they can provide.
   Chapter 1 presents a fundamental review of the relationships between nanotechnology
and lignocellulosic (forest) biomass. The focus of this chapter involves the principles of
nanotechnologies that meet sustainable development, green chemistry, and green engi-
neering. The reader will soon come to realize the tremendous prospects trees possess as
photochemical factories that use air, sunlight and water to produce nanostructured mate-
rials that are used as building blocks for their own construction. Even more important
are the opportunities these resources present for the production of sustainable, renew-
able, recyclable and environmentally friendly products to meet the needs of modern-day
society. Wegner and Jones explain the vision for the nanotechnology of forest products
which encompasses the entire range of values that wood-based lignocellulosic materials
are capable of providing.
   While Chapter 1 makes clear the bright future of forest nanotechnology, Haigler and
Roberts in Chapter 2 review the biogenesis of some of the most important components
in biomass nanotechnology, namely, cellulose nanofibrils. This includes a marvelous
process whereby ß-1,4-linked glucan chains form long, semi-crystalline fibrils with
nanoscale lateral dimensions. The surface interactions of such components of cellu-
lose with other molecules are major determinants of its role as a scaffold for deposition
of other wall components. Later in Chapter 2, the regular association of cellulose
nanofibrils is presented as the product of one of nature’s most remarkable biological
nanomachines, a cellulose synthesis complex. This sets the stage for the advantageous
manipulation of cellulose properties in next-generation biomass plants and, possibly,
synthesis of cellulose in cell-free systems.
   In Chapter 3 Beecher, Hunt and Zhu discuss the tools that are available to unveil
the basic characterization of biomass components at the nanometer scale. This is
important because biomass is a difficult substrate to analyze at the nanoscale, yet
it is a necessity to fully appreciate the unique features of the plant cell walls. A
challenge here is that compared with most other organic macromolecules, the poly-
mers in lignocellulosic biomass form highly interconnected structures that are soft,
hydrophilic, and nonconducting. The basic interactions of water and biomass are there-
fore introduced and methods to evaluate nanoscale accessibility and reactivity of the cell
xii   Preface

wall presented. Additional reviews on the measurement of cellulose crystallinity and
microscopic and spectroscopic methods useful for the study of biomass at the nanoscale
are also discussed.
   One of the tools discussed in Chapter 3, namely, atomic force microscopy (AFM), is
employed by Song, Li, Hinestroza and Rojas in Chapter 4 to unveil surface interactions
involving cellulose. This chapter focuses on nanoscale surface phenomena in cellu-
lose thin films and discusses applications in the area of adsorption and friction. Thus,
after learning key characterization methods, the reader is exposed to a few examples
involving adsorbed polymers and surfactants to modify nanofilms of cellulose. This is
done to illustrate the unique possibilities available to alter or regulate surface properties
(such as surface energy, molecular assembly and composition) so as to modify adhesion,
colloidal stabilization, friction, and heterogeneous reactions. Two important additional
tools, namely, the Quartz Crystal Microbalance (QCM) and the Surface Plasmon Res-
onance (SPR) are conveniently introduced since they will be used in other chapters of
this book. Some of the advantages of these techniques include the possibility to obtain
fundamental information such as affinity of adsorbing molecules to the substrate, vis-
coelasticity of adsorbed layers, kinetics of adsorption and desorption, and the thickness
of the adsorbed layer as well as the amount of coupled water in adsorbed film. Finally,
lateral force microscopy is presented as a useful tool used to directly measure friction
on polymeric surfaces.
                         o                      a
   In Chapter 5, Lingstr¨ m, Johansson and W˚ gberg direct our attention to an additional,
very relevant application of cellulose surface modification: the buildup of polyelectrolyte
multilayers for uses in fiber engineering. Polyelectrolyte multilayers are used in several
applications, but the focus here is to engineer fiber surfaces at the nanoscale to enhance
interfiber bonding. The authors examine the relationship between the properties of the
formed multilayers and the enhanced adhesion by using the AFM colloidal probe tech-
nique. Overall, true nanoscale surface engineering is put to the test through macroscale
phenomena relevant to fiber processing.
   As cellulose was studied in the previous chapters, it is natural that the other impor-
tant components of the cell wall are also discussed in terms of their nanotechno-
logical relevance. We refer to the case of hemicelluloses and lignin. In Chapter
                                 ¨
6, Tammelin, Paananen and Osterberg revisit surface and interfacial phenomena, this
time using adsorbed hemicelluloses. This is highly relevant because there is growing
interest in the use of such polymers as a byproduct of the forest industries. There-
fore, the interfacial behavior of hemicelluloses is discussed to advance our under-
standing of the formation of films of these polysaccharides on cellulose. The authors
direct us to a specific application involving the adsorption of dissolved hemicellulose
fractions isolated from wood pulp as well as pure galactoglucomannan, pure pectin,
and pure xylan on cellulose nanofilms. With the combined results from QCM and
AFM experiments, the reader will deepen their understanding of the adsorption behav-
ior of different hemicelluloses and the properties of hemicellulose films on cellulose
surfaces.
   Lignin, in addition, is discussed by Notley and Norgren in Chapter 7 as a functional
biomaterial with potential in surface chemistry and nanoscience. This is very relevant
because of the large amount of this polymer that is produced in nature and industrial
                                                                               Preface   xiii

processing, with the vast majority used by humans as a fuel or in low value-added
applications. As with all aspects of nanoscience, molecular interactions are of great
importance, whether considering lignin as a polymer in solution or in the solid state, and
hence this chapter discusses the topochemical and interfacial properties of lignin. It is
concluded that lignin can provide many opportunities in nanotechnology.
   The panoply of nanoscopic renewable materials available in our research arsenal for
nanoscience/technology applications is not just limited to cellulosics and lignin, but
also includes chitin. Chitin is among the most abundant biopolymers in the biosphere
today, but only recently have we begun to realize its potential as a valuable biomaterial,
especially in the areas of structural and functional nanocomposites and biomedical appli-
cations such as wound healing, antibacterial activity, and joint lubrication. In Chapter 8
Goodrich, Bhattacharya and Winter describe nanocomposites based on chitin that give
an example of superior mechanical properties. Indeed, nanocomposites have become an
especially fertile avenue for advancements in materials research. For example, Laborie
in Chapter 9 demonstrates very elegantly how the smallest organisms amongst us, bac-
teria, can provide us with a very pure form of cellulose that not only has interest in
its own right, but can be easily combined with a number of other biomaterials (such as
xyloglucan, mannan, pectin, and lignin) to provide unique architectures; for example,
bacterial cellulose and lignin can be very effectively co-located into a bacterial cellulosic
mat matrix so that a better understanding of how cellulose and lignin deposit and arrange
can be obtained.
   The theme of composites pervades our text; in point of fact, the development of
new structures that are compatibilized or complement one another for improvements
in overall properties is an aspect of nanoscience and nanotechnology that cannot be
overemphasized. Nature has efficiently demonstrated the motif of well arranged and
ordered components in living, higher order structures as witnessed in the cell wall of
plants. In these assemblies, crystalline, hydrophilic units of cellulose are interspersed
with amorphous, hydrophobic lignin units which as a whole nevertheless maintain home-
ostasis and functionality. Simonsen and Habibi have chosen in Chapter 10 to explore
the nanoscopic, well ordered cellulose domains of a cellulosic array as part of an effort
to describe the inclusion of these crystalline structures into polymeric matrices. The lit-
erature is replete with work on cellulosic nanocrystals which have become an important
paradigm in the field of renewable materials that attempt to supplant petroleum-based
materials. Simonsen and Habibi have provided a concise overview and account of the
general area of cellulose nanocrystals, more specifically on their composites with various
polymers and their transport properties.
   We then venture to two elegant and sophisticated applications for the renewable mate-
rials of note. We are first presented in Chapter 11 with work from McCullen, Hanson,
Lucia and Loboa that explores the engineering of renewable materials for advanced
applications such as scaffolds for human tissue growth. Finally in Chapter 12, we con-
clude with the manipulation of cellulose nanocrystals by Shin and Exarhos to demonstrate
their ability to act as templates and reducing agents for synthesizing a host of nanoscopic
materials.
   Our sincere hope is that this volume serves as a useful platform and launching point
for all teachers and researchers who wish to begin harnessing the nanoscopic power of
xiv   Preface

renewable materials. Albert Einstein once said, ‘We shall require a substantially new
manner of thinking if mankind is to survive.’ We believe that the materials that nature
offers us provide us with that possibility – it is our duty, therefore, to investigate the
new paradigm of the nanoscience and technology of renewable biomaterials.

                                                  Lucian A. Lucia and Orlando J. Rojas
                      Acknowledgements


We would like to gratefully acknowledge a number of colleagues both at NC State
University and the Cellulose & Renewable Materials (CELL) Division of the American
Chemical Society who gave us their support during the production of this book. Without
their valuable input and suggestions, this book would not have been possible; we are
especially grateful to the CELL Division for allowing us to chair a symposium on this
topic several years ago in Atlanta, the fruits of which are collected in this archival
publication. Finally, we are pleased to acknowledge all contributing authors and their
institutions; they certainly place a very high value on ‘forest biomaterials’ research and
technology and have provided the inspiration to execute this work.
                           Contributors


James F. Beecher, US Forest Service, Forest Products Laboratory, Madison, USA

Deepanjan Bhattacharya, Eastman Chemical Company, CE Process Chemistry Group,
Kingsport, USA

Gregory J. Exarhos, Interfacial Chemistry & Engineering, Pacific Northwest National
Laboratory, Richland, USA

Jacob D. Goodrich, Eastman Chemical Company, CE Process Chemistry Group,
Kingsport, USA

Youssef Habibi, Department of Forest Biomaterials, North Carolina State University,
Raleigh, USA

Candace H. Haigler, Department of Crop Science and Department of Plant Biology,
North Carolina State University, Raleigh, USA

Ariel D. Hanson, Joint Department of Biomedical Engineering, UNC-Chapel Hill &
North Carolina State University, Raleigh, USA

Juan P. Hinestroza, Department of Fiber Science & Apparel Design, Cornell University,
Ithaca, USA

Christopher G. Hunt, US Forest Service, Forest Products Laboratory, Madison, USA

Erik Johansson, Department of Fibre and Polymer Technology, KTH – Royal Institute
of Technology, Stockholm, Sweden

E. Philip Jones, Americas Paper Division, Imerys, Roswell, USA

Marie-Pierre G. Laborie, Department of Civil and Environmental Engineering, Wood
Materials and Engineering Laboratory, Washington State University, Pullman, USA

Yan Li, Department of Fiber Science & Apparel Design, Cornell University, Ithaca,
USA
xviii   Contributors

                 o
Rikard Lingstr¨ m, Department of Fibre and Polymer Technology, KTH – Royal
Institute of Technology, Stockholm, Sweden

Elizabeth G. Loboa, Joint Department of Biomedical Engineering, UNC-Chapel Hill &
North Carolina State University, Raleigh, USA

Lucian A. Lucia, Department of Wood and Paper Science, North Carolina State
University, Raleigh, USA

Seth D. McCullen, Joint Department of Biomedical Engineering, UNC-Chapel Hill &
North Carolina State University, Raleigh, USA

Magnus Norgren, Department of Fibre and Polymer Technology, KTH – Royal Institute
of Technology, Stockholm, Sweden

Shannon M. Notley, Department of Applied Mathematics, Research School of Physics
and Engineering, Australian National University, Canberra, Australia

         ¨
Monika Osterberg, Laboratory of Forest Products Chemistry, Helsinki University of
Technology, Espoo, Finland

Arja Paananen, VTT Technical Research Centre of Finland, FI-02044 VTT, Finland

Alison W. Roberts, Department of Biological Sciences, University of Rhode Island,
Kingston, USA

Orlando J. Rojas, Department of Forest Biomaterials, North Carolina State University,
Raleigh, USA

Yongsoon Shin, Interfacial Chemistry & Engineering, Pacific Northwest National Lab-
oratory, Richland, USA

John Simonsen, Department of Wood Science & Engineering, Oregon State University,
Corvallis, USA

Junlong Song, Department of Forest Biomaterials, North Carolina State University,
Raleigh, USA

Tekla Tammelin, Laboratory of Forest Products Chemistry, Helsinki University of Tech-
nology, Espoo, Finland

        a
Lars W˚ gberg, Department of Fibre and Polymer Technology, KTH – Royal Institute
of Technology, Stockholm, Sweden
                                                                 Contributors   xix

Theodore H. Wegner, US Forest Service, Forest Products Laboratory, Madison, USA

William T. Winter, Cellulose Research Institute and Department of Chemistry, SUNY
College of Environmental Science and Forestry, Syracuse, USA

J. Y. Zhu, US Forest Service, Forest Products Laboratory, Madison, USA
                                                          1
        A Fundamental Review
     of the Relationships between
  Nanotechnology and Lignocellulosic
               Biomass

                                 Theodore H. Wegner and E. Philip Jones




1.1     Introduction

At first glance, the relationship between nanotechnology and lignocellulosic biomass
may seem to be unconnected or at best tenuously connected. It is important to recognize
that, at a fundamental level, lignocellulosic biomass is made up of nanometer-size con-
stitutive building block units that provide valuable properties to wood and other types
of renewable lignocellulosic and cellulosic biomaterials. Other composite biomaterials,
such as bone, teeth, and seashells, have been found to owe their high strength and
optical properties to the nanometer dimensions of their building blocks (Sarikaya et al.
2003). Similarly, the nanometer dimensions of the cellulose, lignin and other compo-
nents provide the origin for the unique properties of wood and a host of wood-based
products including paper, paperboard, oriented strandboard, glulam beams, etc. (Klemm
et al. 2005). For example, paper represents a material produced from fibers that have
been ‘pulped’ and refined to liberate fibrils, microfibrils/nanofibrils, and nanocrystalline
cellulose that are responsible for its inherent strength and performance (Brown et al.
1987). While the relative mass of the nanofibrils and nanocrystalline cellulose are small
their surface area is large and by number they represent an enormous fraction which has
significant consequences.

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
The contribution of Dr Wegner has been written in the course of his official duties as US government employee and is
classified as a US Government Work, which is in the public domain in the United States of America.
2   The Nanoscience and Technology of Renewable Biomaterials

   Nanotechnology holds great promise of revolutionizing materials use in the 21st
century, while lignocellulosic and like-derived biomass provide the key materials plat-
form for the sustainable production of renewable, recyclable, and environmentally prefer-
able goods and products to meet the needs of people in our modern society (Saxton
2007). Nanotechnology can be used to tap the enormous undeveloped potential that
trees possess – as photochemical ‘factories’ that produce rich sources of raw mate-
rials using sunlight and water. The merging of nanotechnology and lignocellulosic
biomass utilization is vital in sustainably meeting the needs of people for food, cloth-
ing, shelter, commerce, and the array of products and goods needed for quality of
life considerations both in meeting creature comfort needs but also ecological needs.
It is critically important to move forward nanotechnology involving renewable bio-
materials by: exploiting wood as an important sustainable and renewable industrial
nanomaterial; enabling other nanomaterials to be used in conjunction with lignocellu-
losic products to impart greater functionality; reducing materials use in producing, for
example, wood-based products; and reducing the environmental footprint for producing
such materials and products.
   The concepts of sustainability and sustainable development provide a convenient
contextual framework for examining the importance of the interrelationship of nan-
otechnology and biomass. Sustainability is many times viewed as a desired goal of
development and environmental management. The term ‘sustainability’ has been used in
a variety of disciplines and in numerous contexts, ranging from the concept of maximum
sustainable yield in lignocellulosic biomass management to the vision of a sustainable
society with a sustainable economy. The meaning of the term is strongly dependent
on the context in which it is applied and on whether its use is based on a social, eco-
nomic, or ecological perspective. Sustainability may be defined broadly or narrowly,
but any useful definition must explicitly specify the context as well as the temporal and
spatial scales being considered. Although societies differ in their conceptualizations of
sustainability, indefinite human survival requires basic support systems which can be
maintained only by a healthy environment and a sustainable use of resources. The def-
inition of sustainability is generally that defined by the 1987 Brundtland Commission
for sustainable development – meeting the needs of the present without compromising
the ability of future generations to meet their own needs (Brundtland 1987). Other
definitions include those of the World Business Council for Sustainable Development
who defines sustainable development as forms of progress that meet the needs of the
present without compromising the ability of future generations to meet their needs. With
respect to lignocellulose and lignocellulosic products (e.g. forests and forest products),
sustainability can be framed as asking whether those that come after us will be able
to enjoy the same or better values and benefits from lignocellulosics as we do today
(Society of American Foresters 2003) As we move forward in providing the goods
and services needed by the billions of people in our world, we must seek to be good
stewards of ecosystems locally, regionally, and globally; minimize the environmental
footprint of our modern society; and allow for raising the living standards and quality of
life for everyone. We must strive to achieve the preceding without hindering economic
and technological growth, development, and progress or hindering the ability of future
generations to meet their needs.
                                               Nanotechnology and Lignocellulosic Biomass – Relationships   3

1.2   Use of Lignocellulosic-based Materials

Matos and Wagner reported on the use of raw materials in the United States and noted a
trend away from renewable materials in the first half of the 20th century (Figure 1.1) as
population increased and the US economy moved from an agricultural to an industrial
base (Matos 1998). In the latter half of the century there was a large increase in raw
materials use as population continued to increase more rapidly and the economy began
a shift toward a service-based economy (Figure 1.2) (Matos and Wagner 1998). These
trends resulted in significant changes in the mix of raw materials used (Sznopek and
Brown 1998, Matthews and Hammond 1999).


                                        100
                                               Renewable materials

                                         80
            on a Per-weight Basis
             Percentage of Total,




                                         60

                                                                    Nonrenewable materials
                                         40


                                         20


                                          0
                                          1900 1910 1920 1930 1940 1950 1960 1970 1980        1995

          Figure 1.1 Renewable materials use in the United States, 1900–1995.



                                      800
                                                Industrial minerals
                                                Recycled metals
                                                Primary metals
                                                Nonrenewable organics
                                      600
                Million Metric Tons




                                                Recycled paper
                                                Primary paper
                                                Wood products
                                                Agriculture
                                      400



                                      200



                                         0
                                         1900 1910 1920 1930 1940 1950 1960 1970 1980        1995

                                      Figure 1.2 Materials use in the United States, 1900–1995.
4     The Nanoscience and Technology of Renewable Biomaterials

    It is noteworthy to observe in Figure 1.2 that the only renewable raw materials used in
significance tonnages to be readily visible are lignocellulosic-based materials (i.e. wood
products, primary paper and recycled paper). These raw materials represent approxi-
mately 20% of the materials consumed. Hence, when it comes to use of renewable
raw materials to produce products, lignocellulosic-based materials are extremely impor-
tant and represent the key renewable raw material link to sustainability. The use of
lignocellulosic-based materials provides the opportunity to produce functional materials
sustainably for an array of end uses from environmentally preferable or, in the worst
case, environmentally benign materials that have been commonly used by mankind for
millennia. Carbon dioxide (CO2 ) from the atmosphere under the effect of photosyn-
thesis in the tree produces the lignocellulosic materials we recognize as wood. In the
US, about 700 million tons (dry basis) of lignocellulosic forest biomass accumulate
annually. As a result, the standing stock of timber in the US continues to grow and is
currently over 20 billion dry tons. About 300 million tons of this is harvested annu-
ally, leaving a very large amount of biomass potentially available for conversion into
a variety of new products, energy, or chemicals. Indeed, the rates of current harvest
levels are not sufficient to contain the ravages of forest wildfires. Additionally, it has
been shown that it will be possible to increase production rate sustainably to levels of
1 billion tons per year through the application of advanced silviculture practices and
genetics to wood-based plantations on a portion of the forest lands in the US (Perlack
et al. 2006).
    Worldwide, forests provide a vast timber resource that is geographically and geopo-
litically dispersed among 150 countries (United Nations 2005). These 150 countries
account for 97.5% of the world’s forests. Globally, approximately 3.87 billion hectares
(ha) are covered by forests; out of a total land mass of 13.06 billion ha. The forests
of the world contain over 386 billion m3 of standing timber with annual use being on
the order of approximately 3.8 billion m3 per year. Globally, the gross value-added
by the forestry sector in 2000 (including forestry, logging and related activities, the
manufacturing of wood, wood products, paper and paper products) is estimated at about
US$354 billion, or about 1.2% of the world’s gross domestic product. The importance
of wood in the economy of the US and North America can not be understated. With
approximately 226 million ha of forestland, the US produces about 25% of the world’s
industrial roundwood. Together the US and Canada produce approximately 40% of the
world’s industrial roundwood.


1.3    Green Chemistry and Green Engineering

The use of lignocellulosic-based materials to produce products that meet the needs of
people in a sustainable and ecologically preferable manner is (1) based upon the efficient
use of solar energy and CO2 and (2) in keeping with the principles of both Green Chem-
istry and Green Engineering (Anastas and Warner 1998). Our current industries using
lignocellulosic products have evolved over many years. The technologies involved min-
imize waste and produce safe materials with minimal hazardous by-product generation.
After use, spent products can be recycled or will degrade with minimal environmental
consequences. The lignocellulosic forest products industry has installed tens of billions
                           Nanotechnology and Lignocellulosic Biomass – Relationships      5

of dollars of capital equipment to be in compliance with stringent environmental rules,
regulations and quality standards in its role as a responsible and trusted producer of
materials for our modern society. Nanotechnology, as it is envisioned for application
to lignocellulosic products, is only expected to further enhance industry’s ability to
produce consumer products from lignocellulosic-based materials in a safe, sustainable
manner in harmony with the principles of both Green Chemistry and Green Engineering.
As nanotechnology with respect to lignocellulosic-based materials moves forward, it is
important to know and adhere to the currently defined principles of both Green Chem-
istry and Green Engineering (Jenck et al. 2004, Schmidt 2007). The Principles of Green
Chemistry are as follows:
1.    Prevent waste – design chemical syntheses to prevent waste, leaving no waste to
      treat or clean up.
2.    Design safer chemicals and products – design chemical products to be fully effective,
      yet have little or no toxicity.
3.    Design less hazardous chemical syntheses – design syntheses to use and generate
      substances with little or no toxicity to humans and the environment.
4.    Use renewable feedstocks – use raw materials and feedstocks that are renewable
      rather than depleting. Renewable feedstocks are often made from agricultural prod-
      ucts or are the wastes of other processes; depleting feedstocks are made from fossil
      fuels (petroleum, natural gas, or coal) or are mined.
5.    Use catalysts, not stoichiometric reagents – minimize waste by using catalytic reac-
      tions. Catalysts are used in small amounts and can carry out a single reaction many
      times. They are preferable to stoichiometric reagents, which are used in excess and
      work only once.
6.    Avoid chemical derivatives – avoid using blocking or protecting groups or any tem-
      porary modifications if possible. Derivatives use additional reagents and generate
      waste.
7.    Maximize atom economy – design syntheses so that the final product contains the
      maximum proportion of the starting materials. There should be few, if any, wasted
      atoms.
8.    Use safer solvents and reaction conditions: Avoid using solvents, separation agents,
      or other auxiliary chemicals. If these chemicals are necessary, use innocuous
      chemicals.
9.    Increase energy efficiency – run chemical reactions at ambient temperature and pres-
      sure whenever possible.
10.   Design chemicals and products to degrade after use – design chemical products to
      break down to innocuous substances after use so that they do not accumulate in the
      environment.
11.   Analyze in real time to prevent pollution – include in-process real-time monitoring
      and control during syntheses to minimize or eliminate the formation of byproducts.
12.   Minimize the potential for accidents – design chemicals and their forms (solid, liquid,
      or gas) to minimize the potential for chemical accidents including explosions, fires,
      and releases to the environment.
As mentioned, the principles of Green Engineering are also important in taking advantage
of nanotechnology with lignocellulosic products and will lead to more socially acceptable
6     The Nanoscience and Technology of Renewable Biomaterials

materials and products derived from trees. The Principles of Green Engineering1 are as
follows (Schmidt 2007):
1. Engineer processes and products holistically, use systems analysis, and integrate envi-
   ronmental impact assessment tools.
2. Conserve and improve natural ecosystems while protecting human health and
   well-being.
3. Use life-cycle thinking in all engineering activities.
4. Ensure that all material and energy inputs and outputs are as inherently safe and
   benign as possible.
5. Minimize depletion of natural resources.
6. Strive to prevent waste.
7. Develop and apply engineering solutions, while being cognizant of local geography,
   aspirations, and cultures.
8. Create engineering solutions beyond current or dominant technologies; improve, inno-
   vate, and invent (technologies) to achieve sustainability.
9. Actively engage communities and stakeholders in development of engineering solu-
   tions.
As we apply nanotechnologies to the lignocellulosic products industry, we will need to
be cognizant of how the applications of these new technologies adhere to and advance
Green Chemistry and Green Engineering principles.



1.4     Nanotechnology

The ability to see materials at or near atomic dimensions and to measure physical
properties at these scales has enabled the emergence of a discipline now known as
Nanotechnology. At these scales and up to approximately 100 nm unusual properties
are often encountered. In addition many fundamental properties are driven by processes
scaled at the 10s of nm dimension. Many ‘natural products’ with valuable properties
such as silk, wool, nacre, wood and clay have building blocks that are 1 to 10s of nm
in dimension and owe their valuable properties to these nanometer-scale building blocks
(Roco 2003). Table 1.1 shows a short compilation of some of the key physical properties
and their dimensional dependencies.
   Already, there are over 700 nanomaterial-containing products available in the market
place, including coatings, computers, clothing, cosmetics, sports equipment, and medical
devices (Langsner 2006). The estimated global market for nanotechnology enabled
products was approximately US$9.4 billion in 2005, over US$10.5 billion in 2006, and
projected to grow to over US$25 billion by 2011 (Lux Research Inc. 2004, Hullmann
2006, Technology Transfer Center 2007). Nanomaterials – particularly nanoparticles
and nanocomposites – currently account for over 85% of the market. Currently used
nanomaterials include carbon nanotubes, carbon black fillers, nanocatalyst thin films,
nanodimensional additives, and nanoscale sensors.
   1
     As developed by more than 65 engineers and scientists at the Green Engineering: Defining the Principles Conference,
held in Sandestin, Florida in May 2003.
                             Nanotechnology and Lignocellulosic Biomass – Relationships   7

             Table 1.1 Characteristic lengths in solid-state science model.
             Property                                            Scale length
             Mechanics
               Dislocation interaction                            1–1000 nm
               Grain boundaries                                   1–10 nm
               Crack tip radii                                    1–100 nm
               Nucleation/growth defect                         0.1–10 nm
               Surface corrugation                                1–10 nm
             Supramolecules
               Kuhn length                                        1–100 nm
               Secondary structure                                1–10 nm
               Tertiary structure                                10–1000 nm
             Electronics
               Electronic wavelength                             10–100 nm
               Inelastic mean free path                           1–100 nm
               Tunneling                                          1–10 nm
             Magnetics
               Domain wall                                       10–100 nm
               Spin-flip scattering length                         1–100 nm
             Optics
               Quantum well                                       1–100 nm
               Evanescent wave decay length                      10–100 nm
               Metallic skin depth                               10–100 nm
             Source: Murday (2002); Pritkethly (2003).




   To date, major national research and development efforts have generally been focused
on how nanotechnology can improve efficiency in manufacturing, energy resources
and utilization, reduce environmental impacts of industry and transportation, enhance
healthcare, produce better pharmaceuticals, improve agriculture and food production,
and expand the capabilities of information technologies. Breakthroughs in nanoscale
science and engineering are seen as a foundation for systemic economic progress.
Nanotechnology is expected to lower raw materials costs in some industries; dramat-
ically improve productivity in others; create some entirely new industries; and increase
demand for some goods while lowering demand for others. Over the course of the 21st
century we will transition from the relatively crude and unsophisticated technologies
society depends upon today to highly efficient and environmentally friendly nanotech-
nologies. Industries that have been regarded as traditional, with few new scientific
challenges, are re-emerging as exciting new areas. The challenges involved with devel-
oping and industrially applying nanotechnology are enormous and it is only now that
we have the scientific tools to address biomaterials such as wood and paper. Maxi-
mizing human benefit will require the development of transformational tools that can be
shared across scientific disciplines and industries such as: new scientific instrumentation;
overarching theoretical concepts; methods of interdisciplinary communication; and new
techniques for production such as those bridging the gap between organic and inorganic
materials.
8     The Nanoscience and Technology of Renewable Biomaterials

   As we move forward, it is vitally important to use common nomenclature and defini-
tions with respect to nanotechnology (American Society for Testing and Materials 2006).
The following are the definitions generally used within the nanotechnology community:
• Nanoparticles – a particle with one or more dimensions at the nanoscale;
• Nanoscale – having one or more dimensions of the order of 100 nm or less;
• Nanoscience – the study of phenomena and manipulation of materials at atomic,
  molecular and macromolecular scales, where properties differ significantly from those
  at a larger scale;
• Nanotechnology – the design, characterization, production and application of struc-
  tures, devices and systems by controlling shape and size at the nanoscale;
• Nanostructured – having a structure at the nanoscale;
• Engineered nanoparticles – nanoparticles manufactured to have specific properties or
  a specific composition;
• Nanofiber – nanoparticles with two dimensions at the nanoscale and an aspect ratio
  of greater than 3:1;
• Quantum dot – a nanoscale particle that exhibits size-dependent electronic and optical
  properties due to quantum confinement;
• Nanocomposites – composites in which at least one of the phases has at least one
  dimension on the nanoscale;
• Nanophase – discrete phase, within a material, which is at the nanoscale;
• Bottom-up processing/manufacturing – additive processing/manufacturing to create
  nanostructures from atoms and molecules;
• Nanowire – a wire with diameter of the order of nanometers. Alternatively, nanowires
  can be defined as structures that have lateral size constrained to tens of nanometers
  or less and an unconstrained longitudinal size. At these scales, quantum mechanical
  effects are important – hence such wires are also known as quantum wires. Many dif-
  ferent types of nanowires exist, including metallic (e.g. Ni, Pt, Au), semi-conducting
  (e.g. Si, InP, GaN) and insulating (e.g. SiO2 , TiO2 ). Molecular nanowires are com-
  posed of repeating molecular units either organic (e.g. DNA) or inorganic (e.g. Mo6
  S9-x Ix ). Nanowires can be coiled and stretched to reach full length.
• Nanoribbon – a nanoribbon has a flat profile rather than the cylindrical profile of a
  nanowire. The thickness is generally of the order of tens of nanometers or less,
  while the width can be of the order of 10 to 100 nanometers and it has unconstrained
  longitudinal size. Like a nanowire, nanoribbons may be coiled and stretched to reach
  full length.



1.5    Nanotechnology-enabled Product Possibilities

By exploiting the full potential of lignocellulosic materials at the nanoscale, nanotech-
nology can provide benefits that extend well beyond fiber production and new materials
development into the areas of energy production, storage, and utilization. For example,
nanotechnology may provide new approaches for obtaining and utilizing energy from
sunlight – based on the operation of the plant cell. Novel new ways to produce energy
and other innovative products and processes from this renewable resource base will help
                          Nanotechnology and Lignocellulosic Biomass – Relationships      9

address some major issues facing many nations, including energy security, global climate
change, air and water quality, and global industrial competitiveness. Other potential uses
for nanotechnology include developing intelligent wood- and paper-based products with
an array of nanosensors built in to measure forces, loads, moisture levels, temperature,
pressure, chemical emissions, detect attack by wood decay fungi, etc. Building function-
ality onto lignocellulosic surfaces at the nanoscale could open new opportunities for such
things as pharmaceutical products, self-sterilizing surfaces, and electronic lignocellulosic
devices (Atalla et al. 2005). Use of lignocellulosic biomass nanodimensional building
blocks will enable the assembly of functional materials and substrates with substantially
higher strength properties, which will allow the production of lighter-weight products
from less material and with less energy requirements. Nano-biomaterials could replace
a wide range of materials such as metals and petroleum-based plastics in the fabrica-
tion of products. Significant improvements in surface properties and functionality will
be possible, making existing products much more effective and allowing the develop-
ment of many more new products. Nanotechnology can be used to improve processing
of wood-based materials into a myriad of products by improving water removal and
eliminating rewetting; reducing energy usage in drying; and tagging fibers, flakes, and
particles to allow customized property enhancement in processing. The exact economic
impacts and opportunities for wood as a nanomaterial are unknown, but it is expected
that nanomaterials and nano-enabled products will grow to exceed US$1 trillion per year
as the technology is further developed and is widely applied commercially during the
21st century (National Research Council 2006).
   Nanotechnology can also play an important role in the production of liquid biofuels
from lignocellulosic biomass. For example, nanoscale cell walls structures within trees
could be manipulated so they are more easily disassembled into constitutive materials
for liquid fuels production whether through conversion by fermentation, gasification, or
catalysis. Another approach would be to use nanocatalysis to break down recalcitrant
cellulose. Recalcitrant cellulose is in the order of 15–25% of the carbohydrate frac-
tion of wood and failure to convert this to sugars reduces fermentation ethanol yields.
In this approach, nanocatalysts would need to be transported to the reaction sites on
the solid substrate recalcitrant cellulose in order to produce water soluble polyol reac-
tion products. In most catalysis schemes, the reactants are brought to the catalyst.
In this case the catalyst needs to be brought to the solid substrate reaction sites and
water soluble reaction products need to be generated in order to permit recovery of the
catalysts. Other possibilities for nanotechnology approaches in biofuel production are
through development of engineered nanoscale enzymes or systems of enzymes (including
glycol hydrolases, expansins, and lignin degrading enzymes) for improved conversion
efficiency. Tree biology could be engineered so that enzymes and enzyme systems are
created and stored/sequestered in the living tree until harvest and then be activated for
engineered woody biomass self-disassembly. Lastly, another concept would be to create
new symbiotic nanoscale biological systems which work together to create ethanol or
other biofuels.
   Cellulose, while at times referred to as a nanofibril, does not differ much from a
coiled nanoribbon. Nanoribbons have been developed specifically as optical wave-
guides for channeling optical and visible light. Nanodimensional cellulose has already
been used as a template to form nanoribbons of Antimony (III) oxide that can then be
10    The Nanoscience and Technology of Renewable Biomaterials

used as an electrical wire (Ye et al. 2006). Nanocomposites for self-cleaning textiles
as well as solar cell applications have also been proposed. CdS nanowire has been
made using a nanocellulose derivative. Other applications could be possible if one were
able to combine other inorganics with cellulose (Venkataramanan and Kawanami 2006).
Cellulose has also been used for electrical devices (including artificial muscles), due
to its piezoelectric nature (Kim and Yun 2006). Termed smart cellulose, ‘electroactive
paper’ (EAPap) is a chemically treated paper with thin electrodes on both sides. When
electrical voltage is applied on the electrodes, the EAPap bends. Natural nanocellu-
lose has also been found to form layer by layer films with antireflective properties.
Multiwalled nanoribbon cellulose has also been used for wound dressing (Brown Jr.
et al. 2007).


1.6   Wood Nanodimensional Structure and Composition

Wood is a cellular hierarchical biocomposite (Figure 1.3) made up of cellulose, hemicel-
lulose, lignin, extractives and trace elements. Wood like many other biological tissues
including bones and teeth are hierarchically structured composites in order to provide
maximum strength with a minimum of material. At the nanoscale level, wood is a
cellulosic fibrillar composite. Wood is approximately 30–40% cellulose by weight
with about half of the cellulose in nanocrystalline form and half in amorphous form
(Figure 1.3g).
   Cellulose (Figure 1.3h) is the most common organic polymer in the world representing
about 1.5 × 1012 tons of the total annual biomass production. Cellulose is the major
carbohydrate component of wood along with the hemicelluloses (20–35% by weight).
Lignin, extractives, and trace amounts of other materials make up the remaining portion
of wood. Cellulose is expressed from enzyme rosettes as 3–5 nm diameter fibrils that
aggregate into larger microfibrils up to 20 nm in diameter (Figure 1.3g and 1.3f; see also
Chapter 2 of this book for more information on the cellulose biomachine). These fibrils
self-assemble in a manner similar to liquid crystals leading to nanodimensional and larger
structures seen in typical plant cell walls (Neville 1993, de Rodriguez et al. 2006). The
theoretical modulus of a cellulose molecule is around 250 GPa, but measurements for
the stiffness of cellulose in the cell wall are around 130 GPa. This means that cellulose
is a high performance material comparable with the best fibers technology can produce
(Vincent 2002).
   Because wood has a hierarchical structure, advances in separation techniques are
goaled at leading to the commercial production and use of multiple nanoscale architec-
tures namely nanocrystalline cellulose, nanofibrils, and nanoscale cell wall architectures
(Figure 1.3g and 1.3f) (Cash 2003). Nanofibrils in their simplest form are the elemen-
tary cellulosic fibrils shown in Figure 1.3g containing both crystalline and amorphous
segments and can be hundreds to a thousand or more nanometers long. Nanoscale cell
wall architectures are the larger nanodimensional structures depicted in Figures 1.3g
and 1.3f that are composed of multiple elementary nanofibril arrangements. Nanocrys-
talline cellulose is the liberated crystalline segments of elementary nanofibril crystalline
cellulose fibrils after the amorphous segments have been removed – usually via treat-
ment with strong acids at elevated temperature. Nanocrystalline cellulose is in the range
                                                   Nanotechnology and Lignocellulosic Biomass – Relationships                   11

                        Tree              Transverse Section                Growth Ring             Cellular Structure




      m


                                                                           mm
                                                     cm

                                                                                                                   500 µm
                                (a)                  (b)                               (c)                   (d)
      Cellulose                             Fibril Structure               Fibril-Matrix        Cell Wall Structure
               O
                   O
                           H
                                              Micro-Fibril                  Structure
                        OH
          H
                           H
                            H
       CH2OH
          H            OH                                                                                                  S3
          H
                   OH
                                                                                                                           S2
          HO           O
                           H
                                                                                                                           S1
          H
           H
                        CH2OH
                                            Amorphous                                                                      P
               OH
                   O
                           H                                                                                               ML
                           H
               O       OH
                                                             Crystalline
          H
                           H
       CH2OH
          H
                   O
                       OH
                                      Elementary
          H
                                         Fibrils
          HO           O
                           H                                                                                       25 µm
           H
          H             CH2OH
               OH          H
                O
                                1 nm                         10 nm                     300 nm
               (h)                                 (g)                           (f)                   (e)


        Figure 1.3 Wood hierarchical structure: from tree to cellulose (Moon 2006).

of 100–300 nm long. Nanocrystalline cellulose is anywhere from a tenth to a quarter
of the strength of carbon nanotubes (Xanthos 2005, Samir et al. 2005). Nanocrystalline
cellulose can also be referred to as nanowhiskers.
   The hierarchical structure of wood, based on its elementary nanofibrilar components,
leads to the unique strength and high performance properties of different species of wood.
While a great deal of valuable study has led to an understanding of many mechanisms
relating to the properties of wood and paper, the overall complexity of wood’s structure
has limited discovery. Today we have the tools used in other areas of nanotechnology
to look at structures down to the atomic scale. While this is fueling discovery in a
wide range of biomimetic materials, studies on wood are only now beginning. Simpler
structures found in seashells, insect cuticles and bones are being understood as relating
to their hierarchical structures. (Aizenberg et al. 2005) and we are poised for these
techniques to be applied to lignocellulosic-based products.


1.7   Nanomanufacturing

The value chain for lignocellulosic-based nanomaterials (Figure 1.4) is the same as for
any other materials – regardless of dimension (Hollman 2007, Langsner 2005). It is
based upon being able to profitably produce and sell products in the marketplace. While
the focus of nanotechnology-related research may seem to be on nanoscale properties
of materials, it is the nanotechnology-enabled macroscale end products that are most
12   The Nanoscience and Technology of Renewable Biomaterials

                        Value Chain for Nanotechnology-based Materials




            Nanomaterial(s)           Intermediate(s)         Nano-enabled Product(s)



                                      Increasing Value


               Figure 1.4 Value chain for nanotechnology-based materials.

important. Therefore, nanotechnology must be viewed as an enabling technology versus
an end in itself (Sixth Framework Programme 2005). To most expeditiously make
scientific and technology advancements, the focus for nanotechnology research must
always have an end use product application in mind. Examples of nanotechnology
enabled end use products include dimensional lumber with built in nanosensors to record
and react to static and dynamic loading; multifunctional siding materials that generate
electricity, are self-cleaning and self-sterilizing, and never need painting; and smart paper
that functions as a digital processor and accepts downloaded information. In the short
term, nanotechnologies for lignocellulosic products will likely be in the areas of barrier
coatings; architectural coatings; and preservative treatments.
   As nanotechnology science and technology develops during the 21st century, the first
applications will be use of passive nanostructures where the nanomaterial itself remains
static once it is encapsulated into the product. The second generation nanotechnology
enabled products will be in the area of active nanostructures where nanostructures change
their state during use by responding in predicable ways to the environment around them.
Third generation products are expected to be systems of nanosystems where assemblies
of nanotools self-assemble and work together to achieve a final goal. Lastly, molecular
nanosystems will be developed where through the intelligent design of molecular and
atomic devices there will be unprecedented understanding and control over the basic
building blocks of all natural and manmade things.
   Much of the current research focus in nanotechnology has been on measuring the prop-
erties of materials at the nanoscale with much focus on semiconductor materials, carbon
nanotubes, and medical applications – especially for diagnostics, cancer treatment, and
delivery of pharmaceuticals to targeted locations within the human body. Much less
emphasis has been placed upon other materials. Biological materials (e.g. wood and
plant materials) have received much less attention despite their many advantages such
as being able to self-assemble, being sustainable, and being ecologically preferable.
   The area of nanomanufacturing science and technology has also not received sufficient
attention despite its being one of the most critical pathways to applying the benefits of
nanotechnology. It is absolutely critical to build the nanomanufacturing science and
technology base to the point where nanomaterial(s) exhibiting unique nanoscale prop-
erties can: (1) routinely be placed into components or systems, (2) retain and combine
their unique nanoscale properties in a matrix of other materials and (3) result in superior
and controllable composites performance (NSET 2007, Department of Energy 2007).
                         Nanotechnology and Lignocellulosic Biomass – Relationships       13

   The extraordinary properties that make nanotechnology so important often also lead to
great difficulties in producing, separating, purifying, consolidating, handling, and mea-
suring nanomaterials. In addition, capturing and retaining nanoscale properties in the
final manufactured macroscale products also pose major obstacles to building products
from nanomaterials. Overcoming technical barriers to achieving cost-effective manufac-
ture of nanomaterials with unique properties and subsequently efficiently and effectively
capturing those properties in producing consumer end use products represent a number
of difficult tasks and can best be guided through researchers working with industrial
partners who can help guide the research efforts into the most economically viable path-
ways and relatively quickly determine if the proposed solution makes economic sense.
Nanomanufacturing technology development will require overcoming major barriers in
materials production and manufacturing and process control as well as predictive mod-
eling. For example, manufacturing products from nanomaterials is challenging because
it has been observed that nanopowders, -solids, and -suspensions have a high propen-
sity to agglomerate; have highly reactive surfaces; and have a fundamental tendency to
change properties with time, temperature, and handling conditions. Equally challenging
is that once nanomaterials are embedded into fibers, sheets, tubes, bars, or other forms,
there is limited technology to join these into useful forms without altering the properties
at the joint or interface. Additionally, when morphology is important to nano-enabled
product performance, it is difficult to obtain this quality throughout the module or body.
A listing of generic technical barriers includes the following (NSET 2007, Department
of Energy 2007):
• being able to commercially and reproducibly manufacture uniform, high quality, con-
  sistent nanomaterials in high volume;
• difficulty in developing economically-viable and scalable unit operations and incorpo-
  ration of nanomaterials into products make many nanomaterials prohibitively expen-
  sive for many applications due to high capital costs and low production volumes;
• difficulty or inability to retain nanomaterial functionality as the material is incorporated
  into products;
• difficulty in incorporating and controlling admixes of nanomaterials into other bulk
  materials;
• process-monitoring tools tailored for analyzing the unique characteristics and satisfy-
  ing the process control needs of producing nanomaterials are lacking; for example,
  real-time, in-line measurement techniques are needed;
• predictive models of nanomaterials behavior are needed for correlations between nano-
  materials properties and end-use performance as a cost-effective aid to design of
  nanomanufacturing processes.
In developing the needed nanomanufacturing technologies, greater industrial influence
and awareness also serves to help guide research into the highest priority and most pro-
ductive areas. For example, production of nano-enable composites is an area that is a
high priority for a number of industry sectors in addition to the forest products industry.
Without developing the science and technology for nanomanufacturing and successfully
incorporating nanomaterials into macroscale products for consumers, nanotechnology
will be primarily a laboratory curiosity. We must be able to reliably, reproducibly, and
cost-effectively produce composite matrices of bulk materials and nanomaterials that
14   The Nanoscience and Technology of Renewable Biomaterials

effectively combine the properties that the individual nanomaterial components and bulk
matrix materials possess. This requires developing a whole new array of nanomanu-
facturing technologies and the fundamental science that underlies them. For example,
when nanomaterials are used to produce products, they need to be able to be controllably
dispersed or mixed into other materials and retain their functionality in the bulk matrix.
Following is a listing of science-based needs with respect to using nanomaterials for com-
mercially producing composite matrices across a number of industries product sectors
including the forest products industry and when producing or using lignocellulosic-based
nanomaterials.
• Develop the science and technologies needed to control and manipulate dispersion of
  nanomaterials into a matrix of other materials.
• Develop the tools needed to adequately and easily measure and characterize nanoma-
  terial dispersion and mixing with other materials into a matrix to include degree of
  nanomaterial dispersion/aggregation.
• Determine how to overcome the deleterious effects of increasing production scale on
  dispersion/mixing of nanomaterials into a matrix of other materials.
• Develop robust online, real-time, in-situ characterization tools and methodologies to
  characterize dispersion and mixing of nanomaterials into a bulk matrix of other mate-
  rials (polar and nonpolar liquids, suspensions, solids and gases).
• Preserve the functionality of nanomaterials (e.g. strength; optical; magnetic, elec-
  trical/electronic; thermodynamic, chemical reactivity, catalysis, etc.) when they are
  incorporated into other materials.
• Develop the science needed to overcome the deleterious effects of high temperature
  processing on admixed nanomaterial properties.
• Understand the interactions between nanofibrils and bulk matrix materials that are
  most important to nanofibril reinforcement to include nanofibril morphology (e.g.
  size, shape aspect ratio, etc.), nanofibril loading level, and surface energies.
• Learn how to control nanofibril orientation in matrices bulk materials.
• Understand how varying composite synthesis methodologies (e.g. extrusion, solvent
  casting, high shear mixing) impact matrix properties.
• Measure the rheological properties of mixtures of nanomaterials (nano-, micro-, and
  macroscales) and bulk matrix materials and the effects on dispersion and mixing.
• Determine methodologies to adequately characterize nano-enable composite matrices.
• Determine the impact of aging and storage on nanocomposites properties.
• Develop multiscale (macro-, micro-, nano-) models that allow the prediction of the
  properties of composite matrices incorporating nanomaterials (to include maximum
  theoretical nanomaterial influence on matrix properties) and allow use of micro- and
  macroscale tests to correlate with nanoscale dispersion/mixing.
• Characterize nanoscale architectures of nanomaterials interacting with bulk matrix
  materials.
• Develop nondestructive quality control testing methods for composites containing
  nanomaterials.
• Develop a database of standardized nanomaterials/matrices properties.
• Develop process control tools for producing nanomaterial/bulk matrix composites.
                        Nanotechnology and Lignocellulosic Biomass – Relationships    15

1.8   Nanotechnology Health and Safety Issues

Environmental, health, and safety issues related to nanomaterials and nanotechnology
have been under explored but have more recently received much public attention (Davies
2006, Greenwood 2007, National Research Council 2006). Nanomaterials are present in
our daily lives (e.g. dust, smoke, ash, soot, etc.) and human exposure to nanodimen-
sional materials has occurred throughout human history. For example, nanomaterials are
produced by combustion of fuels and even by volcanic eruptions. The concerns for the
environmental, health, and safety aspects of nanotechnology arise from the production of
new engineered nanomaterials with unique properties (Friends of the Earth 2008). The
aim in using engineered nanomaterials to produce products must be to maximize benefits
while guarding against potential harm, based on a realistic assessment of technical facts
in the light of human values. Understanding the health risks and risks to the environment
or ecosystem that may result from exposure to or introduction of engineered nanoscale
materials, nanostructured materials, or nanotechnology-based devices is an extremely
important consideration in moving nanotechnology forward (International Risk Gover-
nance Council 2007, NSET 2008). This is not only true for wood-based nanomaterials
but also for nanomaterials and devices from other industry sectors that are incorporated
into forest products. An array of concerns arise with respect the effects of exposure
of nanomaterials and nanoproducts on human health and the environment. Included in
these concerns are the following items:
• determining the toxicology of nanomaterials and nanoproducts to humans and in the
  environment;
• determining the mechanisms for uptake and the biokinetics of nanomaterials in organ-
  isms and the human body;
• understanding transport, transformation, and the fate of nanomaterials and nanoprod-
  ucts in air, water, and soil to include mechanisms and routes of exposure;
• understanding dose metrics on humans and animals of nanomaterials and nano-
  products;
• implementing effective protection and long-term exposure safety measures for workers
  handling and working with nanomaterials and nanoproducts;
  – understanding the properties of nanomaterials as they relate to the effectiveness of
    personal protective equipment;
  – developing sensors and monitors to sample the workplace environment to determine
    workers exposure to nanomaterials and nanoproducts;
  – development of methods to control exposure when working with nanoproducts;
• developing life cycle analyses of nanomaterials and nanoproducts;
• measuring and characterizing nanomaterial key properties such as size, surface area,
  bioactivity, etc.;
• developing scientifically sound in vivo and in vitro protocols and models to understand
  nanomaterial interactions at the molecular and cellular level.
At first glance – for lignocellulosic-based nanomaterials – one would probably expect
that such materials being produced by living organisms should not pose an environ-
mental, health or safety problem. The concern about environmental, health and safety
16    The Nanoscience and Technology of Renewable Biomaterials

issues, however, goes to the heart of the definition of nanotechnology. If nanoscale
properties of wood-based nanomaterials exhibit new and unique properties that depend
upon size, then the environmental, health and safety impacts cannot necessarily be
assumed as known based upon existing information for wood micro- and macroscale
particles. For wood-based nanomaterials, we do not have a large body of risk assess-
ment data. The information that is available on the environmental, health, and safety
risks for lignocellulosic-based nanomaterials to date does not necessarily mean these
lignocellulosic-based nanomaterials pose a risk but there is no clear evidence to rule
such concerns out either. Clearly, more information needs to be developed. The best
way to obtain reliable data is to work with the larger nanotechnology community dealing
with the environmental health and safety of nanomaterials as they develop scientifically
sound protocols and procedures. In the meantime, for researchers and others working
in the area of nanotechnology, sound chemical laboratory practices should be employed,
using nanomaterials in fume hoods or glove boxes, using respirators or at minimum dust
masks, and disposing of nanomaterials in a manner equivalent to hazardous materials
disposal.


1.9   Instrumentation, Metrology, and Standards for Nanotechnology

Instrumentation, metrology, and standards for nanotechnology are critical components in
the chain from discovery of nanomaterials to commercialization of nano-enabled prod-
ucts. Today’s array of metrology tools has been developed to meet the current needs
of exploratory nanoscale research, primarily for inorganic materials and we are near-
ing the limits of their resolution and accuracy. While the tools currently available will
continue to evolve, they are not expected to meet all the future metrology require-
ments of nanoscale research, development, and deployment. Instrumentation to probe
the nanoscale requires revolutionary developments in addition to evolutionary advances
in measurement schemes and instruments (NSET 2006). The immediate tasks at hand
are to adapt currently available nanoscale metrology and instrumentation to biological
materials and obtain artifact-free property measurements. In addition, it is important to
be able to measure the nanoscale properties of wood and wood-based materials in-situ
and to relate wood and wood-based material properties within the context of wood’s
hierarchal composite structure. Instrumentation, metrology, and standards priorities for
nanotechnology include development of:
• next generation techniques, tools, and instruments that provide a major leap forward
  with regard to exceeding today’s spatial and temporal resolution limits; major empha-
  sis needs to be placed on advancing the state-of-the-art of microscopies and analytical
  instrumentation such as with scanning probe microscopes, scanning and transmis-
  sion electron microscopes, and electron, neutron, and photon spectroscopic techniques
  adapted to biological materials; the effects of electrons, neutrons, and photons on bio-
  logical materials can be quite different than with inorganic materials such as metals
  and this can lead to artifacts in the measurements;
• enabling full three-dimensional mapping of biological and nonbiological nanomate-
  rials using instrumentation combining subnanometer spatial resolution with chemical
  specificity and volumetric detection;
                          Nanotechnology and Lignocellulosic Biomass – Relationships              17

• enhancement of national and international nanometrology infrastructure supporting
  commercial manufacture of advanced products;
• standardization of measurement techniques, nomenclature, and testing methodologies
  to facilitate assurance of safety and efficacy of nanoproducts, and their effective reg-
  ulation and production;
• development and verification of advanced simulation, visualization, and data analysis
  techniques and supporting standards for biological materials;
• development of nomenclature to define the growing number of nanostructures that
  is succinctly, precisely, and compatibly related to that of conventional molecular
  nomenclature.


1.10    A Nanotechnology Agenda for the Forest Products Industry

In its first step toward reaching the goals of applying nanotechnology in the forest
products industry, a workshop to develop a vision, explore opportunities, and determine
research needs was convened in October 2004. The American Forest and Paper Asso-
ciation’s (AF&PA) Agenda 2020 Technology Alliance, the Technical Association of the
Pulp and Paper Industry, US Department of Energy (DOE) and the US Department of
Agriculture (USDA), Forest Service sponsored the conference. AF&PA Agenda 2020 is
a special project of the American Forest and Paper Association, the national trade asso-
ciation of the forest products industry. It is an industry-led partnership with government
agencies and departments such as the USDA Forest Service, DOE, the National Science
Foundation, the National Institute for Standards and Technology, etc. and academia.
The overall goal of the AF&PA Agenda 2020 Technology Alliance is to create options
to meet industry’s competitive challenges while contributing solutions to strategic
national needs associated with energy, the environment, and the economy by addressing
shared industry and national strategic goals; developing research, development and
deployment (RD&D) initiatives; provide the foundation for new technology-driven
business models; and leverage collaborative partnerships to drive innovation in the
forest products industry’s processes, materials, and markets. It is vitally important
to remember that close collaboration with the forest products industry is critical for
advancing nanotechnology. This is because the range and magnitude of benefits offered
by nanotechnology science and engineering research and development can only be
realized if the technologies are accepted and implemented (i.e. deployed) by the
industry.
   Over 110 leading researchers with diverse expertise from industry, government lab-
oratories, and academic institutions from North America and Europe attended the nan-
otechnology for the forest products industry workshop. Coming out of this workshop
was a document entitled Nanotechnology for the Forest Products Industry-Vision and
Technology Roadmap (Atalla et al. 2005). The stated vision for nanotechnology for the
forest products industry is as follows:

   To sustainably meet the needs of present and future generations for wood-based materials and
   products by applying nanotechnology science and engineering to efficiently and effectively
   capture the entire range of values that lignocellulosic materials are capable of providing.
18   The Nanoscience and Technology of Renewable Biomaterials

Workshop participants next identified the unique properties and characteristics of wood
lignocellulosic biopolymers that make them an exciting avenue for nanotechnology
research, including:
1. lignocellulosic biopolymers are some of the most abundant biological raw materials,
   have a nanofibrilar structure, have the potential to be made multifunctional, and can
   be controlled in self-assembly;
2. new analytical techniques adapted to biomaterials are allowing us to see the structure
   of wood in new ways;
3. lignocelluloses as nanomaterials and their interaction with other nanomaterials are
   largely unexplored.
Nanotechnology research and development strategies were also discussed and encom-
passed the following two broad approaches (Atalla et al. 2005):
1. Nanotechnologies and nanomaterials developed through nanotechnology research and
   development (R&D) efforts in other industry sectors will be adopted and deployed into
   materials, processes and products used in or produced by the current forest products
   industry. The expected gains of this R&D strategy direction were in improving
   existing products and processes – with some minor-to-moderate modifications and
   additions.
2. Nanotechnology R&D will develop completely new materials or product platforms
   using the improved knowledge of nanoscale structures and properties of the ligno-
   cellulosic wood-based materials used in the forest products industry. This direc-
   tion potentially will lead to radically different products, processing techniques, and
   material applications as the nanoscale properties of lignocellulose and its nanoscale
   architecture have not been exploited to any great degree.
The research challenges associated with these two broad strategies were identified and
span a range of scientific focus areas to include:
• developing fundamental understanding of nanomaterials and analytical tools for mea-
  suring properties at the nanoscale;
• developing new nanoscale building materials;
• developing nanotechnology for manufacturing applications;
• creating nanomaterials by design.
‘Nanomaterials by Design’ is a uniquely solutions-based research goal. As described
in the nanomaterials roadmap developed by the chemicals industry, ‘nanomaterials
by design’ refers to the ability to employ scientific principles in deliberately creating
structures (e.g., size, architecture) that deliver unique functionality and utility for tar-
get applications (Chemical Industry Vision2020 2003). This research area focuses on
the assembly of building blocks to produce nanomaterials in technically useful forms,
such as bulk nanostructured materials, dispersions, composites, and spatially resolved,
ordered nanostructures. It yields a new set of tools that can provide nearly limitless
flexibility for precisely building material functions around end-use applications. Such
a powerful, function-based design capability holds the potential to solve critical, unmet
needs throughout society. Techniques being developed in the areas of self-assembly
and directed self-assembly will allow us to use the building blocks available in the
                        Nanotechnology and Lignocellulosic Biomass – Relationships      19

forest products industry to manufacture materials with radically different performance
properties.
   The following R&D focus areas were initially selected on the basis that they (1) pro-
vide the best path forward for a nanotechnology roadmap by identifying the underlying
science and technology needed, and (2) foster essential interactions among visionary,
interdisciplinary research and technology leaders from industry, academia, research insti-
tutions, and government (Atalla et al. 2005).
1. Polymer composites and nano-reinforced materials – combine wood-based materials
   with nanoscale materials to develop new or improved composite materials with unique
   multifunctional properties.
    • Develop and investigate novel materials with enhanced properties (e.g. films, coat-
      ings, fillers, matrices, pigments, additives, and fibers – especially lignocellulosic
      nanofibrils.
    • Develop and investigate novel materials for processing equipment.
    • Develop and understand the interrelationships between nanoscale material charac-
      teristics and the resulting product end use property improvements.
    • Determine the best way to implement new materials.
    • Develop economic and life-cycle models for forest-based nanoscale materials and
      products.
2. Self-assembly and biomimetics – use the natural systems of woody plants as either the
   source of inspiration or the template for developing or manipulating unique nano-,
   micro-, and macroscale polymer composites via biomimicry and/or direct assembly
   of molecules.
    • Develop a technical platform enabling self-assembly of paper products and other
      lignocellulosic materials at the nanoscale.
    • On existing lignocellulosic substrates create novel, functional, self-assembling sur-
      faces.
    • Develop a fundamental understanding of molecular recognition in plant growth and
      cell wall self-assembly to create new or enhance existing products.
    • Learn to characterize self-assembled natural and synthetic material and to integrate
      micro- and nanoscale organization in products.
3. Cell wall nanostructures – manipulate cell wall nanostructure of woody plants in order
   to modify or enhance their physical properties and create wood and wood fibers with
   superior manufacturability or end-use performance.
    • Investigate the process of formation of cellulose nanofibrils, including genetic,
      biochemical, cellular, and biophysical regulation.
    • Characterize the processes that regulate the formation of the other constituents of
      the cell wall and the manner in which they are coupled with the deposition of
      cellulose.
    • Determine the manner in which the processes of assembly and consolidation are
      guided by the expression of genomic information, the biophysical interactions of
      the synthesized molecules, and the emerging mechanical properties.
    • Apply new instrumentation methods to study the cell wall native state without
      significantly altering its structures.
    • Develop cell walls as models and materials for nanoscale assembly.
20   The Nanoscience and Technology of Renewable Biomaterials

4. Nanotechnology in sensors, processing and process control – use nonobtrusive,
   nanoscale sensors for monitoring and control during wood and wood-based materials
   processing to provide data on product performance and environmental conditions
   during end use service, and to impart multifunctional capabilities to products.
    • Identify microbial species or chemical/optical/physical agents that are unique
      fingerprints or signatures of food spoilage, medical contamination, or product
      degradation, and develop methodologies for incorporating these agents into
      nonobtrusive, low-cost, robust nanosensors for food and medical packaging
      materials.
    • Investigate genetic and chemical modifications of wood lignocellulose materials to
      enable basic sensing capabilities and self regulation (e.g. for moisture, temperature,
      volatile organic compounds (VOCs)).
    • Investigate and develop paper and wood product coating technology and coating
      materials that can deploy nanosensors to these products through mechanical or
      chemical means.
    • Study and develop methods to synthesize data from arrays of nanosensors in order
      to generate useful information for action or process control.
    • Develop cost-effective, efficient, environmentally preferable and highly selective
      nanostructured catalysts for disassembling wood and lignocellulose.
    • Carry out research on the use of nanomaterials in conjunction with unit operations
      processing wood and wood-based materials.
5. Analytical methods for nanostructure characterization – adapt existing analytical tools
   or create new tools (e.g. chemical, mechanical, electrical, optical, and magnetic)
   that accurately and reproducibly measure and characterize the complex nanoscale
   architecture and composition of wood and wood-based lignocellulosic materials.
    • Create and maintain a compendium of available analysis tools.
    • Develop techniques and tools to measure hemicellulose polymer structure and prop-
      erties at the nanoscale.
    • Develop techniques and tools to measure lignin structure and properties at the
      nanoscale.
    • Develop methodologies and instrumentation to determine cell wall morphology and
      measure properties at the nanoscale.
    • Develop and deploy new collaborative strategies for analysis involving multiple
      techniques.
6. R&D collaboration to include the National Nanotechnology Initiative (NNI) and its
   centers – this area emphasized the importance of collaboration and cooperation among
   researchers from various disciplines and organizations, including universities, research
   institutes, national laboratories, and government agencies and departments. Linkages
   were needed to be made between research communities of the forest products sec-
   tor and the broader community of nanotechnology researchers in order to capture
   synergies, enhance accomplishments, and avoid needless duplication of facilities and
   efforts. Identified research entities that need to be engaged include:
    • individual researchers;
    • researchers with differing disciplines;
    • basic and applied researchers and research teams;
                        Nanotechnology and Lignocellulosic Biomass – Relationships       21

   • research institutions including universities, research institutes, and national labora-
     tories;
   • industry, universities, research institutions, and federal agencies and departments;
   • all of the previous groups from countries around the world.
In moving ahead in the area of nanotechnology, the forest products industry must seize
the opportunity to link with larger nanotechnology research and industrial communities
such as the ongoing efforts of the National Nanotechnology Initiative (NNI). The NNI
is a visionary R&D program that coordinates the activities of 25 Federal departments
and agencies and a host of collaborators from academia, industry, and other organiza-
tions. The goals of the NNI are to maintain a world class research and development
program aimed at realizing the full potential of nanotechnology; facilitate transfer of new
technologies into products for economic growth, jobs, and other public benefit; develop
educational resources, a skilled workforce, and the supporting infrastructure and tools
to advance nanotechnology; and support responsible development of nanotechnology.
By linking with communities such as the NNI, the forest products industry would be
able to expand its knowledge of nanotechnology, pool its resources with those of others
pursuing common R&D goals, and advance its own agenda.
   The forest products industry nanotechnology roadmap provides a starting point for
focusing the many potential and diverse efforts in nanotechnology for the forest prod-
ucts industry and also serves to further engaging key stakeholders and stakeholder groups
in dialogue, consensus building, and partnership building. The following are some of the
key stakeholder groups such as primary forest products industry producers, converters,
suppliers, and collective industry groups such as AF&PA; Federal departments and agen-
cies (e.g. the USDA Forest Service, USDA Cooperative State Research, Education and
Extension Service, Department of Energy and its national laboratories, National Science
Foundation, and National Institute of Science and Technology); University and Research
Institute/Laboratory Communities (nationally and internationally). A critical step in mov-
ing nanotechnology for the forest products sector forward is to gain consensus among
stakeholders on what the specific focus should be for the short term, mid term, and long
term. It is important that efforts be focused on high-impact, high-priority activities that
will be the most critical to commercial producers of nanomaterials and nanoproducts.


1.11   Forest Products Industry Technology Priorities

The AF&PA Agenda 2020 Technology Alliance has now gone further to identify and
select six high priority thematic areas for further study that are thought to be the key
to re-inventing the forest products industry in the US. The forest products industry has
developed a Forest Products Industry Technology Roadmap. The Forest Products Indus-
try Technology Roadmap provides a framework for reinvigorating the industry through
technological innovations in processes, materials, and markets. These innovations are
aimed at three necessary ingredients for creating a healthy future for the US-based forest
products industry (American Forest and Paper Association 2006):
1. achieving operational excellence in the production of existing and new products;
2. developing new value streams from wood resources; and
22   The Nanoscience and Technology of Renewable Biomaterials

3. assuring an ecologically sustainable, affordable domestic supply of wood and fiber
   feedstock.
The roadmap’s purpose is to provide the research community, and their funding organiza-
tions, with information on the technical challenges and research needs that are considered
priorities by the US forest products industry. The roadmap’s goal is to stimulate col-
laborative, precompetitive research, development, and deployment that will provide the
foundation for new technology-driven business models that enable the industry to meet
competitive challenges, while also contributing solutions to strategic national needs.
It envisions that the revitalized forest products industry will be built on four corner-
stones:
• significantly improved productivity through lower costs and higher yields for raw
  materials and manufactured products;
• upgraded technical skills of the workforce;
• a stream of new biomass-derived products and materials, including electric power,
  liquid transportation fuels, polymers and composites, and industrial chemicals;
• adding value to society by reducing emissions and effluents and by providing essential
  products from renewable and sustainable raw materials.
The industry roadmap also envisions that it will use and rely heavily on emerging
technologies, such as biotechnology and nanotechnology, coupled with breakthrough
advances in manufacturing process technologies to create and capture value from both
new and existing product streams efficiently, cleanly, and economically. Further, the
roadmap strategy identifies what the forest products industry sees as its inherent strengths:
stewardship of an abundant, renewable, and sustainable raw material base and a manu-
facturing infrastructure that can process wood resources into a wide variety of consumer
products. The industry also views itself as uniquely positioned to move into new, growth
markets centered on bio-based ‘green’ products. The seven industry technology goals
or technology platforms for its reinvention are as follows:
• Advancing the forest ‘bio-refinery’ – transform existing manufacturing infrastructure
  to develop geographically distributed production centers of renewable ‘green’ bioen-
  ergy and bioproducts. Double the return on net assets of existing forest products
  manufacturing plants by applying technologies that extract new value prior to pulping
  and produce new, commercially attractive products and power from wood residuals
  and spent pulping liquors.
• Sustainable forest productivity – develop and deploy wood production systems that
  are ecologically sustainable, socially acceptable, energy-efficient, and economically
  viable to enhance forest conservation and the global competitiveness of forest product
  manufacturing and biorefinery operations in the US.
• Breakthrough manufacturing technologies – develop and apply ‘breakthrough’
  approaches that can achieve revolutionary changes in the manufacturing process to
  significantly lower energy and materials costs by reducing raw material, fiber, and
  energy use and by enhancing fiber functionality.
• Advancing the wood products revolution – revolutionize housing and construction by
  creating superior, low-cost, high-value, sustainable wood products and wood-based
  building systems.
                         Nanotechnology and Lignocellulosic Biomass – Relationships    23

• Next generation fiber recovery and utilization – make recycled fiber interchangeable
  with virgin fiber with respect to product quality, functionality, and availability by
  improving the quality and quantity of recovered fiber and improving process tech-
  nologies at recycling mills.
• Positively impacting the environment – develop and deploy an optimum mix of
  in-process and add-on technologies that will enable continued improvement of the
  industry’s environmental performance.
• Technologically advanced workforce – provide training and education needed to ensure
  that new and existing technologies chosen to create the forest products industry of the
  future are operated by a technically superior workforce.
It is important to note that the AF&PA Agenda 2020 Technology Alliance views nan-
otechnology and biotechnology as means to achieving its technological goals and not
as ends in themselves. The forest products industry views implementing its technology
roadmap will require efforts in all parts of the research, development and deployment
(RD&D) continuum, from concept generation to technology deployment. A strong focus
on deployment is also a key aspect of the implementation strategy. The range and mag-
nitude of benefits offered by the roadmap’s research platforms can only be realized if
the technologies are accepted and implemented by the industry. However, the indus-
try views that it is imperative that fundamental scientific research be tapped to explore
the rich set of opportunities offered by the rapidly advancing fields of nanotechnology
and biotechnology in order to achieve breakthroughs in sustainable forestry, feedstock
processing and conversion, and end-product properties.


1.12 Nanotechnology Priority Areas to Meet the Needs of the Forest
     Products Industry

Building off The Forest Products Industry Technology Roadmap, an AF&PA Agenda
2020 task group analyzed where nanotechnology could be expected to make major con-
tributions to achieving the goals in the industry technology platform areas given in the
previous section. The following six areas were deemed to be the highest priorities:
•   achieving lighter weight, higher strength materials;
•   production of nanocrystalline cellulose and nanofibrils from wood;
•   controlling water/moisture interactions with cellulose;
•   producing hyperperformance nanocomposites from nanocrystalline cellulose;
•   capturing the photonic and piezoelectric properties of lignocelluloses;
•   reducing energy usage and reducing capital costs in processing wood to products.
Descriptions of these six areas are as follows (http://www.nanotechforest.org).

1.12.1 Achieving Lighter Weight, Higher Strength Materials
The objective is to improve strength/weight performance of paper and paperboard by
at least 40% using one or several nanotechnology-based approaches. It was thought
that a 40% improvement level was not attainable with any currently known technology
and would require one or several breakthroughs in the three key areas of: (1) strength,
24   The Nanoscience and Technology of Renewable Biomaterials

(2) optical properties and (3) surface enhancement. It was envisioned that nanotech-
nology could generate these breakthroughs where conventional approaches have been
lacking. These enabling and precompetitive technology breakthroughs in the three key
areas can then be leveraged separately or in combination by the industry to generate
competitive advantage. The solutions would allow radical reduction of raw material
use by the industry and its customers; provide opportunities to develop and market new
and advanced products with superior performance; and ultimately allow the industry
to develop new and unique materials for markets outside the pulp and paper industry.
Reduced grammage (basis weight) of paper and paperboard products will substantially
reduce wood consumption and the volume of material processed in the pulp and paper
industry, with proportional energy reductions and environmental impact. It will also
reduce the mass of nonrecoverable paper ending up in landfills. Furthermore, it would
provide opportunities to replace nonrenewable materials in a wide range of markets
with sustainable materials made from cellulose-based alternatives. The physical and
chemical properties of the cellulose fiber network in paper and board have been studied
extensively over the past 50 years and vast amounts of information on the subject can
be found in the literature. In essence the strength of the network is governed by the
bond strength, fiber strength, fiber size and shape, effect of any additives or fillers and
uniformity of material distribution. While commercial strength enhancement chemicals
are effective to a point, these technologies are not capable of leveraging the inherent
strength of cellulose nanofibrillar material, which approaches that of steel. In addition
many biologically derived materials of high strength are made up of building blocks that
are noncovalently bonded. They rely on the shear large number of points of contact to
build strength and provide mechanisms for energy dissipation, i.e. crack termination.
Therefore, it is an opportunity in using nanotechnology to diminish or close the gap
between actual current network strength, and the orders of magnitude higher strength of
the basic cellulose structure building blocks. Both nature and science have accomplished
some impressive results in strength development using very small amounts of materials
on the nanoscale.
   As part of previous research in this area, extensive modeling and theoretical back-
ground results have been accomplished, and it is likely that this information can be
used as a starting point for developing the theoretical foundation for a nanotechnology-
derived strength enhancement. There is a need as part of this priority area to develop
tailored modeling capabilities and theory to predict and elucidate strength effects of
nanoscale-level modifications to the network structure and effects of nanodimensionally
sized additives. This enhanced modeling package should be used as part of a first step to
develop a theoretical perspective on the levels and kinds of enhancements to the struc-
ture that will be needed to meet research objectives. Such nanotechnology solutions will
allow a 40% reduction of basis weight of current products and establish the precompeti-
tive platform for development of new and stronger materials from cellulosic fibers. The
preferred solution will allow the industry to continue using current production assets,
but this should not be a constraint on the development work. There is a great deal of
value in solutions that involve significant modifications to infrastructure as well. Those
solutions that allow significant simplifications of the current assets are of special inter-
est. Data are readily available in the literature to understand what the property effects
of reduced basis weights will be with the limitations of current technology. It is known
                        Nanotechnology and Lignocellulosic Biomass – Relationships      25

that deficiencies will arise in strength, optical properties, and probably surface quality
as basis weight is reduced. It is envisioned that solutions would likely be developed to
address each of these deficiencies separately but compatibly is necessary as a set of pre-
competitive, enabling technologies. The solutions will need to be combined in various
ways to generate actual improved products, so adequate coordination will be necessary
in order to assure that solutions in strength, optical properties and surface enhancement
are compatible. With solutions in place, companies with access to the new technologies
can leverage them in order to enter new markets, reduce cost, or in other ways generate
competitive advantage. As a secondary objective, the industry is interested in using
improved fibers and networks to access opportunities in other markets.

1.12.2 Production of Nanocrystalline Cellulose and Nanofibrils from Wood
The objectives in this area are the liberation and use of nanocrystalline cellulose and
nanofibrils derived from lignocellulosic feedstock. Part of nanotechnology-based solu-
tions in this area is the need to identify more commercially attractive methods to
liberate nanodimensional materials. Nanotechnologies using noncovalent disassembly
and reassembly nanofractionalization is a concept worth pursuing. The entropic effects
in the assembly and disassembly of nanomaterials in forest need to be understood. The
use of nanocatalysis (e.g. delignification) for separations is a promising concept that
should be explored. Once liberated, the nanomaterials must be adequately character-
ized, stabilized, and the nanomanufacturing and macromanufacturing technologies be
developed to allow incorporation of nanocrystalline cellulose and nanofibrils into exist-
ing forest products industry allocations as well as new applications. There is a lack of
established methods and technology in the Forest Products Industry to do any of the pre-
ceding. Success in this area allows the Forest Product Industry the opportunity to become
a major supplier of nanoparticles for a wide range of industries. Because of the tonnages
of wood available for processing, commercial production would be both sustainable and
renewable as well as create an industrially significant supply. Nanocrystalline cellulose
and nanofibrils could be extracted from currently underutilized feedstocks, such as forest
residuals and sorted wood wastes. In addition, these nanodimensional cellulosic mate-
rials would likely not have any deleterious environmental, health and safety issues as
cellulose is a biological material that is the world’s most abundant polymer and enjoys
the label of generally regarded as safe.
   There are a wide range of cellulose surface modification technologies available and it
should be possible to impart multifunctional properties and characteristics to nanocrys-
talline cellulose and nanofibrils. Additional nanotechnology research needs include
(1) identifying and isolating other commercially viable nanomaterials, in addition to
nanocrystalline cellulose and nanofibrils, present in biomass; (2) determining the effects
of species, age, growth conditions, juvenile wood, mature wood, reaction wood etc., on
nanocrystalline cellulose/nanofibril properties and morphology; (3) developing new and
modified metrologies to characterize nanomaterials derived from biological materials;
and (4) identifying new high-value applications.
   In the production of nanocrystalline cellulose or nanofibrils, it is important that a
consistent high quality nanomaterial product be able to be produced that does not differ
in with respect to important properties such as composition, diameter, aspect ratio, shape,
26   The Nanoscience and Technology of Renewable Biomaterials

and surface properties. For example, nanocrystalline cellulose is roughly 3–5 nm in
diameter and hundreds of nanometers in length. While nanocrystalline cellulose has the
potential to be produced in extremely large volumes, the utility of the nanocrystalline
cellulose for commercially desirable products will greatly depend upon its uniformity of
size, composition, structure, and surface functionality. The properties of nanocrystalline
cellulose must not differ from one lot or batch to another. Cost-effective methodologies
must be developed to liberate, fractionate, and separate cellulosic nanomaterials into
uniform, reproducible cohorts that can be easily dispersed for fabrication of macroscale
products. Isolation of nanocrystalline cellulose and nanofibrils is an important area for
research and development because current techniques appear to be lacking. For example,
hydrolysis of wood with strong acids to liberate nanocrystalline cellulose does not appear
to be an environmentally or economically friendly process and ultrasonic disintegration
has shown only partial success. In addition, real-time, inline measurement techniques
are needed to monitor and provide reproducible control of properties such as particle
size and distribution. Predictive models of nanomaterials behavior are also needed
in order to correlate nanomaterials’ properties and end-use performance requirements.
Such predictive models are critically important to cost effectively determine macroscale
properties from constitutive bulk matrix and admixed nanomaterials properties without
having to do costly and time-consuming trial and error experimentation for product
development.

1.12.3 Controlling Water/Moisture Interactions with Cellulose
This nanotechnology priority area is aimed at the very broad area of understanding and
controlling lignocellulosic/water interactions. A primary goal is to develop a substantial
knowledge base which will enable us to advantageously alter lignocellulosic/water inter-
actions to produce new and improved products and achieve more efficient and effective
processes. Because of the almost universal influence of the relationship between water
and lignocellulosics, this priority area is closely tied to many of the technology plat-
form areas goals for research, development and demonstration expressed in The Forest
Products Industry Technology Roadmap. The specific objectives in this nanotechnology
area are to (1) develop an extensive knowledge base of the interactions of water and
lignocellulosics at the nanoscale and (2) influence and modify these relationships with
the goal of producing new as well as improved existing products and processes.
   Virtually all aspects of lignocellulosic-based products and the processes by which
they are made are impacted by the relationship between water and the lignocellulosic
components of the products. The response of cellulose, hemicelluloses and lignin to
moisture (both liquid and vapor) is due almost entirely to the super molecular structure
of the biopolymers and the nanoscale structures of the lignocellulosic composites that
comprise the wood fiber. Factors such as extractives content and location also play
a role. However, most of the response to moisture depends on characteristics of the
nanoscale structures in the fiber walls. Elementary nanofibrils, which have cross-section
dimensions of about 3–5 nm are composed of cellulose polymer chains arranged in
ordered (crystalline) and less ordered (amorphous) regions. The nature of these structures
greatly influence the way in which the woody plant fiber responds to moisture. Gaining
an understanding of these interactions and learning how to manipulate the structures
                        Nanotechnology and Lignocellulosic Biomass – Relationships     27

at the nanoscale will enable us not only to decrease the negative impact of moisture
on woody materials, but perhaps allow us to turn what is currently perceived as a
disadvantage into an advantage. In addition to the size of the nanofibrils, the angle of
orientation of the fibrillar bundles of nanofibrils relative to the long axis of the fiber
plays a major role in the dimensional stability of the fiber in response to moisture. The
degree of crystallinity (i.e. the ratio of the ordered regions to the amorphous regions
in the microfibrils) will also impact the response to moisture. Because of inter- and
intra-chain hydrogen bonding, crystalline regions are less accessible to moisture. These
characteristics, which vary with species, can be genetically manipulated within a given
species. Changing the conditions under which the trees are grown and even changing
the drying conditions, as moisture is removed from the fiber during processing, can also
impact features such as the degree of crystallinity. By studying and understanding the
nature of bonding within paper and wood structures at the nanoscale, it may be possible
to modify how each composite material responds to moisture. The ability to modify and
control mechanosorptive behavior may lead to improvements in existing products and
many potential new products based on the lignocellulosic biomass resource, in addition
to greatly improving the efficiencies of the processes by which current products are made.
Durability of wood and paper products is closely tied to their response to moisture as
well. An understanding of the interactions between moisture and woody materials at the
nanoscale may permit the development of new and innovative technologies which will
decrease or even eliminate degradation.
   Control or modification of surfaces of composites based on lignocellulose using
nanocoatings or impregnation of nanoparticles could be used to provide physical/
chemical barriers to prevent or control the transfer of moisture. In addition, modification
of the topography and surface chemistry could be used to control attractive and repulsive
forces between cellulosics and other materials thus enhancing or decreasing wetting
and adhesion. For example, this could be used to increase the specific bond strength
of an interfiber bond thus permitting a lighter paper sheet with strength and optical
properties, equivalent to a heavier weight sheet.
   Very large amounts of water must be handled in the making and drying of products
made from the forest. Such activities account for a very high percentage of the costs
of production. Using nanotechnology, the nature of the interactions between the ligno-
cellulosics and water can be manipulated to improve drainage during formation of the
paper and increase the efficiency of drying of both wood and paper. This could take the
form of nanomaterials that modify fiber surfaces or change the viscosity of water. Such
materials could also be used as coatings on paper machine wires and press felts thus
enhancing drainage rates of liquid water. They might also be used on paper machine
dryer felts and dryer cans to improve heat transfer, making drying more efficient.
   Understanding and manipulating the interactions between water and wood/paper will
permit huge reductions in energy and water usage in processes by which products are
made from these complex materials. It most likely will result in the more economical
use of the raw materials in a broad base of new and existing products. It may also
enable the substitution of products based on a sustainable renewable resource for some
of the products derived from a more limited and less environmentally friendly material
such as petroleum. The relationships between water and lignocellulosic materials have
been studied extensively and a great body of literature exists. However, relatively little
28   The Nanoscience and Technology of Renewable Biomaterials

effort has been directed toward the interactions that occur at the interfaces of the two
materials. These interfacial areas can be defined a number of ways. One such interface
is the surface of the wood itself. Another would be the wood or wood pulp fiber and
a third would be the interfaces that exist between water and the within the fiber walls.
The latter can even be broken down to the amorphous areas of the cellulose bundles and
the crystalline areas. The arrangement, size and degree of ‘crystallinity’ of the cellulose
bundles at the nanoscale can greatly influence the behavior of the materials in response
to water and water vapor at the macrolevel. Characterization of the interfaces, and the
relationships between water (liquid and vapor) and lignocellulosics at the nanoscale can
be achieved utilizing a combination of the newer available tools (e.g. atomic force
microscopy (AFM), scanning tunneling electron microscopy (STEM), etc.) and standard
technologies (e.g. inverse gas chromatography (IGC), transmission electron microscopy
(TEM), scanning electron microscopy (SEM), electron spectroscopy for chemical analy-
sis (ESCA), auger electron spectroscopy (AES), ultraviolet photoemission spectroscopy
(UPS) and time-of-flight secondary-ion mass spectroscopy (ToF-SIMS)). Older meth-
ods such as electro-kinetic analysis (EKA), water vapor sorption, differential scanning
calorimetry (DSC) and microcalorimetry are also very useful, especially when studying
the surface charge and thermodynamics in the interfacial regions.
   The scope of the proposed research could include the influence of species, location
in the tree, site, process conditions, and the response of products in use to water and
water vapor. By developing a fundamental knowledge base at this level, tools may be
made available to allow companies in the forest products industry to ‘design’ trees for
the properties needed in a range of products, to produce them by utilizing processes
that are much more efficient in terms of the consumption of energy, water and raw
materials; and may even make available a renewable, sustainable resource to the devel-
opment of products currently made from less environmentally friendly materials. The
primary output from work on lignocellulosic/water interactions at the nanoscale will be
a knowledge base or ‘toolbox’ from a materials science perspective rather than a wood
products or papermaker’s point of view. However, it will provide developers of prod-
ucts and processes in these disciplines the tools needed to improve current products and
processes as well as those needed for the development of entirely new products from
the forest.
   Understanding and characterizing the interfaces in cellulose fibers at the nanoscale are
the first step toward modifying the fiber and enhancing its properties as a building block
for many products (existing and new). Using the water/lignocellulosic interaction (liquid
and vapor) as a probe will enable the understanding of the surface energies of the fiber and
permit us to more effectively add coatings (nanolayers), or other surface modifications; as
well as derivatize cellulose to meet new and existing product requirements (e.g. strength
enhancement, adhesion, and hydrophilic/hydrophobic properties). This knowledge base
will be applicable to virtually all products derived from woody and nonwoody plants
wherever water and lignocellulosics interact. With the exception of entirely new product
streams, the information developed will permit more effective utilization of the current
assets in the wood products and paper industries.
   Outputs from this nanotechnology priority area would include the following: (1) a
package of fundamental knowledge relating to cellulose/water interfacial interactions
at the nanoscale; (2) a model based on the fundamental information developed;
                        Nanotechnology and Lignocellulosic Biomass – Relationships     29

and (3) a knowledge base which relates these interactions to more applied areas
of adsorption/desorption, drying, dimensional stability strength/weight relationships,
surface modification, product durability, and process improvements. The majority of the
work considered to be precompetitive would include fundamental studies that relate to
fiber/water or other lignocellulose/water interactions. These would relate to areas such
as the impact of the nanostructure and properties such as the degree of crystallinity,
dimensions of microfibrils and microfibril angle. The surface chemistry and topography
of these materials at interfaces between water and the lignocellulosics at the nanoscale
will have a big impact on how the materials respond to moisture. Heats of immersion
or wetting, energies of adsorption of vapor, and surface free energy of the materials
are all impacted by the natural nanostructures. In addition to the cellulose portion
itself, hemicelluloses, lignin, extractives and trace minerals etc. will also influence the
response of these materials to water or water vapor. Fundamental studies characterizing
these materials based on surface science would provide a base to move into applications
that take advantage of the properties at the nanoscale.

1.12.4 Producing Hyperperformance Nanocomposites from Nanocrystalline
       Cellulose
In addition to the wood-based composites, paper and paperboard can be considered
to be a form of nanocomposite as they are made up of components that are essentially
nanodimensional. Most work, to date, has been the result of empirical formulation where
wood or pulp fibers have been mixed together with other components to make useful
functional materials. Cellulose is a material which has unique tensile properties. In its
pure form it can create fibers that are as strong or stronger then Kelvar (Cellulose =
70 to 137 GPa, Kevlar = 100 GPa). It is desired to form composites in which cellulose
provides its maximum tensile strength. Other properties of interest include formability
and geometrical complexity at very small scale, unique physical properties, surface
smoothness, biomedical compatibility, and ability to reinforce polymer foams.
   It is also desired though the use of nanomaterials, and chemistry to either form or
reform cellulose fibers in a variety of matrixes in which the cellulose can contribute
its full modular strength to the matrix (Podsiadlo 2007). It has been postulated that
the structure of wood is the result of the cellulose nanofibrils forming liquid crystal
arrays under the influence of the hemicellulose (Vincent 2002). This represents a form
of self-assembly that we would like to capture in order to produce new materials with
high strength at lightweight. The interactions are typically noncovalent, such as hydro-
gen bonding and Van der Waals forces but, because of the extremely small size the
interactions add up to provide a high degree of strength.
   Other potential avenues can also be investigated to accomplish this. These include:
(1) modification of the side chains of inorganic compounds, such as siloxanes, silanes,
or sodium silicates to link the cellulose fibers through Si-OH bonds forming an organic/
inorganic matrix; (2) growing the cellulose from bacteria; or an enzyme engine such
that the cellulose forms in a matrix; (3) a nanostructure template and nanocatalysts
could be used to help structure the matrix and increase the rate of formation of the
cellulose fibers within the structured matrix; (4) disassembly of plants with enzymes/
30   The Nanoscience and Technology of Renewable Biomaterials

chemistries that allows for the separation of cellulose from lignin without mechani-
cal action; (5) development of systems that simulate the growth of cellulose in trees
or plants that can be accomplished on a industrial scale; (6) dissolution of cellulose
into ionic liquids with precipitation of cellulose into a continuous fiber or incorpo-
ration of threads or honeycomb weaves of cellulose into a variety of different mate-
rial matrixes; (7) reacting wood pulp fibers in a solvent medium that does not fully
penetrate the fibers followed by hot-pressing the partially modified pulp fibers at ele-
vated temperature to form a semi-transparent polymer sheet that is a nanocompos-
ite of cellulose esters and unmodified cellulose; (8) use of cellulose nanocrystals for
reinforcement of other matrix materials; extreme refining of cellulose fiber resulting
in increasing Canadian Standard Freeness (Roman and Winter 2006): and (9) mod-
ification of the side chains of cellulose to further enhance self assembly (Gray and
Roman 2006).
   A variety of understanding and characterization techniques would need to be estab-
lished to accomplish the above. These include: (a) understanding cell wall formation
in tree and plants; (b) development of the appropriate inorganic chemistry for linking
cellulose; (c) understanding of cellulose chemistry and the sheet layering of cellulose
to establish pathways by which cellulose could be modified to enhance self assembly;
(d) understanding and modeling the formation of cellulose from glucose or other simple
sugars by bacteria; (e) understanding of the effect of a variety of enzymes on the struc-
ture and tensile strength of cellulose; (f) understanding of the chemistry of cellulose and
manipulation of its precipitation based on its solubility in various liquids and subsequent
processing; and (g) effects of enzymes and extreme refining conditions on cellulose and
cellulose composites.
   Use of cellulose in a variety of different matrixes will be dependent on the interactiv-
ity of the matrix material with cellulose and lignocellulose surface chemistry. Wetting
and surface area play key roles in the formation of high-strength interfaces between the
matrix, matrix components and cellulose. Nanomaterials can provide unique levels of
surface area for the formation of chemical bridges between the cellulose, the matrix and
other fillers used. The strength of cellulose composites is influenced by the chemical
interface and cellulose particle geometry. Interfacial interactions are governed by adhe-
sion, water sorption, durability, and processing of the material. Cellulose derivatives
can also be combined with nanomaterials and used in conjunction with cellulose fibers,
or other fibers to form nanocomposites (Choi and Simonsen 2006).

1.12.5 Capturing the Photonic and Piezoelectric Properties of Lignocelluloses
Many grades of paper require using higher grammage (basis weights) than needed, not
because of strength property end use requirement but because of the need to achieve
sufficient opacity. Our target to use nanotechnology to help overcome this technical
barrier and allow lower grammage sheets to be used in printing and writing allocations
will result in paper with lower opacity and the likelihood of it not being fit for use.
While this would allow savings by permitting raw material reductions in both fiber and
coating, we need to avoid the loss of optical performance of the paper by building
‘optical band gap’ coatings to enhance opacity. Nanotechnology could also provide the
ability to produce high sheet brightness with no fluorescence and could eliminate the
                         Nanotechnology and Lignocellulosic Biomass – Relationships      31

requirement to bleach pulp to make white paper. The color gamut in the final printed
image could be greatly improved by allowing for ‘pure’ pastel shades.
   Currently the Kubelka-Munk approach of deriving apparent light scatter and absorp-
tion coefficients is useful for characterizing materials (Jones 2004). In addition, Mie
theory uses fundamental Maxwell equations to describe the way light is scattered from
particles and is useful for predictions of optimal sizes for light scattering units. Regular
arrangement of these units can give rise to reinforcement of light interactions and is
called a ‘photonic effect’. For example, photonic band-gaps are structures that prevent
the passage of light. Photonic properties have been shown to be possible using ‘standard’
materials and producing structures with regularities that provide photonic effects. Natu-
ral materials such as butterfly wings, seashells (abalone) and insect cuticles demonstrate
effective optical barriers with minimal materials (Vukusic, Hallam and Noyes 2007).
These are effects different from those described by Kubelka-Munk, Mie or Raleigh
scattering.
   It has been demonstrated that it is possible to make photonic structures that stop narrow
wavelength band (e.g. latexes, Stober Silica – Synthetic Opals, and block co-polymers).
The challenge is to make a structure with a range of sizes that has an effect over a
very broad bandwidth and therefore, appears white. We are looking for high (close to
100%) opacity with high whiteness/brightness with minimal amounts of materials. More
efficient optical performance with minimal weight is required at all grade levels but
especially at the ultra-light grade where opacity decreases rapidly with weight. If we
can make coated paper in the same weight range as tissue paper we can gain the benefits
of high distribution costs and consequently limit competition from far afield. This will
revitalize production units serving local areas. As a subgoal we expect to achieve:
1.   a range of colors through unique structures (permanently stable unlike dyes);
2.   improved gloss and appearance through control of unique structures;
3.   pearlescent/iridescent effects;
4.   control of ink interactions;
5.   applications for security/ticket stock;
6.   brightness unachievable today without using optical brightening agents (OBA) for
     enhanced image quality.
Electromechanical coupling effects in wood date back to 1950 when Bazhenov reported a
piezoelectric response in wood (Bazhenov 1961). In 1955, Fukada also showed how the
piezoelectric coefficients of wood were related to oriented cellulose crystallites (Fukada
2000). Piezoelectricity, a linear coupling between electrical and mechanical properties,
is displayed by crystal structures that lack a center of symmetry (noncentric symmetric).
Cellulose in wood is piezoelectric due to the internal rotation of polar atomic groups
associated with asymmetric carbon atoms providing the noncentric symmetry. Shear
piezoelectricity in wood varies depending on the type of wood, orientation of wood
samples, moisture, and temperature and is comparable to that of quartz. Despite these
early studies the potential of cellulose as smart lightweight material that can be used as
a sensor and an actuator has not been investigated. Kim et al. have shown that it is
possible to take advantage of this noncentric symmetry feature of cellulose to develop
electro-active devices (Kim 2006). It is envisaged that, as we develop self-assembling
32   The Nanoscience and Technology of Renewable Biomaterials

cellulosic materials, we will be able to take advantage of these piezoelectric properties
to build in greater functional performance.
   Smart paper and packaging materials including radio frequency identification (RFID)
and integrated moisture, impact and biological/chemical sensors, require paper substrates
with new physical and chemical specifications. Moreover, advanced devices may require
the ability to print much smaller and more uniform features onto paper substrates. How-
ever, several areas of advancement are needed. For example, printed electronics on
paper will place new constraints on paper substrate frequency response and conductiv-
ity. The complex dielectric constant and dielectric loss tangent performance will need
to be addressed to accommodate different frequency regimes. Radio frequency identi-
fication (RFID), for example, operates in the ∼1–50 MHz range, but development of
systems operating in the 300–500 MHz range is also underway. The specific applica-
tion will drive the final specifications, but process, material and coating technologies
capable of supporting device operation in the 1–50 MHz and 50–500 MHz should be
explored.
   Electronic devices such as printed interconnects, resistors, reactive components and
even active electronic and optoelectronic devices operating at high frequencies will
require small printed feature dimensions and film thicknesses produced with better unifor-
mity and reproducibility than is currently achievable, thus placing an additional constraint
on the surface morphology of the paper substrate. Printed features, such as interconnect
lines, may exhibit feature dimensions ranging from <10 to 100 microns in width and
from <1 to 10 microns in thickness. The roughness and porosity of the paper sub-
strate may be limiting in these applications, creating opportunities for new fillers, fiber
materials, and fiber assembly or coatings to improve the structure of the substrate. We
expect to improve the particle size and shape distributions of selected building blocks.
This may be the extension of the engineered mineral pigments emerging currently or the
development of new synthetic materials from wood fiber.
   In order to achieve the most benefit from the opportunities to modify optical and elec-
trical properties, we need to be able to select with precision the needed building blocks
and then control their assembly into useful structures. This is probably the most challeng-
ing area. Paper coatings currently use additives and components that are applied in shear
fields and dried under precise conditions to control the migration and positioning of the
components to best advantage. Much of this has been derived empirically with informa-
tion inferred from bulk measurements. Similarly in papermaking, pulp refining has been
tuned to select the most useful building blocks and drainage, retention, and formation
aids are used to control the structure development in the shear and compressive fields
that occur on the paper machine in papermaking. Again many of these developments
have been empirical but have benefited from recent microcopy developments.
   The new disciplines of soft matter physics and nanotechnology are leading people to
the study of self-assembling systems and offer the opportunity to leverage the findings
into the world of forest products. Some of the areas that should be of value are:
1. block copolymer reactions;
2. drying moderated assembly;
3. hydrophobic/hydrophilic assembly;
                         Nanotechnology and Lignocellulosic Biomass – Relationships      33

4. mineral based liquid crystal assemblies;
5. layer by layer assemblies of polyelectrolyte coatings.
The forest products industry will also be able to take advantage of the substantial appli-
cation equipment industry that has developed a number of novel machines for making
paper and applying coatings at very high speeds such as:
1.   metered size press;
2.   spray applicator;
3.   multilayer curtain coater;
4.   new applications devices yet to be determined.

1.12.6 Reducing Energy Usage and Reducing Capital Costs in Processing Wood
       to Products
Priority goals for the application of nanotechnology in the forest products industry are the
reduction of energy consumption as well as reducing capital costs. These are priority
goals because conversion of wood into lignocellulosic products – lumber, engineered
wood, composites, pulp, and paper – uses considerable amounts of energy and is quite
capital intensive. This is because of: (1) the large tonnage of forest product used
annually in the US – over 205 million metric tons; (2) variability of wood as a raw
material; (3) the need for water removal from the final product; (4) many sequential
processing steps, (5) the amounts of water used in processing, (6) the need to deal with
byproduct waste streams, and (7) the fact that many of the conversion technologies have
their origins dating back from many decades to even centuries ago.
   Overall energy consumption is between 2.891 EJ (2,740 trillion BTUs) and 3.511
EJ (3,272 trillion BTUs). This level of consumption represents 12–16% of US man-
ufacturing energy demand – depending upon the literature source cited. Energy use is
the second or third largest cost factor for the industry, especially as fuel and electricity
prices continue to rise. Although the industry as a whole self-generates almost 50% of its
energy needs from on-site combustion of biomass and pulping by-products, the industry
still ranks as the country’s fourth largest consumer of fossil energy (American Forest
and Paper Association 2006). Paper and paperboard production accounts for the largest
share of energy use (approximately 78%) in the industry, mainly due to the amount of
energy required to evaporate the large quantities of water used to form the pulp slurry
and the paper web. Pulping (7%), engineered wood products/composites (7%), sawn
lumber (5%), and preservative treated and other lumber production (3%) uses compar-
atively less energy. However, because of the large tonnages of materials involved, the
amount of energy used is substantial.
   Capital costs are also a major problem for the forest products industry. For example,
the pulp and paper industry ranks as one of the most capital-intensive industries in the
nation. Paper machines are by far the largest and most expensive capital component.
Pulping and bleaching equipment and chemical recovery plants also represent a large
share of installed capital due to their size and complexity. By-product waste stream
abatement and control are also a significant and recurring capital expense industry-wide.
   The objectives for applying nanotechnology in reducing energy and capital costs are
to employ nanomaterials in forest products processing in order to reduce manufacturing
34   The Nanoscience and Technology of Renewable Biomaterials

costs by both reducing the amount of energy consumed during processing and capital
equipment required. Nanotechnology applications can take the forms of: nanocata-
lysts to reduce the temperatures and time needed to delignify wood in pulping; low
corrosion nanocoatings and nanomaterials to prolong the life of capital equipment; nan-
odimensional tags/markers for fiber separations; nano-inspired products that help with
water removal on paper machines (drainage wires, vacuum boxes, wet presses, and dry-
ers), kilns, and hot presses; and robust nanodimensional sensors (temperature, pressure,
tensile/compressive forces, etc.) that can be used to monitor and optimize processing
conditions as well as reduce/eliminate off specification product productions; etc.
   Fiber, energy, and chemicals rank as the highest nonlabor operating cost items or
categories at most pulp and paper mills. The ratio of costs will vary among different types
of mills, but a typical integrated mill producing 1500 tons of kraft pulp per day will spend
about US$45–60 million for wood, US$30 million for chemicals, and US$15–20 million
for purchased energy each year. Energy reduction goals for pulping and papermaking
are to reduce pulping process energy consumption by at least 33% and produce the same
or better quality fiber at 5–10% higher yield; reduce energy consumed in the process
of increasing black liquor solids (kraft pulping) by at least 50%; develop lower-cost
technology to replace the current (energy and capital intensive) causticizing process;
reduce energy consumed in the paper machine wet end by at least 33%; reduce the
energy consumed in paper dewatering, pressing, and drying by at least 50%; and reduce
energy and produce same or better-quality paper products by using: (a) nanocoating
pigments and (b) three times the nonfiber filler content.
   Drying is the most energy-intensive process employed in most pulp and paper mills,
consuming between 4.6 to 9.2 GJ/metric ton (4 to 8 million BTU/ton) of pulp, depend-
ing on the paper grade. The amount of water removed by drying is determined by
the efficiency of the nonthermal water removal processes (i.e., drainage, vacuum dewa-
tering and wet pressing). As an approximation, every 1% increase in sheet solids as
the sheet passes to the dryer section effects a 4% savings in dryer energy use. Paper-
making is a complex operation requiring tight control to produce the expected level
of quality. The huge quantity of water that must be removed and the required level
of precision drive capital and operating costs. Nanotechnology needs for paper and
paperboard drying are to develop cost-effective nano-inspired technologies that reduce
the energy consumed in web/paper/paperboard dewatering, pressing, and drying by at
least 50% via (1) developing next generation one-way water removal wet presses that
employ felts that prevent/eliminate sheet rewetting, allow higher press nip forces, and
extended nip lengths/dwell times to achieve significantly higher solids content of the
paper/paperboard web entering the dryer section and (2) developing next-generation tech-
nologies that improve energy transfer to the web/sheet and water/water vapor removal
for drying.
   Energy and chemical usage varies with the pulping processing used. In the US,
kraft pulping (both bleached and unbleached) is by far the largest pulping process
used with over 45 million metric tons (50 million tons) of pulp produced annually.
Semi-chemical, chemi-thermomechanical, Thermomechanical, and refiner mechanical
pulps are employed but the tonnages produced are much, much less. In kraft chemical
pulping, wood chips are heated to 160–180 ◦ C (320–356 ◦ F) at a liquor to wood ratio
of about 3.5 to 1 using sodium hydroxide and sodium sulfide and held for a period of
                        Nanotechnology and Lignocellulosic Biomass – Relationships     35

time ranging from about 30 to 45 minutes depending upon the degree of delignification
desired. The resulting pulp exiting the digesters is diluted and washed with water to
remove the black liquor (spent pulping chemicals and dissolved wood solids). Large
quantities of water are used, resulting in dilute black liquor – typically between 12 and
18% solids. The diluted black liquor must be concentrated before it can be efficiently
burned in a recovery boiler to produce energy and recover the chemicals for reuse in
the pulping process. The recovery of chemicals to form fresh pulping liquors is a vital
part of the pulping operation. The energy generated by burning the black liquor in the
recovery boiler is used in pulping and papermaking operations and significantly reduces
purchased energy requirements. The black liquor is concentrated to 70–78% solids
in steam-heated, multiple-effect evaporators. The evaporators are typically the second
largest energy users in pulp mills and the largest source of steam consumption, at around
4.4 GJ (4.2 million BTU) of steam per metric ton of pulp.
   Causticizing is a multistep process in chemical recovery process chain aimed at regen-
erating the original pulping liquor (‘white’ liquor) from the molten smelt of inorganic
chemicals (sodium sulfide and carbonate) exiting the chemical recovery boiler. The pro-
cess starts by dissolving the molten salts in water to form an aqueous solution called
‘green liquor.’ Lime (calcium oxide) is then mixed with the green liquor in a slaker to
form sodium hydroxide and calcium carbonate. The chemical reactions that form the
white liquor are completed in a series of reactors called ‘causticizers’. The spent car-
bonate sludge (‘lime mud’) from the slaker and causticizers is removed in a clarification
step, thickened, washed, and calcined in a lime kiln to recover calcium oxide for reuse
in the slaking process. Causticizing is an old technology that has not benefited from
innovation in many years. It is extremely capital intensive and suffers from very high
operating and maintenance costs. The lime kiln in particular is a high-energy user and
prone to maintenance problems. Lime kilns are typically fueled by oil or natural gas
and represent one of the largest consumers of purchased energy in the pulp mill. Lime
kilns and causticizers can also be a production bottleneck, limiting the mill’s production
capacity.
   Nanotechnology opportunities/needs in pulping are to: (1) use nanotechnology to
reduce steam use for black liquor evaporation to achieve energy savings; (2) develop
low corrosion nanocoatings and nanomaterials to prolong the life of capital equipment
especially in bleach plants; (3) develop cost-effective alternative nanocatalyzed, simpler
means of regenerating white liquor, that regenerates sodium hydroxide in the recovery
boiler and smelt-dissolving tank; (4) develop new nanocatalysis techniques to rapidly
delignify wood (e.g. in 10 minutes or less) at lower temperatures (i.e. below the boiling
point of water so as to not require pressurized vessels) that would also allow for easier
separation of spent pulping liquor components, easier solids concentration, and easier;
and (5) develop new nanocatalysis techniques for separating wood cell wall constituents
without altering native structures of wood constitutive components (i.e. hemicellulose,
cellulose, and lignin).
   The kraft pulp industry has traditionally been a source of odorous emissions (primar-
ily methylmercaptans) that, although not a health risk, are regarded as a nuisance in
nearby communities. In recent years, the industry has made great progress by installing
state-of-the art systems to reduce in-mill sources of the odors. However, odors from
wastewater treatment operations continue to be a cause for community concern in many
36   The Nanoscience and Technology of Renewable Biomaterials

locations. Nanotechnology needs in this area are to develop cost-effective methods to
reduce or eliminate odorous kraft emissions beyond the mill property.
   The US pulp and paper industry discharges about 45 m3 /ton (12,000 gallons/ton) of
wastewater. Although this is a significant improvement over decades past, the indus-
try is still among the largest industrial water consumers. Developing ways to reduce
water use in the mill and/or to recycle the process water within the mill for reuse
would significantly lower wastewater treatment costs. Low-quality thermal energy in
the wastewater could potentially be captured in the water recycling process and the
industry’s water use and effluent discharges would be greatly reduced. Almost all of the
wastewater generated from the pulp and paper industry is treated in wastewater treat-
ment plants made up of settling ponds (primary treatment) and biological purification
(secondary treatment). The treatment process involves multiple steps and generally
requires significant amounts of electrical power, adding to manufacturing costs and
resulting in emissions from power production. In addition, many of these systems
are reaching the ends of their useful lifetime and will need to be renovated or replaced
at significant capital expense to the industry. Nanotechnology needs in wastewater
treatment are to (1) develop alternative methods for wastewater treatment that are less
energy- and capital-intensive than current biological effluent treatment systems and (2)
develop low corrosion nanocoatings and nanomaterials to prolong the life of capital
equipment.
   Current recycling mills are complex, require numerous separate unit operations, and
are energy inefficient and costly to operate. These operations need to be streamlined to
improve operating efficiency, lower capital costs, and reduce energy and water consump-
tion. Technologies to improve yield, recover all on-grade fibers, and tolerate or remove
contaminants are also needed. Because of the variability in recovered paper collection
and sorting systems across the US, gross contaminants and out-of-spec fiber contin-
ues to reach the recycle mill. Moreover, lower quality and rejected fibers – including
shorter, inferior fibers from recycled paperboard imported from Asia – are also gener-
ating an increased volume of solid waste in US recycled fiber mills. Overall recycling
goals focus on sorting and recycling mill wet-end equipment and processes such as
pulpers, screens, cleaners, and flotation devices to significantly improve paper fiber
recovery, fiber utilization, and energy efficiency in order to reduce fiber yield loss by
50%; improve contaminant removal by two-thirds; reduce overall costs by as much as
US$40 per ton; reduce energy use by 50%; and reduce water use by 50%. Nanotech-
nology needs are to: (1) develop functional nanomaterials to enable paper and fiber
tagging; (2) use nanomaterials to facilitate ink removal (i.e. de-inking) and contaminant
removal; (3) develop low corrosion nanocoatings and nanomaterials to prolong the life
of capital equipment; and (4) develop nanomaterials to improve recyclability of paper
and paperboard products.
   Processing wood products requires large amounts of energy, and represents the single
highest wood processing cost. Drying processes account for the largest share of energy
consumption. Energy used to cure wood composites and to dry lumber and other wood
products in kilns accounts for 50–80% of the manufacturing energy consumed in these
operations. More efficient wood drying and curing processes, better technologies for
utilizing sawmill residues for energy, and methods for utilizing low-grade energy from
available engines and motors could significantly reduce the purchased energy-intensity of
                        Nanotechnology and Lignocellulosic Biomass – Relationships     37

wood products mills. A large amount of energy is also used in pollution control devices
to control emissions of volatile organic compounds (VOCs) and hazardous air pollutants
(HAPs). New regulations may require increased use of energy-intensive regenerative
thermal oxidizers to further control releases of VOCs and HAPs from mill operations.
The use of fossil fuels to power thermal oxidizers for VOC and HAP emissions control
has significantly increased natural gas consumption. Development of energy-efficient
and cost-effective technologies to reduce VOC and HAP emissions is needed. Advances
could include more energy-efficient pollution control technologies, methods to reduce
VOC and HAP precursors in the wood itself, or technologies to trap and purify specific
VOCs suitable for sale as byproduct chemicals. Overall goals for wood products are
to reduce capital and operating costs for wood products manufacturing by improving
energy efficiency and reducing emissions control costs while providing greater flexibility
to customize products for end-users. Nanotechnology needs are to: (1) reduce VOC and
HAP emissions from manufacturing wood-based products by 90%; (2) use nanoscale
materials and technology to improve conversion efficiencies of wood products; (3) use
nanocoatings and nanocatalysis to decrease emissions to indoor air from wood-based
products by 50%; (4) investigate ways to use nanotechnology and nanomaterials to
enhance and increase the efficiency of drying wood and wood-based materials in kilns and
presses; (5) increase marketable chemical byproducts of wood by 10%; and (6) employ
robust nanodimensional sensors (temperature, pressure, tensile/compressive forces, etc.)
to monitor and optimize processing conditions and improve conversion yields as well as
reduce/eliminate off-specification product productions; etc.


1.13 Summary

Nanotechnology is an emerging area of science and technology that will revolution-
ize materials use in the 21st century. Over the course of this century, we will move
from many of the relatively crude and unsophisticated technologies on which we cur-
rently depend and replace them with highly efficient and environmentally friendly nan-
otechnologies that meet the desired goals, guidelines, and principles of sustainability,
sustainable development, green chemistry, and green engineering. For the forest (ligno-
cellulosic) products industrial sector, nanotechnology will be used to tap the enormous
undeveloped potential that tree’s possess – as photochemical ‘factories’ that produce rich
sources of raw materials using sunlight and water. Lignocellulosic biomass resources
provide a key materials platform for the sustainable production of renewable, recy-
clable, and environmentally preferable raw materials for producing goods and products
to meet the needs of people. Lignocellulosics provide a vast material resource and are
geographically dispersed. Humankind has done an excellent job of capturing the val-
ues that wood can provide at the macro- to microscales, but the values of wood and
wood-based materials at the nanoscale are virtually untapped. The vision for nanotech-
nology in the forest products is to sustainably meet the needs of present and future
generations for wood-based materials and products by applying nanotechnology science
and engineering to efficiently and effectively capture the entire range of values that
wood-based lignocellulosic materials are capable of providing. In addition, the ligno-
cellulosic products industry sees its inherent strengths as including: (1) stewardship of
38   The Nanoscience and Technology of Renewable Biomaterials

an abundant, renewable, and sustainable raw material base; (2) a manufacturing infras-
tructure that can process wood resources into a wide variety of consumer products; and
(3) being uniquely positioned to move into new, growth markets centered on bio-based
environmentally preferable products. Nanotechnology, as it is envisioned, will further
enhance the industry’s ability to produce new high performance consumer products from
lignocellulosic-based materials in a safe and sustainable manner.
   While the focus of nanotechnology research may seem to be on determining
the nanoscale properties of materials, it is really achieving nanotechnology-enabled
macroscale end products that are most important. Nanotechnology must be viewed as
an enabling technology versus an end in itself. To most rapidly make scientific and
technology advancement, the focus for nanotechnology research must have an end
use application or product in mind. The range and magnitude of benefits offered by
nanotechnology science and engineering research and development can only be realized
if the technologies are accepted and deployed by industry to produce economically-
viable products that consumers want and need. Therefore, it is important that research
efforts be focused on high-impact, high-priority activities that will be the most critical
to commercial producers of nanomaterials and nanoproducts. It is absolutely critical to
build our nanomanufacturing science and technology knowledge base so nanomaterials
exhibiting unique nanoscale properties can be controllably placed into components
or systems, retain and combine their unique nanoscale properties in a matrix of
other materials, and result in superior products. Industrial input and influence on
nanotechnology science and engineering serves to help guide activities into the highest
priority and most productive areas. The US forest products industry has identified six
priority nanotechnology application areas. These six areas are to use nanotechnology to
(1) achieve lighter weight, higher strength materials; (2) produce nanocrystalline fibrils
from wood; (3) control water interactions with cellulose; (4) produce hyperperformance
nanocomposites using nanocrystalline cellulose fibrils; (5) capture the photonic and
piezoelectric properties of lignocelluloses; and (6) reduce energy usage and capital
costs in processing wood to products. Lastly, understanding the health risks and taking
appropriate action to mitigate risks to health, safety, and the environment that result
from exposure to or introduction of engineered nanoscale materials, nanostructured
materials, and nanotechnology-based devices is an extremely important consideration
in responsibly moving nanotechnology forward – for both wood-based nanomaterials
and nanomaterials and devices that are produced by other industry sectors that are
incorporated into forest products.


References

Aizenberg, J., Weaver, J., Thanawala, M., Sundar, V., Morse, D., and Fratzl, P. (2005)
  Skeleton of Euplectella sp.: structural hierarchy from the nanoscale to the macroscale,
  Science, 309, July 8: 275–8.
American Forest and Paper Association Agenda 2020 Technology Alliance (2006)
  http://www.agenda2020.org, Forest Products Industry Technology Roadmap, July.
American Society for Testing and Materials (2006) Standard Terminology Relating to
  Nanotechnology, E 2456–06.
                        Nanotechnology and Lignocellulosic Biomass – Relationships    39

Anastas, P. and Warner, J. (1998) Green Chemistry: Theory and Practice, Oxford
  University Press, New York, NY.
Atalla, R., Beecher, J., Jones, P., Wegner, T., et al. (2005) Nanotechnology for the For-
  est Products Industry Vision and Technology Roadmap, http://www.nanotechforest.org,
  TAPPI Press, Atlanta, GA.
Bazhenov, V.A. (1961) Piezoelectric Properties of Woods, Consultants Bureau, New
  York.
Brown, B.J., Hanson, M.E., Liverman, D.M., and Merideth Jr., R.W. (1987) Global
  sustainability: toward definition, Environmental Management, Springer, New York,
  NY, Vol. 11, No. 6, November: 713–19.
Brown Jr., M., Czaja, W., Jeschke, M., and Young, D. (2007) Multiribbon Nanocellulose
  as a Matrix for Wound Healing, US Patent 20070053960, March.
Brundtland, G.H. (1987) Our Common Future, World Commission on Environment and
  Development, Oxford University Press, New York.
Cash, M., et al. (2003) Derivatized Microfibrillar Polysaccharide, United States Patent
  6,602,994 B1, August 5.
Chemical Industry Vision2020 Technology Partnership (2003) Chemical Industry R&D
  Roadmap for Nanomaterials by Design from Fundamentals to Function, http://www.
  ChemicalVision2020.org, December.
Choi, J.Y. and Simonsen, J. (2006) Cellulose Nanocrystal-filled carboxymethyl cellulose
  nanocomposites, J Nanoscience and Nanotechnology, 6, (3): 633–9.
Davies, J. (2006) Managing the effects of nanotechnology, Resources for the Future.
  RFF Press, Washington, DC.
Department of Energy, Energy Efficiency and Renewable Energy (2007) Nanomanufac-
  turing Roadmap for Energy Efficiency, November.
Friends of the Earth (2008) Out of the Laboratory and onto Our Plates, Nanotechnology
  in Food & Agriculture, March, http://nano.foe.org.au.
Fukada, E. (2000) History and recent progress in piezoelectric polymers, IEEE Transac-
  tions on Ultrasonics, Ferroelectrics, and Frequency Control . 47(6): 1277–90.
Gray, D. and Roman, M. (2006) Self assembly of cellulose nanocrystals. ACS Sympo-
  sium Series, Vol. 938, Cellulose Nanocomposites: Processing, Characterization and
  Properties, Chapter 3.
Green Engineering (2003) Defining the Principles Conference, Sandestin, Florida, May
  2003.
Greenwood, M. (2007) Thinking Big about Small, Creating an Effective Oversight System
  for Nanotechnology, Woodrow Wilson International Center for Scholars, March.
Hollman, M. (2007) Nanomaterial Forecast: Volumes and Applications, International
  Council on Nanotechnology Nanomaterial Environmental Health and Safety Research
  Needs Assessment, http://www.luxresearchinc.com, January 9.
Hubbe, M.A. (2006) Bonding between cellulosic fibers in the absence and presence of
  dry-strength agents – a review, Bioresources, (2), 281–318.
Hullmann, A. (2006) The Economic Development of Nanotechnology: An Indicators
  Based Analysis, http://cordis.europa.eu./nanotechnology, November 28.
International Risk Governance Council (2007) Nanotechnology Risk Governance,
  Geneva, Switzerland, www.irgc.org.
40   The Nanoscience and Technology of Renewable Biomaterials

Jenck, J., Agterberg, F., and Droescher, M. (2004) Products and processes for a sustain-
  able chemical industry: a review of achievements and prospects, Green Chemistry, 6:
  544–56.
Jones, J.P.E. (2004) Continued evolution of coated paper, board will rely on formula-
  tion ‘building blocks’: the ability to control coating pigment structure is vital in the
  development of paper and paperboard grades for increasing demanding end users, Pulp
  & Paper, Jan.: 24–8.
Kim, J., and Yun, S. (2006) Discovery of cellulose as a smart material, Macromolecules
  39: 4202–6 (Piezo section).
Klemm, D., Fink, H.-P., et al. (2005) Cellulose: Fascinating Biopolymer and Sustain-
  able Raw Material, Angewandte Chemie International Edition, 44: 3358–93.
Langsner, H. (2005) Nanotechnology Non-traditional Methods for Valuation of Nanotech-
  nology Producers, Innovest Strategic Value Advisors, Inc. New York, NY. August 29.
Lux Research Inc. (2004) Statement of Findings: Sizing Nanotechnology’s Value Chain,
  Lux Research Inc., New York, NY.
Matos, G. and Wagner, L. (1998) Consumption of materials in the United States
  1900–1995, Annual Rev. of Energy Environ. 23: 107–22.
Matthews, E. and Hammond, A. (1999) Critical Consumption Trends and Implications
  Degrading Earth’s Ecosystems, World Resources Institute, Washington, DC.
Moon, R.J. (2008) Nanomaterials in the forest products industry. McGraw-Hill Yearbook
  in Science & Technology, Chicago, IL, pp. 226–9.
Murday, J.S. (2002) The coming revolution: science and technology of nanoscale struc-
  tures. The AMPTIAC Newsletter , Spring, 66: 5–12. http://ammtiac.alionscience.com/
  pdf/AMPQ6 1.pdf.
Nanoscale Science, Engineering and Technology Subcommittee (NSET) (2006) Instru-
  mentation and Metrology for Nanotechnology.
Nanoscale Science, Engineering and Technology Subcommittee (NSET) (2007) Manu-
  facturing at the Nanoscale, Report of the 2002–2004 National Nanotechnology Initia-
  tive Workshops.
Nanoscale Science, Engineering and Technology Subcommittee (NSET) (2008) Strategy
  for Nanotechnology-Related Environmental, Health and Safety Research, February.
National Research Council (2006) A Matter of Size: Triennial Review of the National
  Nanotechnology Initiative, National Academies Press.
National Science and Technology Council Committee on Technology, Interagency Work-
  ing Group on Nanoscience (1999) Engineering and Technology (IWGN) Nanotech-
  nology Research Directions, IWGN Workshop Report Vision for Nanotechnology R&D
  in the Next Decade, September.
Neville, A.C. (1993) Biology of Fibrous Composites: Development beyond the Cell Mem-
  brane, Cambridge University Press, New York, NY.
Perlack, R. et al. (2006), Biomass as Feedstock for a Bioenergy and Bioproducts Indus-
  try: The Technical Feasibility of a Billion-Ton Annual Supply, ORNL/TM-2005/66,
  April 2006.
Podsiadlo, P. et al. (2007) Ultrastrong and stiff layered polymer nanocomposites, Science
  318, October: 80–3.
Pritkethly, M.J. (2003) NanoToday, Dec: 36–42.
                       Nanotechnology and Lignocellulosic Biomass – Relationships   41

Roco, M. (2003) Nanotechnology: convergence with modern biology and medicine,
  Current Opinion in Biotechnology 14: 337–46.
de Rodriguez, N., Thielemans, W., and Dufresne, A. (2006) Sisal cellulose whiskers
  reinforced polyvinyl acetate nanocomposites, Cellulose 13: 261–70.
Roman, M. and Winter, W.T. (2006) Cellulose nanocrystals for thermoplastic reinforce-
  ment: Effect of filler surface chemistry on composite properties. ACS Symposium
  Series, Vol. 938: Cellulose Nanocomposites: Processing, Characterization and Prop-
  erties, Chapter 8.
Samir, M.A.S.A., Alloin, F., and Defresne, A. (2005) Review of recent research in
  cellulosic whiskers, their properties and their application in nanocomposites field,
  Biomacromolecules 5: 612–26.
Sarikaya et al. (2003) Molecular biomimetics: nanotechnology through biology, Nature
  Materials 2, September: 577–85.
Saxton, J. (2007) Nanotechnology: The Future is Coming Sooner Than You Think , Joint
  Economic Committee, United States Congress, http://www.house.gov/jec/, March,
  p. 21.
Schmidt, K. (2007) Green Nanotechnology It Is Easier than You Think , Woodrow Wilson
  International Center for Scholars, Washington DC, April.
Sixth Framework Programme (2005) Nanomaterial Roadmap 2015 Overview on Promis-
  ing Nanomaterials for Industrial Applications, Steinbeis-Europa-Zentrum, Karlsruhe,
  Germany, September.
Society of American Foresters (2003) Perspectives on America’s Forests, Multiple Per-
  spectives on the National Report on Sustainable Forests – 2003 . Society of American
  Foresters, Bethesda, MD.
Sznopek, J. and. Brown, W. (1998) Materials Flow and Sustainability, US Geological
  Survey, Fact Sheet FS-068-98, June.
Technology Transfer Center (2007) Government Funding, Companies, and Applications
  in Nanotechnology Worldwide 2007 , http://www.nano.org.uk/reports.htm.
United Nations Food and Agricultural Organization (2005) State of the World’s
  Forest – 2005 .
Venkataramanan, N and Kawanami, H. (2006) Green synthetic protocol for metal-oxide
  nanowires with natural cellulose, Kagaku, Kogakkai Shuki Taikai Kenkyu Happyo Koen
  Yoshishu 38: K323.
Vincent, J. (2002) Survival of the cheapest, Materials Today, Elsevier Science Ltd,
  December: 28–41.
Vukusic, P., Hallam, B. and Noyes, J. (2007) Brilliant whiteness in ultrathin beetle
  scales, Science, 315: 348.
World Business Council for Sustainable Development, http://www.wbcsd.org.
Xanthos, M. (2005) Modification of polymer mechanical and rheological properties with
  functional fillers, Chapter 2, Functional Fillers for Plastics, M. Xanthos, ed. John
  Wiley & Sons-VCH, GmbH & Co KGaA, p. 21.
Ye, C., Wang, G., Kong, M., and Zhang, L. (2006) Controlled synthesis of Sb2o3
  nanoparticles, nanowires, and nanoribbons, Journal of Nanomaterials, Vol. 2006.
                                                        2
      Biogenesis of Cellulose Nanofibrils
        by a Biological Nanomachine

                               Candace H. Haigler and Alison W. Roberts



2.1     Introduction

Cellulose is synthesized by diverse organisms including prokaryotes, protists, animals,
and plants. However, cellulose achieves its natural dominance within plants, where
ß-1,4-linked glucan chains form long, semi-crystalline fibrils with nanoscale lateral
dimensions. Although these fibrils have been conventionally called ‘microfibrils’, the
term ‘nanofibril’ may be a more appropriate name in the age of nanoscience and nan-
otechnology (1). This term would reflect the fibril width (ranging between ∼1.5 and
25 nm in different cells) and the importance of surface properties in the chemistry and
biological roles of cellulose. A main feature of nanomaterials (with 1–100 nm dimen-
sions) is unique properties that: (a) often arise from a high surface-to-volume ratio; and
(b) bridge between the molecular mechanics that applies to the molecular scale and the
Newtonian physics that applies to larger objects (2, 3). The surface interactions of cellu-
lose with other molecules are major determinants of its role as a scaffold for deposition
of other wall components and the coherence and physical properties of the composite
cell wall.
   In the plant cell wall, the cellulose nanofibrils are commonly 2–6 nm in diameter, with
the larger nanofibrils usually occurring in secondary walls. All plant cells contain about
∼15% cellulose in the thin, extensible primary walls that surround growing cells. In this
role, the cellulose nanofibrils are able to constrain the direction of plant cell expansion
(as driven by isodiametric turgor pressure). This function derives from the high breaking
strain energy (5 − 50 × 106 J/m3 ) and tensile strength (∼GN/m2 ) of cellulose fibrils, in
the same range as high tensile steel. In proportion to its density, which is lower than steel,

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
44    The Nanoscience and Technology of Renewable Biomaterials

cellulose is the strongest and highest energy storing material known (4). After expansion
ceases, some plant cells deposit thick (or secondary) walls, which often contain higher
percentages of cellulose. For example, the secondary walls within the water-conducting
and supportive xylem tissue contain 40–50% cellulose. These thick walls confer bulk
material strength to plant cells and tissues, as well as supporting other functions such
as water conduction. The accumulation of xylem becomes most pronounced in large
woody trees, which, together with the abundance of aquatic algae, results in cellulose
being the most abundant renewable biomaterial.
   In other resonances with nanoscience, cellulose nanofibrils are produced in associa-
tion with one of nature’s most remarkable biological nanomachines, a cellulose synthesis
complex (CSC), which is the focus of this chapter. (Historically, other names for such
complexes have been microfibril terminal complex (TC) or ‘rosette’ for the particular
CSC of land plants and their close relatives (see below). Recently, the acronym CSC has
sometimes been used to denote ‘cellulose synthase complex’, but we prefer ‘cellulose
synthesis complex’ to allow the possibility that other proteins besides cellulose synthase
(CS, or CesA in plants) will be identified as part of this complex.) It has also been noted
that the plant CSC functions within a ∼100 nm planar area bridging the cortical cyto-
plasm, plasma membrane, and exoplasmic space (or the area where cellulose nanofibrils
crystallize before they are integrated into the cell wall) (5).
   Given the importance of cellulose in nature as well as the energy and material needs
of human civilization, it is astonishing that we understand so few details about the
mechanistic operation of the CSC. The purpose of this chapter is to summarize cur-
rent knowledge and remaining questions about how the CSC of land plants acts as a
nanomachine to produce cellulose I fibrils.


2.2   Background

Before 1999, there were several major advances in our understanding of plant cellulose
biogenesis. The reviews of Delmer (1999) and Tsekos (1999) can be consulted for fur-
ther details and primary citations on the background findings that are summarized below.
Freeze fracture transmission electron microscopy (FF-TEM) generates micrographs of
metal replicas of cryo-fractured, shadowed, membranes including their embedded pro-
teins. Beginning in the 1970s, FF-TEM was used to see high resolution views of large,
distinctive, protein aggregates (CSCs) in the plasma membranes of cells that synthe-
size cellulose. Often the CSCs were at the termini of cellulose fibril impressions in
the plasma membrane. However, the geometry of individual CSCs and the arrange-
ment within the membrane of multiple CSCs varied between organisms, and especially
between major lineages. As observations of CSCs in various organisms accumulated,
it became apparent that cellulose nanofibril lateral dimensions, extent of crystallinity,
modes of crystallization (to form cellulose Iα or Iß, see below), and manner of possible
macrofibril assembly varied in parallel with CSC geometry and, occasionally, higher
order aggregation of CSCs.
   The ability to interfere with cellulose crystallization in vivo in both prokaryotic and
eukaryotic cells by addition of cellulose-binding molecules demonstrated that there was a
temporal lag between polymerization and crystallization and led to the unifying principle
                     Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine      45

that the aggregation of CS proteins was an essential part of natural cellulose I biogenesis.
The aggregation of multiple polymerization sites in the plasma membrane allows the
synthesis of numerous ß1,4-glucan chains in a spatially compact area, which facilitates
cellulose I crystallization with parallel chain conformation by reducing the chances of
individual chain folding. FF-TEM imaging of unusual CSC arrays during secondary
wall deposition in a green alga, Micrasterias denticulata, showed that multiple CSCs
could contribute to the biogenesis of a single cellulose nanofibril, the dimensions of
which were determined by the geometric arrangement of the CSCs. The density of
CSCs is also higher in xylem cells synthesizing cellulose-rich secondary walls, which
have thicker cellulose nanofibrils compared to cells synthesizing primary walls.
   Although these cell biological observations made a compelling case that CSCs did
facilitate cellulose nanofibril biogenesis, it was not until 1990 and afterwards that
this hypothesis was proven through identification of CS genes. After identification in
1990 of a prokaryote CS gene in Gluconacetobacter xylinus (then called Acetobacter
xylinum), plant CesA genes encoding structurally similar proteins were identified in 1996
among genes expressed during secondary wall deposition in cotton fiber. An essential
role for CesA proteins in plant cellulose synthesis was demonstrated genetically in
1998 through mutant analysis in Arabidopsis. Although there was substantial sequence
divergence between major lineages, all of the CS proteins were classified as UDP-
glucose: 1,4-ß-D-glucosyltransferase enzymes in glycosyltransferase family 2 (GT2).
Glycosyltransferases are enzymes that transfer sugar moities from activated sugar
donors, thereby forming a glycosidic bond in a short oligosaccharide or long chain
polysaccharide. The CSs are large membrane spanning proteins, for example, the
Arabidopsis CesA proteins range from 985 to 1088 amino acids and ∼110–120 kDa
mass. The fundamental advances summarized above set the stage for probing more
deeply into the nanoscale structure and function of the CSC during the last decade.


2.3   CesA Protein is a Major Component of the Plant CSC

Land plants as well as their charophyte algal progenitors (collectively called the strepto-
phytes) have a rosette-type CSC with hexagonal symmetry. In 1999, immunolabeling in
conjunction with FF-TEM showed that the rosette CSC contained CesA protein (6). The
rosette region of the CSC is about 25 nm in diameter and contains six intramembrane
particles. Each of the six subunits is ∼8 nm in diameter, reflecting an aggregate of CesA
proteins (Figure 2.1). It is important to recognize that the rosettes visualized by FF-TEM
are only the signatures on the cleaved inner leaflet of the plasma membrane of aggregates
of CesA transmembrane helices (TMH) traversing the membrane. The catalytic loop is
believed to exist in the cytoplasm based on protein structure prediction algorithms and
the cytoplasmic location of substrate UDP-glucose (7). The predicted proximity of adja-
cent TMH suggests that very little of the protein exists on the exoplasmic surface of the
plasma membrane.
   Figure 2.2 shows schematic diagrams of the domain structure of CS proteins of organ-
isms from diverse lineages (Gluconacetobacter xylinus (a eubacterium), Anabaena vari-
abilis (a photosynthetic cyanobacterium), Dictyostelium discoideum (a protistan social
amoeba), and Gossypium hirsutum (a vascular plant)). All known CS proteins share
46   The Nanoscience and Technology of Renewable Biomaterials




Figure 2.1 Transmission electron micrograph of rosette CSCs within the plasma membrane
of Zinnia elegans cells that were depositing secondary walls during tracheary element
differentiation in suspension culture. The specimen was prepared by cryo-fracture followed
by shadowing with platinum and carbon, and the inner leaflet of the cleaved plasma
membrane is shown. The six subunits of the rosette CSCs (circled) reflect aggregates of CesA
transmembrane helical domains traversing the membrane. The CSCs in the main micrograph
reflect the typical maximum density seen during secondary wall deposition in primary xylem
elements. The inset in the upper left shows one CSC at higher magnification. The main
micrograph and the inset show, respectively, the results of unidirectional or rotary shadowing
by platinum/carbon. The bars in the main micrograph and the inset represent 30 nm and
10 nm, respectively.


predicted N- and C-terminal transmembrane domains (TMD) where 2–6 TMHs pass
through the membrane. In total, there are 8–9 TMH in CS proteins, with fewer near
the N-terminus, more at the C-terminus, and a long intervening hydrophilic region (the
catalytic region). The U1–U4 regions are 5–7 amino acids in length with 70–80% amino
acid conservation in all species; these are involved in the catalytic function (see below).
Of the examples in Figure 2.2, only G. hirsutum is known to have rosette CSCs: A. vari-
abilis has an unknown CSC structure; the D. discoideum CSC can exist in linear or block
like arrays; and the G. xylinus CSC forms an extended linear array. Correspondingly,
only the plant-type CESA proteins, e.g. from G. hirsutum, are known to include an
N-terminal, cysteine-rich, zinc-binding domain (ZnBD) and a variable ‘class specific
region’ (CSR) between U2 and U3 (8). The sequence of the insertion between the U1
and U2 domains (designated the CR-P, for ‘conserved region-plant’) is also unique to
plant-type CesAs (9). Although other CS proteins from D. discoideum (10) and certain
cyanobacteria such as A. variabilis (11) have insertions in the same area, these are not
clearly homologous to the CR-P region.
   The known or possible functions of these protein domains will be discussed further
below. In general, the absence of the ZnBD and the CR-P and CSR regions in all
non-plant CS proteins demonstrates that they are not needed to carry out ß-1,4-glucan
polymerization. Instead, they could modulate the additional functions associated with
                     Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine     47




Figure 2.2 Protein structure diagrams showing domains within representative CS pro-
teins from Gossypium hirsutum (U58283.1, vascular plant type), Anabaena variabilis
(YP− 322086.1, cyanobacterial type), Dictyostelium discoideum (AF163835.1, protistan type),
and Gluconoacetobacter xylinus (CAA38487.1, eubacterial type). The diagrams are aligned
at the U3 region. Domains shown by color blocks are: zinc-binding domain (Zn, yellow);
transmembrane domains (TMD, black), each of which contains multiple TMH; CS protein
conserved regions (U1-U4, red); conserved region plant (CR-P, green); and class-specific
region (CSR, purple). The diagrams are to scale, with 50 amino acids included in the length
indicated in the KEY (upper right).


plant vs., for example, bacterial CSCs, i.e., assembly into ‘rosette’ geometry, co-assembly
with other isoforms, interaction with other proteins in the plant cell cortex, and movement
in an oriented manner within the plasma membrane.
   Mutant analysis has shown that several other proteins are required for cellulose bio-
genesis, as judged by cellulose deficiencies and/or arrested or abnormal development in
mutants of Arabidopsis and a few other species. Several of these, including Korrigan,
the COBRA-type proteins, and Kobito, are membrane spanning proteins that may be
integrated within or near the CSC (12). This possibility exists even for proteins not
yet shown to exist in the plasma membrane by fluorescence technology, which may not
resolve proteins integrated within the membrane in groups of fewer than 10 molecules
(13). Other proteins not yet identified may also function along with the CSC to accom-
plish cellulose biogenesis. In the discussion below, we will emphasize CesA since it is
the only proven component of the plant CSC.


2.4   The Functional Operation of the CSC

In the absence of a real mechanistic understanding or even identification of all of the
components, we expect the plant CSC to carry out several activities. As the CSC
mediates cellulose biogenesis to build the cell wall it must: assemble with genetically
determined morphology; stabilize in operational form in the plasma membrane; acquire
UDP-glucose substrate; polymerize glucose with ß-1,4-linkage; operate so that fibrils
48   The Nanoscience and Technology of Renewable Biomaterials

emerge outside the plasma membrane; control cellulose chain length and fibril size;
possibly control cellulose crystallization; and move in the plasma membrane to spin out
cellulose fibrils. We will consider each of these functions in order to illustrate both the
remarkable capabilities of this cellular fibril-spinning nanomachine and the extensive
lack of mechanistic information that currently exists. Readers wishing to consider any
of these functions in detail are encouraged to read the cited references as well as recent
reviews (12–16).

2.4.1 Assemble with Genetically Determined Morphology
As reviewed by Tsekos (1999), there are a variety of CSC morphologies in nature, and
it is evident that these are genetically determined. The rosette CSC is known so far
to exist only in the land plants and the charophycean algae. Its existence is correlated
with an increase in the number of CesA genes in these lineages, although there may not
be a causal relationship (see the section on phylogenetic analysis below). In flowering
plants (angiosperms), there is experimental evidence that different types of CesA pro-
teins (isoforms) are required for cellulose synthesis and CSC assembly (see review (12)).
Genetic analysis of mutants with cellulose deficient phenotypes in Arabidopsis showed
that AtCesA1, 3, and 6 (or 6-like proteins) are all required for primary wall cellulose
synthesis (17–21), whereas AtCesA4, 7, and 8 are all required for secondary wall cellu-
lose synthesis in xylem and fibers (22–24). Various phylogenetic analyses (25–28) group
the CesA genes from angiosperms into six clades (or sequence groups with a common
evolutionary origin) (Figure 2.3). Each of these six clades includes one of the three
Arabidopsis CesA genes required for primary wall synthesis/CSC assembly (AtCesA1,
3, or 6-like; defining clades P1, P2, or P3, respectively) or one of the three CesA genes
required for secondary wall synthesis/CSC assembly (AtCesA4, 7, or 8 ; defining clades
S1, S2, of S3, respectively). The six clades also include orthologs from other angiosperm
taxa, and mutant analysis in other angiosperm species is generally consistent with the
results from Arabidopsis. (Orthologs are genes in two or more species that evolved from
a common ancestor and typically retain the same function.)
   Doblin et al. (2002) (29) proposed a modification of a previous model (30) that
explains the geometry of rosette CSCs as a function of the inter- and intra-particle inter-
action between three distinct CesA subunits that associate with each other through distinct
binding sites. For both the primary and secondary wall cases, there is accumulating evi-
dence that each member of a CesA triad plays a distinct role in the assembly of rosettes
(17–24, 30–32). First, inhibition of cellulose crystallization in the temperature-sensitive
Arabidopsis cesA1 (rsw1 ) mutant, which has a single amino acid substitution in the cyto-
plasmic loop, was accompanied by depletion of rosette CSCs in the plasma membrane
(33). When grown at the restrictive temperature, AtCesA1, 3, and 6 behaved indepen-
dently during immunoprecipitation experiments rather than in a coordinated ∼840 kDa
complex as was found in plants grown at the permissive temperature (32). Second,
specific association between CesA subunits from both Arabidopsis and cotton has been
demonstrated in vitro (17, 22–24, 32, 34). The ZnBD is able to mediate the in vitro
association of cotton CesA proteins either as homo- or heterodimers, directly implicating
this feature in rosette assembly. For cotton GhCesA1, dimerization is regulated by the
redox state of the ZnBD (34). Third, interaction in vivo between all possible pairs of
                      Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine         49




Figure 2.3 Majority consensus bootstrap tree (heuristic search method, 1000 replicates) of
CesA protein sequences within several plant lineages. The tree is rooted with CesA1 from
Mesotaenium caldariorum (Mc, AAM83096), a green alga with rosette CSCs (pictured at left).
Bootstrap values above 50 are shown. Sequences were aligned using the PRANK algorithm
(default values). By reference to Arabidopsis, clades containing primary cell wall CesAs (P1-P3)
are boxed in pink, orange, or green, and those containing secondary cell wall CesAs (S1-S3)
are boxed in yellow, blue, and red. Pictured at the top and bottom right, respectively, are
micrographs of secondary cell walls (in a scanning electron microscope view of cultured xylem
elements) and primary cell walls (in a thin section of plant tissue viewed in TEM). Sequences
included in the tree are from Physcomitrella patens (Pp, DQ902545, DQ902546, DQ902547,
DQ902548, DQ902549, DQ902550, DQ902551), Selaginella moellendorfii (Sm, JGI protein
IDs:141535, 73698, 75715, 163575), Pinus taeda (Pta, AY789650, AY789651, AY789652),
Pinus radiata (Pr, AAQ63935.1), Oryza sativa (Os, AAT77342, AAP21426, BAD30574,
BAD87094, AAO41140, XM− 477282, AAP54202, BAC57282, DAA01752), and Arabidopsis
thaliana (At, TAIR locus IDs: At4g32410, At4g39350, At5g05170, At5g44030, At5g09870,
At5g64740, At5g17420, At4g18780, At2g21770, At2g25540).


primary wall CesAs 1,3, and 6 have been demonstrated by bimolecular fluorescence
complementation (17). Fourth, CSC assembly and secretion depends on the presence of
a complete triad of CesA proteins coexpressed in developing xylem of Arabidopsis (24).
For primary wall CesA proteins, the region from the start of the second TMH to the
C-terminus, helps to determine which site in the CSC is accessible to a chimeric CesA
(a fusion of parts of two native CesAs) (20).
50   The Nanoscience and Technology of Renewable Biomaterials

2.4.2 Stabilize in Operational Form in the Plasma Membrane
The assembly of the rosette CSCs occurs in the plant endomembrane system. The
organized rosette was seen by FF-TEM in cells synthesizing secondary walls within distal
areas of Golgi cisternae, as well as in small vesicles free in the cytoplasm and apparently
fusing with the plasma membrane (1). When one member of the triad of secondary
cell wall CesA proteins is absent, the remaining two CesA proteins accumulate in the
Golgi (24). Fluorescently labeled AtCesA6 has also been observed to traffic through the
Golgi compartment during primary wall synthesis (36). Presently we do not know how
CSC activity is prevented in vascular plants prior to its incorporation into the plasma
membrane.
   Specific factors also maintain the stability of preassembled rosettes after their insertion
into the plasma membrane, as indicated by FF-TEM images of cultured plant cells that
were synthesizing secondary walls when they were treated with a cellulose synthesis
inhibitor. In this case, assembled rosette CSCs were observed in the membrane, but they
appeared to expand in diameter through separation of the six subunits, then disappear
quickly as organized entities (35). In the absence of immunolabeling, it was not possible
to discern the fate of individual subunits after the CSC dispersed. Similarly, another
cellulose synthesis inhibitor caused detectable fluorescently labeled AtCesA6 to clear
from the membranes of cells synthesizing primary walls (36).
   Older cell biological evidence and recent biochemical experiments with GhCesA1
from cotton fiber show that the half-life of a CesA in the plasma membrane is <30 min,
which is unusually short for a membrane protein (37). The functional significance of this
short half-life is not yet known. Although CesA proteins could be recycled between the
plasma membrane and the endomembrane system, recent data show that they can also
be proteolytically degraded. Targeting for degradation can be enhanced for AtCesA7 by
phosphorylation within the catalytic domain (38), and monomeric ZnBDs (and, poten-
tially, intact CesA proteins) are also more rapidly degraded (37). It is not known how
rapid turn-over in CesA proteins regulates the outcomes of CSC function in vivo.

2.4.3 Acquire UDP-Glucose Substrate
As in bacteria, UDP-glucose is the immediate substrate for plant cellulose synthesis.
Common catalytic function within the GT2 family is reflected by the conservation of
three aspartate (D) residues and a QxxRW motif embedded within the conserved U1,
U2, U3 and U4 regions (7); (Figure 2.2). (QxxRW signifies five amino acids: glutamine,
two variable residues, arginine, and tryptophan.) The heterologously expressed cotton
fiber CesA protein (GhCesA1) bound UDP-glucose (9), and, using UDP-glucose sub-
strate, glucan polymerization was initiated from a putative primer (sitosterol-glucoside)
by membranes from yeast expressing the GhCesA1 gene (39). Cleavage of the UDP
moiety releases energy that can be used for addition of a liberated glucose residue to a
growing ß-1,4-linked glucan chain.
   There is evidence that some cells with secondary walls, such as cotton fiber and xylem
sclerenchyma, preferentially accept UDP-glucose generated by sucrose synthase even
though UDP-glucose is usually found within a free pool in the cytoplasm (40). Despite
its name, sucrose synthase most commonly degrades sucrose to release UDP-glucose
and fructose in heterotrophic cells. In cells synthesizing cellulose-rich secondary walls,
                     Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine     51

sucrose synthase is abundantly located just below the plasma membrane where the CSC
is located (41, 42). The relationship of sucrose synthase to CesA may have become more
specialized in wood and cotton fibers, which synthesize massive amounts of cellulose,
because Arabidopsis growth was unaltered (under normal conditions) when six sucrose
synthase isoforms were down-regulated singly or in pairs of related genes in Arabidopsis
(43). However, a mechanism for preferential transfer of UDP-glucose from sucrose
synthase to at least some isoforms of CesA is not known. High concentrations of
sucrose promote association of sucrose synthase with membranes (44), which establishes
a mechanistic basis for the idea that high rate cellulose synthesis supported by sucrose
synthase would occur more frequently under ideal growth conditions when abundant
fixed carbon is available (45).

2.4.4 Polymerize Glucose with ß-1,4-Linkage
Although CS proteins in different lineages have low overall amino acid sequence homol-
ogy, they share similar structure and common domains associated with their catalytic
function as glycosyltransferases in the GT2 family that are capable of adding glucose
derived from UDP-glucose to a growing polymer. CesA is thought to polymerize glucose
in a processive manner, meaning that it remains bound to the product during successive
polymerization steps. To account for multiple CesA isoforms in most land plants and to
rationalize some data, it has been proposed that certain CesAs may generate a short-chain,
possibly lipid-linked, primer, with other CesAs perhaps facilitating the polymerization
of high molecular weight cellulose (39, 46). Data consistent with a lipid-linked glucan
primer as part of cellulose synthesis are still reported (see discussion in (32)), but the
actual biochemical product in vivo of any of the plant CesA isoforms is unproven.
   The ß-1,4 linkage distinguishes cellulose from other glucose homopolymers, includ-
ing starch (with α-1,4 glucan backbone) and callose (with ß-1,3 glucan backbone) and
generates the flat-chain conformation that facilitates crystallization of extended parallel
chains into cellulose I. This linkage implies, however, that alternate glucose residues are
flipped 180 ◦ relative to each other so that the actual repeating unit is cellobiose. This
poses special challenges for polymerase activity. One possible explanation is that for
each forming polymer chain, paired active sites from two CS isoforms exist in the CSC
to hold glucose in opposite orientation (8, 31, 47, 48). Data consistent with this idea
exist for the analogous systems of chitin and hyaluronan synthase (reviewed in (49)).

2.4.5 Operate so that Fibrils Emerge Outside the Plasma Membrane
Since substrate UDP-glucose is acquired in the cytoplasm, it is possible that the aggre-
gated TMH of the CesA protein(s) create a pore in the membrane to allow elongating
glucan chains to exit on the cell wall side. Data to demonstrate the transport mechanism
for polymerized ß-1,4 glucan are lacking.

2.4.6 Control Cellulose Chain Length
Cellulose chain length is determined by the degree of polymerization (DP) and is vari-
able during development as shown by analysis of single-celled cotton fiber. During
primary wall deposition, cotton fiber cellulose has a wider length distribution, averaging
52   The Nanoscience and Technology of Renewable Biomaterials

∼4000 DP. During secondary wall deposition, the distribution is more narrow, averaging
∼10,000 DP (50), with each 2000 glucose units contributing to ∼1 µm of chain length.
The degree of polymerization of cellulose within crystallites is positively correlated with
cotton fiber strength (51), and similar effects are likely to occur for individual cellu-
lose nanofibrils within plant cell walls. It is possible that the different CesA isoforms
employed for primary vs. secondary wall synthesis help to control the degree of poly-
merization. For example, differences in the CSR region occur between the primary and
secondary cell wall CesA proteins (52), and these are likely to affect aspects of protein
function that are as yet unknown. Control of glucan chain length could occur directly
via limitation of the average time of continuous polymerization by a single CesA protein
or be controlled indirectly via different CesA lifetime in the plasma membrane (7); see
also details about CesA lifetime (above).

2.4.7 Control Cellulose Nanofibril Diameter
Nanofibril size can be regulated by the number of CesA proteins aggregated together
in the plasma membrane, which in turn determines the number of ß-1,4-glucan chains
that are likely to coalesce without interference from other molecules. Nanofibril cross-
sectional dimensions range from 2 to 25 nm and have been correlated with CSC geom-
etry or higher order aggregation in Gluconoacetobacter xylinum (53), Dictyostelium
discoideum (54), various lineages of algae (55), and tunicates (56). In eukaryotic
D. discoideum, the ability of only one CesA protein (57) to assume different aggrega-
tion states and produce cellulose fibrils with different sizes has been demonstrated (54).
The number of CesA proteins in a rosette CSC is still unknown, but current data sug-
gest that it may be less than the 36 subunits commonly modeled (29) and diagrammed
in Figure 2.3. Recent analysis of extracted, hydrated small cellulose nanofibrils with
cellulose IV crystallinity (indicative of good longitudinal but poor lateral chain order)
indicated that they were ∼2.4–3.2 nm in diameter, which is consistent with 15–25 total
chains (58). Especially given the possibility of fibril coalescence during sample prepa-
ration, such data argue against a fundamental unit of plant cellulose synthesis with 36
chains. Therefore, it may be more likely that 3–4 CesA proteins exist within each
subunit of the rosette CSC, for a total of 18–24 CesAs. This possibility is consistent
with rough estimates of space filled by TMH in the plasma membrane, which predicted
4 CesAs within each of six subunits of the rosette CSC (59). The minimum crystal-
lite size for cellulose I was estimated to be 2.8 × 2.8 nm, which would be contained
within a ∼3.5 nm fibril, including less ordered surface chains that may be present in
the cellulose nanofibrils of vascular plants (60). Therefore, one rosette CSC may not
by itself produce cellulose I fibrils. Instead, fibril coalescence could occur before final
crystallization, although rosette CSCs have not been seen in geometric arrays or tightly
packed aggregates even during secondary wall synthesis in plants.
   Prior to crystallization, other matrix polymers with ß-1,4-glycan backbones such as
xyloglucan and xylan are able to interact with cellulose and limit the extent of its higher
order coalescence and crystallization. The degree to which this occurs would vary in
parallel with the type and amount of matrix polymers in the cell wall space; for example
larger cellulose nanofibrils and crystallites are produced in tension wood that are severely
depleted in matrix components (61).
                     Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine       53

2.4.8 Control Crystallization?
Niklas (1992) points out that the parallel chain conformation within cellulose nanofibrils
is an essential basis of the high elastic modulus and tensile strength of cellulose within
cell walls. As explained before, the organized CSC is a necessary facilitator of cellulose
I crystallization through its extrusion of numerous ß-1,4-glucan chains in close prox-
imity. In addition, the nanostructure of the CSC seemingly regulates fine differences
in the allomorph of cellulose I synthesized by different species. Spectral analysis via
solid-state 13 C NMR showed that native celluloses include species-specific mixtures of
two allomorphs, Iα and Iß , which differ primarily in their hydrogen bonding patterns.
Typically, higher plants with rosette type CSCs produce 60–80% of the Iß form despite
variances in the size and crystallinity of nanofibrils (62, 63). Currently, it is not known if
or how differences in CSC geometry lead to such subtle differences in chain associations
and crystallinity.
   Other proteins may also impact cellulose crystallinity, including specialized endo-ß-
1,4-glucanases (i.e. CMCax or Korrigan) associated with cellulose synthesis in prokary-
otes and plants (64–66). It is not known how an endo-ß-1,4-glucanase is functionally
integrated with the operation of the CSC.

2.4.9 Move in the Plasma Membrane as it Spins out Cellulose Nanofibrils
Both live cell imaging of labeled CesA protein (36) and biophysical modeling (67)
supported the early hypothesis that the force of crystallization causes the CSC to move
in the plasma membrane as it spins cellulose fibrils. Microscopy showed an average rate
of movement for labeled CesA protein of 330 nm/min during primary wall synthesis in
Arabidopsis hypocotyls. The role of the cytoskeleton in determining the orientation of
this movement, possibly via a feedback mechanism also involving pre-existing nanofibrils
as controlling entities, is beyond the scope of this chapter.



2.5   Phylogenetic Analysis

Phylogenetic analysis is valuable for assessing essential elements and possible points of
functional variation in the CSC nanomachine. Phylogenetic analysis is a computational
method allowing the estimation of evolutionary relationships among groups of divergent
CesA gene sequences. When sequence relationships are viewed in the context of the
timing of evolution of different groups, plant morphology, and CSC structure, we can
generate hypotheses about control of basic function and possible functional diversification
of the CSC. This approach is the most powerful in the context of fully sequenced
genomes that allow complete gene families to be identified. Therefore, the discussion
below emphasizes plant species with sequenced genomes, and this approach will become
even more informative as the number of sequenced genomes increases.

2.5.1 Possible Functional Diversification of CS Proteins
The presence of orthologs, from other seed plants, of the functionally distinct Arabidopsis
CesAs within clades P1-3 and S1-3 indicates that heterotrimeric rosette CSCs specialized
54   The Nanoscience and Technology of Renewable Biomaterials

for primary and secondary cell wall biosynthesis are common among angiosperms. For
example, orthologs of AtCesA 4, 7 , and 8 are expressed in wood and cotton fibers
during secondary wall deposition (9, 68, 69). The use of a different family of CesA
genes for primary vs. secondary wall synthesis could cause differences in the fine
structure of cellulose nanofibrils that are not yet fully known. Phylogenetic analyses
(cited above) generally agree that the divergence of primary and secondary wall types
of CesAs occurred prior to the diversifications that produced the three different primary
and secondary CesAs. Therefore, rosettes composed of P1-P3 or S1-S3 CesA subunits
evolved independently in the context of both primary and secondary wall deposition.
Among the sequenced and annotated angiosperm genomes, CesA genes in the S clades
are less diversified than those in the P clades, which is consistent with unknown greater
constraints on functionality in the S clades.
   The P3 clade of Arabidopsis is unique in having four members (AtCesA2 , 5 , 6 and
9) with origins that can be traced to the two most recent duplications of the Arabidop-
sis genome (70). Selective gene retention in this lineage is consistent with selection
favoring functional specialization, and there is evidence for tissue-specific roles of the
AtCesA6-like genes in Arabidopsis (17, 71). Whereas AtCesA1 and AtCesA3 null muta-
tions are embryo lethal, mutations in CesA2,5,6 and/or 9 yield phenotypes related to
cellulose deficiency in certain organs. However, the genome duplications that gave rise
to these paralogs occurred after the divergence of the crucifer lineage. (Paralogs are
genes within a genome that share a common ancestor but differ in function.) There-
fore, the particular case of functional specialization among AtCesA2,5,6 , and 9 cannot
be generalized to other angiosperms groups. Accounting for all 10 of the Arabidopsis
CesA genes, AtCesA10 arose from a recent duplication of AtCesA1 , and it has limited
expression within the Arabidopsis plant.
   As the CesA gene families in other species are understood in more detail, similar
cases of diversification and functional specialization may well be discovered. Addi-
tional research is needed to reveal whether CesA diversification within specific clades
and taxa underpin adaptive changes in the regulation of cellulose synthesis and/or the
nanoscale properties of cellulose that enhance fitness in relation to particular growth
habits and/or ecological niches. For example, among the sequenced and annotated
angiosperm genomes, Populus trichocarpa has the most numerous CesA genes, with
18 (25). We can speculate that 18 P. trichocarpa CesA genes (vs. 10 in Arabidopsis)
might support the massive scale of cellulose synthesis in wood as well as developmental
specializations required for large trees to survive stress, such as the ability to synthesize
cellulose-rich tension wood with larger cellulose nanofibrils upon bending (61).
   The CesA gene families of nonvascular plants (the moss, Physcomitrella patens) and
early-divergent vascular plants (the lycophyte or spikemoss, Selaginella moellendorfii )
can shed light on the significance of CesA diversification in seed plants. Phylogenetic
analyses support the contention that the common ancestor of P. patens and S. moellen-
dorfii shared a single CesA gene (27), as did the common ancestor of lycophytes and
seed plants (A. Roberts, unpublished). Although S. mollendorfii is a vascular plant, it
has no orthologs of angiosperm P1-P3 or S1-S3 CesA genes. Therefore, divergence of
the P and S CesA clades was not a precondition for evolution of vascular cells with sec-
ondary walls. The number of CesA genes increased and S1-S3 clades diverged before the
divergence of gymnosperms and angiosperms within the euphyllophyte lineage, which
                     Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine     55

includes these groups and ferns (26). However, additional research would be required
to evaluate whether the S1-S3 protein types were correlated with any changes in the
nanoscale structure of cellulose.
   Clearly, the basic structure or catalytic function of the rosette CSC do not depend on
the P1-P3 or S1-S3 clades per se since neither P. patens nor S. mollendorfii has members
of these clades. Heterotrimeric rosettes cannot be ruled out in early land plants, however,
because P. patens (27) and S. mollendorfii (A. Roberts, unpublished) have seven and
four functional CesA genes, respectively. If the heterotrimeric condition exists in these
descendents of early land plants, it must have evolved independently. This case would
imply that the heterotrimeric CSC evolved independently at least three times, which
would be consistent with it having strong adaptive advantages. Alternatively, monomeric
rosette CSCs in P. patens and/or S. mollendorfii would suggest that heterotrimeric CSCs
evolved later, probably aiding the structural and/or developmental complexity arising
during the radiation of land plants through serving a regulatory function or conferring
subtle functional differences.



2.6   Conclusion

Further research will reveal how the detailed molecular structure of the CSC governs
cellulose structure at the nanoscale and also set the stage for advantageous manipula-
tion of cellulose properties in next-generation biomass plants and, possibly, synthesis of
cellulose in cell-free systems. Fully understanding the structure and mechanistic oper-
ation of the cellulose synthesizing nanomachine is surely one of the grand challenges
presented to us by the natural world. Solving the puzzle of its structure and operational
mechanisms will have positive impacts on a sustainable future through allowing full
exploitation of renewable cellulose to meet our material and energy needs.


References

1. Haigler, C.H.; Brown, R.M., Jr., Transport of rosettes from the Golgi apparatus to the
   plasma membrane in isolated mesophyll cells of Zinnia elegans during differentiation
   to tracheary elements in suspension culture. Protoplasma 1986, 134, 111–20.
2. Roco, M.C., Nanoparticles and nanotechnology research. J Nanoparticle Res 1999,
   1, 1–6.
3. Roco, M.C., Nanotechnology: convergence with modern biology and medicine.
   Curr. Opin. Biotechnol. 2003, 14(3), 337–46.
4. Niklas, K.J., Plant Biomechanics. University of Chicago Press: Chicago, 1992.
5. Lindeboom, J.; Mulder, B.M.; Vos, J.W.; Ketelaar, T.; Emons, A.M., Cellulose
   microfibril deposition: coordinated activity at the plant plasma membrane.
   J. Microsc. 2008, 231(2), 192–200.
6. Kimura, S.; Laosinchai, W.; Itoh, T.; Cui, X.; Linder, C.R.; Brown, R.M., Jr.,
   Immunogold labeling of rosette terminal cellulose-synthesizing complexes in the
   vascular plant Vigna angularis. Plant Cell 1999, 11, 2075–85.
56     The Nanoscience and Technology of Renewable Biomaterials

7.    Delmer, D.P., Cellulose biosynthesis: Exciting times for a difficult field of study.
      Annu. Rev. Plant Physiol. Plant Mol. Biol. 1999, 50, 245–76.
8.    Vergara, C.E.; Carpita, N.C., b-D-glycan synthases and the CesA gene family:
      lessons to be learned from the mixed-linkage (1→3),(1→4)b-D-glucan synthase.
      Plant Mol. Biol. 2001, 47, 145–60.
9.    Pear, J.R.; Kawagoe, Y.; Schreckengost, W.E.; Delmer, D.P.; Stalker, D.M., Higher
      plants contain homologs of the bacterial celA genes encoding the catalytic subunit
      of cellulose synthase. Proc. Natl. Acad. Sci. USA 1996, 93, 12637–42.
10.   Blanton, R.L.; Fuller, D.; Iranfar, N.; Grimson, M.J.; Loomis, W.F., The cellulose
      synthase gene of Dictyostelium. Proc. Natl. Acad. Sci. USA 2000, 97, 2391–6.
11.   Nobles, D.R.; Romanovicz, D.K.; Brown, R.M., Jr., Cellulose in cyanobacteria.
      Origin of vascular plant cellulose synthase? Plant Physiol . 2001, 127, 529–42.
12.   Taylor, N.G., Cellulose biosynthesis and deposition in higher plants. New Phytol .
      2008, 178(2), 239–52.
13.   Somerville, C., Cellulose synthesis in higher plants. Annu. Rev. Cell Dev. Biol .
      2006, 22, 53–78.
14.   Mutwil, M.; Debolt, S.; Persson, S., Cellulose synthesis: a complex complex. Curr.
      Opin. Plant Biol . 2008, 11(3), 252–7.
15.   Joshi, C.P.; Mansfield, S.D., The cellulose paradox – simple molecule, complex
      biosynthesis. Curr. Opin. Plant Biol . 2007, 10(3), 220–6.
16.   Saxena, I.M.; Brown, R.M., Jr., Cellulose biosynthesis: current views and evolving
      concepts. Ann. Bot. (Lond) 2005, 96(1), 9–21.
17.   Desprez, T.; Juraniec, M.; Crowell, E.F. et al., Organization of cellulose synthase
      complexes involved in primary cell wall synthesis in Arabidopsis thaliana. Proc.
      Natl. Acad. Sci. USA 2007, 104(39), 15572–7.
18.   Desprez, T.; Vernhettes, S.; Fagard, M.; et al., Resistance against herbicide isoxaben
      and cellulose deficiency caused by distinct mutations in same cellulose synthase
      isoform CESA6. Plant Physiol . 2002, 128(2), 482–90.
19.   Burn, J.E.; Hocart, C.H.; Birch, R.J.; Cork, A.C.; Williamson, R.E., Functional
      analysis of the cellulose synthase genes CesA1 , CesA2 , and CesA3 in Arabidopsis.
      Plant Physiol . 2002, 129(2), 797–807.
20.   Wang, J.; Howles, P.A.; Cork, A.H.; Birch, R.J.; Williamson, R.E., Chimeric pro-
      teins suggest that the catalytic and/or C-terminal domains give CesA1 and CesA3
      access to their specific sites in the cellulose synthase of primary walls. Plant Physiol.
      2006, 142(2), 685–95.
21.   Persson, S.; Paredez, A.; Carroll, A.; et al., Genetic evidence for three unique
      components in primary cell-wall cellulose synthase complexes in Arabidopsis. Proc.
      Natl. Acad. Sci. USA 2007, 104(39), 15566–71.
22.   Taylor, N.G.; Laurie, S.; Turner, S.R., Multiple cellulose synthase catalytic subunits
      are required for cellulose synthesis in Arabidopsis. Plant Cell 2000, 12, 2529–39.
23.   Taylor, N.G.; Howells, R.M.; Huttly, A.K.; Vickers, K.; Turner, S.R., Interactions
      among three distinct CesA proteins essential for cellulose synthesis. Proc. Natl.
      Acad. Sci. USA 2003, 100, 1450–5.
24.   Gardiner, J.C.; Taylor, N.G.; Turner, S.R., Control of cellulose synthase complex
      localization in developing xylem. Plant Cell 2003, 15, 1740–8.
                     Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine      57

25. Djerbi, S.; Lindskog, M.; Arvestad, L.; Sterky, F.; Teeri, T.T., The genome sequence
    of black cottonwood (Populus trichocarpa) reveals 18 conserved cellulose synthase
    (CesA) genes. Planta 2005, 221, (5), 739–46.
26. Nairn, C.J.; Haselkorn, T., Three loblolly pine CesA genes expressed in developing
    xylem are orthologous to secondary cell wall CesA genes of angiosperms. New
    Phytol . 2005, 166(3), 907–15.
27. Roberts, A.W.; Bushoven, J.T., The cellulose synthase (CESA) gene superfamily of
    the moss Physcomitrella patens. Plant Mol. Biol. 2007, 63, 207–19.
28. Tanaka, K.; Murata, K.; Yamazaki, M.; Onosato, K.; Miyao, A.; Hirochika, H.,
    Three distinct rice cellulose synthase catalytic subunit genes required for cellulose
    synthesis in the secondary wall. Plant Physiol . 2003, 133(1), 73–83.
29. Doblin, M.S.; Kurek, I.; Jacob-Wilk, D.; Delmer, D.P., Cellulose biosynthesis in
    plants: from genes to rosettes. Plant Cell Physiol. 2002, 43, 1407–20.
30. Scheible, W.-R.; Eshed, R.; Richmond, T.; Delmer, D.; Somerville, C., Mod-
    ifications of cellulose synthase confer resistance to isoxaben and thiazolidinone
    herbicides in Arabidopsis Ixr1 mutants. Proc. Natl. Acad. Sci. USA 2001, 98,
    10079–84.
31. Perrin, R.M., Cellulose: how many cellulose synthases to make a plant? Curr. Biol .
    2001, 11, R213–R216.
32. Wang, J.; Elliott, J.E.; Williamson, R.E., Features of the primary wall CESA complex
    in wild type and cellulose-deficient mutants of Arabidopsis thaliana. J. Exp. Bot.
    2008, 59(10), 2627–37.
33. Arioli, T.; Peng, L.; Betzner, A.S. et al., Molecular analysis of cellulose biosynthesis
    in Arabidopsis. Science 1998, 279, 717–20.
34. Kurek, I.; Kawagoe, Y.; Jacob-Wilk, D.; Doblin, M.; Delmer, D., Dimerization
    of cotton fiber cellulose synthase catalytic subunits occurs via oxidation of the
    zinc-binding domains. Proc. Natl. Acad. Sci. USA 2002, 99, 11109–14.
35. Kiedaisch, B.M.; Blanton, R.L.; Haigler, C.H., Characterization of a novel cellulose
    synthesis inhibitor. Planta 2003, 217(6), 922–30.
36. Paredez, A.R.; Somerville, C.R.; Ehrhardt, D.W., Visualization of cellulose synthase
    demonstrates functional association with microtubules. Science 2006, 312(5779),
    1491–5.
37. Jacob-Wilk, D.; Kurek, I.; Hogan, P.; Delmer, D.P., The cotton fiber zinc-binding
    domain of cellulose synthase A1 from Gossypium hirsutum displays rapid turnover
    in vitro and in vivo. Proc. Natl. Acad. Sci. USA 2006, 103(32), 12191–6.
38. Taylor, N.G., Identification of cellulose synthase AtCesA7 (IRX3) in vivo phospho-
    rylation sites – a potential role in regulating protein degradation. Plant Mol. Biol.
    2007, 64(1–2), 161–71.
39. Peng, L.; Kawagoe, Y.; Hogan, P.; Delmer, D., Sitosterol-b-glucoside as primer for
    cellulose synthesis in plants. Science 2002, 295, 147–50.
40. Amor, Y.; Haigler, C.H.; Johnson, S.; Wainscott, M.; Delmer, D.P., A membrane-
    associated form of sucrose synthase and its potential role in synthesis of cellulose
    and callose in plants. Proc. Natl. Acad. Sci. USA 1995, 92(20), 9353–7.
41. Salnikov, V.V.; Grimson, M.J.; Seagull, R.W.; Haigler, C.H., Localization of
    sucrose synthase and callose in freeze-substituted secondary-wall-stage cotton fibers.
    Protoplasma 2003, 221(3–4), 175–84.
58   The Nanoscience and Technology of Renewable Biomaterials

42. Salnikov, V.V.; Grimson, M.J.; Delmer, D.P.; Haigler, C.H., Sucrose synthase
    localizes to cellulose synthesis sites in tracheary elements. Phytochem. 2001, 57(6),
    823–33.
43. Bieniawska, Z.; Paul Barratt, D.H.; Garlick, A.P. et al., Analysis of the sucrose
    synthase gene family in Arabidopsis. Plant J . 2007, 49(5), 810–28.
44. Hardin, S.C.; Duncan, K.A.; Huber, S.C., Determination of structural requirements
    and probable regulatory effectors for membrane association of maize sucrose syn-
    thase 1. Plant Physiol . 2006, 141, (3), 1106–19.
45. Haigler, C.H.; Ivanova-Datcheva, M.; Hogan, P.S.; Salnikov, V.V.; Hwang, S.;
    Martin, K.; Delmer, D.P., Carbon partitioning to cellulose synthesis. Plant Mol.
    Biol . 2001, 47(1–2), 29–51.
46. Read, S.M.; Bacic, T., Prime time for cellulose. Science 2002, 295, 59–60.
47. Saxena, I.M.; Brown, R.M., Jr.; Dandekar, T., Structure – function characterization
    of cellulose synthase: relationship to other glycosyltransferases. Phytochem. 2001,
    57(7), 1135–48.
48. Carpita, N.; Vergara, C., A recipe for cellulose. Science 1998, 279(5351), 672–3.
49. Merzendorfer, H., Insect chitin synthases: a review. J. Comp. Physiol. [B] 2006,
    176(1), 1–15.
50. Timpa, J.D.; Triplett, B.A., Analysis of cell-wall polymers during cotton fiber devel-
    opment. Planta 1993, 189, 101–8.
51. Benedict, C.R.; Kohel, R.J.; Jividen, G.M., Crystalline cellulose and cotton fiber
    strength. Crop Sci . 1994, 34(1), 147–51.
52. Vergara, C.E.; Carpita, N.C., Beta-D-glycan synthases and the CesA gene fam-
    ily: lessons to be learned from the mixed-linkage (1 – >3),(1 – >4)beta-D-glucan
    synthase. Plant Mol. Biol. 2001, 47(1–2), 145–60.
53. Haigler, C.H.; White, A.R.; Brown, R.M., Jr.; Cooper, K.M., Alteration of in vivo
    cellulose ribbon assembly by carboxymethylcellulose and other cellulose derivatives.
    J. Cell Biol . 1982, 94(1), 64–9.
54. Grimson, M.J.; Haigler, C.H.; Blanton, R.L., Cellulose microfibrils, cell motility,
    and plasma membrane protein organization change in parallel during culmination in
    Dictyostelium discoideum. J. Cell Sci . 1996, 109, 3079–87.
55. Tsekos, I., The sites of cellulose synthesis in algae: Diversity and evolution of
    cellulose-synthesizing enzyme complexes. J. Phycol. 1999, 35, 635–55.
56. Kimura, S.; Itoh, T., Cellulose synthesizing terminal complexes in the ascidians.
    Cellulose 2004, 11(3–4), 377–83.
57. Chisholm, R.L.; Gaudet, P.; Just, E.M. et al., dictyBase, the model organism
    database for Dictyostelium discoideum. Nucleic Acids Res. 2006, 34(Database
    issue), D423–7.
58. Kennedy, C.J.; Cameron, C.J.; Sturcova, A. et al., Microfibril diameter in celery
    collenchyma cellulose: X-ray scattering and NMR evidence. Cellulose 2007, 14,
    235–46.
59. Bowling, A.J.; Brown, R.M., Jr., The cytoplasmic domain of the cellulose-
    synthesizing complex in vascular plants. Protoplasma 2008, 233(1–2), 115–27.
60. Chanzy, H.; Imada, K.; Vuong, R., Electron diffraction from the primary walls of
    cotton fibers. Protoplasma 1978, 94, 299–306.
                    Biogenesis of Cellulose Nanofibrils by a Biological Nanomachine   59

61. Muller, M.; Hori, R.; Itoh, T.; Sugiyama, J., X-ray microbeam and electron diffrac-
    tion experiments on developing xylem cell walls. Biomacromolecules 2002, 3(1),
    182–6.
62. Atalla, R.H.; Vanderhart, D.L., Native cellulose: A composite of two distinct crys-
    talline forms. Science 1984, 223(4633), 283–5.
63. VanderHart, D.L.; Atalla, R.H., Studies of microstructure in native celluloses using
    solid-state carbon-13 NMR.. Macromolecules 1984, 17, 1465–72.
64. Rudsander, U.J.; Sandstrom, C.; Piens, K. et al., Comparative NMR analysis of cel-
    looligosaccharide hydrolysis by GH9 bacterial and plant endo-1,4-beta-glucanases.
    Biochemistry 2008, 47(18), 5235–41.
65. Rudsander, U.J. Functional studies of a membrane-anchored cellulase from poplar.
    Dissertation, Royal Institute of Technology, Stockholm, Sweden, 2007, http://www.
    diva-portal.org/kth/theses/abstract.xsql?dbid=4520.
66. Yasutake, Y.; Kawano, S.; Tajima, K. et al., Structural characterization of the Ace-
    tobacter xylinum endo-ß-1,4-glucanase CMCax required for cellulose biosynthesis.
    PROTEINS: Structure, Function, and Bioinformatics 2006, 64, 1069–77.
67. Diotallevi, F.; Mulder, B., The cellulose synthase complex: a polymerization driven
    supramolecular motor. Biophys J . 2007, 92(8), 2666–73.
68. Djerbi, S.; Aspeborg, H.; Nilsson, P. et al., Identification and expression analysis
    of genes encoding putative cellulose synthases (CesA) in the hybrid aspen, Populus
    tremula (L.) X P. tremuloides (Michx.). Cellulose 2004, 11(3–4), 301–12.
69. Haigler, C.H.; Zhang, D.; Wilkerson, C.G., Biotechnological improvement of cotton
    fiber maturity. Physiol. Plant. 2005, 124, 285–94.
70. Bowers, J.E.; Chapman, B.A.; Rong, J.; Paterson, A.H., Unravelling angiosperm
    genome evolution by phylogenetic analysis of chromosomal duplication events.
    Nature 2003, 422(6930), 433–8.
71. Persson, S.; Caffall, K.H.; Freshour, G. et al., The Arabidopsis irregular xylem8
    mutant is deficient in glucuronoxylan and homogalacturonan, which are essential
    for secondary cell wall integrity. Plant Cell 2007, 19, 237–55.
                                                          3
       Tools for the Characterization
    of Biomass at the Nanometer Scale

                         James F. Beecher, Christopher G. Hunt and J.Y. Zhu



3.1     Introduction

To take advantage of nanoscale features in plant cell walls and create our own nanos-
tructures based on plant biomass, we must make reliable measurements at the nanoscale.
Although nanoscale measurement methods have expanded in recent years, not all these
techniques are useful for soft, hydrophilic, nonconducting biomass specimens. Here we
discuss those methods with the potential to be particularly useful in studying nanoscale
properties of plant biomass.
   In contrast to most engineering materials, plant biomass structure changes with water
availability. Water swells biomass, creating pores that transport enzymes and reagents
into and out of the cell wall during processing. Therefore we begin with a description
of basic interactions of water and biomass. Nanoscale accessibility and reactivity of the
cell wall are often critical to bioprocessing, so we discuss several methods of evaluating
these properties. This chapter also describes methods to measure cellulose crystallinity,
because crystallinity affects properties and crystallites are an interesting material in them-
selves. Finally the chapter reviews microscopic and spectroscopic methods useful for
the study of biomass at the nanoscale.


3.2     Water in Biomass

Water has a profound effect on the nanoscale structure of plant cell walls. Nanoscale
pores in wet spruce wood, a representative biomass, commonly contain 0.3 g water

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
The contribution of Dr Beecher, Dr Hunt and Dr Zhu has been written in the course of their official duties as US government
employees and is classified as a US Government Work, which is in the public domain in the United States of America.
62    The Nanoscience and Technology of Renewable Biomaterials

per gram of biomass (1, 2). When dried normally, wood typically contains only ca.
2% void space (3), a typical value for amorphous polymers. Clearly water is acting
as a plasticizer, forcing itself between the molecular chains of the polymer, causing
swelling and softening. This ‘bound water’ acting as biomass plasticizer has higher
density, lower freezing temperature, and lower vapor pressure than water at standard
temperature and pressure (1). Other solvents swell wood also, but hydrogen bond donors
are the most effective (4). Not all components of wood have the same moisture affinity:
lignin, cellulose, and hemicellulose absorb 0.60, 0.92, and 1.56 times as much water,
respectively, as an equal weight of dry wood (2), and no water enters the interior of
cellulose crystals.
   Fiber saturation point (FSP) is the point where biomass becomes saturated with bound
water. Below FSP, any water removed must come from between polymer chains within
the cell wall. This creates enormous surface tension forces, which collapse the hydrated
layers. Compared to never-dried biomass, air-dried biomass has high density, nonporous
cell walls, and modified molecular conformations.
   Mechanical action and chemical treatments can change FSP. Breaking chemical
crosslinks within the fiber makes it easier for water to push into the spaces between
polymer chains and generally results in higher FSP. Because water cannot enter
cellulose crystals, increasing cellulose crystallinity leads to lower FSP. Extractives and
lignin tend to displace water in the cell wall, so removing these usually increases FSP.
Hysteresis is also evident: the FSP of never-dried wood is 10% to 20% higher than
after the first drying (1, 5).


3.3   Measurement of Specific Biomass Properties

3.3.1 Pore Structure and Accessibility
Processes for utilizing biomass often depend on mass transfer into and out of cell walls
and the accessibility of cell wall components. Bioconversion of lignocellulose through
enzymatic saccharification depends on enzyme accessibility to cellulose. Pores within
cell walls provide a route for enzymes to access cellulose inside the cell wall. As a
result, porosimetry, or determining pore size and structure, of biomass is important.
This section reviews several techniques developed for this task.

3.3.1.1 Porosimetry by Solute Exclusion
The solute exclusion technique was initially created to determine the accessibility of
macromolecules to fibrous substrates in water-swollen state (6). Later the technique was
applied to wood and wood pulp (7–11). The technique is based on the accessibility
of solute molecules (probe molecules) to the substrate pores of different sizes. The
pore is considered accessible when the pore is connected to the bulk water and large
enough to hold the probe molecule (Figure 3.1). Solute exclusion experiments can be
conducted using the following procedure. A known volume and concentration of a probe
molecule solution is added into a swollen substrate immersed in an excess of water. After
thorough mixing, the probe molecule solution is diluted by the water contained in the
initial substrate. If all pores are accessible to the probe molecule, then all water in the
                      Tools for the Characterization of Biomass at the Nanometer Scale                63


                                                                  WATER


              SUBSTRATE

                                                                  Closed Inaccessible Pore


                   Open Pore
                                     ADDING               Partially Accessible Bottle-necked Pore
                                    SOLUTION




 CASE I                            CASE II                       CASE III




                     Accessible Pore Water                   Inaccessible Pore Water


Figure 3.1 Pore water in biomass substrate is inaccessible when the probe molecule cannot
enter the pore.


initial substrate will contribute to the dilution. The water present in the pores that is
not accessible to the probe molecules will not contribute to dilution. As a result, the
measured concentration of the probe molecule in the final substrate mixture depends on
the pore size and volume distribution. Using a set of different solute (probe molecule)
solutions with various molecule sizes, the substrate pore size and volume distribution
can be determined by
                                    w                 w
                           cf =           · ci =                · ci                                (3.1)
                                  w + wac        w + (q − win )
where
        cf and ci are the final and initial probe concentrations,
        w is the weight of the probe molecule solution,
        wac and win are the weight of water accessible and inaccessible to the probe
                 molecule in the wet pulp, respectively, and
        q is the weight of water in the wet pulp (wac = q − win ).

Therefore, win , the pore water per gram of biomass that is inaccessible to a probe
molecule, is
                                   win   w+q     w    ci
                          Win =        =     1−     ×                                               (3.2)
                                   p      p     w+q   cf
where p is the weight of dry biomass sampled.
64   The Nanoscience and Technology of Renewable Biomaterials

   Accurate concentration measurement is critical when using the solute exclusion tech-
nique. Probe molecule concentrations have been measured by interference refractometer
(8), spectropolarimeter (12, 13), and HPLC using a refractive index detector (14).
   The validity and accuracy of the solute exclusion technique for pore size and volume
distribution measurements are based on two assumptions: (1) the concentration of a
probe molecule in the accessible pores is the same as in bulk solution surrounding the
pulp specimen, and (2) probe molecules can fully penetrate the pore to get full access
to the pore water. To meet these two assumptions, the probe molecules should not
adsorb on nor chemically react with the substrate. They should be spherical and must
be available in a large variety of highly monodisperse molecular sizes. Cross-linked
dextrans (10) and poly(ethyleneglycol)s (11) were originally proposed. Both of these
probe types have seen continued use (12–15). Gel permeation chromatography indi-
cated that complete penetration of a pore is not possible (16). The concentration of
the probe molecule in pores is shown both theoretically (17, 18) and experimentally
(19–21) to vary with pore shape and relative size of the pore and probe molecule.
Probe molecules with low hydrogen bonding capability have limited potential to access
water in pores, and electrostatic charge on the probe can effect the ability to penetrate
pores (22).
   Other problems with determining pore size and volume distribution with the solute
exclusion technique are the ‘ink-bottle’ effect and osmotic pressure (23) (Figure 3.1).
Probe molecules can be excluded from the water in pores with a narrow opening but
wider space inside the pore. If the pore substrate contains ionized groups, even nonionic
solutes can be excluded from pores by osmotic pressure. In this case, using molecules
that interact and adsorb on fibers has been suggested as they can be forced to enter pores
(24). Cell walls appear to have a lamellar structure (25), and the simple slit model for
cell wall pores is a reasonable assumption (8, 10).
   In view of the discussion above, we should emphasize the terms ‘effective pore size’
and ‘accessible water’ when measured by the solute exclusion technique. The solute
exclusion technique is a valuable tool for determining the total pore volume accessible
to a probe molecule of given size. It can also quantify relative changes in the porous
structure from various treatments and between different substrates. Solute exclusion is
not an acceptable tool for the determination of absolute pore size and volume distribution.
In this sense this technique fits the needs for practical applications in bionanosensing,
such as determining enzyme accessibility to substrates and comparing the effectiveness of
various mechanical and chemical pretreatment processes and substrates for lignocellulose
bioconversion.
   Despite the shortcoming of the solute exclusion technique, it is very useful for under-
standing the effects of molecular size on accessibility (26, 27). In addition, a very useful
pore structure model is based on the results of the method (10).

3.3.1.2 Porosimetry by Differential Scanning Calorimetry
The differential scanning calorimetry (DSC) technique for pore size distribution mea-
surements is based on the principle that water contained inside pores has a lower freezing
point than that of bulk water. This technique, called thermoporosimetry, was initially
developed for measuring pore size distribution in other materials (28–30) but has been
                      Tools for the Characterization of Biomass at the Nanometer Scale     65

successfully applied to pulp fibers (31, 32). The relations between the specific melting
enthalpy and pore size are described by the Gibbs-Thompson equation:

                                            −Vm σls
                                      r=           T
                                                                                         (3.3)
                                             Hm ln T0

where
        r is the radius of cylindrical pore,
        Vm is the specific molar volume of ice,
          Hm is the specific melting enthalpy of water,
        σls is the surface energy at the ice-water interface,
        T0 is the melting temperature of water at normal pressure, and
        T is the melting temperature corresponding to Hm .

  The pore volume Vr corresponding to the pore size at a measured melting enthalpy is
the volume occupied by the frozen water melted in DSC measurements:

                                               Ht
                                    Vr =                                                 (3.4)
                                              Hm · ρt

where Ht is the energy absorbed when the ice in a frozen specimen is completely
melted (the area under dynamic endothermic curve obtained from DSC measurement)
and ρt is the water density at the melting temperature.
   One problem with thermoporosimetry is the hysteresis observed when the freezing
and melting cycle is repeated several times. The following thermal sequence has been
developed to minimize hysteresis: cooling to −45 ◦ C, heating at 5 ◦ C/min to −30 ◦ C
and holding for 5 min, then cooling to −45 ◦ C (31). The cycle was repeated with the
isothermal melting set point at −8, −5, −3, −2, −1, −0.6, −0.4, −0.2, −0.1, and
0 ◦ C for the calibration specimen. The final temperature scan was a dynamic mea-
surement from −30 ◦ C to +15 ◦ C used to determine Ht . The specimen is entirely
frozen and melted only twice in this sequence. Other thermoporosimetry issues to
keep in mind are that osmotic pressure can also cause melting point depression and
that the pore size distribution may change with freezing or with raising temperature to
where enzyme accessibility to lignocellulose substrate is of great interest. Furthermore,
some enclosed pores contribute to the measured pore volume but cannot be accessed by
enzymes (33).

3.3.1.3 Porosimetry by NMR
Water inside the cell wall has a much slower diffusion rate than free water. In nuclear
magnetic resonance (NMR) measurements, this diffusion rate influences the T2 relaxation
time. Therefore, NMR T2 relaxation time is useful for quantifying the amount of water
in different environments such as the cell wall and cell lumen. This technique has been
used to characterize pore water in various cellulosic fibers (34–36), native wood (37),
pretreated corn stover (38), enzymatically hydrolyzed cellulose (39), and model plant
cell walls (40).
66   The Nanoscience and Technology of Renewable Biomaterials

3.3.1.4 Porosimetry by Gas Absorption
The gas absorption technique for pore measurement is well established and sound. It
requires dry specimens so is useful for dry biomass applications but has limited utility for
understanding wet processing such as enzymatic hydrolysis of lignocellulose in biorefin-
ing. The technique is based on the Brunauer-Emmett-Teller (BET) equation (41), where
the volume of gas absorbed in a monolayer, vm , is related to the ratio of the equilibrium
(P ) and saturation (P0 ) pressures and the total gas volume absorbed, υ:

                                 1          c−1        P          1
                                          =                  +                          (3.5)
                          v[(P0 /P ) − 1]   vm c       P0        vm c

Therefore vm and c (BET constant) can be obtained from fitting Equation (3.5) to the
experimental data of P /P0 and v. The specific surface is then evaluated using

                                                  (vm N s)
                                   SBET,total =                                         (3.6)
                                                     V
where N is Avogadro’s constant, s is adsorption cross section (area per adsorbed
molecule), and V is molar volume of the adsorption gas. Water sorption on cellu-
lose powder was studied by comparing the adsorption of water and nitrogen with BET
theory (42).

3.3.1.5 Enzymatic Hydrolysis Rate
The rate of enzymatic degradation of cellulose is especially important for those attempt-
ing to use biomass-derived glucose for fermentation. This assay typically applies a
standard cellulase to biomass and measures the rate of cellulose depolymerization.
Reducing aldehydes produced by the depolymerization are commonly detected by color
reactions with reagents such as Cu(II) or ferrocyanide. For further discussion, we sug-
gest a recent review (43) because there are too many variations on this method to discuss
here. Access of the enzymes to the substrate is critical to hydrolysis, and this accessi-
bility can be elegantly assayed (43). Caution should be used when comparing cellulase
activities because the size of the substrate, in addition to porosity, can influence observed
cellulase activity when diffusion is limiting (27).

3.3.2 Cellulose Crystallinity
Cellulose crystallinity is often measured because cellulose crystals (crystallites, whiskers)
have a slow degradation rate, high strength, high aspect ratio, and ability to form chiral
nematic liquid crystals. Assessing the mass fraction and size of crystals is commonly
accomplished by NMR or X-ray diffraction, but Raman and infrared (IR) methods can
also be used. Crystal size can be increased by allowing structures to relax at high tem-
perature and humidity, such as during kiln drying of wood, and decreased by mechanical
damage, such as ball milling or swelling. Although we discuss cellulose crystals, a more
proper term might be ‘ordered domains’ because they do not technically meet all the
requirements of a perfect crystal. Not only is there a large number of defects in the
directions perpendicular to the cellulose chain, but there appears to be a slow twist to
                       Tools for the Characterization of Biomass at the Nanometer Scale     67

the entire structure. The structure of cellulose, the various crystalline forms, and the
issue of crystallinity are discussed in Chapter 6.
   Spectroscopic methods of measuring cellulose crystallinity (NMR, Raman, IR) are
typically used with pure cellulose samples because hemicellulose and amorphous cellu-
lose produce very similar signals, whereas lignin is much different. The X-ray diffraction
method, by contrast, can be used on native biomass samples because the lignin, hemi-
cellulose, and amorphous cellulose all contribute to a broad background peak, whereas
the cellulose crystals produce a sharper peak.

3.3.2.1 X-ray
Cellulose crystallinity index (CrI), or percentage by mass of crystalline cellulose in
a sample, is determined by diffracting or reflecting Cu Kα X-rays off of randomly
oriented, relatively pure cellulose specimens. The original method is still widely used.
In this method, the peak intensity of the 002 peak (I002 , 2 = 22.6◦ ) diffracted from
the crystalline cellulose is compared with the intensity of the reflection from amorphous
cellulose, (Iamorph ), which has a very broad peak at 18◦ (44).
                                         I002 − Iamorph
                                 CrI =                  × 100                             (3.7)
                                              I002
Segal (44) also discusses the effect of specimen packing, density, orientation, and other
factors that all must be controlled to get reproducible results.
   Crystal width can be calculated using the Scherrer equation on the peak assigned to
(002) planes:

                                     D = Kλ(B cos )

where D is crystal thickness, λ the radiation wavelength, the diffraction angle, and B
the full width of the diffraction peak measured at half-maximum height. The correction
factor, K, is typically set at 0.9 (45).
   Because of the small crystal size and subsequent broad peaks, numerous efforts have
been made to improve quantification. These range from simply taking areas rather
than heights for the crystalline and amorphous reflections (46) to modeling the entire
diffraction pattern (47). If treatments partially dissolve cellulose, then different crystal
forms, such as cellulose II, and their reflections are likely to appear upon solvent removal.
The exact peak positions are reported to vary by more than 0.5◦ .

          13
3.3.2.2        C NMR
Solid-state 13 C NMR is often used to assay the amount of cellulose in crystalline vs.
amorphous form. The anhydroglucose carbon 4 (C4) peak is shifted 89 ppm when
inside the crystal and 86 ppm when on the crystal surface or in amorphous cellulose
and hemicellulose (45, 48). In highly crystalline specimens, the C6 peak at 65 ppm
can be compared with the amorphous peak at 60–62 ppm. Band assignments have
been made for all the components (49), and the effect of some factors such as residual
hemicellulose and crystal polymorph (Iα vs. Iβ ) on the observed crystallinity has been
described (50, 51).
68    The Nanoscience and Technology of Renewable Biomaterials

3.3.2.3 Raman and Infrared Spectroscopy
Raman and IR spectroscopy have both been suggested as ways to determine the ratio of
amorphous to crystalline cellulose I. In Raman, a linear correlation has been observed
between X-ray crystallinity and I1481 /(I1481 + I1462 ), where the peaks at 1481 cm−1 and
1462 cm−1 are assigned to crystalline and amorphous cellulose, respectively (52). Using
IR, the height or area of the crystalline band at 1280 cm−1 is compared with the relatively
constant band at 1200 cm−1 (53). In both methods the intensity of the bands need to be
determined through mathematical deconvolution of the spectra because of overlap. The
correspondence of Raman-derived values and X-ray crystallinity values is better when
the Raman crystallinity is calibrated on a contiguous set of cellulose samples rather than
cellulose from widely varying sources. Also, there is some indication that the Raman
method is not sensitive above crystallinity index of 75. The FTIR method does not
account for any potential changes in the ratio of cellulose Iα to Iβ but otherwise is a
reasonable method for routine characterization.



3.4   Microscopy and Spectroscopy

So far we have described a variety of techniques for measuring specific attributes of
biomass. The remainder of this chapter will give an overview of microscopic and spec-
troscopic techniques useful in generating images and chemical information, respectively,
from biomass samples.

3.4.1 Specimen Preparation
Biomass has several characteristics that can make it difficult to analyze. We have already
discussed water interactions and how the methods of freezing or removing water from
biomass can alter nanostructure. The modest pyrolysis temperature of 300 ◦ C (54) for
cellulose, as well as long experience with organic substrate damage by electrons and
x-rays shows that biomass constituents are susceptible to change during observation by
high energy probes. While the aromatic nature of lignin distinguishes it from carbohy-
drates, cellulose and hemicellulose are chemically very similar and so are difficult to
differentiate.
   The obvious approach to analyzing biomass structure by selectively removing partic-
ular components has serious problems. Removing one polymer from a plant cell wall
almost always causes chemical and physical changes in the polymer removed, as well
as the residual material. For example, the relatively gentle acid chlorite delignification
procedure only removes about half of lignin before damage to hemicellulose becomes
apparent (55). Amorphous cellulose and hemicellulose are so chemically similar that
it is very difficult to dissolve one without disturbing the natural state of the other. In
addition, chemical crosslinks (lignin-carbohydrate complexes, or LCC) between lignin
and hemicellulose are well documented and hinder the removal of hemicellulose (56).
These chemical considerations, as well as the interpenetrating nature of the cellulose,
hemicellulose, and lignin polymer networks, makes it very difficult to analyze the native
structure of any one of the wood components by itself.
                     Tools for the Characterization of Biomass at the Nanometer Scale   69

   Many techniques require preparing dry, flat or thin specimens for characterization.
Preparing these without introducing artifacts is a challenge because biomass is naturally
wet and fibrous. Confirmation from multiple independent techniques is often necessary
before one can be confident that the observation is not an artifact of specimen prepara-
tion. We suggest an excellent review of microscopic techniques used to probe cell wall
structure (57).

3.4.1.1 Drying Issues
Although drying invariably changes biomass structure, understanding the nature of those
changes can minimize the impact and guide interpretation of results. Analysis of biomass
in the wet state is usually preferred, but some analytical techniques require high vacuum.
Even the simple task of getting the dry weight of a specimen can ruin the specimen
for other analyses, as the act of drying causes irreversible changes in cell wall nano-
structure.
   Several methods have been used to preserve structures during water removal, and many
books are devoted to the techniques, as biological specimens for electron microscopy
have always faced this problem (58, 59). Two principal mechanisms cause specimen
alteration during drying: changing solvent properties and surface tension. Although
surface tension can be avoided, biomass structures surrounded by air or a different
solvent have a different energy, and so have different stability, than in water. A stable
molecular conformation in water could be unstable after water removal. Therefore, even
the best water removal methods have the potential to modify specimens.
   The simplest drying strategy is simply to let water evaporate, for example in an oven
at 105 ◦ C. This is an easy method but causes the pores to collapse from surface tension.
To avoid this, specimens are either frozen while wet or the solvent is exchanged. Solvent
exchange is often followed by critical point drying or embedding.
   In freeze drying, specimens are frozen while wet and placed under vacuum so that ice
sublimes, completely avoiding surface tension. Standard freezing techniques cause ice
crystal formation that partially dehydrates the specimen, causing some fiber shrinkage,
collapse, and artifacts from crystal formation (60). Because of the interaction with
wood polymers, the freezing point of some bound water is as low as −30 ◦ C (31),
confirming empirical observations that specimens must be kept extremely cold throughout
the entire freeze-drying process. Partial freeze drying is also common, especially in
electron microscopies with a cold stage. In this technique, some surface water is allowed
to sublime by heating the specimen to ca. −80 ◦ C, and then the specimen is cooled again
to prevent further water loss during imaging.
   Water can also be removed by deep freezing followed by flooding the specimen with
dry acetone or ethanol (59, 61). This is often followed by embedding procedures, carried
out at low temperature up to the point of polymerizing the embedding medium.
   Under proper conditions, damage by ice crystal formation can be avoided by producing
vitreous ice (ice without crystals). Vitreous ice is formed when cooling rates approach
106 ◦ C/s,61 and the water is diluted with 10–15% of a solute such as sucrose (62, 63).
Samples kept at high pressure during freezing can have vitreous layers ca. 10 times
thicker than samples frozen at atmospheric pressure (200+ vs. 20 µm) (62). Fast
freezing is achieved by placing thin specimens in contact with liquids or plates cooled
70     The Nanoscience and Technology of Renewable Biomaterials

Table 3.1 Affinity and swelling power of different solvents toward biomass (4, 65).
Solvent            Liquid-holding capacity                Liquid-holding capacity               Tangential swelling
                      of α-Cellulose (%)a                    of sulfite pulp (%)b                of spruce wood (%)c
Water                           85.0                                  403.0                                8.4
Methanol                        60.5                                  253.6                                8.2
Ethanol                         41.7                                  109.4                                7.0
Acetone                         13.4                                   47.6                                5.7
a
  α -Cellulose is pure cellulose.
b
  Sulfite pulp contains some hemicellulose and lignin residue; cellulose is less crystalline than α -Cellulose.
c
  Tangential swelling is relative to oven dry.




by liquid nitrogen (59, 61, 62). While this kind of super-fast cooling is essential for live
cells, this method is rarely used when preparing biomass samples, even though there is
evidence that slow or normal freezing changes pore structure (31).
   Another approach to avoiding liquid:vapor surface tension during drying is to use
critical point drying (CPD). In this method, water is replaced by a transitional solvent
(ethanol or acetone), and then CO2 . When CO2 is pressurized to the critical point, there
is no surface tension between the liquid and gas phases, allowing evaporation of all the
liquid with no surface tension (64). If proteins and lipids are present, fixation before
CPD is suggested.
   The primary problem with either CPD or low-temperature embedding is the need
to change solvent. Besides the obvious problem of dissolving cellular components,
changing solvents changes the stability of solvated structures, resulting in bulk swelling
and shrinking and unknown changes to nanostructures. The extent of biomass swelling
in a few typical solvents is given in Table 3.1.
   Exposing wood to a solvent may change some aspects of the nanostructure, but it
can also be useful in revealing nanoscale features. Removing extractives (resins, waxes,
gums, fats) by acetone or ethanol, or ethanol/toluene is common as these extractives can
interfere with chemical analysis. Tokareva showed how various specimen preparation
procedures, including extraction by acetone or critical point drying, improved their ability
to see fine structure such as cellulose macrofibrils (66).

3.4.1.2 Microtoming
Preparing 20- to 30-nm cross sections of homogeneous materials with a microtome
is technically challenging, but the heterogeneity and fibrous nature of biomass makes
producing these specimens from biomass even more difficult. The microtome knife is
really initiating a crack that will grow along the path of least resistance. Biomass is
difficult to microtome without distortion because fibrous structures and very nonuni-
form mechanical properties of different cell wall components redirect the progress of
the crack.
   Distortions introduced by microtoming have been systematically examined (67), and
means to minimize distortion with diamond and glass knives have been reported (68,
69). H. Sitte has extensively studied microtomy and directed the design of commercial
instruments and has thoroughly reviewed microtomy practice (70).
                     Tools for the Characterization of Biomass at the Nanometer Scale   71

3.4.1.3 Focused Ion Beam Cutting
Focused ion beams (FIBs) are often used in the semiconductor electronics industry, but
the process may be especially useful for cutting specimens with phases of very different
hardness, which is a particular problem when preparing plant cell walls. Typically FIB
cuts specimens with a 100-nm wide-beam of 30-keV gallium ions. In the usual process,
stair-stepped depressions are carved into the specimen with a rastered ion beam on either
side of a thin (∼100 nm) section to be studied. Finally, the edges of the thin section are
cut free from the specimen, which is recovered and mounted on a transmission electron
microscopy (TEM) grid (71–73).
   While cutting the specimen, these ions also cause a number of artifacts from ion
implantation and heat (74). Specific results depend upon material composition and ion
energy, but the ion implantation depth is usually about 20 nm and atom displacements
are confined to about 30 nm for gallium ions in low atomic number materials. The depth
of the damage layer is strongly dependent upon ion accelerating voltage and the incident
ion milling angle (75). Because almost all the ion kinetic energy is converted to heat
(76), organic and polymeric materials can experience temperatures up to 500 ◦ C at the
etching site. However, FIB cutting does not necessarily overheat specimens; vitreously
frozen water has been prepared without heat-induced ice crystal formation (77). New
developments in cluster ions, discussed later under ‘Secondary Ion Mass Spectrometry,’
promise to produce even less damage in biomass specimens.
   Most published examples of FIB involve inorganic semiconductor materials (78, 79)
or inorganic composites (80), but use with soft materials is also possible. Human hair
and housefly eye (81) represent biological specimens. Photographic film (79) and a toner
particle (82) have also been successfully sectioned. The toner particle is an important
example because contrary to microtomy sections, filler within the toner was not damaged
or displaced in FIB sections. FIB is also used to decompose gases such as W(CO)6 to
tungsten metal at the specimen surface for protection, to minimize charge accumulation,
or to weld a section to a micromanipulator.

3.4.2 Scanning Probe Microscopies
Scanning probe microscopy (SPM) techniques, especially tapping mode atomic
force microscopy (AFM), are the most frequently used techniques for characterizing
nanometer-scale structures. These techniques move a probe along the specimen surface
to determine topography, material properties, or chemical structure. SPM techniques
are reported in well over 10,000 research papers each year, but the results must be
interpreted with caution. Like all microscopy techniques, with enough images it
is possible to find exactly what is of interest, even if it is an artifact of specimen
preparation or not representative.
   The early use of AFM by Hanley and Gray et al. to describe wood cell structure
illustrates some of the specimen preparation problems and limitations (83, 84). Cell
structures were subjected to physical and chemical treatment in the preparation process.
Although such methods are commonly used, whether the resulting specimen is an accu-
rate representation of native plant morphology is questionable. The AFM images in the
paper illustrate a problem common to all SPM: the observed surface is a convolution of
the actual surface and the shape of the probe tip.
72   The Nanoscience and Technology of Renewable Biomaterials

   Interpretation of high-resolution SPM images requires a thorough understanding of
probe/specimen interactions. Quantitative modeling of the contrast mechanism is encour-
aged. The use of other microscopy techniques along with SPM is beneficial in two ways:
(1) examination at larger scale will establish a context for high-resolution studies and
help to select representative fields and (2) other imaging methods create different types
of artifacts and so can be used to confirm observations.
   Recent reviews of scanning probe microscopies abound (85, 86). One excellent
and comprehensive review by an SPM pioneer offers a good place to begin learning
of the promise and pitfalls of SPM (87). This review begins with scanning tunnel-
ing microscopy (STM), follows its evolution, and concludes with the measurement of
mechanical properties utilizing nanoindentation methods.

3.4.2.1 Atomic Force Microscopy
Atomic force microscopy (AFM) is easy to use and is the most frequently used SPM
method for describing molecular solids or biological specimens. In its original mode,
contact AFM, the stylus tip was maintained in contact (or near contact) with the specimen
as in a miniature profilometer. To minimize specimen damage, most current work uses
tapping mode, where the stylus and supporting cantilever are set into vibration near their
resonant bending frequency (nominally ∼100 kHz).
   In tapping mode, the AFM tip makes only intermittent contact with the specimen
surface, but the tip/specimen interactions alter the amplitude, resonance frequency, and
phase angle of the oscillating cantilever. Amplitude modulation mode (AM-AFM) is an
excellent mode for specimens in air or liquids. In AM-AFM, the oscillation frequency
of the tip is kept constant, while the amplitude of vibration reveals topography and the
phase shift between driving force and oscillation reveals interaction forces dependent on
specimen viscoelastic properties (primarily stiffness) and adhesion between the tip and
specimen. The phase information in AFM has been used to measure material properties
of specimens such as elastic modulus, hardness, and material boundaries (88). A good
understanding of the mechanisms involved is important because the interpretation of
phase information is not always clear (87, 89). An exhaustive review of dynamic AFM
including theory and operation is available (90).
   Interactions between the AFM tip and specimen arise from many different mechanisms,
including van der Waals attraction, electrostatic, friction, viscoelastic, and wetting forces.
Which of these forces are dominant is not always clear. One frequent effect that is
not always anticipated is the condensation of water on the specimen surface about the
tip. This results in a capillary force large enough to dominate the probe/specimen
interaction. Purging the sample chamber with nitrogen gas diminishes this condensation
effect enough to image most samples but does not eliminate it. The most frequent practice
for high-resolution studies is to work in ultrahigh vacuum (usual for atomic solids) or
under fluids (usual for biological specimens). Static charges on tip or specimen can be
avoided by placing an ionization source (an alpha particle emitter, for example) near the
specimen.
   A review of the 20-year history of developments in atomic force microscopy along
with the physics involved is recommended (91). This work also views scanning tunneling
microscopy, subatomic imaging, atom manipulation, and future projections. An excellent
                      Tools for the Characterization of Biomass at the Nanometer Scale      73

guide to AFM for organic and biological materials has been written by pioneers in this
field (92). This book was written for biologists but contains detailed practical information
useful to chemists and material scientists working with soft and molecular materials.
   A few examples of AFM studies of nanostructure in biomass are given here. Two
related papers treat the subject of specimen roughness and its effect on pull-off force
and the use of soft colloidal probes to minimize the effect of roughness on force mea-
surement (93, 94). The van der Waals forces between regenerated cellulose surfaces in
an aqueous environment with low pH and high ionic strength to suppress charge effects
is described by Notley (95), while Zauscher studied cellulose surface interactions in
various electrolyte concentrations (96).
   A critical comparison of the AM (amplitude modulation) tapping mode and the FM
(frequency modulation) noncontact tapping mode for imaging soft matter is available
(97). In the FM mode, the tip exerts a very gentle force on soft materials and provides
height measurement close to the true value. In the AM mode, the tip typically exerts
a stronger force on soft materials and causes their deformation, especially in the liquid
environment.

3.4.2.2 Force Spectroscopy
In addition to the topographic image, force spectroscopy can provide chemical infor-
mation about a specimen from the forces that occur as the AFM probe approaches and
retracts from the surface (98, 99). Leite and Herrmann provide an introduction and
extensive review of this subject with a particular emphasis on adhesion phenomena
(100). Force curves can be recorded at many locations on the specimen surface. These
experiments can measure attractive and adhesion forces, the location and area of contact,
and the mechanical properties (modulus and plasticity) of the specimen.
   Chemical force microscopy (CFM) uses chemically modified AFM tips to expand
the range of possible interactions. Once demonstrated (101), CFM has been applied to
characterize cell surfaces (102–104) and intermolecular interactions of cellulose surfaces
(96, 105). A thorough and readable introduction to CFM, along with many applications,
is suggested (106).
   Nanotribology, or nanoscale friction, can also be studied using SPM (107, 108). An
in-depth review of friction force measurement applied mostly to atomic solids was pre-
pared by Carpick and Salmeron (109). The nonvertical component of stylus motion is
often attributed to friction or viscosity, but as Carpick points out, ‘If the sample surface is
not flat, the surface normal force will have a component directed laterally and will result
in contrast in the lateral force image.’ Despite these problems, valuable information on
friction at cellulose surfaces has been obtained with AFM (105, 110, 111).
   A novel AFM design called torsional harmonic cantilever (THC) allows the mapping
of interaction forces and topology across a surface in a short time (112). Previous meth-
ods make time-consuming point-by-point measurements by approaching and retracting
the tip. By placing the tip off-center on the cantilever, every time the tip taps the sur-
face, the cantilever experiences a torque. The torsional oscillations are at a much higher
frequency than the flexural resonance of the cantilever, allowing them to be analyzed
separately to determine a force. As a demonstration of the capabilities of this approach,
Sahin et al. measured the mechanical properties of blend of poly(methyl methacrylate)
74   The Nanoscience and Technology of Renewable Biomaterials

and polystyrene as it was heated (112). These rigid polymers become rubbery and flex-
ible above their glass transition temperatures. The different phases were readily imaged
while the Young’s moduli of the components were measured. The loading forces are
estimated at a modest 10 nN with a few nanometer indentation of the specimen.

3.4.2.3 Probe Shape
In all scanning probe microscopies, the shape and size of the probe tip is important
because the SPM image is a convolution of the probe tip shape and the morphology
of the specimen. The tip of the probe is often described as a hemisphere having a
particular radius, though real probe tips are more complex and change by wear during
the experiment. Therefore characterizing the tip is an important step when obtaining
nanoscale morphological information (113).
   Carbon nanotubes with diameters near 1 nm have been suggested as the ultimate high-
resolution probe. These probes are currently very expensive and fragile, and the data
can be difficult to analyze when the nanotube is not precisely perpendicular to the
specimen surface (114). Multiwalled carbon nanotubes used as probes in tapping mode
AFM are more rugged and therefore easier to use than the single-walled nanotubes, but
data interpretation is more difficult because of the complex mechanics of multiwalled
nanotubes (115).

3.4.2.4 Near-Field Scanning Optical Microscopy
The spatial resolution attainable with conventional optical techniques is limited to about
half the wavelength of the light source used. For visible radiation, this results in a
theoretical resolution limit of 200–300 nm. Higher resolution can be obtained with
near-field scanning optical microscopy (NSOM) by illuminating a specimen through a
small aperture (approximately 50 nm) positioned within one aperture diameter of the
specimen. An extensive and detailed review of NSOM is recommended (116).
   NSOM tips are typically used in tapping AFM mode by synchronizing the detection
and tip vibration. Many examples involve illuminating specimens with NSOM and
recording fluorescent emission from the specimen. Spectra from single molecules have
been measured using this approach. One of the current limitations of NSOM is the high
temperature (up to 500 ◦ C) developed at the end of the scanning tip because most of the
radiant energy is absorbed by the conductive coating that defines the aperture.

3.4.2.5 Nanoindentation
Although AFM has been used to measure material properties of specimens, such as elastic
modulus, hardness, and the identification of phases, the method has serious limitations.
More rigorous and quantitative material property measurements can be obtained at a
sacrifice of speed and some spatial resolution using a nanoindenter (87, 117). With
a nanoindenter, the force on the probe and the position of the probe are measured
independently. This is not the case with an AFM probe, for which the force on the
stylus point is determined by the deflection of the cantilever, which is related to the
position relative to the specimen. The nanoindenter probe moves only vertically, whereas
an AFM stylus is always tilted from the vertical and often twisted by lateral force.
                     Tools for the Characterization of Biomass at the Nanometer Scale    75

However, nanoindenter probes are on the order of 50 nm in tip radius, compared with
approximately 10 nm for an AFM. A review of the relevant contact mechanics has been
published (118), emphasizing the assumptions underlying and restricting the application
of most commonly used models and their implications for nanoscale force measurements.
   Nanoindentation offers an excellent way to transition between the micron scale of
optical microscopy and the nanometer scale of AFM. It has been used to quantitatively
measure static (119, 120) and dynamic (121) mechanical properties on the nanometer
scale. One of the more difficult aspects of nanoindentation experiments is preparing a
relatively smooth surface free of artifacts.
   The earliest applications of nanoindentation in biomass compared the hardness and
Young’s modulus of spruce tracheid secondary walls (122) with the lignin-rich cell corner
middle lamella (123). Since then, nanoindentation has been used to characterize cell wall
development (124), the effect of anisotropy (125), microfibril angle (126), and pyrolysis
(127) on mechanical properties. Nanoindentation has also been useful in understanding
adhesive–wood interactions (128–130).
   The original nanoindentation theory, which was used in the previously mentioned
studies, was derived assuming the indentation of an isotropic elastic half-space. Almost
all biological specimens will violate this assumption. For example, a 1-µm-diameter
nanoindent placed on the transverse plane of a cell wall of wood will be in relatively
close proximity to structural heterogeneities, such as the empty lumen or middle lamella,
because the cell wall is typically only 5 µm wide. Previous researchers minimized this
effect by filling the lumina with epoxy to support the cell walls during testing, but whether
epoxy changes the properties of the cell walls is not certain. Methods to prepare wood
specimens without any embedment and to account for structural compliances that result
from nearby structural heterogeneities have since been developed (131, 132).

3.4.2.6 Scanning Thermal Microscopy
AFM and nanoindenter tips can also be used to measure thermal properties of materials.
An electrically heated AFM probe with a tip radius of about 20 nm can be used to
simultaneously heat and image the specimen surface. When the material below the tip
reaches a phase transition, it softens and the probe penetrates the specimen. This provides
the nanoscale equivalent of a bulk thermomechanical analysis experiment, where phase
transition temperatures Tg or Tm are measured (133, 134). The thermal conductivity and
surface temperature of the specimen can be measured as well. Nanothermal analysis has
been used to investigate adhesive penetration and modification in wood cell walls (135).

3.4.3 Focused Beam Microscopies
Focused beam microscopies differ from SPM in that a beam of particles or rays, rather
than solid objects, are scanned across the specimen. In most cases, the absorption
or scattering of the incident beam is monitored, but particles or radiation generated
by the specimen can also be analyzed. The major experimental artifact generated by
focused beams is specimen damage. Scientists must be aware that the beams used almost
always have enough energy to break chemical bonds in biological specimens, commonly
resulting in mass loss and chemical modification.
76   The Nanoscience and Technology of Renewable Biomaterials

3.4.3.1 Scanning Electron Microscopy
Scanning electron microscopy (SEM) provides high-resolution topographic information
with little specimen preparation. SEM cannot differentiate carbon, nitrogen, and oxygen
so provides little chemical information about biomass unless higher atomic weight ele-
ments are present in the specimen. In SEM, a narrow beam of electrons is scanned
across a specimen while the intensity of reflected or ejected electrons provides the
image. SEMs require a vacuum and some means to dissipate the charge of electrons
that lodge in the specimen. Traditionally, charge was dissipated by coating specimens
with conductive materials. Microscopes that tolerate higher pressure, called variable
pressure or ‘atmospheric’ SEMs, were developed to image specimens in a more nat-
ural state, and have the added advantage that ionized gas surrounding the specimen
often carries away accumulated charge. With a tiny aperture over the specimen to
limit evaporation rate, some variable pressure SEMs can image wet specimens at room
temperature.
   Another option that allows SEM of wet biomass is the cold stage, which keeps the
specimen very cold (down to −120 ◦ C). At this temperature, ice-filled specimens can
be imaged under high vacuum, and the temperature, pressure, and time on the stage
can be adjusted to allow a controlled amount of freeze drying of the specimen surface.
Additionally, specimens analyzed at cryogenic temperatures typically show less evidence
of beam damage (136).
   Biomass specimens may benefit from low voltage (0.5–5 keV) SEM operation
(LVSEM), which often affords good image contrast on uncoated specimens and
minimizes charging and damage (137). LVSEM can also be used to optimize the
difference in secondary electron emission between polymeric materials of different
composition (138). LVSEM images represent only a shallow surface layer because the
penetration depth of these electrons is limited (137).
   Energy dispersive X-ray analysis (EDX) uses X-rays stimulated by SEM electrons
to make maps of elemental distribution. Resolution is typically a few microns, and
sensitivity is far better for heavy atoms than for oxygen and carbon (137). Gases in the
specimen chamber also scatter electrons, so variable pressure or ‘environmental’ EDX
analyses almost always has more background than measurements at high vacuum.

3.4.3.2 Helium Ion Microscopy
A helium ion microscope is analogous to an SEM except that helium ions, rather than
electrons, bombard the specimen. Helium ion microscopes provides a brighter imag-
ing beam with better spatial resolution and different elemental sensitivity than SEM
(139). Spatial resolution (approximately 0.5 nm) is partly the result of a much smaller
beam/substrate interaction volume for helium than for electrons. Image contrast mech-
anisms are different than in SEM, which can provide new opportunities for image
enhancement.
   Although high-atomic-number ions (such as cesium ions) can damage specimens by
sputtering, the low-atomic-number helium ions have low sputtering probability. As
with an SEM, images can be generated from backscattered ions, whose probability of
                     Tools for the Characterization of Biomass at the Nanometer Scale    77

backscattering is dependent upon atomic number. Helium ions can penetrate deeply
into materials, effectively creating another contrast mechanism. To date no studies of
biomass have been reported using a helium ion microscope.

3.4.3.3 X-ray Beam Probes
Scanning transmission X-ray microscopy STXM measures the absorption or deflection
of an X-ray beam by a specimen. STXM uses a synchrotron X-ray source to gener-
ate up to 10-nm, 0.3-eV resolution images of specimens 100–150 nm thick. Though
specimen degradation is still an issue, X-rays are less damaging than are the electrons
used in transmission electron microscopy. Contrast is developed by selecting X-rays
of different wavelengths. This technique has been used to describe the morphology of
polymer composites (140). Developing contrast between cellulose and hemicellulose
may be difficult, though a recent study of mixtures of ethylene–butene copolymer with
ethylene–octene copolymer distinguished the separate phases of these polymers differing
only in the length of the side chain (141).
   X-ray photoelectron spectroscopy (XPS), originally known as ESCA, is a common
technique for measuring the presence and oxidation state of atoms at the specimen
surface. XPS uses a monochromatic X-ray to eject an electron in the specimen, whose
kinetic energy is determined by the difference between X-ray energy and binding energy
of the electron. Ejected electrons rarely escape from more than 10 nm below the surface.
XPS is an ultrahigh-vacuum technique that typically probes approximately a square
centimeter of surface but can have up to 20 nm resolution when coupled to a focused
X-ray source, such as a synchrotron (142). This method is sensitive enough to determine
the oxidation state of carbon, so it can distinguish lignin from carbohydrate, and has been
applied extensively in probing chemical modification of surfaces. XPS data should be
interpreted cautiously because the escape depth, and therefore sensitivity, may change
with chemical composition of the surface, and the incident X-rays and ejected electrons
may cause chemical reactions in the organic substrate. Also, localized areas can become
charged and therefore eject electrons with different energy than the rest of the specimen.

3.4.3.4 Secondary Ion Mass Spectrometry
Secondary ion mass spectrometry (SIMS) produces good chemical specificity with a
resolution of 50 nm to 1 µm. A focused ion beam is directed to a solid surface,
removing material in the form of neutral and ionized atoms and molecules. The ions
are accelerated into a mass spectrometer and separated according to their mass-to-charge
ratio. This is inherently a surface probe with 1- to 10-nm depth resolution. The sensitivity
to different chemical species varies greatly because only ions are detected, whereas
most of the products of sputtering are neutral fragments (143). SIMS is inherently an
ultrahigh vacuum technique requiring flat specimens and some means to eliminate charge
accumulation.
   The use of cluster ions, that is, ions of high molecular weight such as Csn , SF5 ,
C60 , and Aun , to bombard the specimen generally produces better SIMS spectra for
molecular solids than the original SIMS method, which uses gallium ions (144, 145).
78   The Nanoscience and Technology of Renewable Biomaterials

Cluster ions produce greater useful signal intensity and sputter rate while limiting damage
and penetration depth. Cluster ions may be used for analysis at low ion current (static
SIMS) or can systematically remove layers from the specimen at higher ion currents
than gallium, as shown on poly(methyl methacrylate) (146). Because many fragments
have no charge, a second ion beam can be used to ionize fragments (144). Winograd
reviewed the recent development of cluster ion mass spectrometry (147). Some focused
ion guns allow SIMS imaging. An Au3 ion source provides a high-brightness beam with
a 200-nm spot size, whereas C60 and gallium ions have been focused to about 2 µm and
50 nm, respectively.
   Specimen bombardment with C60 may produce a larger spot size than other ions but
has many desirable properties. C60 produces little roughening during erosion experiments
and has larger usable mass range and sensitivity than do gallium ion beams. C60 ions
have increased secondary ion yields and fewer low-mass fragments but no increase in
damage with ion energy up to 120 keV (148). Langmuir-Blodgett films sputtered with
C60 ions produced close to two orders of magnitude more characteristic secondary ions
than gallium (149), apparently from increased sputter yield. Carbohydrate films doped
with peptides bombarded with C60 ions produced high-quality time-of-flight secondary
ion mass spectra, even with ion doses 100 times greater than those used for gallium
bombardment (150).
   The combination of SIMS and AFM is useful because SIMS produces chemical images
and AFM provides topology and other material properties. Zhu used this combination
to examine penetration of gold atoms through alkanethiolate self-assembled monolayers
(151). Wucher used this pair of techniques to produce a three-dimensional representation
of a 300-nm-thick peptide-doped carbohydrate film with a combination of imaging and
etching (152).

3.4.4 Transmission Electron Microscopy
Transmission electron microscopy (TEM) is analogous to transmission light microscopy
except that electrons, rather than photons, are passed through the specimen. TEM can
provide images with subnanometer resolution and so has been extensively used by biol-
ogists for nearly 50 years. The most serious limitations of TEM are that specimens must
resist electron damage and usually have thickness less than approximately 100 nm so
that only single-electron scattering events are likely.
   Carbon replicas have long been a standard means of studying biomass with TEM
(153–155). For example, the highest resolution images of cellulose fibril morphology
were produced by carbon replicas of developing wood cells at a stage prior to the
incorporation of lignin (156). In this example, cells were frozen quickly to avoid ice
crystal formation and cleaved at −150 ◦ C. The exposed surfaces were coated with carbon
and shadowed with platinum; then the biomass was dissolved with concentrated sulfuric
acid and the replica examined by TEM.
   TEM specimens are also made by cutting 30- to 60-nm sections of material. Before
cutting, samples are embedded in resin, which holds the specimen together. Of all spec-
imen preparation techniques, embedding in ice by fast freezing and keeping at −100 ◦ C
throughout specimen preparation and imaging produces TEM specimens most similar to
the original biomass.
                    Tools for the Characterization of Biomass at the Nanometer Scale   79

   One problem with TEM is that the electron scattering cross section shows little dif-
ference, and hence contrast, between carbon, oxygen, and nitrogen. One approach to
developing image contrast is to preferentially label a component with stains contain-
ing high-atomic-number elements. Lignin has been labeled using bromine, potassium
permanganate, and osmium tetroxide (57). Our current knowledge of hemicellulose
deposition is from TEM studies where hemicellulose has been labeled with gold-tagged
antibodies (157). Use of antibodies tagged with nanometer-sized gold particles is routine
in biological microscopy (158).
   Another problem of TEM is damage from the electron beam. Embedding medium, a
model organic substrate, shrank 5% laterally and 25% in thickness during the first 5 min
of TEM exposure, or 13,000 electrons per nm2 (159). Shrinkage and degradation of
specimens should be considered whenever evaluating TEM images of biomass.
   The electron energy loss spectrum (EELS) is commonly used to determine the valence
state of atoms in TEM specimens. Electrons passing through a specimen lose energy by
ionization of specimen atoms, and these quantized losses are characteristic of different
elements. Thus electrons interacting with carbon and oxygen lose 285 eV and 532 eV,
respectively (160). The valence state of the atom can be determined by small shifts in
this value.

3.4.4.1 Electron Tomography
In contrast to conventional two-dimensional TEM, electron tomography (ET) using
brightfield TEM produces a three-dimensional rendering of a volume. ET has been exten-
sively used to study cell structure (161, 162). To determine a unique three-dimensional
representation, approximately 100 images must be obtained using 1◦ to 5◦ tilt increments
over a large angular range, such as ±70◦ .
   The large number of images needed could subject a specimen to severe electron
damage, so several techniques have been adopted to minimize electron exposure (see
Transmission Electron Microscopy, above). Low-dose microscopy techniques use auto-
matic focusing and drift correction on a specimen field adjacent to the one being imaged.
Because many images will be summed to make the final composite, each image can tol-
erate slightly higher noise than standard images and therefore less electron exposure. A
case has been made that the total electron exposure needed for a tilt series is about the
same as that for a single two-dimensional image (163). In any case, the electron flux
should be kept within a few thousand electrons per square nanometer (159). Another
method of minimizing damage from electron exposure is to maintain the specimen at
cryogenic temperature (136). Cryo-electron tomography has been useful in describing
cellular structure and the structure of cell components (164–166).
   Electron tomography has revealed the nanometer-level organization of cellulose
microfibrils in the S2 layer of the cell wall and the arrangement of some residual
lignin and hemicellulose about the cellulose microfibrils in radiata pine (167). The
cell sections were partly delignified with peracetic acid, dehydrated, treated with
multiple heavy metal stains, and embedded in Spurr’s epoxy. Although some structural
distortion may have been introduced by the treatment, the results are a major step
toward understanding wood cell wall structure.
80    The Nanoscience and Technology of Renewable Biomaterials

3.4.4.2 Ultrafast Electron Microscopy
Nobel laureate Ahmed H. Zewail and associates developed ultrafast electron microscopy
(UEM) which enables imaging of kinetic processes at atomic resolution (168). A con-
ventional TEM was modified so that a femtosecond pulsed laser stimulated one electron
emission per laser pulse. A version capable of imaging wet specimens is under construc-
tion (169). Although the resolution is the same as for conventional TEM, the time resolu-
tion provided by this new technology promises to shed new light on dynamic processes.


3.5   Summary

Biomass is a difficult substrate to analyze at the nanoscale. It is a nanoscale, interpene-
trating, lightly crosslinked network of three different types of polymer. Compared with
most inorganics, the polymers of biomass are strongly affected by water, unstable, and
difficult to distinguish from each other.
   Because the natural state of biomass is in water and because of the strong interaction
with water, native structures should be characterized in water when possible. Removing
water from biomass causes structural changes that are only partially reversible and not
well understood. The nanostructure of biomass will be different depending on the amount
of water present during analysis and how water has been removed (i.e. its history).
   Because biomass is organic, many analytical techniques, particularly focused beam
microscopies, can easily damage specimens. Most particle or electromagnetic beams with
high resolution have the potential to vaporize or initiate chemical changes in the biomass
substrate. The probe beam can also cause accumulation of charge on the nonconducting
biomass substrate, which can directly effect measurements in some methods, as well as
cause further chemical changes.
   Atomic force microscopes are the most common scanning probe microscopes, because
they are so versatile and easy to use. They were originally developed to measure
nanoscale topography, but a variety of techniques now provide information on chemical
and material properties as well. More specialized instruments, such as nanoindenters
and near-field scanning optical microscopes, may be less versatile but generally provide
‘cleaner’ information that requires the user to make fewer assumptions during analysis.
   In short, the characterization of nanostructure in biomass is challenging. Like most
scientific problems, however, proper choice of analytical techniques and specimen prepa-
ration, as well as a skeptical approach to data interpretation, continue to provide new
understanding of this fascinating system.


References

1.    Stamm A. Wood and Cellulose Science. New York: The Ronald Press Co 1964.
2.    Berry S, Roderick M. Plant-water relations and the fibre saturation point. New
      Phytologist. 2005;168:25–37.
3.    Kellog RM, Wangaard FF. Variation in the cell wall density of wood. Wood and
      Fiber. 1969;1:180–204.
4.    Mantanis GI, Young RA, Rowell RM. Swelling of Wood, Part II Swelling in
      Organic Liquids. Holzforschung. 1994;48(6):480–90.
                     Tools for the Characterization of Biomass at the Nanometer Scale   81

5.    Spalt H. The Sorption of Water Vapor by Domestic and Tropical Woods. Forest
      Products Journal . 1957;7(10):331.
6.    Aggebrandt L, Samuelsson O. Penetration of water soluble polymers into cellulose
      fibers. J Appl Polym Sci . 1964;8(6):2701–2812.
7.    Tarkow W, Feist W. The superswollen state of wood. Tappi Journal . 1968;51(2):
      80–3.
8.    Stone JE, Scallon AM. The effect of componenet removal upon the porous structure
      of the cell wall of wood. II. Swelling in water and the fiber saturation point. Tappi
      Journal . 1967;50(10):496–501.
9.    Stone JE, Scallon AM. The effect of component removal upon the porous structure
      of the cell wall of wood. III. A Comparison between the sulphite and kraft process.
      Pulp and Paper Magazine of Canada. 1968;69(12):69–74.
10.   Stone JE, Scallan AM. A structural model for the cell wall of water swollen wood
      pulp fibres based on their accessibility to macromolecules. Cellulose Chem Tehnol.
      1968;2:343–58.
11.   Tarkow H, Feist W, Southerland CF. Interaction of wood with polymeric materials.
      Penetration versus molecular size. Forest Products J . 1966;16(10):61–5.
12.   Grethlein HE. The effect of pore size distribution on the rate of enzymatic hydrol-
      ysis of cellulosic substrates. Bio/Technology. 1985;3(2):155–60.
13.   Grous WR, Converse AO, Grethlein HE. Effect of steam explosion pretreatment on
      pore size and enzymatic hydroxlysis of poplar. Enzyme and Microbial Technology.
      1986;8(5):274–80.
14.   Ishizawa CI, Davis MF, Schuell DF, Johnson DK. Porosity and its effect on the
      digestibility of dilute sulfuric acid pretreated corn stover. J Agric and Food Chem.
      2007;55(7):2575–81.
15.   Lin JK, Ladisch MR, Patterson JA, Noller CH. Determining pore size distribution
      in wet celllulose by measuring solute exclusion using a differential refractometer.
      Biotechnology and Bioengineering. 1987;29(8):976–81.
16.   Altgelt KH, Segal L. Gel Permeation Chromatography. New York: Marcel Dekker
      1971.
17.   Alince B, Comments on porosity or swollen pulp fibers analyzed by solute-
      exclusion. TAPPI J. 1991;74(11):200.
18.   Casassa EF, Tagami. An equilibrium theory for exclusion chroatography of
      branched and linear polymer chains. Macromolecules. 1969;2(1):14.
19.   Moore JC, Arrington MC. Symposium on Macromolecular Chemistry; 1966; Tokyo;
      1966. p. VI–107.
20.   Day JC, Alince B, Robertson AA. Interaction of polymers in solution with porous
      solids. I. Penetration of porous glass by dextran. Can J Chem. 1978;56(23):2951.
21.   Day JC, Alince B, Robertson AA. The characterization of pore systems by macro-
      molecular penetration. Cellul Chem Technol . 1979;13(3):317.
22.   Rowland SP, Bertoniere NR. Some interaction of water-soluble solutes with cellu-
      lose and Sephadex. Textile Research Journal . 1976;46(10):770–5.
23.   Lindstrom T. Paper Structure and Properties. New York: Marcel Dekker 1986.
24.   Alince B. Comments on porosity or swollen pulp fibers analyzed by solute exclu-
      sion. Tappi Journal . 1991;74(11):200–2.
25.   Kerr AJ, Goring DA. Lamellation of hemicellulsoe in the fiber wall fo birch wood.
      Wood Sci 1977;9(3):136–9.
82    The Nanoscience and Technology of Renewable Biomaterials

26. Tanaka M, Ikesaka M, Matsuno R, Converse AO. Effect of pore size in sub-
    strate and diffusion of enzyme on hydrolysis of cellulosic materials with cellulases.
    Biotechnology and Bioengineering. 1988;32(5):698–706.
27. Sangseethong K, Meunier-Goddik L, Tantasucharit U, Liaw ET, Penner MH. Ratio-
    nale for particle size effect on rates of enzymatic saccharification of microcrystalline
    cellulose. Journal of Food Biochemistry. 1998;22(4):321–30.
28. Rennie GK, Clifford J. Melting of ice in porous solids. Journal of the Chemi-
    cal Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases.
    1977;73:680–9.
29. Homshaw LG. Calorimetric Determination of Porosity and Pore Size Distribu-
    tion (PSD): Effect of Heat on Porosity, Pore Form, and PSD in Water-Saturated
    Polyacrylonitrile Fibers. Journal of Colloid and Interface Science. 1981;84(1):
    127–40.
30. Ishikiriyama K, Todoki M, Motomura K. Pore size distribution (PSD) measure-
    ments of silica gels by means of differential scanning calorimetry. I: Optimization
    for determination of PSD. Journal of Colloid and Interface Science. 1995;171(1):
    92–102.
31. Maloney T. Hydration and swelling of pulp fibers measured with differential scan-
    ning calorimetry. Nordic Pulp and Paper Research. 1998;13:31–6.
32. Maloney TC, Paulapuro H. The formation of pores in the cell wall. Journal of
    Pulp and Paper Science. 1999;25(12):430–6.
33. Park S, Venditti RA, Jameel H, Pawlak JJ. Changes in pore size distribution during
    the drying of cellulose fibers as measured by differential scanning calorimetry.
    Carbohydrate Polymers. 2006;66(1):97–103.
34. Froix MF, Nelson R. Interaction of Water with Cellulose from Nuclear Magnetic-
    Resonance Relaxation-Times. Macromolecules. 1975;8(6):726–30.
35. Li T-Q, Henriksson U, Odberg L. Determination of pore volume in cellulose fibers
    by the pulsed gradient spin-echo NMR technique. Journal of Colloid and Interface
    Science. 1995;169(2):376–9.
36. Li TQ, Haggkvist M, Odberg L. Porous structure of cellulose fibers studied by
    Q-space NMR imaging. Langmuir. 1997 Jun;13(13):3570–4.
37. Menon RS, Mackay AL, Hailey JRT, Bloom M, Burgess AE, Swanson JS. An Nmr
    Determination of the Physiological Water Distribution in Wood During Drying.
    Journal of Applied Polymer Science. 1987 Mar;33(4):1141–55.
38. Ishizawa CI, Davis MF, Schell DF, Johnson DK. Porosity and Its Effect on the
    Digestibility of Dilute Sulfuric Acid Pretreated Corn Stover. J Agric Food Chem.
    2007;55(7):2575–81.
39. Felby C, Thygesen LG, Kristensen JB, Jørgensen H, Elder T. Cellulose-water inter-
    actions during enzymatic hydrolysis as studied by time domain NMR. Cellulose.
    2008:1–8.
40. Rondeau-Mouro C, Defer D, Leboeuf E, Lahaye M. Assessment of cell wall
    porosity in Arabidopsis thaliana by NMR spectroscopy. International Journal of
    Biological Macromolecules. 2008;42(2):83–92.
41. Brunauer S, Emmet PH, Teller E. Adsorption of gases in multimolecular layers.
    J Am Chem Soc. 1938;60:309.
                     Tools for the Characterization of Biomass at the Nanometer Scale     83

42.   Strømme M, Mihranyan A, Ek R, Niklasson GA. Fractal Dimension of Cellulose
      Powders Analyzed by Multilayer BET Adsorption of Water and Nitrogen. Journal
      of Physical Chemistry B. 2003;107(51):14378–82.
43.   Percival Zhang YH, Himmel ME, Mielenz JR. Outlook for cellulase improvement:
      Screening and selection strategies. Biotechnology Advances. 2006;24(5):452–81.
44.   Segal L, Creeley JJ, Martin AEJ, Conrad CM. An emperical method for estimating
      the degree of crystallinity of native cellulose using the x-ray diffractometer. Textile
      Research Journal . 1959;29:786.
45.   Duchemin BJCZ, Newman RH, Staiger MP. Phase transformations in microcrys-
      talline cellulose due to partial dissolution. Cellulose. 2007;14(4):311–20.
46.   He J, Cui S, Wang SY. Preparation and crystalline analysis of high-grade bamboo
      dissolving pulp for cellulose acetate. Journal of Applied Polymer Science. 2008;
      107(2):1029–38.
47.   Thygesen A, Oddershede J, Lilholt H, Thomsen AB, Stahl K. On the determina-
      tion of crystallinity and cellulose content in plant fibres. Cellulose. 2005;12(6):
      563–76.
48.   Atalla RH, VanderHart DL. The role of solid state 13C NMR spectroscopy in
      studies of the nature of native celluloses. Solid State Nuclear Magnetic Resonance.
      1999;15(1):1–19.
49.   Larsson PT, Hult EL, Wickholm K, Pettersson E, Iversen T. CP/MAS 13C-NMR
      spectroscopy applied to structure and interaction studies on cellulose I. Solid State
      Nuclear Magnetic Resonance. 1999;15(1):31–40.
50.   Liitia T, Maunu SL, Hortling B. Solid state NMR studies on cellulose crystallinity in
      fines and bulk fibres separated from refined kraft pulp. Holzforschung. 2000;54(6):
      618–24.
51.   Maunu S, Liitia T, Kauliomaki S, Hortling B, Sundquist J. 13C CPMAS NMR
      investigations of cellulose polymorphs in different pulps. Cellulose. 2000;7(2):
      147–59.
52.   Schenzel K, Fischer S, Brendler E. New method for determining the degree of cel-
      lulose I crystallinity by means of FT Raman spectroscopy. Cellulose. 2005;12(3):
      223–31.
53.   Hulleman SHD, Van Hazendonk JM, Van Dam JEG. Determination of crystallinity
      in native cellulose from higher plants with diffuse reflectance Fourier transform
      infrared spectroscopy. Carbohydrate Research. 1994;261(1):163–72.
54.   Alemdar A, Sain M. Isolation and characterization of nanofibers from agricul-
      tural residues - Wheat straw and soy hulls. Bioresource Technology. 2008;99(6):
      1664–71.
55.   Ahlgren PA, Goring DAI. Removal of Wood Components During Chlorite Delig-
      nification of Black Spruce. Canadian Journal of Chemistry. 1971;49.
56.   Lawoko M, Henriksson G, Gellerstedt G. New method for quantitative prepa-
      ration of lignin-carbohydrate complex from unbleached softwood kraft pulp:
      Lignin-polysaccharide networks I. Holzforschung. 2003;57(1):69–74.
57.   Duchesne I, Daniel G. The ultrastructure of wood fibre surfaces as shown by a
      variety of microscopical methods – A review. Nordic Pulp and Paper Research
      Journal . 1999;14(2):129–39.
84    The Nanoscience and Technology of Renewable Biomaterials

58.   Hall JL, Hawes C. Electron Microscopy of Plant Cells. San Diego, CA: Academic
      Press 1991.
59.   Dykstra M, Reuss L. Biological Electron Microscopy Theory, Techniques, and Trou-
      bleshooting. 2nd ed. New York: Kluwer 2003.
60.   de Silveira G, Forsberg P, Conners TE. Scanning Electron microscopy: A Tool for
      the Analysis of Wood Pulp Fibers and Paper. In: Conners TE, Banerjee S, eds.
      Surface Analysis of Paper. Boca Raton, FL: CRC Press 1995.
61.   Ruzin S. Plant Microtechnique and Microscopy. New York: Oxford University
      Press 1999.
62.   Sartori N, Richter K, Dubochet J. Vitrification depth can be increased more than
      10-fold by high-pressure freezing. Journal of Microscopy. 1993;172(1):55–61.
63.   Al-Amoudi A, Chang JJ, Leforestier A, McDowall A, Salamin LM, Norlen LPO,
      et al . Cryo-electron microscopy of vitreous sections. EMBO Journal . 2004;23(18):
      3583–8.
64.   Weatherwax, Caulfield D. Cellulose Aerogels: An improved method for preparing
      a highly expanded form of dry cellulose. Tappi Journal . 1971;54(6):985–6.
65.   Li TQ, Hendriksson U, Klason T, Odberg L. Water diffusion in wood pulp cellulosic
      fibers studied by means of the pulsed gradient spin-echo method. J Colloid and
      Interface Sci . 1992;154(2):305–15.
66.   Tokareva EN, Fardim P, Pranovich AV, Fagerholm HP, Daniel G, Holmbom B.
      Imaging of wood tissue by ToF-SIMS: Critical evaluation and development of
      sample preparation techniques. Applied Surface Science. 2007;253(18):7569–77.
67.   Jesior J-C. How to avoid compression II. The influence of sectioning conditions.
      J Ultrastructure. 1986;95.
68.   Matzelle TR, Gnaegi H, Ricker A, Reichelt R. Characterization of the cutting edge
      of glass and diamond knives for ultramicrotomy by scanning force microscopy
      using cantilevers with a defined tip geometry. Part II. Journal of Microscopy. 2003;
      209(2):113–17.
69.   Matzelle TR, Kruse N, Reichelt R. Characterization of the cutting edge of glass
      knives for ultramicrotomy by scanning force microscopy using cantilevers with a
      defined tip geometry. Journal of Microscopy. 2000;199(3):239–43.
70.   Sitte H. Advanced instrumentation and methodology related to cryoultramicrotomy:
      a review. Scanning Microscopy Supplement. 1996;10.
71.   Giannuzzi LA, Stevie FA. A review of focused ion beam milling techniques for
      TEM specimen preparation. Micron. 1999;30.
72.   Giannuzzi LA, Stevie FA. Introduction to Focused ion Beams: Instrumentation,
      Theory, Techniques, and Practice. New York: Springer 2005.
73.   Overwijk MHF, van den Heuvel FC, Bulle-Lieuwma CWT. Novel scheme for
      the preparation of transmission electron microscopy specimens with a focused ion
      beam. J Vac Sci Technol B . 1993;11(6):2021–4.
74.   Melngailis J. Focused ion beam technology and applications. J Vac Sci Technol B .
      1987;5(2):469–95
75.   McCaffrey JP, Phaneuf MW, Madsen LD. Surface damage formation during
      ion-beam thinning of samples for transmission electron microscopy. Ultrami-
      croscopy. 2001;87(3):97–104.
                    Tools for the Characterization of Biomass at the Nanometer Scale   85

76.   Volkert CA, Minor AM. Focused ion beam microscopy and micromachining. MRS
      Bulletin. 2007;32(5):389–95.
77.   Marko M, Hsieh C, Moberlychan W, Mannella CA, Frank J. Focused ion beam
      milling of vitreous water: Prospects for an alternative to cryo-ultramicrotomy of
      frozen-hydrated biological samples. Journal of Microscopy. 2006;222(1):42–7.
78.   De Veirman A, Weaver L. The use of focused-ion-beam machine to prepare trans-
      mission electron microscopy samples of residual photoresist. Micron. 1999;20:
      213.
79.   Phaneuf MW. Applications of focused ion beam microscopy to materials science
      specimens. Micron. 1999;30(3):277–88.
80.   Kim ST, Dravid VP. Focused ion beam sample preparation of continuous fibre-
      reinforced ceramic composite specimens for transmission electron microscopy.
      Journal of Microscopy. 2000;198(2):124–33.
81.   Ishitani T, THirose H, Tsuboi H. Focused-ion-beam digging of biological speci-
      mens. J Electron Microsc. 1995;44.
82.   Mayer J, Giannuzzi LA, Kamino T, Michael J. TEM sample preparation and FIB-
      induced damage. MRS Bulletin. 2007;32(5):400–7.
83.   Hanley SJ, Giasson J, Revol J-F, Gray DG. Atomic force microscopy of cellulose
      microfibrils: comparison with transmission electron microscopy. Polymer 1992;
      33(21).
84.   Hanley SJ, Gray DG. Atomic force microscope images of black spruce wood
      sections and pulp fibers. Holzforschung. 1994;48(1).
85.   Poggi MA, Gadsby ED, Bottomley LA, King WP, Oroudjev E, Hansma H. Scan-
      ning probe microscopy. Analytical Chemistry. 2004;76(12):3429–44.
86.   Meyer E, Jarvis SP, Spencer ND. Scanning probe microscopy in materials science.
      MRS Bulletin. 2004;29(7):443–5.
87.   Colton RJ. Nanoscale measurements and manipulation. J Vac Sci Technol B . 2004;
      22(4):1609.
88.   Bischel MS, Vanlandingham MR, Eduljee RF, Gillespie Jr JW, Schultz JM. On
      the use of nanoscale indentation with the AFM in the identification of phases in
      blends of linear low density polyethylene and high density polyethylene. Journal
      of Materials Science. 2000;35(1):221–8.
89.   Raghavan D, Gu X, Nguyen T, VanLandingham M, Karim A. Mapping poly-
      mer heterogeneity using atomic force microscopy phase imaging and nanoscale
      indentation. Macromolecules. 2000;33(7):2573–83.
90.   Garcia R, Perez R. Dynamic atomic force microscopy methods. Surface Science
      Reports. 2002;47.
91.   Giessibl FJ, Quate CF. Exploring the nanoworld with atomic force microscopy.
      Physics Today. 2006(December).
92.   Morris VJ, Kirby AR, Gunning AP. Atomic Force Microscopy for Biologists. Lon-
      don: Imperial College Press 1999.
93.   Tormoen GW, Drelich J. Deformation of soft colloidal probes during AFM pull-off
      force measurements: Elimination of nano-roughness effects. Journal of Adhesion
      Science and Technology. 2005;19(3–5):181–98.
86     The Nanoscience and Technology of Renewable Biomaterials

94.    Tormoen GW, Drelich J, Nalaskowski J. A distribution of AFM pull-off forces
       for glass microspheres on a symmetrically structured rough surface. Journal of
       Adhesion Science and Technology. 2005;19(3–5):215–34.
95.    Notley SM, Pettersson B, Wagberg L. Direct measurement of attractive van der
       Waals’ forces between regenerated cellulose surfaces in an aqueous environment.
       Journal of the American Chemical Society. 2004;126(43):13930–1.
96.    Zauscher S, Klingenberg DJ. Normal forces between cellulose surfaces measured
       with colloidal probe microscopy. Journal of Colloid and Interface Science. 2000;
       229(2):497–510.
97.    Yang CW, Hwang IS, Chen YF, Chang CS, Tsai DP. Imaging of soft matter
       with tapping-mode atomic force microscopy and non-contact-mode atomic force
       microscopy. Nanotechnology. 2007;18(8).
98.    Dufrene YF. Atomic force microscopy, a powerful tool in microbiology. J Bacte-
       riology. 2002;184(19):5205.
99.    Van der Aa BC, Michel RM, Asther M, Zamora MT, Rouxhet PG, Dufrene YF.
       Stretching cell surface macromolecules by atomic force microscopy. Langmuir.
       2001;17(11):3116–19.
100.   Leite FL, Herrmann PSP. Application of atomic force spectroscopy (AFS) to studies
       of adhesion phenomena: A review. Journal of Adhesion Science and Technology.
       2005;19(3–5):365–405.
101.   Frisbie CD, Rozsnyai LF, Noy A, Wrighton MS, Lieber CM. Functional group
       imaging by chemical force microscopy. Science. 1994;265(5181):2071–4.
102.   Ahimou F, Denis FA, Touhami A, Dufrene YF. Probing microbial cell surface
       charges by atomic force microscopy. Langmuir. 2002;18(25):9937–41.
103.   Ong YL, Razatos A, Georgiou G, Sharma MM. Adhesion forces between E. coli
       bacteria and biomaterial surfaces. Langmuir. 1999;15(8):2719–25.
104.   Frederix PLTM, Hoogenboom BW, Fotiadis D, Muller DJ, Engel A. Atomic force
       microscopy of biological samples. MRS Bulletin. 2004;29(7):449–55.
105.   Bastidas JC, Venditti R, Pawlak J, Gilbert R, Zauscher S, Kadla JF. Chemical force
       microscopy of cellulosic fibers. Carbohydrate Polymers. 2005;62(4):369–78.
106.   Vezenov DV, Noy A, Ashby P. Chemical force microscopy: Probing chemical ori-
       gin of interfacial forces and adhesion. Journal of Adhesion Science and Technology.
       2005;19(3–5):313–64.
107.   Burns AR, Huston JE, Carpick RW, Michalske TA. Molecular level friction as
       revealed with a novel scanning probe. Langmuir. 1999;15(2922).
108.   Caprick RW, Eriksson MA. Measurements of in-plane material properties with
       scanning probe microscopy. MRS Bulletin. 2004;29(7):472–7.
109.   Carpick RW, M. S. Scratching the Surface: Fundamental investigation of Tribology
       with Atomic Force Microscopy. Chem Rev . 1997;97:1163–94.
110.   Garoff N, Zauscher S. The influence of fatty acids and humidity on friction and
       adhesion of hydrophilic polymer surfaces. Langmuir. 2002;18(18):6921–7.
111.   Zauscher S, Klingenberg DJ. Friction between cellulose surfaces measured with
       colloidal probe microscopy. Colloids and Surfaces A: Physicochemical and Engi-
       neering Aspects. 2001;178(1–3):213–29.
                    Tools for the Characterization of Biomass at the Nanometer Scale   87

112. Sahin O, Magonov S, Su C, Quate CF, Solgaard O. An atomic force microscope tip
     designed to measure time-varying nanomechanical forces. Nature Nanotechnology.
     2007;2(8):507–14.
113. Villarrubia JS. Algorithms for scanned probe microscope image simulation, surface
     reconstruction, and tip estimation. Journal of Research of the National Institute of
     Standards and Technology. 1997;102(4):425–54.
114. Snow ES, Campbell PM, Novak JP. Atomic force microscopy using single-wall C
     nanotube probes. Journal of Vacuum Science and Technology B: Microelectronics
     and Nanometer Structures. 2002;20(3):822–7.
115. Lee SI, Howell SW, Raman A, Reifenberger R, Nguyen CV, Meyyappan M.
     Nonlinear tapping dynamics of multi-walled carbon nanotube tipped atomic force
     microcantilevers. Nanotechnology. 2004;15(5):416–21.
116. Dunn RC. Near-filed scanning optical microscopy. Chem Review . 1999;99:2891.
117. Bhushan B, Kulkarni AV, Bonin W, Wyrobek JT. Nanoindentation and picoindenta-
     tion measurements using a capacitive transducer system in atomic force microscopy.
     Philosophical Magazine A: Physics of Condensed Matter, Structure, Defects and
     Mechanical Properties. 1996;74(5):1117–28.
118. Unertl WN. Implications of contact mechanics models for mechanical properties
     measurements using scanning force microscopy. Journal of Vacuum Science &
     Technology A: Vacuum, Surfaces, and Films. 1999;17(4):1779–86.
119. Doerner MF, Nix WD. A method for interpreting the data from depth-sensing
     indentation instruments. J Mater Res. 1986;1(4):601–9.
120. Oliver WC, Pharr GM. Improved technique for determining hardness and elastic
     modulus using load and displacement sensing indentation experiments. Journal of
     Materials Research. 1992;7(6):1564–80.
121. Syed Asif SA, Wahl KJ, Colton RJ, Warren OL. Quantitative imaging of nanoscale
     mechanical properties using hybrid nanoindentation and force modulation. Journal
     of Applied Physics. 2001;90(3):1192–1200.
122. Wimmer R, Lucas BH, Tsui TY, Oliver WC. Longitudinal hardness and Young’s
     modulus of spruce tracheid secondary walls using nanoindentation technique. Wood
     Science and Technology. 1997;31(2):131–41.
123. Wimmer R, Lucas BN. Comparing mechanical properties of secondary wall and
     cell corner middle lamella in spruce wood. IAWA Journal . 1997;18(1):77–88.
124. Gindl W, Gupta HS, Grunwald C. Lignification of spruce tracheid secondary
     cell walls related to longitudinal hardness and modulus of elasticity using nano-
     indentation. Canadian Journal of Botany. 2002;80(10):1029–33.
125. Gindl W, Schoberl T. The significance of the elastic modulus of wood cell walls
     obtained from nanoindentation measurements. Composites Part A: Applied Science
     and Manufacturing. 2004;35(11):1345–9.
126. Gindl W, Gupta HS, Schoberl T, Lichtenegger HC, Fratzl P. Mechanical properties
     of spruce wood cell walls by nanoindentation. Applied Physics A: Materials Science
     and Processing. 2004;79(8):2069–73.
127. Zickler GA, Schoberl T, Paris O. Mechanical properties of pyrolysed wood: A
     nanoindentation study. Philosophical Magazine. 2006;86(10):1373–86.
128. Konnerth J, Gindl W. Mechanical characterisation of wood-adhesive interphase cell
     walls by nanoindentation. Holzforschung. 2006;60(4):429–33.
88   The Nanoscience and Technology of Renewable Biomaterials

129. Konnerth J, Jager A, Eberhardsteiner J, Muller U, Gindl W. Elastic properties of
     adhesive polymers. II. Polymer films and bond lines by means of nanoindentation.
     Journal of Applied Polymer Science. 2006;102(2):1234–9.
130. Gindl W, Gupta HS. Cell-wall hardness and Young’s modulus of melamine-
     modified spruce wood by nano-indentation. Composites Part A: Applied Science
     and Manufacturing. 2002;33(8):1141–5.
131. Jakes JE. A proposed method for accounting for edge effects and structural com-
     pliance in nanoindentation of wood [M.S.]: University of Wisconsin-Madison;
     2007.
132. Jakes JE, Frihart CR, Beecher JF, Moon RJ, Stone DS. Experimental method to
     account for structural compliance in nanoindentation measurements. Journal of
     Materials Research. 2008;23(4):1113–27.
133. King WP, Kenny TW, Goodson KE, Cross G, Despont M, Durig U, et al . Atomic
     force microscope cantilevers for combined thermomechanical data writing and read-
     ing. Applied Physics Letters. 2001;78(9):1300–2.
134. Grandy D, Kjoller K. The identification of particle in a polymer film using nano-
     thermal analysis. Microscopy Today. 2006;14(4):58–60.
135. Konnerth J, Harper D, Lee SH, Rials TG, Gindl W. Adhesive penetration of wood
     cell walls investigated by scanning thermal microscopy (SThM). Holzforschung.
     2008;62(1):91–8.
136. McIntosh JR. Electron microscopy of cells: A new beginning for a new century.
     Journal of Cell Biology. 2001;153(6).
137. Goldstein JI, Newbury DE, Echlin P, Joy DC, Fiori C, Lifshin E. Scanning Electron
     Microscopy and X-ray Microanalysis: Plenum Press 1984.
138. Berry VK. Characterization of polymer blends by low voltage scanning electron
     microscopy. Microscopy. 1988;10:19.
139. Postek MT, Vladar AE, Kramar J, Stern LA, Notte J, McVey S. The helium ion
     microscope: A new tool for nanomanufacturing. Proceedings of SPIE - The Inter-
     national Society for Optical Engineering; 2007.
140. Ade H, Zhang X, Cameron S, Costello C, Kirz J, Williams S. Chemical con-
     trast in X-ray microscopy and spatially resolved XANES spectroscopy of organic
     specimens. Science. 1992;258(5084):972–5.
141. Appel G, Koprinarov I, Mitchell GE, Smith AP, Ade H. X-Ray spectromicroscopy
     of branched polyolefin blends. Annual Meeting of the American Physical Society
     2002.
142. Fulghum JE. Recent developments in high energy and spatial resolution analysis
     of polymers by XPS. J Electron Spectr. 1999;100.
143. Briggs D, Seach MP. Practical Surface Analysis, Ion and Neutral Spectroscopy:
     John Wiley & Sons, Ltd. UK. 1992.
144. Gillen G, King L, Freibaum B, Lareau R, Bennett J, Chmara F. Negative cesium
     sputter ion source for generating cluster primary ion beams for secondary ion mass
     spectrometry analysis. Journal of Vacuum Science and Technology, Part A: Vacuum,
     Surfaces and Films. 2001;19(2):568–75.
145. Postawa Z, Czerwinski B, Szewczyk M, Smiley EJ, Winograd N, Garrison
     BJ. Enhancement of sputtering yields due to C60 versus Ga bombardment of
                     Tools for the Characterization of Biomass at the Nanometer Scale   89

       Ag{111} as explored by molecular dynamics simulations. Analytical Chemistry.
       2003;75(17):4402–7.
146.   Wagner MS. Impact Energy Dependence of SF5+-Induced Damage in Poly(methyl
       methacrylate) Studied Using Time-of-Flight Secondary Ion Mass Spectrometry.
       Analytical Chemistry. 2004;76(5):1264–72.
147.   Winograd N. The magic of cluster SIMS. Analytical Chemistry. 2005;77(7).
148.   Fletcher JS, Conlan XA, Jones EA, Biddulph G, Lockyer NP, Vickerman JC.
       TOF-SIMS analysis using C60. Effect of impact energy on yield and damage.
       Analytical Chemistry. 2006;78(6):1827–31.
149.   Sostarecz AG, McQuaw CM, Wucher A, Winograd N. Depth profiling of
       Langmuir-Blodgett films with a buckminsterfullerene probe. Analytical Chemistry.
       2004;76(22):6651–8.
150.   Cheng J, Winograd N. Depth profiling of peptide films with TOF-SIMS and a C60
       probe. Analytical Chemistry. 2005;77(11):3651–9.
151.   Zhu Z, Daniel TA, Maitani M, Cabarcos OM, Allara DL, Winograd N. Control-
       ling gold atom penetration through alkanethiolate self-assembled monolayers on
       Au{111} by adjusting terminal group intermolecular interactions. Journal of the
       American Chemical Society. 2006;128(42):13710–19.
152.   Wucher A, Cheng J, Winograd N. Protocols for three-dimensional molecular imag-
       ing using mass spectrometry. Analytical Chemistry. 2007;79(15):5529–39.
153.   Cote WA, Koran Z, Day AC. Replica techniques for electron microscopy of wood
       and paper. Tappi Journal . 1964;47(8):477.
154.   Norberg PH. A method for electron microscopy observation of wet wood fibre
       surfaces. Svensk Papperstidning. 1968;71(23).
155.   Kimura S, Laosinchai W, Itoh T, Cui X, Linder CR, Malcolm Brown Jr R. Immuno-
       gold labeling of rosette terminal cellulose-synthesizing complexes in the vascular
       plant Vigna angularis. Plant Cell Physiol. 1999;11(11):2075–85.
156.   Itoh T. Deep-etching electron microscopy and 3-dimensional cell wall architecture.
       In: Chaffey N, ed. Wood Formation in Trees. London: Taylor & Francis 2002.
157.   Baba K, Sone Y, Misaki A, Hayashi T. Localization of xyloglucan in the macro-
       molecular complex composed of xyloglucan and cellulose in the pea stems. Plant
       Cell Physiol . 1994;35:439.
158.   Baschong W, Stierhofr Y-D. Preparation, use, and enlargement of ultrasmall gold
       particles in immunoelectron microscopy. Micros Res Techn. 1998;42:66.
159.   Luther PK, Lawrence MC, Crowther RA. Method for monitoring the collapse of
       plastic sections as a function of electron dose. Ultramicroscopy. 1988;24(1):7–18.
160.   Leapman RD, Hunt JA. Compositional imaging with electron energy loss spec-
       troscopy. Microscopy. 1992;22(1).
161.   Frank J. Three-Dimensional Electron Microscopy of Macromolecular Assemblies.
       2nd ed. New York: Oxford University Press 2006.
162.   Koster AJ, Grimm R, Typke D, Hegerl R, Stoschek A, Walz J, et al . Perspectives
       of molecular and cellular electron tomography. Journal of Structural Biology.
       1997;120(3):276–308.
163.   McEwen BF, Downing KH, Glaeser RM. The relevance of dose-fractionation
       in tomography of radiation-sensitive specimens. Ultramicroscopy. 1995;60(3):
       357–73.
90   The Nanoscience and Technology of Renewable Biomaterials

164. Medalia O, Weber I, Frangakis AS, Nicastro D, Gerisch G, Baumeister W. Macro-
     molecular architecture in eukaryotic cells visualized by cryoelectron tomography.
     Science. 2002;298(5596):1209–13.
165. Beck M, Forster F, Ecke M, Plitzko JM, Melchior F, Gerisch G, et al. Nuclear pore
     complex structure and dynamics revealed by cryoelectron tomography. Science.
     2004;306(5700):1387–90.
166. Kurner J, Frangakis AS, Baumeister W. Cryo-electron tomography reveals the
     cytoskeletal structure of Spiroplasma melliferum. Science. 2005;307(5708):
     436–8.
167. Xu P, Donaldson LA, Gergely ZR, Staehelin LA. Dual-axis electron tomogra-
     phy: A new approach for investigating the spatial organization of wood cellulose
     microfibrils. Wood Science and Technology. 2007;41(2):101–16.
168. Lobastov VA, Srinivasan R, Zewail AH. Four-dimensional ultrafast electron
     microscopy. Proceedings of the National Academy of Sciences of the United States
     of America. 2005;102(20):7069–73.
169. Park HS, Baskin JS, Kwon OH, Zewail AH. Atomic-scale imaging in real and
     energy space developed in ultrafast electron microscopy. Nano Letters. 2007;7(9):
     2545–51.
                                                        4
  Tools to Probe Nanoscale Surface
 Phenomena in Cellulose Thin Films:
      Applications in the Area
     of Adsorption and Friction

                 Junlong Song, Yan Li, Juan P. Hinestroza and Orlando J. Rojas



4.1     Introduction

Surfaces and interfaces play important roles in defining material interactions. Several
developments in science and technology highlight the importance of interfaces in appli-
cations involving material functionalization, coatings, colloidal stability, etc. (Karim and
Kumar 2000). In many cases, the interfacial properties are more relevant than the nature
and composition of the bulk phases and ultimately define the molecular behavior of the
system.
  The ‘thickness’ of a boundary between two phases, if possible to define, is expected to
be extremely narrow. The interface between (bio)polymers or that for a polymer-coated
substrate and the surrounding medium typically entails a ‘soft’ layer with molecular or
nanoscale dimensions. The use of adsorbed polymers and surfactants to modify solid
surfaces offers unique possibilities to alter or regulate their properties, including surface
energy, molecular assembly and composition, among others. In order to effectively
or permanently modify the interfacial properties the adsorbing material (or adsorbate)
has to bind to some degree or extent to the respective surface. Therefore, adsorption is
fundamental in many important applications, particularly in the general fields of adhesion,
colloidal stabilization, friction, and heterogeneous reactions.



The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
92    The Nanoscience and Technology of Renewable Biomaterials




                                                             D



Figure 4.1 Schematic illustration of polymers adsorbing from solution onto a surface.
D is the average thickness of the adsorbed polymer layer, the value of which depends on the
method used to measure it.

   Adsorption results as a consequence of the balance between surface energy and the
nature of the adsorbing species. While the conformation of a polymer in solution depends
on solvency and polymer chain composition and architecture, at an interface the polymer
can be perturbed by the interaction of its segments with the surface (see Figure 4.1).
When this interaction involves attractive chemical or physical forces the resulting adsorp-
tion is classified as chemisorption or physisorption, respectively (Eisenriegler 1993).
   Macromolecules possess a broad diversity of properties that are often related to their
dissociation ability in aqueous solution. As such they are classified into ionic (also known
as polyelectrolytes) and nonionic polymers. Ionic polymers are also classified into simple
polyelectrolytes, with either positive or negative charged groups, and polyampholytes,
which contain both positive and negative charged groups.
   Polymer adsorption has been extensively studied from theoretical and experimental
perspectives. In this chapter, we will first describe the adsorption of a relevant type of
charged polymer onto cellulose surfaces. We will then review aspects related to boundary
lubrication in the case of adsorbed nonionic polymer on the same substrates. Finally,
we will present a brief account on the techniques used to study polymer adsorption
and lubrication. Specifically, we will discuss two tools to determine the extent and
dynamics of polymer and surfactant adsorption: The quartz crystal microbalance QCM
and the surface plasmon resonance SPR techniques. We will also discuss the use of
lateral force microscopy LFM as a useful approach to investigate friction phenomena.
This information presented in this chapter will be helpful to appreciate other chapters
covering specific aspects of surface modification (including hemicellulose adsorption and
polymer multilayers). Complementary tools for nanoscale characterization of biomass
was discussed in chapter 3.


4.2   Polyampholytes Applications in Fiber Modification

Hydrosoluble polymers are commonly used in industry. Among these, amphoteric
macromolecules or polyampholytes have been employed in papermaking to modify cellu-
losic fibers thereby enhancing inter-fiber bonding. Generally speaking, a polyampholyte
is defined as charged macromolecule carrying both acidic and basic groups (Dobrynin,
                Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films     93

Colby et al. 2004). These polymers find application in several other fields including
colloid stabilization, wetting, lubrication and adhesion (Mazur, Silberberg et al. 1959;
Bratko and Chakraborty 1996; Jeon and Dobrynin 2005; Sezaki, Hubbe et al. 2006a,
2006b; Song, Wang et al. 2006; Wang, Hubbe et al. 2006, 2007; Hubbe, Rojas et al.
2007a, 2007b).
   Under appropriate conditions the acidic and basic groups in polyampholytes disso-
ciate in aqueous solution producing ionic groups and their respective counterions. If
the ionic groups on the polymer chain are weak acids or bases, the net charge of the
polyampholytes can be changed by varying the pH of the aqueous medium. At the
isoelectric point (IEP), the number of positive and negative charges on the polyion is
the same, giving a net charge of zero. In the vicinity of the isoelectric pH, the polymers
are nearly charge-balanced and exhibit the unusual properties of amphoteric molecules.
At conditions of high charge asymmetry (far above or below the isoelectric pH),
these polymers exhibit a simple polyelectrolyte-like behavior (Gutin and Shakhnovich
1994; Kantor and Kardar 1995; Ertas and Kantor 1996; Hwang and Damodaran
1996; Long, Dobrynin et al. 1998; Lee and Thirumalai 2000; Yamakov, Milchev
et al. 2000; Dobrynin, Colby et al. 2004; Jeon and Dobrynin 2005; Lord, Stenzel
et al. 2006).
   As fiber recycling increases more interesting and new polymer molecular architectures
have been proposed as means to improve product strength from loses (especially in tensile
and burst strengths) due to reuse (Nazhad and Paszner 1994; Nazhad 2005). After
extensive recycling fibers may not longer be useful without the addition of chemical
additives.
   While several polymer chemistries are used in the applications explained above,
polyampholyte treatments may be less common. To our knowledge, the first report
on the application of polyampholytes to enhance strength of paper was published in
1977 by Carr, Hofreiter et al. (Carr, Hofreiter et al. 1977). In this seminal report,
starch-based polyampholytes were prepared using xanthating cationic cornstarch deriva-
tives, which had either tertiary amino [−CH2 CH2 N(C2 H5)2 ] or quaternary ammonium
[−CH2 CHOHCH2 N+ (CH3 )3 ] groups attached to the macromolecule. Anionic xanthate
groups were then introduced into the cationic starch amines. The substitution degree of
the obtained derivatives ranged from 0.023 to 0.33 for the amine cation and 0.005 to
0.165 for the xanthate anion. This work demonstrated that wet-end additions of a starch
polyampholyte was effective in improving both wet and dry strengths, exceeding those
given by either cationic or anionic starch polyelectrolytes. For a given amine degree of
substitution (DS), there was a charge ratio of A (amine, positive)/X (xanthate, negative)
at which point each polyampholyte gave a well-defined maximum value for wet strength.
This A/X ratio was about 1 for tertiary amine with a low DS (DS of 0.023, 0.035, and
0.06) but was about 2 to 3 for tertiary amines with a high DS of 0.33 (see Figure 4.2).
The authors also found that polyampholytes with quaternary amines substitution were
slightly more effective than those with tertiary amines.
   Recently fully synthetic polyampholytes were systematically investigated in our lab-
oratories with the aim of enhancing paper strength (Sezaki, Hubbe et al. 2006a, 2006b;
Song, Wang et al. 2006; Wang, Hubbe et al. 2006, 2007; Hubbe, Rojas et al. 2007a,
2007b). The employed polyampholytes were prepared by free-radical polymerization
of cationic monomer N-[3-(N ,N -dimethylamino)propyl]acrylamide (DMAPAA), a
94   The Nanoscience and Technology of Renewable Biomaterials



                                           1600                               DS 0.33




                 Wet breaking length (m)
                                           1200                     DS 0.06

                                                            DS 0.035
                                            800


                                            400
                                                            DS 0.023

                                              0.00   0.04    0.08      0.12   0.16      0.20
                                                              Xanthate DS

Figure 4.2 Wet strength (wet breaking length) of paper treated with xanthated starch amine
having various tertiary amine and xanthate degrees of substitution (DS). The paper samples
were prepared from unbleached kraft furnish treated with 3% XSA, oven dry pulp basis, at
pH 7.0. Reproduced with permission from Carr, Hofreiter et al. (1977).



tertiary amine, anionic monomer methylene butanedioic acid (or itaconic acid,
IA), and neutral acrylamide (AM) monomer. Some of the advantages of synthetic
polyampholytes include higher charge densities; simple control of the molecular weight
and charge ratio of cationic and anionic groups; uniform molecular weight distribution
(lower degree of polydispersity), etc. The superior dry strength of polyampholytes
over simple polyelectrolytes was reported in several publications (Sezaki, Hubbe et al.
2006a, 2006b; Song, Wang et al. 2006; Wang, Hubbe et al. 2006; Hubbe, Rojas et al.
2007a, 2007b; Wang, Hubbe et al. 2007). Under the experimental conditions used,
polyampholytes were applied at 1% addition level on bleached hardwood kraft fibers.
Paper’s breaking length increased 20–50% compared with control experiments (see
Figure 4.3). An interesting observation was the fact that the strength of the paper
increased as the charge density increased, reaching a maximum for polyampholytes of
intermediate charge density. After reaching a maximum strength value, the strength
decreased as highly charged polyampholytes were employed. A near neutral pH was
found to be optimum condition to maximize strength performance. This interesting
behavior could be explained by the fact that close to the iso-electric point (IEP) of the
polyampholytes, a maximum efficiency for adsorption is achieved and bonding between
fibers is promoted.
   Despite the fact that a number of theoretical and computational efforts have been
reported (Gutin and Shakhnovich 1994; Kantor, Kardar et al. 1994; Kantor and Kardar
1995; Bratko and Chakraborty 1996; Ertas and Kantor 1996; Schiessel and Blumen 1996;
Srivastava and Muthukumar 1996; Lee and Thirumalai 2000; Yamakov, Milchev et al.
2000), there is still a lack of experimental data regarding the dynamics of adsorption,
and interactions of polyampholytes at the nanoscale. Understanding such phenomena
will lead to new functional formulations and improved performance of fibers after sur-
face modification. In this chapter we will revisit the issue of polyampholyte adsorption
                Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films                   95

                                         5.0
                                                                                          pH = 4
                                               Bleached HW Kraft Fibers
                                                                                          pH = 5
                                         4.5
                                                                                          pH = 8.5


                  Breaking length (km)
                                         4.0

                                         3.5

                                         3.0

                                         2.5

                                         2.0




                                                                                         t.


                                                                                                .
                                                                           8
                                                                    4
                                                             2




                                                                                  16




                                                                                              An
                                                     ol




                                                                                       Ca
                                                                         mp
                                                                  mp
                                                           mp
                                                 ntr




                                                                                mp
                                                                        PA
                                                                 PA
                                                          PA
                                               Co




                                                                               PA
                                                           Polymer (1% treatment level)

Figure 4.3 Effect of macromolecular composition and pH on the tensile strength of
polymer-treated bleached kraft fibers at 1000 µS/cm conductivity. Polyampholytes denoted
as ‘PAmp 2, 4, 8, 16’ correspond to polymers of increased charge density (with the ratio of
anionic-to-cationic groups kept constant) while ‘Cat’ and ‘An’ correspond to the respective
single cationic and anionic polyelectrolytes (with same molecular masses). These polymers
were based on cationic DMAPAA (tertiary amine), anionic itaconic acid (IA) and neutral
acrylamide (AM) (see text for more details). Reproduced with permission from Song, Wang
et al. (2006).

in the context of adsorbed nanolayers with high viscoelasticity to enhance fiber bond-
ing. This phenomenon can only be explored with some of the tools described in later
sections.


4.3 Cellulose Thin Films

Studies at the nanoscale usually involve substrates that are limited to surrogates of cel-
lulose fibers. This is because the intrinsic complexity of natural fibers, which includes
chemical and topographical heterogeneities that prevents derivations of cause-effect rela-
tionships. A common approach is to use cellulose thin films as model for cellulose. There
is an abundance of literature about this topic and the reader is referred to the review
by Konturri et al. for an excellent account on the subject (Kontturi, Tammelin et al.
2006). Here we limit ourselves to spin coated films of cellulose prepared on silica or
gold substrates according to a procedure reported elsewhere (Gunnars, Wagberg et al.
2002; Falt, Wagberg et al. 2004) and modified slightly as follows (Song, Liang et al.
2009): Cellulose solution was prepared by dissolving microcrystalline Avicel cellulose
in 50%wt water/N-methylmorpholine-N-Oxide (NMMO) at 115 ◦ C. Dimethyl Sulfoxide
(DMSO) was added to adjust the concentration (0.05%) and the viscosity of the cellulose
suspension. Polyvinylamine (PVAm) was used as anchoring polymer of the cellulose
film. Silica or gold substrates were immersed in PVAm for 20 min followed by wash-
ing with water and drying with a gentle nitrogen jet. The cellulose solution was then
96     The Nanoscience and Technology of Renewable Biomaterials

spin-coated (Laurell Technologies model WS-400A-6NPP) by depositing 50–100 µl on
the PVAm-modified substrates at 5000 rpm for 40 seconds. We found these conditions
as optimal for obtaining robust, smooth films. The cellulose-coated substrates were
removed from the coater and then immersed in water during four hours and placed
in an oven for two hours at 80 ◦ C. The substrates were then washed thoroughly with
water, dried with a nitrogen jet and stored at room temperature in a clean chamber for
further use. An AFM image of the obtained films as well as a typical height profile
are shown in Figure 4.4. Because of the chemical homogeneity and flat topography
such thin films of cellulose are useful as platform for nanoscale studies that involve
Surface Plasmon Resonance, Quartz Crystal Microbalance as well as Lateral Force
Microscopy.


 µm
 4.5                                               8
 4.0                                               7
 3.5                                               6
 3.0                                               5
                                              nm




 2.5                                               4
 2.0                                               3
 1.5                                               2
                                                   1
 1.0
                                                   0
 0.5
   0
           1.0   2.0         3.0   4.0   µm
                       (a)                                        (b)

 µm
 4.5
 4.0
 3.5
 3.0
 2.5
 2.0
 1.5
 1.0
 0.5
   0
           1.0   2.0         3.0   4.0   µm
                       (c)

Figure 4.4 5 × 5 µm non-contact mode AFM height (a), corresponding section analysis (b),
and phase (c) images of cellulose thin film on a silica wafer. The film is about 20 nm thick
(obtained by ellipsometry) with an RMS roughness of ca. 2 nm.
                Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films    97

4.4   Friction Phenomena in Cellulose Systems

Friction is an important surface phenomenon that is strongly influenced by molecu-
lar adsorption. Inter-fiber friction plays an important role in flocculation and network
strength of paper (Zauscher and Klingenberg 2001). Relevant work related to the mea-
surement of friction in cellulose systems can be found in several references (Bogdanovic,
Tiberg et al. 2001; Zauscher and Klingenberg 2001; Theander, Pugh et al. 2005;
Stiernstedt, Brumer et al. 2006; Stiernstedt, Nordgren et al. 2006).
   Friction, lubrication and wear have long been of both technical and practical interest
since the operation of many mechanical systems depends on these surface phenomena
(Dowson 1998). They have received increased attention in response to the inordinate
waste of resources that has resulted from high friction and wear. In fact, estimates
indicate that proper attention to tribology issues could lead to economic savings up to
1.3 to 1.6% of the Gross National Product (Jost 1990). Beyond industrial applications
tribology is critical in the performance of body implants, cell adhesion, and interfacial
phenomena in composite materials.
   Fibrous polymeric materials go through different processing stages including pretreat-
ment, dyeing, printing and finishing before they are finally assembled into end products
(woven and nonwoven webs, composites, etc.). Machinery and equipment are inevitably
involved in handling fibers at high rates of deformation. Fibers and related materials are
also subjected to destructive abrasive forces that may result in mutual abrasion between
fibers and/or between fibers and equipment surfaces. In order to control friction and
reduce wear between fibers and between fibers and solid surfaces, surface modifica-
tion treatments are necessary. Fiber finishes are commonly used during the production
of many different fiber grades (Proffitt and Patterson 1988) and a myriad of different
finishing formulations exists depending on the intended use of the fibers and the fiber
processing operation conditions. In general four general classes of boundary lubricants
can be identified:
1. high molecular weight, water dispersible products – significantly reduce abrasion
   damage to fibers in aggressive processes and seem to function most effectively in
   dynamic, higher speed situations;
2. waxy materials – traditional boundary lubricants that function in both low speed (fiber
   to fiber) and high speed (fiber to metal, fiber to ceramic) processing conditions;
3. low molecular weight polymers that have high affinity for the surface of the fiber
   and tend to self-assemble depending on the chemical interactions with the modified
   substrate;
4. silicone based materials – tend to have high affinity for the surface of many of the
   fiber forming polymers.
Recent technological developments in fiber processing trend towards higher speed pro-
cessing making the dynamics of the adsorption process and the durability of the adsorbed
layer even more relevant. A need to continuously develop high performance finishes for
surface modification is required in order to meet the increasing requirements of modern
fiber processing operations.
98    The Nanoscience and Technology of Renewable Biomaterials

4.5   Lubrication

Lubrication phenomena are involved when a finish or lubricant is applied to (moving)
objects as means to reduce friction between them. Amonton’s law was proposed in the
17th century in order to analytically describe sliding friction at the macroscopic scale
(Dowson 1998):
                                                                µ = Ff /N              (4.1)
where µ is the coefficient of friction, a dimensionless scalar value that describes the
ratio of the force of friction between two bodies, Ff , the force pressing them together
and the normal force applied, N . From a macroscopic perspective, µ is a constant
related to the nature of both contacting objects. The frictional force (Ff ) is independent
of the apparent contact surface. The Amonton equation can be applied in many cases
at the macroscopic scale and for sliding objects directly in contact. However, simple
experimental observation has shown that frictional forces do depend on the contact area,
the surface roughness as well as the chemical nature of the sliding substances.
   When dealing with fluid lubricants the situation becomes more complicated since the
gap between the two moving objects may vary. The friction coefficient may depend
on the gap between the sliding surfaces as well as the sliding speeds or shear rates.
According to Hamrock (Hamrock, Schmid et al. 2004), four different regimes of fluid
film lubrication can be defined, i.e. boundary, mixed, elasto-hydrodynamic and hydrody-
namic regimes. These regimes depend on a liquid film parameter, . A plot of friction
coefficient as a function of is illustrated by the Stribeck curve (Figure 4.5). The film
parameter, , represents the minimum film thickness separating the two surfaces and
can be quantified by using Equation (4.2):
                                                                = V × ηb /P            (4.2)



                                               Boundary lubrication
                                                       a
                     Friction Coefficient




                                                            b

                                                           Elastohydrodynamic
                                                           lubrication (EHL)




                                                                        Hydrodynamic
                                                                        lubrication


                                                                   Film parameter, Λ

                                            Mixed lubrication

Figure 4.5 Stribeck curve displaying the different regimes of lubrication. Figure redrawn
from Hamrock, Schmid et al. (2004).
                Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films      99

where V is the speed of the moving (sliding) material (for example fiber); ηb is the bulk
viscosity of the lubricant and P is the pressure applied between the two sliding surfaces.
   In full-film lubrication (aka hydrodynamic lubrication) the surfaces are separated by
a thick lubricant film. Ideally there is no wear of the solid surfaces and the friction
is determined by the rheology, surface chemistry, and intermolecular forces of the bulk
lubricant. During boundary lubrication regime the load is carried by the surface asperities
and the lubricant film and the friction behavior is determined by the dynamic properties
of the boundary film. In the intermediate mixed region both the bulk lubricant and
the boundary film do play key roles. Under these conditions the properties of the
adsorbed components and the chemistry and dynamics of the interfacial region between
the tribosurfaces are of utmost importance.
   In the Stribeck curve, the bulk viscosity ηb applies to all the cases considered, from
wide to narrow gaps between the sliding surfaces. However, in reality, the local or
microscopic effective viscosity ηeff may be quite different from the bulk viscosity ηb
especially in the case of very confined systems of ultra narrow gaps (Cho, Cai et al.
1997).
   Luengo, Israelachvili and Granick proposed a set of improved Stribeck-type curves
that are based on experimental data typical in engineering conditions. The corresponding
generalized map of friction force against sliding velocity in various tribological regimes
were also discussed (Luengo, Israelachvili et al. 1996). In the boundary layer film ηeff
is noted to be much higher that the bulk value, ηb . As the shear rate increases a point
is reached where the effective viscosity starts to drop with a power-law dependence
on the shear rate. As the shear rate further increases, a second Newtonian plateau is
encountered. At higher loads ηeff continues to grow with load and transition to sliding
at high velocity is discontinuous and usually of the stick-slip type. While this chapter
covers the general topic of adsorption and lubrication, our emphasis in the next sections
will be the chemistry and adsorbed layer state of polymeric surfactants. Issues related
to roughness, asperities and others are not considered here.


4.6   Boundary Layer Lubrication

In the boundary lubrication regime, the load is carried by a lubricant thin film. A typical
lubricant film usually has a thickness of 100 nm or lower, i.e., only several hundreds of
molecules thick (Guddati, Zhang et al. 2006; Guo, Li et al. 2006; Izumisawa and Jhon
2006). Studying the structure of lubricant thin films and how the molecules organize
during the lubrication process is of utmost importance. In this regime physisorption (as
opposed to chemisorption) is a dominant effect since during fiber processing the lubricant
film is not always intended to be retained onto the surface (in some cases the lubricant
on fiber surfaces could interfere with successive processes or uses of the fiber). The
robustness or strength of adsorbed layer of lubricants during fiber processing is an issue
that has not been addressed systematically.

4.6.1 Thin Films: Property Changes and Transitions
As discussed above, the properties of lubricant thin films change depending on their
distance from the surface. When the thickness of the adsorbed film is comparable to the
100    The Nanoscience and Technology of Renewable Biomaterials


                                     Solid     Boundary     Bulk liquid
                                               liquid

                    • Viscosity
                    • Elasticity
                    • Relaxation
                      time
                                                            Continuum
                                                            properties




                                       Å           nm           µm

Figure 4.6 Schematic diagram of how the effective viscosity, elasticity and relaxation change
with thickness of a lubricant film. Reprinted from Cho, Cai et al. Copyright (1997) with
permission from Elsevier.


dimensions of the lubricant molecules, the properties of the thin film are quite different
than those of the bulk medium (Cho, Cai et al. 1997). As shown in Figure 4.6, the
effective viscosity, elasticity and relaxation time increase with diminishing thickness
and diverge when the film thickness is sufficiently small. At these dimensions classical
continuum considerations, which can be applied to the bulk phase, do not hold for
thin films.
   The diffusion coefficient of finish molecules in thin films also diverges when compared
with that in the bulk. Mukhopadhyay et al. (Mukhopadhyay, Zhao et al. 2002) found
that the molecular diffusion coefficient decreases exponentially from the edges towards
the center in systems under Hertzian contact. Hertzian contact is an ideal model to
describe deformation and lubrication. In Hertzian contact only small deformation occurs
in the contact areas as contacting bodies are elastic. Granick et al. (Mukhopadhyay, Bae
et al. 2004; Granick and Bae 2006) studied the influence of shear behavior on polymer
interfacial diffusion. According to their results shear did not substantially modified the
Brownian diffusion.
   Phase behaviors of lubricants may change in confined conditions and that is one of the
main reasons why properties of thin films differ from those of the bulk. Confinement-
induced phase states of lubricant layers could change from liquid-like to an amorphous
state and then to a solid-like state (Yoshizawa, Chen et al. 1993). While low friction
is exhibited by solid-like and liquid-like layers, high friction is exhibited by amor-
phous layers. A change of some controlling variables such as temperature and humidity
may shift the phase status from the solid-like towards the amorphous or liquid-like
states. Confinement-induced solidity of lubricant was observed by Denirel and Granick
(Demirel and Granick 2001) by placing octamethyl cyclotetrasiloxane (OMCTS) liquids
between two rigid mica plates and decreasing their spacing below ca. 10 molecular
dimensions of the lubricant. This phenomena was also observed by Israelachvili and
coworkers (Israelachvili, Luengo et al. 1996; Luengo, Schmitt et al. 1997) by shearing
polybutadiene (PBD) of 7000 Daltons. They found that at low shear rates PBD exhibited
bulk-like properties in films thicker than 200 nm while in thinner films (200–220 nm)
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films       101

the shear viscosity ηeff and moduli G and G became quite different from those of
the bulk. On entering the tribology regime (film thickness <30 nm) PBD exhibited
highly nonlinear behavior and yield points indicative of phase transitions to ‘glassy’
or ‘solid-like’ states. Klein et al. (Klein and Kumacheva 1998) discovered that the
transition between liquid-like behavior and a solid-like phase of the liquids under pro-
gressive confinement take place abruptly at a distance of few molecular layers. The
films that are thinner behaved in a solid-like fashion and they required a critical stress to
shear them.

4.6.2 Orientation of Lubricant Films
Why can lubricants reduce friction? How do lubricant molecules work and behave under
shear? These questions are currently being investigated by several groups. Lubricant
molecules organize themselves under shear as illustrated in Figure 4.7 by Yoshizawa et al.
(Yoshizawa, Chen et al. 1993). A critical velocity Vc * exists; if the sliding velocity of
two surfaces are below Vc * a polymeric lubricant film exhibits amorphous structure and
the polymer chains interplay and entangle with each other. In this case high friction is
produced (static-kinetic sliding). This phenomenon supports experimental observations
in which chain interdigitation was found to be an important molecular mechanism giving
rise to ‘boundary’ friction and adhesion hysteresis of monolayer-coated surfaces. If the
sliding velocity of two surfaces is above the critical velocity, polymer chains will be
aligned or ‘combed’ by shear into an ordered conformation and therefore will result in
very low friction (superkinetic sliding).
   The phenomenon of shear–induced alignment of lubricant molecules has been vali-
dated by a number of experiments. For example, Frantz and co-workers (Frantz, Perry
et al. 1994) adsorbed polyisoprene onto a single solid surface and found that the back-
bone of the polymer oriented in the direction of flow. They also found that the extent
of orientation increased with increasing molecular weight. The structure of the lubri-
cant, such as chain length (Frantz, Perry et al. 1994), packing densities (Ruths 2003;
Ruths, Alcantar et al. 2003), and nature of the polymer (brush-like (Zappone, Ruths
et al. 2007) or grafted polymer (Urbakh, Klafter et al. 2004) and chain ends (Chen,
Maeda et al. 2005)) have been found to influence molecular alignment of the lubricant
under shear.
   Within these investigations, the work of Urbakh et al. (Urbakh, Klafter et al. 2004)
is very significant. They used grafted polyelectrolytes, hyaluronan and hylan, to mimic


                                       V<Vc*                         V>Vc*



                                               D



                    Static-kinetic sliding         Superkinetic sliding

Figure 4.7 Lubricant molecules organized by shear. Reprinted from Yoshizawa, Chen et al.
Copyright (1993) with permission from Elsevier.
102   The Nanoscience and Technology of Renewable Biomaterials

cartilage lubrication. These polysaccharides (outermost cartilage layer) were not expec-
ted to be the responsible molecule for the great lubricity of cartilage. However, the
authors found that they may contribute to the loadbearing and wear protection in these
surfaces. Their study showed that a low coefficient of friction is not a requirement for,
or necessarily a measure of, wear protection.


4.7 Techniques to Study Adsorption and Friction Phenomena

A common function of thin films in boundary and mixed lubrication regimes is to offer
friction reduction and wear protection. A better understanding of thin film lubrication
will improve our knowledge of how lubricants work and this knowledge can be used to
develop superior lubricant formulations and help to predict tribological failures.
   In the last few decades, rapid advancements in analytical instrumentation and tech-
niques as well as the expansion in computing power have offered an unprecedented
opportunity to unveil the behavior of lubricant polymers under boundary lubrication con-
ditions (at the atomic/molecular or nano levels). For example, atomic force microscopy
(AFM) with lateral force capabilities can measure the friction between a substrate and
a sharp tip with contact areas of a few to several hundred atoms. In fact, the lateral
resolution of LFM can be less than an atomic spacing (Behary, Ghenaim et al. 2000;
Breakspear, Smith et al. 2003). The surface force apparatus (SFA) can measure the
                                                                            ˚
forces between atomically flat surfaces as their separation is varied with Angstrom level
resolution. The friction and adhesion can be studied as a function of the chemistry
and thickness of the material between the surfaces (Hu and Granick 1998; Sulek and
Wasilewski 2006; Drummond, Rodriguez-Hernandez et al. 2007; McGuiggan, Gee et al.
2007; Zappone, Ruths et al. 2007; Zhang, Hsu et al. 2007). Computer simulation has
also played an important role in interpreting and explaining the findings from these exper-
imental methods. Computer simulations and theoretical investigations have shed much
light on the molecular details underlying both structural and dynamic behavior of liquids
in the highly confined regime (Akagaki and Kato 1988; Kong, Tildesley et al. 1997).
   From a molecular perspective lubricant molecules adsorb on a metal or organic surface
as ordered or oriented chains. The interactions of solid surfaces and lubricant films could
be categorized as physical adsorption or chemical reaction (Hsu 2004). As the thickness,
the adsorption mass and structure of the adsorbed layer are crucial to the performance
of lubrication (Rabinowi 1967; Grudev and Bondaren 1973; Visscher and Kanters 1990;
Gilmour, Paul et al. 2002) in situ techniques that can measure these phenomena are
needed. Surface Plasmon Resonance (SPR) and Quartz Crystal Microbalance (QCM)
are well-established noninvasive methods capable of providing a wealth of information
about interfacial phenomena in situ, in real time and in fluid media (Stockbridge 1966;
Nomura, Okuhara et al. 1981; Nomura and Okuhara 1982; Kanazawa and Gordon 1985a,
1985b; Johannsmann, Mathauer et al. 1992; Liedberg, Nylander et al. 1995; Rodahl,
Hook et al. 1995; Rodahl and Kasemo 1996a, 1996b; Mak and Krim 1997; Homola,
Yee et al. 1999; Bailey, Kanazawa et al. 2001; Bruschi and Mistura 2001; Bailey,
Kambhampati et al. 2002; Wang, Mousavi et al. 2003; Krim, Abdelmaksoud et al.
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films    103

Table 4.1 General comparison between QCM and SPR techniques.
Instrument                        QCM-D                                  SPR
Principle           Piezoelectric/electromechanical           Optical
Resolution          Few ng/cm2 in water
Detection range     The detection range varies from           ∼300 nm (related to the
                      nanometers to micrometers,               wavelength of the
                      depending on the viscoelasticity of      probing light)
                      the adsorbed film. In pure water it
                      is approximately 250 nm.
Information         – Adsorbed mass                           – Total adsorbed mass
   provided         – Adsorption kinetics                     – Adsorption kinetics
                    – Dissipation                             – Refractive index adjacent
                                                                to metal surface


2004; Lundgren, Persson et al. 2006). Ellipsometry is another powerful technique that
can be applied to measure the mass and thickness of adsorbed layers (Fukuzawa, Shimuta
et al. 2005). Even though friction cannot directly be measured with these techniques,
they can be instrumental in finding a relationship between the extent of adsorption (and
viscoelasticity of the adsorbed layer in the case of QCM with dissipation monitoring,
QCM-D) and lubrication (as measured by LFM, SFA and others).
   Table 4.1 compares these two techniques. QCM-D systems are more sensitive to water-
rich and extended layers, while the SPR system is favored for compact and dense layers.
The reason for this difference is due to the different physical principles by which the
coupled mass is measured. The mass-uptake estimated from SPR data is based on the
difference in refractive index between the adsorbed materials and water displaced upon
adsorption. Therefore water associated with the adsorbed materials, i.e. hydration water,
is essentially not included in the mass determination. In contrast, changes in frequency
acquired with QCM-D are affected by the coupled water arising from hydration, the
viscous drag and/or entrapment in cavities in the adsorbed film. In QCM-D measure-
ments the layer is essentially sensed as a ‘hydrogel’ composed of the macromolecules
and coupled water. Changes in the QCM dissipation (D) signals can be related to the
shear viscous losses induced by the adsorbed layers. These viscous losses can provide
information to identify structural differences between adsorbed systems, or structural
changes in a given type of molecule during the adsorption process. By applying proper
interpretation models one can therefore decouple the effect of associated solvent. A more
detailed account of the principles involved in QCM and SPR is given in the next section.


4.8   Surface Plasmon Resonance (SPR)

A surface plasmon is a electromagnetic wave occurring at the interface between a metal
and a dielectric material (Liedberg, Nylander et al. 1995). Surface plasmons are excited
when the energy of the photon electrical field is tuned to a specific value at which it can
104    The Nanoscience and Technology of Renewable Biomaterials



                                                      Detector




                                                                               Intensity
        Light                                                                                  I    II
                                              Reflected
        source                                             II
                                                   light
                 Polarized                                                                              Angle
                 Incident light                                   II
                                                                 II
                                      Prism




                                                                       refractive index
                                                                                                   II




                                                                          Change in
              Chip with metal
              thin film
                                                                                           I
                                                                                                        Time



                       Flow channel


                    Figure 4.8 Schematics of surface plasmon resonance.

interact with free electrons available in the metal surface. This photon energy is then
transferred to a charge density wave and can be observed as a sharp dip in the reflected
light intensity. The angle at which the sharp dip happens is called ‘SPR angle’. Outside
the metal surface there is an evanescent electric field which decays exponentially. This
evanescent field interacts with the close vicinity of the metal. The SPR signal arises under
conditions of total internal reflection and depends on the refractive index of solutions in
contact with the surface. Molecules in solution exhibit changes in refractive index and
thus give rise to a measurable SPR signal if specific interactions occur. A schematic
illustration of SPR is shown in Figure 4.8.
   The refractive index near the sensor surface changes because of the binding of poly-
mers to the surface. As a result, the SPR angle will change according to the amount
of bound material. The thickness of the adlayer can be estimated from Equation (4.3)
(Bailey, Kanazawa et al. 2001), which assumes that the thickness of the dielectric film
is much smaller than the wavelength of the probing laser:
                 √
              nλ −εm εs (εs − εm )           εf            εm + εs 2
        df =                                                           (sin θc )       (4.3)
                       2π           (εf − εs )(εf − εm )     εm εs
where df is the thickness of adlayer; n is the solvent refractive index; λ is the wavelength
of the incident laser; εf is the dielectric constant of the film; εs is the dielectric constant
of the solvent; εm is the real part of the dielectric constant of the metal; and θc is the
critical resonant angle on the plasmon resonance curve. So for a given system with
known solvent and metal (usually gold), θc is the only variable. Equation (4.3) can be
simplified as:
                                        df = k (sin θc )                                                        (4.4)
where k is a factor that can be obtained after a calibration. In most cases, θc is
very small and there is a linear relationship between the amount of bound material
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films      105

and the shift of the SPR angle (Liedberg, Nylander et al. 1995; Homola, Yee et al.
1999). SPR response values are usually expressed in resonance or refractive index
units. Detection of adsorption of small molecules (less than 200 Dalton) is diffi-
cult in SPR. On the other hand, much larger molecular masses are also difficult to
sense due to limitations in the penetration depth of the evanescent wave. However,
both situations are not relevant in most experimental cases and the linear relation-
ships hold. The reader is referred to a number of excellent review papers that discuss
SPR and its principles of operation (Liedberg, Nylander et al. 1995; Homola, Yee
et al. 1999).


4.9   Quartz Crystal Microbalance with Dissipation (QCM)

A QCM crystal consists of a thin quartz disc sandwiched between a pair of (gold)
electrodes. Due to the piezoelectric properties of quartz, it is possible to excite the
crystal to oscillation by applying an AC voltage across its electrodes.
   The resonant frequency (f ) of the quartz crystal depends on the total oscillating mass,
including water coupled to the resonator. When a thin film is attached to the crystal
its frequency decreases. If the film is thin and rigid, negligible or minimum energy
dissipation occurs and the decrease in frequency is proportional to the mass of the film.
In this case the Sauerbrey relation can be applied (Sauerbrey 1959):
                                  ρq tq f      ρq vq f       c f
                           m=−            =−          2
                                                         =−                           (4.5)
                                    nf0          2nf0          n

      C = typically 17.7 ng Hz−1 cm−2 for a 5 MHz quartz crystal.
      n = 1, 3, 5, 7 is the overtone number.

Because the change in frequency can be detected very accurately the QCM operates as
a very sensitive balance. The quartz crystal microbalance was first used to monitor thin
film deposition in vacuum or gas atmospheres. Later on, it was shown that QCM may
be used in the liquid phase thus dramatically increasing the number of applications. The
Sauerbrey relation was initially developed for adsorption from the gas phase but it is
now extended to liquid media where it holds in most cases. In order to describe soft
adlayers of polymer adsorbing from liquid media, the dissipation value D was introduced.
Rodahl et al. (Rodahl, Hook et al. 1995) extended the use of the QCM technique and
introduced the measurement of the dissipation factor simultaneously with the resonance
frequency by switching on and off the voltage applied onto the quartz. The measured
change in dissipation is originated by changes in the coupling between the oscillating
sensor and its surroundings and it is influenced by the layer’s viscoelasticity and slip
of the adsorbed layer on the surface. The dissipation factor D, is the inverse of the
so-called Q factor and it is defined by:
                                        1     Edisspated
                                  D=      =                                         (4.6)
                                       Q      2π Estored
106    The Nanoscience and Technology of Renewable Biomaterials

where Edissipated is the energy dissipated during one period of oscillation and Estored is the
energy stored in the oscillating system. The resonance frequency is measured when the
oscillator is on and the amplitude A of the oscillation is monitored when the oscillator
is turned off. A can be determined in its decay as an exponentially damped sinusoidal
function:
                             A(t) = A0 e−t/τ sin(ωt + ϕ) + c                            (4.7)
where τ is the decay time, ω is the angular frequency at resonance, φ is the phase angle
and the constant, c, is the offset. The dissipation factor is related to the decay time
through Equation (4.8).
                                               1
                                       D=                                           (4.8)
                                             πf τ
Combining Equations (4.5) and (4.8) the dissipation changes can be expressed as
Equation (4.9). This equation shows that dissipation changes depend not only on the
properties of the adsorbed layer but also the density and viscosity of the solution
(Rodahl and Kasemo 1996a):
                                     √ 1        ηf ρf
                                D= n                                          (4.9)
                                         ρq tq 2πf
Generally, soft adlayers dissipate more energy and thus are of higher dissipation value.
From this point of view, dissipation can be used as an indicator of the conformation of
the adlayer.
  A typical QCM-D system records the signals of fundamental frequency (5 MHz) and
overtones (e.g. 15, 25 and 35 MHz and even high frequencies for newly developed
systems). Each overtone has its own detection range in thickness. Theoretical work by
Voinova and coworkers (Voinova, Rodahl et al. 1999) advanced a general equation to
describe the dynamics of two-layer viscoelastic polymer materials of arbitrary thickness
deposited on solid (quartz) surfaces in a fluid environment as follows:
                                                                            
                       1  η3                              η3  2
                                                                     ηj ω2   
             f ≈−                +         hj ρj ω − 2hj                           (4.10)
                    πρ0 h0  δ3                            δ3    µ2 + ω2 ηj 
                                                                           2
                                   j =1,2                          j
                                                                     
                      1     η                    η3 2     µj ω       
                               3
             D≈                   +        2hj                                     (4.11)
                  2πfρ0 h0  δ3                    δ3   µ2 + ω2 ηj 
                                                                  2
                                    j =1,2               j

where ρ stands for density; h is the thickness; η is the viscosity and δ stands for
                                      2η
the viscous penetration depth (δ = ρω ). The subscript 0, 1, 2 and 3 denote quartz
crystal layer 1, layer 2 and bulk solution respectively. From this model, the shift of
the quartz resonance frequency and dissipation factor strongly depend on the viscous
loading of the adsorbed layers and on the shear storage and loss moduli of the over-
layers. These results can readily be applied to quartz crystal acoustical measurements
of polymer viscoelasticity which conserve their shape under the shear deformations and
do not flow as well as layered structures such as protein films adsorbed from solution
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films      107

onto the surface of self-assembled monolayers. By measuring at multiple frequencies
and applying this model the adhering film can be characterized in detail: viscosity, elas-
ticity and thickness may be extracted even for soft films when certain assumptions are
made.


4.10   Application of SPR and QCM to Probe Adsorbed Films

4.10.1 Monitoring Adsorption and Desorption of Macromolecules
SPR and QCM techniques are useful to determine if a given molecule has affinity or
not with the respective metal/organic/polymeric substrate. They also enable elucida-
tion of how strong the affinity is by measuring the actual kinetics of adsorption and
desorption. For example, in a report about the uptake from an organic solution of
octadecyltrichlorosilane, which is of particular interest for the fabrication of microelec-
tromechanical system devices, the authors used quartz crystal microbalance data to fit
a Langmuir isotherm (Hussain, Krim et al. 2005). In this case the adsorption rate was
written as follows (Equation 4.12):
                                         β
                                 φ(t) = [1 − exp(−αt)]                               (4.12)
                                         α
where φ is the fraction of free active sites on the surface, α = Cb kaf + kar and β =
Cb kaf . Cb is the concentration of adsorbate, while kaf and kar represent the constants of
adsorption and desorption. The parameters α and β can be obtained by fitting frequency
data. Furthermore, from the relation between α and Cb , the values of kar and kaf and
the adsorption equilibrium constant (Keq = kaf /kar ) was calculated as well as the free
energy of adsorption (Equation 4.13):
                                     G = −RT ln Keq                                 (4.13)
   Lubricant degradation can also be measured via QCM. In order to monitor the
degrading process of lubricants at high temperature, Wang et al. (Wang, Mousavi
et al. 2004) used QCM at high temperatures (more than 200 ◦ C) to evaluate the
thermal stability of polyol ester lubricants. Figure 4.9 provides an example that
demonstrates how two lubricants showed different sensitivities to temperature. Here
the lubricants were held in a T-controlled chamber. The lubricants degraded gradually
when they were heated to very high temperature leaving solid residues on the tested
surfaces. The behavior of two commercial-grade pentaerythritol tetrapelargonate
based lubricants, represented by the codes ‘EM’ and ‘AF’ (corresponding to two
commercial lubricant compositions), are shown in this figure. During the first nine
hours, both EM and AF did not change with the thermal treatment indicating that both
lubricants were stable. However, after exposure to high temperatures for nine hours
the frequency of AF decreased rapidly while that of EM barely changed. This behavior
indicated that EM was much more stable than AF at the tested temperature of 200 ◦ C.
QCM can thus provide an integral picture of the thermal stability of lubricants in
real-time.
108   The Nanoscience and Technology of Renewable Biomaterials


                               0
                                                                   EM


                             −500



                  Df (Hz)   −1000


                            −1500


                            −2000                                   AF


                            −2500
                                    0   2   4   6   8 10 12 14 16 18 20 22 24
                                                     Time (hour)

Figure 4.9 Time-dependent frequency change of QCM for EM and AF adsorbed on QCM
crystal at 200 ◦ C. Reprinted with permission from, Mousavi et al. Copyright (2004) American
Chemical Society.


4.10.2 Conformation of Adsorbate Layers Revealed by the QCM-D
Indirect information about the conformation of adsorbed layers can also be derived
from QCM experimental data. For rigid, ultrathin, and evenly distributed adsorbed
layers, the Sauerbrey equation (Sauerbrey 1959) describes successfully the proportional
relationship between the adsorbed mass (m) and the shift of the QCM crystals’ resonance
frequency (f ). Under these conditions, the dissipation value is a constant. It doesn’t
change with time or with increasing adsorbed mass. On the other hand, if the adsorbed
material exhibits a viscoelastic behavior, such as that exhibited by layers of proteins,
substantial deviations from the Sauerbrey equation can occur. Using D– f plots one
can eliminate time as an explicit parameter and as concluded in previous studies (Rodahl
and Kasemo 1996a; Hook, Kasemo et al. 2001; Edvardsson, Rodahl et al. 2005), the
absolute slopes and their gradients can provide information about the kinetic regimes
and the conformational changes of the polymer. The magnitude of the slope provides
an indication on the conformation of the adsorbed layer, lower values indicate a softer
layer. If more than one slope exists it can be concluded that different conformation
states of the adsorbed layer are present during the adsorption process.
   Figure 4.10 shows QCM results (shifts in frequency) for a cellulose-coated sensor
after injection of a high charge density polyampholyte solution using a 1 mM NaCl
background electrolyte solution (130 µl/min flow rate). For comparison, the case of a
silica surface is also included.
   Figure 4.10 illustrates that shifts in frequency upon polyampholyte adsorption on silica
were two times larger than those measured in the case of cellulose films. Also, it is
interesting to note that for both substrates, silica sand cellulose, a small change in the
                 Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films       109

                         40

                         35

                         30

                         25
              −Df (Hz)



                         20

                         15

                         10

                          5                                 Cellulose film
                                                            Silica surface
                          0
                              0   20       40         60        80           100
                                            Time (min)

Figure 4.10 Changes in frequency with high-charge density polyampholyte adsorption,
before and after rinsing, for silica substrate and cellulose film surfaces. Conditions: pH 4.3;
temperature 25 ◦ C; and [NaCl] 1 mM. The polyampholyte was injected at time 10 min and
after an incubation time of ca. 55 min rinsing with background electrolyte was performed.



measured QCM frequency was observed after replacing the polymer solution with the
buffer solution (rinsing). These observations imply that the interactions forces between
the polyampholyte and silica are stronger than for the cellulose film. These results can
be explained by considering electrostatic interactions as the main driving mechanism for
adsorption as both substrates exhibit significantly different surface charge densities. For
long equilibration times small changes in frequency are evident; this is hypothesized
to be the result of polymer reconformation and exchange at the interface, given the
polydisperse nature of this macromolecule.
   The swelling and water-holding ability of adsorbed polyampholyte layers on cellulose
films as a function of ionic strength was evaluated by using D − f plots. Figure 4.11
shows the relation between dissipation and frequency change for the same high charge
density polyampholyte adsorbed on cellulose at different ionic strengths.
   Larger variations in energy dissipation imply more viscoelastic layers. Significant
changes in energy dissipation can be seen for intermediate values of salt concentration
(e.g., 10 and 100 mM). The viscoelastic character of the polyampholyte layers built up
at extreme salt conditions is interpreted as being the result of more rigid structures (e.g.,
at 0.1, 1, and 1000 mM). On the other hand, no major differences are observed on the
state of hydration and extension of the adsorbed layer.

4.10.3 Coupling QCM and SPR Data
While SPR and QCM are often used to monitor adsorption and adsorbed layer dynamics,
each technique has its own strengths and weaknesses. Also, as presented before, they
have assumptions inherent in data collection and analysis (Bailey, Kambhampati et al.
110    The Nanoscience and Technology of Renewable Biomaterials

                                       8
                                                   0.1 mM NaCl
                                       7
                                                   1 mM NaCl
                                       6           10 mM NaCl
                                                   100 mM NaCl
                                       5           1000 mM NaCl
                 DD x 106              4

                                       3

                                       2

                                       1

                                       0
                                           0        5      10        15     20       25    30    35      40
                                                                          −Df (Hz)

Figure 4.11     D − f profiles for polyampholyte adsorption on cellulose surfaces at different
ionic strengths. The high charge density polyampholyte consisted of 20% cationic and 16%
anionic groups.



                                        1.2
                                                   Injection
                                           1        Artifact
                Normalized thickness




                                        0.8

                                        0.6

                                        0.4

                                        0.2
                                                                                          QCM Kinetics
                                           0
                                                                                          SPR Kinetics
                                       −0.2
                                               0               500         1000           1500        2000
                                                                          Time (s)

Figure 4.12 Comparison of adsorption kinetics of a perfluoropolyether lubricant (Fomblin
ZDOL) deposited on silver surfaces as measured by SPR and QCM techniques. Reprinted with
permission from Bailey, Kambhampati et al. Copyright (2002) American Chemical Society.



2002). However, since the two techniques rely on fundamentally different principles
of physics, namely optical and electromechanical, a more complete perspective of the
adsorption phenomena can be achieved by combining them. Figure 4.12 illustrates an
example to demonstrate how QCM and SPR data can be combined to study the kinetics
of adsorption of a thin organic film. In this case both curves agree with each other
very well.
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films      111

   Deviations between the signals in QCM and SPR experiments may indicate that the
film is viscoelastic or that there is some coupled water in the adsorbed layer. By carefully
considering the nature of each measurement it is possible to decouple the viscoelastic
properties and the contributions from coupled water in the film. Below a more detailed
explanation about the role of coupled water is presented. Water can be used as a
boundary lubricant as the fluidity of the hydration layers nanoconfined between two
surfaces significantly differs from the behavior of the water in the bulk (Raviv, Laurat
et al. 2001; Zhu and Granick 2001; Raviv and Klein 2002; Leng and Cummings 2005).
The water coupled with lubricant polymers has the same function, i.e. to protect the
contact surfaces and minimize abrasion. Measurement of the coupled water is not an
easy task since it is difficult to distinguish the coupled from the bulk water. Below
are two alternative ways to decoupled the contribution from water via QCM and SPR
measurements.
   The first approach is to substitute water solvent with D2 O, as reported by Hook and
others (Hook, Kasemo et al. 2001; Craig and Plunkett 2003; Notley, Eriksson et al.
2005). D2 O substitution increases the density and shear viscosity of the bulk liquid and
coupled water by ∼10% and ∼25%, respectively but presumably it doesn’t change any
kinetic and equilibrium state. Therefore, from the slight difference in frequency from
experiments conducted in normal and heavy water, the coupled water fraction can be
obtained through Equation 4.14 (Craig and Plunkett 2003).

                                              fs − fd
                      Sfraction =                                                   (4.14)
                                            ρd            ρs
                                    fs   1−      − fd 1 −
                                            ρp            ρp

Subscript s, d, p represents solvent, deuterated water and polymer respectively. In some
cases where ρp = ρs , Equation (4.14) can be simplified to Equation (4.15):

                                                fs − fd
                                Sfraction =                                         (4.15)
                                                     ρd
                                              fs 1 −
                                                     ρp

Since the difference is very small, only polymers adsorbing in large quantities or carrying
large amounts of coupled water can be analyzed with this approach.
   The second method to decouple hydration from bulk water is by combining QCM and
optical methods, for example SPR or ellipsometry (Hook, Kasemo et al. 2001). The
change in resonant frequency (f ) of the QCM crystal depends on the total oscillating
mass which includes the coupled water. In the case of SPR or ellipsometry water coupled
with adsorbed molecules doesn’t affect the refractive index hence they are not detected
by these optical techniques. Therefore by subtracting the mass determined from SPR or
ellipsometry measurements from that obtained from QCM measurements the contribution
of coupled water can be revealed. Figure 4.13 is an example used here to demonstrate the
combination of QCM and SPR techniques. The polymer tested was a cationic polyamide
(5% cationic groups), with molecular weight ca. 3 million. The surface used in this
experiment was a negatively charged silica surface. The experimental results indicate
that there was around 25% of water in the adsorbed polymer layer.
112    The Nanoscience and Technology of Renewable Biomaterials

                                          250

                                                          QCM
                                          200




                  Asorbed mass (ng/cm2)
                                          150             SPR


                                          100
                                                                     Rinse with water

                                          50


                                           0
                                                0   500     1000      1500    2000      2500
                                                                   Time (s)

Figure 4.13 Decoupling water content through the combination of QCM and SPR mea-
surements. The polymer used in this experiment was a cationic polyamide. The calculated
coupled water determined by this method was found to be 25%.


4.11   Lateral Force Microscopy

Both SPR and QCM allow for real time, in-situ monitoring of adsorption processes.
Although relevant to lubrication phenomena these adsorption techniques do not measure
friction behavior. Lateral Force Microscopy (LFM) is a technique that can directly
measure friction by lateral forces. These direct measurements allow for the evaluation
of lubricants’ performance on specific surfaces with nanoscale resolution. LFM when
used with SPR and QCM techniques could unveil a more comprehensive understanding
of lubrication phenomena.
   LFM is based on scanning probe microscopy and it is one of the few experimental
methods capable of assessing forces at the single contact or atomic level. LFM and
atomic force microscopy (AFM) share the same principles. A typical AFM comprises
three main components: laser source, cantilever and photo-detector (see Figure 4.14).
When an atomic force microscope (AFM) tip slides on a surface it is deformed both in
the vertical and the horizontal directions (Figure 4.14). The force Fn , which is normal
to the surface of the sample, results in vertical bending of the free end of the cantilever.
By contrast, the force Fl , which is parallel to the probed surface and is in the opposite
direction to the sliding direction, leads the cantilever into a twisting motion. A typical
AFM measures only the normal force, Fn . What distinguishes LFM form AFM, as the
name indicates, is that it measures both Fn and Fl .
   In order to precisely detect the forces between the tip and the surface, a laser beam is
reflected off the back of the cantilever onto a quadrant photodiode detector. The output
of the quadrant detector is used to determine the degrees of bending and twisting of the
cantilever. The laser beam method is the most commonly used monitoring technique
as it can achieve a resolution comparable to that of an interferometer while it is also
inexpensive and easy to use. The availability of lateral force microscopy (LFM) has
made it possible to explore friction and wear at the molecular level and to examine the
                Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films        113




                             Laser
                                                              Photo
                                                             Detector




                               Cantilever                   Fn

                                Surface
                                                                 F



                    Figure 4.14 Schematic of lateral force microscopy.




                           Sample
                                       different material

                         LFM Image




                           Sample


                         LFM Image



Figure 4.15 Lateral deflection of the cantilever from changes in surface friction (top) and
from changes in slope (bottom) (redrawn from http://mechmat.caltech.edu/∼kaushik/park/
1-4-0.htm).


effectiveness of a finishing treatment in modifying a specific behavior of the substrate.
LFM has been used extensively to study molecular lubrication phenomena on hard sur-
faces, such as mica, silica, and graphite. Studies on polymer surfaces, relevant to fiber
applications, however, have been limited, primarily due to the fact that polymer sur-
faces deform easily, which adds complexity to the experiment and to the interpretation
of the data.
   Lateral force acting on cantilever usually arises from two sources: changes in surface
friction and changes in slope, as illustrated in Figure 4.15. In the first case, since different
materials provide different friction, the cantilever can experience different twisting extent
114   The Nanoscience and Technology of Renewable Biomaterials

                             80

                             60
                                                                             In air
                             40
                                                                             P65-1
        Lateral force (pA)
                             20
                                                                             P65-2
                              0
                                                                             P65-3
                             −20
                                                                             P65-4
                             −40
                                                                             P65-5
                             −60

                             −80
                                   0   1       2           3       4    5
                                           Surface position (µm)

Figure 4.16 One line scanning profiles for cellulose-coated silica surface while immersed
in a nonionic triblock lubricant (E19P29E19) solutions and in air. P65 is used to indicate
lubricant E19P29E19, which is a triblock copolymer with 19 E groups at both ends and 29 P
groups in the middle. P65-1∼P65-5 represent a series of ethanol solutions with the increase
of ethanol concentration.


even though the surface being measured is topographically smooth. In the second case,
the cantilever may twist when it encounters a steep slope. In order to eliminate the
roughness effect caused by the second case in lubrication, two scans on the same line
(back and forth) are performed on the substrate in order to measure the net effect (Behary,
Ghenaim et al. 2000).
   When a tip in lateral force microscopy is sliding on a surface, lateral force profiles can
be measured both in air (no lubricant applied) and in solution. Figure 4.16 shows lateral
force profiles for a cellulose surface imaged in air and immersed in a solution with
nonionic E-P-E triblock polymeric surfactants (commonly used as lubricant finishes).
Here E and P represent ethylene oxide and propylene oxide groups, respectively. During
these experiments, the lubricant was dissolved in ethanol aqueous solutions at various
levels of ethanol concentration (22, 38, 52, 66, and 87%). It was observed that the friction
forces measured in air were significantly larger than those in the respective solutions,
confirming the lubrication attributes of the polymer. However, the force profiles in
the five solutions were undistinguishable, making the effect of ethanol concentration
unimportant.
   Studies on copolymer adsorption are usually conducted with hydrophobic surfaces
and only a few reports have addressed the case of adsorption on hydrophilic surfaces.
The adsorption behavior of E19P29E19 copolymers on hydrophilic cellulose surfaces is
hereby briefly discussed. It is expected that the self-assembly mechanism of the block
copolymer in the case of cellulose will be different from that exhibited by hydrophobic
surfaces such as propylene or polyethylene. Wu et al. (Wu, Liu et al. 2000) carried
out an AFM study involving triblock copolymer chains on hydrophilic silica surfaces.
They suggested that in the case of hydrophilic surfaces, the E blocks bind the surface
                Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films                      115

                                                 1
                                                                                         RP10E13
                                                                         Cellulose       RP13E17
                                                                                         E26P40E26




                   Log, Friction coefficient
                                                                                         E133P50E133
                                                                                         Air
                                                                                         Water

                                                0.1




                                               0.01
                                                      0   10   20   30    40    50      60   70    80
                                                                    Normal force (nN)

Figure 4.17 The relationship of friction coefficient (COF) and normal force (Fn ) on cellulose
films in air, water and in the presence of four types of nonionic polymers. E: polyethylene
oxide; P: polypropylene oxide; R: alkyl groups.

because the shared hydrophilicity nature of the E blocks and silica surface (affinity
between the E blocks and the silica surface). Consequently the P blocks are repelled
from the surface. A competition between solvency of E segments and the enthalpic
E-to-surface attraction is likely to be present in the case of cellulose substrates. Therefore,
an anchor-buoy-anchor configuration may exist on the hydrophilic cellulose surface.
Molecular self-assembled structures are formed on the interfaces between sliding surfaces
as a result of morphology changes at a nanoscale level. These changes mainly depend on
the chemical nature of the surface and the liquid. In boundary lubrication, it is believed
that surface coatings of organized, molecular liquid films will control friction and reduce
wear in fiber processing.
   Figure 4.17 shows an LFM curve for coefficient of friction at different applied
loads. Under low normal forces the coefficient of friction decreases as the normal
force increases. However, at high normal forces, the value of friction coefficient does
increase. The threshold for this transition was around 30–40 nN. This behavior can
be explained by the fact that lubricant molecules self assemble onto the surface and
form a layer under shear and normal forces. At higher shear rates or normal forces,
the polymer aligns better and forms a more compact structure with a low coefficient
of friction. However, at higher pressures and loadings, the polymer film might be
distorted (molecules can be driven out from the interface) and the tip can make direct
contact with the unlubricated surface thus measuring a higher coefficient of friction.
This phenomenon is especially relevant in the case of sharp LFM tips where even a
normal force of only 30–40 nN can produce a substantially high pressure.


4.12 Summary

In this chapter we discussed the use of QCM and SPR as tools to monitor the adsorp-
tion of molecules on solid surfaces. Some examples were provided with regards to the
116   The Nanoscience and Technology of Renewable Biomaterials

modification of the surface of cellulose thin films via adsorption of polyampholytes and
nonionic polymers. These techniques allow the determination of fundamental informa-
tion, relevant to lubrication phenomena including (1) affinity of adsorbing molecules to
the substrate, (2) viscoelasticity of adsorbed layers, (3) kinetics of adsorption and des-
orption, and (4) thickness of the adsorbed layer as well as the amount of coupled water
in adsorbed films. LFM was presented as a useful tool to directly measure friction on
polymeric surfaces. LFM complements results from the adsorption tests as LFM allows
us to quantify the extent of the adsorption as well as the conformation of adsorbed lay-
ers. Based on information provided via LFM, SPR and QCM, a better understanding of
friction phenomena on cellulosic systems can be achieved. By correlating the structure
and lubricant effect of adsorbates, novel formulations with superior performance can be
tailored. In return one can significantly improve the efficiency of cellulose fiber and
textile processing and improve the quality of products being manufactured.
   Overall, it is concluded that a fundamental understanding of adsorption and friction
behavior can unveil a more complete understanding about boundary lubrication and
nanostructuring phenomena on cellulose systems.


Acknowledgements

Funding supported from the National Textile Center under the Grant number C05-NS09
and the National Research Initiative of the USDA Cooperative State Research, Education
and Extension Service, grant number 2007-35504-18290 is gratefully acknowledged.
Dr Tom Theyson, from Goulston Corp. is acknowledged for his advice and suggestions.


References

Akagaki, T. and K. Kato (1988) Simulation of flow wear in boundary lubrication using
  a Vickers indentation method. Tribology Transactions 31(3): 311–16.
Bailey, L.E., D. Kambhampati, et al. (2002) Using surface plasmon resonance and the
  quartz crystal microbalance to monitor in situ the interfacial behavior of thin organic
  films. Langmuir 18(2): 479–89.
Bailey, L.E., K.K. Kanazawa, et al. (2001) Multistep adsorption of perfluoropolyether
  hard-disk lubricants onto amorphous carbon substrates from solution. Langmuir
  17(26): 8145–55.
Behary, N., A. Ghenaim, et al. (2000) Tribological analysis of glass fibers using atomic
  force microscopy (AFM)/lateral force microscopy (LFM). Journal of Applied Polymer
  Science 75(8): 1013–25.
Bogdanovic, G., F. Tiberg, et al. (2001) Sliding friction between cellulose and silica
  surfaces. Langmuir 17(19): 5911–16.
Bratko, D. and A.K. Chakraborty (1996) A numerical study of polyampholyte configu-
  ration. Journal of Physical Chemistry 100(4): 1164–73.
Breakspear, S., J.R. Smith, et al. (2003) AFM in surface finishing: Part III. Lateral
  force microscopy and friction measurements. Transactions of the Institute of Metal
  Finishing 81: B68–B70.
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films    117

Bruschi, L. and G. Mistura (2001) Measurement of the friction of thin films by means
  of a quartz microbalance in the presence of a finite vapor pressure. Physical Review B
  63(23).
Carr, M.E., B.T. Hofreiter, et al. (1977) Starch polyampholyte for paper. TAPPI Journal
  60(10): 66–9.
Chen, N.H., N. Maeda, et al. (2005) Adhesion and friction of polymer surfaces: The
  effect of chain ends. Macromolecules 38(8): 3491–3503.
Cho, Y.K., L. Cai, et al. (1997) Molecular tribology of lubricants and additives. Tribol-
  ogy International 30(12): 889–94.
Craig, V.S.J. and M. Plunkett (2003) Determination of coupled solvent mass in quartz
  crystal microbalance measurements using deuterated solvents. Journal of Colloid and
  Interface Science 262(1): 126–9.
Demirel, A.L. and S. Granick (2001) Origins of solidification when a simple molecular
  fluid is confined between two plates. Journal of Chemical Physics 115(3): 1498–
  1512.
Dobrynin, A.V., R.H. Colby, et al. (2004) Polyampholytes. Journal of Polymer Science
  Part B-Polymer Physics 42(19): 3513–38.
Dowson, D. (1998) History of Tribology, 2nd Edition. London; New York, Longman.
Drummond, C., J. Rodriguez-Hernandez, et al. (2007) Boundary lubricant films under
  shear: Effect of roughness and adhesion. Journal of Chemical Physics 126(18).
Edvardsson, M., M. Rodahl, et al. (2005) A dual-frequency QCM-D setup operating at
  elevated oscillation amplitudes. Analytical Chemistry 77(15): 4918–26.
Eisenriegler, E. (1993) Polymers near Surfaces: Conformation Properties and Relation
  to Critical Phenomena. Singapore; River Edge, NJ, World Scientific.
Ertas, D. and Y. Kantor (1996) Randomly charged polymers, random walks, and their
  extremal properties. Physical Review E 53(1): 846–60.
Falt, S., L. Wagberg, et al. (2004) Model films of cellulose II – improved preparation
  method and characterization of the cellulose film. Cellulose 11(2): 151–62.
Frantz, P., D. Perry, et al. (1994) Orientation of adsorbed polymer in resonse to sherar-
  flow. Colloids and Surfaces A-Physicochemical and Engineering Aspects 86: 295–98.
Fukuzawa, K., T. Shimuta, et al. (2005) Measurement of thickness of molecularly thin
  lubricant film using ellipsometric microscopy. IEEE Transactions on Magnetics 41(2):
  808–11.
Gilmour, K.R., S. Paul, et al. (2002) The influence of lubricant film thickness on fric-
  tion coefficients during slow speed deep drawing operations. Journal of Tribology-
  Transactions of the ASME 124(4): 846–51.
Granick, S. and S.C. Bae (2006) Open questions about polymer interfacial diffusion.
  Journal of Polymer Science Part B-Polymer Physics 44(24): 3434–5.
Grudev, A.P. and Bondaren. Va (1973) Investigation of dependence of coefficient of
  friction on thickness of lubricant film in cold rolling. Steel in the USSR 3(3): 219–20.
Guddati, S., J. Zhang, et al. (2006) Nanolubrication: Characterization of patterned
  lubricant films on magnetic hard disks. Tribology Letters 21(3): 253–61.
Gunnars, S., L. Wagberg, et al. (2002) Model films of cellulose: I. Method development
  and initial results. Cellulose 9(3–4): 239–49.
Guo, Q., L. Li, et al. (2006) A spreading study of lubricant films via optical surface
  analyzer and molecular dynamics. IEEE Transactions on Magnetics 42(10): 2528–30.
118   The Nanoscience and Technology of Renewable Biomaterials

Gutin, A. M. and E. I. Shakhnovich (1994) Effect of a net charge on the conformation
   of polyampholytes. Physical Review E 50(5): R3322–R3325.
Hamrock, B. J., S. R. Schmid, et al. (2004) Fundamentals of Fluid Film Lubrication.
   New York, Marcel Dekker.
Homola, J., S.S. Yee, et al. (1999) Surface plasmon resonance sensors: review. Sensors
   and Actuators B-Chemical 54(1–2): 3–15.
Hook, F., B. Kasemo, et al. (2001) Variations in coupled water, viscoelastic proper-
   ties, and film thickness of a Mefp-1 protein film during adsorption and cross-linking:
   A quartz crystal microbalance with dissipation monitoring, ellipsometry, and surface
   plasmon resonance study. Analytical Chemistry 73(24): 5796–5804.
Hsu, S.M. (2004) Molecular basis of lubrication. Tribology International 37(7): 553–9.
Hu, Y.Z. and S. Granick (1998) Microscopic study of thin film lubrication and its
   contributions to macroscopic tribology. Tribology Letters 5(1): 81–8.
Hubbe, M.A., O.J. Rojas, et al. (2007a) Charge and the dry-strength performance of
   polyampholytes – Part 2. Colloidal effects. Colloids and Surfaces A-Physicochemical
   and Engineering Aspects 301(1–3): 23–32.
Hubbe, M.A., O.J. Rojas, et al. (2007b) Unique behaviour of polyampholytes as dry-
   strength additives. Appita Journal 60(2): 106–111.
Hussain, Y., J. Krim, et al. (2005) OTS adsorption: A dynamic QCM study. Colloids
   and Surfaces A-Physicochemical and Engineering Aspects 262(1–3): 81–6.
Hwang, D.C. and S. Damodaran (1996) Equilibrium swelling properties of a novel
   ethylenediaminetetraacetic dianhydride (EDTAD)-modified soy protein hydrogel.
   Journal of Applied Polymer Science 62(8): 1285–93.
Israelachvili, J., G. Luengo, et al. (1996) Thin film rheology and tribology of con-
   fined polymer melts. Abstracts of Papers of the American Chemical Society 212:
   104–PMSE.
Izumisawa, S. and M.S. Jhon (2006) Calculation of disjoining pressure for lubricant
   films via molecular simulation. IEEE Transactions on Magnetics 42(10): 2543–5.
Jeon, J. and A.V. Dobrynin (2005) Molecular dynamics simulations of polyampholyte-
   polyelectrolyte complexes in solutions. Macromolecules 38(12): 5300–12.
Johannsmann, D., K. Mathauer, et al. (1992) Viscoelestic properties of thin-films probed
   with a Quatz-crystal resonator. Physical Review B 46(12): 7808–15.
Jost, H. P. (1990) Tribology – origin and future. Wear 136(1): 1–17.
Kanazawa, K.K. and J.G. Gordon (1985a) Frequency of a quartz microbalance in contact
   with liquid. Analytical Chemistry 57(8): 1770–71.
Kanazawa, K.K. and J.G. Gordon (1985b) The oscillation frequency of a quartz resonator
   in contact with a liquid. Analytica Chimica Acta 175(Sep): 99–105.
Kantor, Y. and M. Kardar (1995) Instabilities of charged polyampholytes. Physical
   Review E 51(2): 1299–1312.
Kantor, Y., M. Kardar, et al. (1994) Statistical-Mechanics of Polyampholytes. Physical
   Review E 49(2): 1383–92.
Karim, A. and S. Kumar (2000) Polymer surfaces, interfaces and thin films. Singapore;
   River Edge, NJ, World Scientific.
Klein, J. and E. Kumacheva (1998) Simple liquids confined to molecularly thin layers.
   I. Confinement-induced liquid-to-solid phase transitions. Journal of Chemical Physics
   108(16): 6996–7009.
              Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films    119

Kong, Y.C., D.J. Tildesley, et al. (1997) The molecular dynamics simulation of
  boundary-layer lubrication. Molecular Physics 92(1): 7–18.
Kontturi, E., T. Tammelin, et al. (2006) Cellulose – model films and the fundamental
  approach. Chemical Society Reviews 35(12): 1287–1304.
Krim, J., M. Abdelmaksoud, et al. (2004) Scanning tunneling microscope-quartz crystal
  microbalance studies of ‘real world’ and model lubricants. Dynamics and Friction in
  Submicrometer Confining Systems. 882: 1–18.
Lee, N. and D. Thirumalai (2000) Dynamics of collapse of flexible polyampholytes.
  Journal of Chemical Physics 113(13): 5126–9.
Leng, Y.S. and P.T. Cummings (2005) Fluidity of hydration layers nanoconfined between
  mica surfaces. Physical Review Letters 94(2).
Liedberg, B., C. Nylander, et al. (1995) Biosensing with surface-plasmon resonance –
  how it all started. Biosensors & Bioelectronics 10(8): R1–R9.
Long, D., A.V. Dobrynin, et al. (1998) Electrophoresis of polyampholytes. Journal of
  Chemical Physics 108(3): 1234–44.
Lord, M.S., M.H. Stenzel, et al. (2006) The effect of charged groups on protein inter-
  actions with poly(HEMA) hydrogels. Biomaterials 27(4): 567–75.
Luengo, G., J. Israelachvili, et al. (1996) Generalized effects in confined fluids: New
  friction map for boundary lubrication. Wear 200(1–2): 328–35.
Luengo, G., F.J. Schmitt, et al. (1997) Thin film rheology and tribology of confined
  polymer melts: Contrasts with bulk properties. Macromolecules 30(8): 2482–94.
Lundgren, S.M., K. Persson, et al. (2006) Adsorption of fatty acids from alkane solution
  studied with quartz crystal microbalance. Tribology Letters 22(1): 15–20.
Mak, C. and J. Krim (1997) Quartz crystal microbalance studies of disorder-induced
  lubrication. Faraday Discussions: 389–97.
Mazur, J., A. Silberberg, et al. (1959) Potentiometric behavior of polyampholytes. Jour-
  nal of Polymer Science 35(128): 43–70.
McGuiggan, P.M., M.L. Gee, et al. (2007) Friction studies of polymer lubricated sur-
  faces. Macromolecules 40(6): 2126–33.
Mukhopadhyay, A., S.C. Bae, et al. (2004) How confined lubricants diffuse during
  shear. Physical Review Letters 93(23).
Mukhopadhyay, A., J. Zhao, et al. (2002) Contrasting friction and diffusion in molecu-
  larly thin confined films. Physical Review Letters 89(13).
Nazhad, M.M. (2005) Recycled fiber quality – A review. Journal of Industrial and
  Engineering Chemistry 11(3): 314–29.
Nazhad, M.M. and L. Paszner (1994) Fundamentals of strength loss in recycled paper.
  Tappi Journal 77(9): 171–9.
Nomura, T. and M. Okuhara (1982) Frequency-shifts of piezoelectric quartz crystals
  immersed in organic liquids. Analytica Chimica Acta 142(Oct): 281–4.
Nomura, T., M. Okuhara, et al. (1981) Behavior of a piezoelectric quartz crystal in
  organic-solvents. Bunseki Kagaku 30(6): 417–18.
Notley, S.M., M. Eriksson, et al. (2005) Visco-elastic and adhesive properties of
  adsorbed polyelectrolyte multilayers determined in situ with QCM-D and AFM
  measurements. Journal of Colloid and Interface Science 292(1): 29–37.
Proffitt, T.J. and H.T. Patterson (1988) Oleochemical surfactants and lubricants in the
  textile-industry. Journal of the American Oil Chemists Society 65(10): 1682–94.
120   The Nanoscience and Technology of Renewable Biomaterials

Rabinowi, E. (1967) Variation of friction and wear of solid lubricant films with film
  thickness. ASLE Transactions 10(1): 1 ff.
Raviv, U. and J. Klein (2002) Fluidity of bound hydration layers. Science 297(5586):
  1540–3.
Raviv, U., P. Laurat, et al. (2001) Fluidity of water confined to subnanometre films.
  Nature 413(6851): 51–4.
Rodahl, M., F. Hook, et al. (1995) Quartz-crystal microbalance setup for frequency
  and Q-factor measurements in gaseous and liquid environments. Review of Scientific
  Instruments 66(7): 3924–30.
Rodahl, M. and B. Kasemo (1996a) Frequency and dissipation-factor responses to local-
  ized liquid deposits on a QCM electrode. Sensors and Actuators B-Chemical 37(1–2):
  111–16.
Rodahl, M. and B. Kasemo (1996b) On the measurement of thin liquid overlayers with
  the quartz-crystal microbalance. Sensors and Actuators A-Physical 54(1–3): 448–56.
Ruths, M. (2003) Boundary friction of aromatic self-assembled monolayers: Comparison
  of systems with one or both sliding surfaces covered with a thiol monolayer. Langmuir
  19(17): 6788–95.
Ruths, M., N.A. Alcantar, et al. (2003) Boundary friction of aromatic silane self-
  assembled monolayers measured with the surface forces apparatus and friction force
  microscopy. Journal of Physical Chemistry B 107(40): 11149–57.
Sauerbrey, G. (1959) The use of quartz oscillators for weighing thin layers and for
  microweighing. Zeitschrift Fur Physik 155(2): 206–22.
Schiessel, H. and A. Blumen (1996) Instabilities of polyampholytes in external electrical
  fields. Journal of Chemical Physics 105(10): 4250–6.
Sezaki, T., M.A. Hubbe, et al. (2006a) Colloidal effects of acrylamide polyampholytes –
  Part 2: Adsorption onto cellulosic fibers. Colloids and Surfaces A-Physicochemical
  and Engineering Aspects 289(1–3): 89–95.
Sezaki, T., M.A. Hubbe, et al. (2006b) Colloidal effects of acrylamide polyampholytes.
  Colloids and Surfaces, A: Physicochemical and Engineering Aspects 281(1–3): 74–
  81.
Song, J., J. Liang, et al. (2009) Development and characterization of thin polymer films
  relevant to fiber processing. Thin solid films. 517(15): 4348–54.
Song, J., Y. Wang, et al. (2006) Charge and the dry-strength performance of polyam-
  pholytes. Part 1: Handsheet properties and polymer solution viscosity. Journal of
  Pulp and Paper Science 32(3): 156–62.
Srivastava, D. and M. Muthukumar (1996) Sequence dependence of conformations of
  polyampholytes. Macromolecules 29(6): 2324–6.
Stiernstedt, J., H. Brumer, et al. (2006) Friction between cellulose surfaces and effect
  of xyloglucan adsorption. Biomacromolecules 7(7): 2147–53.
Stiernstedt, J., N. Nordgren, et al. (2006) Friction and forces between cellulose model
  surfaces: A comparison. Journal of Colloid and Interface Science 303(1): 117–23.
Stockbridge, C.D. (1966) Effects of gas pressure on quartz crystal microbalances. Vac-
  uum Microbalance Techniques (5): 147–78.
Sulek, M.W. and T. Wasilewski (2006) Tribological properties of aqueous solutions of
  alkyl polyglucosides. Wear 260(1–2): 193–204.
               Tools to Probe Nanoscale Surface Phenomena in Cellulose Thin Films    121

Theander, K., R.J. Pugh, et al. (2005) Friction force measurements relevant to de-inking
  by means of atomic force microscope. Journal of Colloid and Interface Science 291(2):
  361–8.
Urbakh, M., J. Klafter, et al. (2004) The nonlinear nature of friction. Nature 430(6999):
  525–8.
Visscher, M. and A.F.C. Kanters (1990) Literature-review and discussion on measure-
  ments of leakage, lubricant film thickness and friction of reciprocating elastomeric
  seals. Lubrication Engineering 46(12): 785–91.
Voinova, M.V., M. Rodahl, et al. (1999) Viscoelastic acoustic response of layered
  polymer films at fluid–solid interfaces: Continuum mechanics approach. Physica
  Scripta 59(5): 391–6.
Wang, D., P. Mousavi, et al. (2003) Evaluating thermal degradation of textile finishing
  aids using GC and QCM. Proceedings of the National Conference on Environmental
  Science and Technology, Greensboro, NC, United States, Sept. 8–10, 2002: 243–51.
Wang, D.X., P. Mousavi, et al. (2004) Novel testing system for evaluating the ther-
  mal stability of polyolester lubricants. Industrial & Engineering Chemistry Research
  43(21): 6638–46.
Wang, Y., M.A. Hubbe, et al. (2006) The role of polyampholyte charge density on its
  interactions with cellulose. Nordic Pulp Paper Res. J.: submitted.
Wang, Y., M.A. Hubbe, et al. (2007) Charge and the dry-strength performance
  of polyampholytes. Part 3: Streaming potential analysis. Colloids and Surfaces
  A-Physicochemical and Engineering Aspects 301(1–3): 33–40.
Wu, C.H., T.B. Liu, et al. (2000) Atomic force microscopy study of E99P69E99 triblock
  copolymer chains on silicon surface. Langmuir 16(2): 656–61.
Yamakov, V., A. Milchev, et al. (2000) Conformations of random polyampholytes.
  Physical Review Letters 85(20): 4305–8.
Yoshizawa, H., Y.L. Chen, et al. (1993) Recent advances in molecular-level understand-
  ing of adhesion, friction and lubrication. Wear 168(1–2): 161–6.
Zappone, B., M. Ruths, et al. (2007) Adsorption, lubrication, and wear of lubricin on
  model surfaces: Polymer brush-like behavior of a glycoprotein. Biophysical Journal
  92(5): 1693–1708.
Zauscher, S. and D.J. Klingenberg (2001) Friction between cellulose surfaces measured
  with colloidal probe microscopy. Colloids and Surfaces A-Physicochemical and Engi-
  neering Aspects 178(1–3): 213–29.
Zhang, J., S.M. Hsu, et al. (2007) Nanolubrication: Patterned lubricating films using
  ultraviolet (UV) irradiation on hard disks. Journal of Nanoscience and Nanotechnology
  7(1): 286–92.
Zhu, Y.X. and S. Granick (2001) Viscosity of interfacial water. Physical Review Letters
  8709(9).
                                                        5
                Polyelectrolyte Multilayers
                  for Fibre Engineering

                                      o                             a
                        Rikard Lingstr¨ m, Erik Johansson and Lars W˚ gberg



5.1     Background

The method of forming Polyelectrolyte Multilayers (PEM) was introduced as a general
method in the early 1990s by Decher (1), but the principle was discussed already in
the 1960s by Iler (2). A multilayer is formed by adsorbing an oppositely charged
polyelectrolyte to a charged substrate. The substrate is recharged and in a second step,
an oppositely charged polyelectrolyte can be adsorbed. By repeating this process, layers
consisting of a large number of individual layers can be formed (3).
   In early publications, the recharging of the surface was considered to be a necessary
condition for the formation of PEM (4). Electrokinetic measurements conducted for
PEMs formed from different combinations of polylelectrolytes showed that there was a
change in the sign of the surface potential depending on the polyelectrolyte adsorbed in
the outermost layer (5, 6). Recent investigations have, however, indicated that recharging
is not a prerequisite for the PEM formation (7). This indicates that the entropic gain due
to the release of counter ions and immobilized water molecules is the most important
factor for the PEM formation to take place (8).
   Each adsorption step follows the fundamentals of polyelectrolyte adsorption, but
the mechanisms behind the recharging of the substrate are still not theoretically fully
understood. Different theories have been proposed, based on both equilibrium and
nonequilibrium models, but there is still no general theory available to describe the
PEM formation. Dobrynin et al. (9) have proposed a mechanism based on PEM as a
thermodynamically favourable state, i.e. the PEM-structure that is formed minimises the



The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
124   The Nanoscience and Technology of Renewable Biomaterials

free energy of the system via lateral correlation of adsorbed polyelectrolytes leading to a
recharging of the system. An alternative to this model has been proposed by Cohen Stu-
art (10) arguing that the recharging is due to the kinetic locking of a structure consisting
of loops and tails.
   By using different combinations of polyelectrolytes and nano-particles (4), PEMs
showing very different properties can be formed and the PEM-properties can also be
influenced by varying parameters such as the salt concentration (11), type of salt, tem-
perature (12), molecular mass of the polyelectrolytes (13), type of counter-ions of the
polyelectrolytes (8, 14), and naturally the charge of the polyelectrolytes (4, 15, 16).
   Highly charged polyelectrolytes are known to form thin layers when they are adsorbed
onto a solid substrate. Schlenoff et al. (17) have shown that the thickness of a single
layer of poly(diallyldiamethylammonium chloride) (PDADMAC), adsorbed in the pres-
ence of 0.5 M NaCl, and determined in the dry state by ellipsometry, is less than 1 nm.
PEMs formed from PSS as the anionic polyelectrolyte and a copolymer of PDADMAC
and N-methyl-N-vinylformamide as the cationic polymer were studied by Glinel et al.
(18) for different charge densities, and they reported a decrease in thickness when the
charge density was increased. In another study, Shiratori and Rubner (16) have shown
that polyallylamine (PAH) and poly acrylicacid (PAA), which are both weak polyelec-
                                                                                ˚
trolytes (i.e. sensitive to pH changes) form very thin PEMs (less than 10 A in bilayer
                                               ˚
thickness(dry)) and very thick PEMs (>120 A(dry)) depending on the pH strategies used
the during formation of the PEM. Later investigations using Quartz Crystal Microbalance
(QCM) have indicated that PEMs formed from PAH/PAA at pH 7.5/3.5 are thicker than
layers formed at pH 5/5 and pH 7.5/7.5 (19).
   During the last decade, the LbL method has been developed as a noncomplicated
and general protocol to improve the properties of any solid substrate. PEM treatment
is already in use in several applications such as sensor technology (20) and contact
lens coating (20). Highly efficient membranes (21) can also be formed using the PEM
technique as well as hollow capsules for the controlled release of active chemicals
(22, 23).
   Since PEM treatment influences the properties of the substrate, it can also be used
as a way of improving the adhesion between surfaces. Investigations during the last
decade have shown the potential of the PEM technique as a way of improving the
adhesion between wood fibres and thereby improving paper strength (24). PEMs were
first used on wood fibres in 1998 (25, 26) and improvements were observed in the
tensile strength of papers made from PEM-treated fibres, quantitatively comparable to
those achieved by mechanical beating. Sheets made from fibres carrying PEMs formed
from polyallylaminehydrochloride (PAH) and polyacrylic acid (PAA) (27) showed an
increase in tensile index from about 20 kNm/kg for sheets made from nontreated fibres
to 55 kNm/kg for sheets made from fibres having PEMs formed from 8 individual layers.
PEMs from cationic and anionic starch have also shown very promising results, and a
tensile index higher than 60 kNm/kg has been achieved when fully bleached chemical
softwood fibres were pre-treated with three layers of starch (28).
   One interesting feature is that the tensile strength seems to be dependent on the
polymer adsorbed in the outermost layer (24, 27). Different explanations of this have
been proposed (19), and in this chapter the phenomenon is discussed in terms of the
PEM properties and the wettability of the PEM-treated fibres.
                                     Polyelectrolyte Multilayers for Fibre Engineering   125

5.2   The Formation of PEM on Wood Fibres

The large efficiency of PEM treatment of fibres for preparing strong papers and its
promise as a new tool for fibre engineering naturally raised questions as to whether and
how multilayers were formed on the fibres. It is possible that PEMs are adsorbed only to
the external fibre surface or to both the external surface and other parts of the fibre wall,
or that the added polyelectrolytes form other types of structure on the fibre surface that
are advantageous for the fibre/fibre adhesion. Considering the dimensions of a typical
wood fibre, i.e. cylindrical body with a length of about 2 mm, a diameter of 20 µm
and a wall thickness of about 4 µm, and the difference in morphology between different
fibres, it is very difficult to sample fibres from batches of treated and untreated fibres and
determine the properties of the layers with the aid of e.g. Scanning Electron Microscopy
(SEM). Such a study is further complicated by the fact that the thickness of the layers
of polyelectrolytes formed on the fibres surface are of the order of nm, depending on
the types of polyelectrolyte used and the conditions used during their preparation (17).
Nevertheless, a solution to this problem was essential in order to create a scientific base
for the engineering of fibres with PEM for different products.
    In order to overcome the problem of the difference in morphology between different
fibres, it was decided to use single fibres with the aid of a Cahn balance where only a part
of the fibre was treated with the multilayers (29). This set-up is schematically shown
in Figure 5.1 where one end of the fibre is fastened between two layers of adhesive
tape, attached to the load-cell of the microbalance and the other end is consecutively
dipped into solutions of cationic and anionic polyelectrolyte solutions with the aid of
the moving stage of the microbalance. A careful washing is conducted between each
polyelectrolyte adsorption step.
    After the treatment with a predetermined number of polyelectrolyte layers, the fibre
was removed and both the treated and nontreated part were studied with Environmental
SEM (ESEM) (29). In this way the same fibre can be analysed before and after treatment.
A typical result from such a measurement is shown in Figure 5.2 where an untreated
section of a bleached chemical softwood fibre is shown together with a section that has
been treated with 5.5 bilayers of polyallylamine hydrochloride (PAH) and polyacrylic
acid (PAA) (29).
    As can be seen, the PEM treatment creates a much smoother surface where the small-
scale variations have been reduced to a large extent. From these types of images, it
was concluded that the consecutive treatment with oppositely charged polyelectrolytes
with intermediate washing steps created thin films of PEM on the surface of the fibres.
Similar results have also been presented by Lvov (30) using fluorescently labeled PAH
and fluorescently labeled polystyrene sulphonate (PSS).
    Having established this, the formation of the multilayers can be simply monitored by
adsorption measurements on batches of fibres using standard titration techniques and/or
standard chemical analysis where, for example, the concentration of sulphur in the fibres
after treatment can be used to determine the amount of PSS adsorbed (29). However,
the establishment of adsorption isotherms for the fibres is time-consuming and it gives
little molecular detail about either the kinetics of the adsorption or the properties of the
adsorbed layers. Therefore the PEM-treatment of fibres, in our laboratory, is always
126    The Nanoscience and Technology of Renewable Biomaterials




                                         (a)


                                                                        Adhesive



                                                     Fibre



         Upper level of washing


                                                       Upper level of dipping/treatment




                            Dipping




                                         (b)

Figure 5.1 Schematic representation of how a fibre is placed between two layers of adhesive
tape (a). The fibre is treated in the microbalance with the polyelectrolytes and then with a
subsequent washing stage (b). Since the assembly of adhesive and fibre is attached to the load
cell in the microbalance, it is possible to determine the difference in wetting of the treated
and nontreated parts respectively and to estimate the contact angle between water and the
fibre. Reprinted from (29). Copyright (2006), with permission from Elsevier.
                                     Polyelectrolyte Multilayers for Fibre Engineering   127

                                                             Non-treated




                     Treated



Figure 5.2 ESEM images of sections of an untreated bleached chemical softwood fibre and a
section of the same fibre treated with 5.5 bilayers of PAH/PAA. Reprinted from (29). Copyright
(2006), with permission from Elsevier.

preceded by model experiments by using Stagnation Point Adsorption Reflectometry
(SPAR) (31, 32) where the adsorbed amount in g/m2 can be estimated and by Quartz
Crystal Microbalance (QCM) (19, 33) measurements where the adsorption in g/m2 can
be determined as a combination of the adsorbed amount of polymer and the amount of
water immobilized by the adsorbed layer. The QCM-D technique also makes it possible
to determine the energy dissipation into the system when the driving current to the quartz
crystal is stopped. A high energy dissipation indicates a layer with a low viscosity and
a low elastic modulus (33), and this information is important for the interpretation of
the influence of the multilayers on the adhesion between treated surfaces.
   In both the SPAR and QCM experiments, we have chosen to work with SiO2 surfaces
since these can be prepared in a well-defined way and, as earlier shown (31), these
surfaces can serve as an acceptable model surface for bleached chemical fibres whose
interaction with oppositely charged polyelectrolytes is dominated by charge interactions.
It is thus possible with these model experiments to estimate how the formation of PEM
on fibres depends on the charge density of the surface/polyelectrolyte and also on the
electrolyte concentration in the solution. It is also possible to derive molecular infor-
mation from the model experiments that is difficult, if not impossible, to obtain from
adsorption measurements with fibres. Naturally the extrapolation from SiO2 surfaces
to fibres must be treated with care and with proper consideration of what is known in
the adsorption literature. The ideal situation would be to use model cellulose surfaces
in both these type of model experiments. Developments in our laboratory have shown
that this will be possible provided that the swelling and deswelling of the cellulose
layers upon adsorption can be quantified in a proper way to quantify the adsorption of
polyelectrolytes.
   An example of the usefulness of combining SPAR and QCM-D measurements is
shown in Figure 5.3 (19) where the adsorption of PAH and PAA on SiO2 surfaces is
shown.
   The SPAR data shown in Figure 5.3a indicate that there is a steady build-up of PEM
for each consecutive layer whereas for the PAA there seems to be a small decrease in
the signal after it passes through a maximum for each PAA layer. This can both be due
to desorption of polyelectrolyte or due to a restructuring of the polyelectrolytes in the
128            The Nanoscience and Technology of Renewable Biomaterials



              1.6

              1.4

              1.2
                          PAH       PAA        PAH
                1
 DS/S0 × 10




              0.8

              0.6
                                      Rinse
              0.4

              0.2

                0
                    0              1000              2000                3000           4000           5000
                                                             Time (s)

                                                                  (a)

               50

                                Rinse                                   PAH pH 7.5/PAA pH 7.5

                 0



               −50
  Df (Hz)




              −100       PAH    PAA     PAH



              −150



              −200
                     0          1000          2000      3000             4000       5000        6000
                                                       Time (s)
                                                            (b)

Figure 5.3 Typical results from SPAR and QCM-D measurements as functions of time
with combinations of PAH/PAA low molecular mass polyelectrolytes (15,000/5000) to SiO2
surfaces at a pH = 7.5 for both polyelectrolytes. (a) SPAR data corresponding to the solid
amount of adsorbed polyelectrolyte; (b) QCM-D frequency data corresponding to the adsorbed
amount including immobilised liquid; and (c) QCM-D dissipation data corresponding to the
visco-elastic properties of the adsorbed layers. Data from (19).
                                                 Polyelectrolyte Multilayers for Fibre Engineering   129

                  4.5
                              PAH pH 7.5/PAA pH 7.5

                  3.5


                            PAH   PAA     PAH
                  2.5
      DD × 106




                  1.5             Rinse



                  0.5



                 −0.5
                        0         1000          2000       3000        4000         5000        6000
                                                         Time (s)
                                                            (c)

                                            Figure 5.3 (continued).

adsorbed layers which changes the polarisability of the adsorbed layer. As can be seen
in Figure 5.3b, this decrease cannot be detected in the frequency spectra, indicating that
no desorption is taking place during the adsorption of the PAA in the 2nd, 4th layer
etc. On the other hand, the data in Figure 5.3c show that there is a large decrease in
the dissipation as the PAA is adsorbed. The data in Figures 5.3a–c thus indicate that the
structure of the PEM is highly dependent on the type of polyelectrolyte adsorbed in the
outermost layer of the PEM. If the PAH is in the outermost layer, the PEM has a lower
viscosity and a lower shear modulus than if the PAA is adsorbed in the outermost layer.
As will be shown later, this has a large influence on the effect of the PEM. The exact
reason to this change in properties of the PEM depending of which polyelectrolyte that
is adsorbed in the outermost layer is yet not known.


5.3         Formation of PEM with Different Polyelectrolytes and the Properties
            of the Layers Formed

As mentioned in the earlier section, the adsorption on silicon oxide using Stagnation Point
Adsorption Reflectometry (SPAR) has been utilised in order to establish that polyelec-
trolyte multilayers are formed from PDADMAC/PSS, PAH/PAA and PAH/PEDOT:PSS
and to predict that PEMs are also formed from these polyelectrolytes on wood fibres.
   Figure 5.4 shows the formation of PEMs from the strong polyelectrolyte pair
PDADMAC/PSS on silicon oxide, where the adsorption was conducted at different
concentrations of NaCl (29). Figure 5.4a shows the stepwise increase in adsorption as
polylelectrolytes are consecutively added and a polyelectrolyte multilayer is formed,
and figure 5.4b shows the corresponding saturation signals for these experiments. From
this it was obvious that a larger amount was adsorbed when the NaCl concentration
130   The Nanoscience and Technology of Renewable Biomaterials




                0.5


                0.4


                0.3
        DS/S0




                0.2                                                                no salt
                                                                                   0.01 M NaCl
                                                                                   0.05 M NaCl
                0.1
                                                                                   0.1 M NaCl

                0.0
                      0   2000    4000       6000     8000 10,000 12,000 14,000 16,000
                                                    Time (s)
                                                      (a)



                0.4               0.1 M NaCl, LMw
                                  0.1 M NaCl, HMw
                                  0.05 M NaCl, HMw
                0.3               0.01 M NaCl, HMw
                                  without addition, HMw
       DS/S0




                0.2


                0.1



                0.0


                      0   1   2    3     4     5      6         7   8   9   10 11 12 13
                                              Number of layers
                                                          (b)

Figure 5.4 Reflectometry data showing the relative change in the reflected signal ( S/S0 )
when a PEM was stepwise adsorbed onto SiO2 by consecutively adding PDADMAC and
PSS of molecular weight 30 k/80 k (0.1 M NaCl) and >500 k/1000 k (at 0, 0.01. 0.5
and 0.1 M NaCl. Figure 5.4a shows the signal plotted versus time for the high molecular
PDADMAC/PSS and Figure 5.4b shows the saturation signal for each adsorbed layer, when
the added polyelectrolyte had been adsorbed for 30 s. The measurements were conducted
without any adjustment of the pH which was determined to 5.5-6. Data from (27, 35).
                                         Polyelectrolyte Multilayers for Fibre Engineering   131

was increased. This was expected since previous investigations (34) have shown that
the adsorbed amount usually is increased when the salt concentration is increased. This
is probably due to the adsorption of the polymers with a larger number of loops and
tails due to the higher electrostatic screening when the salt concentration is increased.
   The formation of PEMs from the two weak polyelectrolytes PAH and PAA is highly
sensitive to the pH. By using different combinations of pH during the formation, PEMs
                                             ˚
with thicknesses varying from 10 to 120 A have been formed (16). A large difference
between PEM formation on silicon oxide has also been shown for PAH/PAA at pH
7.5/7.5, 5/5 and 7.5/3.5 (19), the adsorption being greater at pH 7.5/3.5 (Figure 5.5).
Figure 5.6 shows that there is also a significant increase in signal when the PEM is
formed from a combination of high molecular mass PAH/PAA.
   According to the standard procedure in our group for the evaluation of PEM formation,
the PEMs from low molecular mass PAH/PAA (19) and high molecular mass PDAD-
MAC/PSS (35) were also studied with the aid of Quartz Crystal Microbalance with
dissipation(QCM-D), which makes it possible to measure both the frequency change
and simultaneously the energy dissipation in the adsorbed layer, which is (as mentioned
earlier) a measure of the rigidity of the film.
   The dissipation values for PEMs of PAH/PAA adsorbed at pH 7.5/3.5 were signifi-
cantly higher than those for PDADMAC/PSS. When 7 layers of PDADMAC/PSS were
adsorbed, there was an increase of about 1.5 units compared to about 4 units with PAH/
PAA, indicating that PAH/PAA forms layers with a lower shear elasticity and lower vis-
cosity. For both systems there was also a difference in dissipation depending on which
polymer was adsorbed in the outermost layer (see also Figure 5.3), i.e. a higher dissi-
pation was also found when the cationic polyelectrolyte was adsorbed in the outermost


                       0.5


                       0.4

                                                                         pH 7.5/7.5
                       0.3
                                                                         pH 7.5/3.5
               DS/S0




                                                                         pH 5/5
                       0.2


                       0.1


                       0.0

                             0   1   2      3      4      5      6   7     8     9
                                                Number of layers

Figure 5.5 Reflectometry data showing the relative change in the reflected signal ( S/S0 )
when a SiO2 surface was consecutively treated with PAH/PAA (15,000/5000) at a background
electrolyte concentration of 0.01 M NaCl. The PEMs were formed at pH 7.5/7.5, 7.5/3.5 and
5.0/5.0 respectively and the saturation signal for each adsorbed layer was recorded when the
added polyelctrolyte had been adsorbed for 30 s. Data from (27).
132   The Nanoscience and Technology of Renewable Biomaterials

                  0.5


                  0.4
                                                                    15 k/5 k
                                                                    70 k/240 k
                  0.3
                                                                    150 k/750 k
          DS/S0



                  0.2


                  0.1


                  0.0

                        0   1   2       3       4      5      6
                                    Number of layers

Figure 5.6 Reflectometry data showing the relative change in the reflected signal ( S/S0 )
when PEMs were formed from PAH/PAA of molecular mass 15 k/5 k, 70 k/240 k and 150 k/
750 k adsorbed at pH 7.5/3.5. The PEMs were stepwise adsorbed onto SiO2 by consecutively
adding the polyelectrolytes at a background electrolyte concentration of 0.01 M NaCl. The
saturation signal for each adsorbed layer was recorded when the added polyelctrolyte had
been adsorbed for 30 s. Data from (27, 35).


layer for the PDADMAC/PSS system. This indicates a change in structure of the PEM
depending on which polymer was adsorbed in the outermost layer; the structure being
more rigid when the anionic polyelectrolyte was outermost than when PDADMAC was
in the external layer.


5.4   Formation of PEM on Fibres

The adsorption of PEMs on wood fibres can be shown using different techniques
as discussed earlier. For PEMs formed from PDADMAC/PSS, PAH/PAA and PAH/
PEDOT:PSS, the adsorption has been studied by destructive elemental analysis of
nitrogen (PDADMAC and PAH) (27, 29, 36) and sulphur (PSS) (29). showing that
there was an increase in the adsorbed amount with increasing number of adsorbed
layers (Figure 5.7). This clearly shows that PEMs are indeed formed on wood fibres
using these polymer systems.
   In the case of PAH/PAA, these experiments have shown that there was a higher amount
adsorbed in the first layer, and also that this trend was more significant when PEMs were
formed from low molecular mass polymers. This is most probably due to a somewhat
higher degree of interpenetration of polymers into the external parts of the fibre walls
                                                                                   a
in the case of low molecular mass polyelectrolytes, as has also been shown by Gim˚ ker
et al. (37), using fluorescently labelled PAH of different molecular weights.
   The influence of salt concentration on the formation of PEM on wood fibres was also
investigated with the high molecular mass combination of PDADMAC/PSS. As can be
                                                               Polyelectrolyte Multilayers for Fibre Engineering   133


                                      35

            Adsorbed amount (mg/g)    30

                                      25

                                      20

                                      15

                                      10
                                                                                          PAH/PAA 70 k/240 k
                                       5                                                  PAH/PAA 15 k/5 k

                                       0
                                               1       2           3        4        5        6       7
                                                                   Number of layers
                                                                           (a)


                                      14
             Adsorbed amount (mg/g)




                                      12


                                      10


                                       8


                                       6                                                          PDADMAC >500 k
                                                                                                  PDADMAC 30 k
                                       4

                                           0   1   2       3   4       5   6     7    8   9       10 11 12
                                                                   Number of layers
                                                                           (b)

Figure 5.7 Adsorbed amount of PDADMAC (7b, 30 k and >500 k) and PAH (7a, 15 k and
70 k) on wood fibres, determined using nitrogen analysis (ANTEK) and plotted as a function
of the number of layers. The fibres were treated with a background electrolyte concentration
of 0.1 M NaCl for PDADMAC and in 0.01 M NaCl for PAH. Data from (27, 29, 35).


seen in Figure 5.8, a higher electrolyte concentration leads to a higher adsorption. This
is probably due to a more extensive adsorption of PDADMAC and PSS into the porous
fibres due to a coiling of the polyelectrolytes at higher salt concentrations. These results
are also in accordance with earlier published data (38) where it was shown that the
adsorption of PDADMAC increases significantly as the salt concentration is increased.
This was explained as being due both to a penetration of the polyelectrolyte into the fibre
wall, making a larger surface area available for the polyelectrolyte, and to an increased
134   The Nanoscience and Technology of Renewable Biomaterials


                                             14           PDADMAC, without addition
                                                          PDADMAC, 0.01 M NaCl




                    Adsorbed amount (mg/g)
                                             12           PDADMAC, 0.05 M NaCl
                                                          PDADMAC, 0.1 M NaCl
                                                          PSS, 0.1 M NaCl
                                             10

                                              8

                                              6

                                              4

                                              2

                                                  0   1   2    3    4   5   6   7     8   9   10 11
                                                                   Number of layers

Figure 5.8 Amount of PDADMAC and PSS adsorbed per gram of fibres, determined by
nitrogen and sulphur analysis, respectively. The adsorption measurements were conducted
without any extra addition of NaCl and with the addition of 0.01, 0.05, and 0.1 M NaCl. The
fibres analysed with respect to adsorbed PSS were treated with 0.1 M NaCl. Data from (29).


overcharging of the fibre surface due to the formation of large tails and loops of the
polyelectrolyte on the fibre surface. This increased adsorption in the initial layer then
propagates throughout the consecutive layers of the PEM.
   Polyelectrolyte titration has also been used in order to study the formation of PEMs
from high and low molecular mass PDADMAC/PSS, and the results were similar to
those obtained using elementare analysis. In the case of one layer of PDADMAC, it
has also been shown that an increase in adsorption time from 10 to 30 minutes did not
result in any increase in the adsorbed amount.
   Having adsorption data both for fibres and for SiO2 surfaces with the same polyelec-
trolyte systems, it is possible to compare the adsorption on wood fibres to the adsorption
on silicon oxide as the substrate. To be able to make this comparison, the S/S0 sig-
nal from SPAR was converted to the amount of adsorbed polymer, the adsorption in
equivalents/m2 onto fibres being calculated assuming a specific surface area of the fibres
of 1.37 m2 /g (39). This value is probably low since it corresponds to the external
part of the fibre and the PDADMAC molecules probably have access to a somewhat
larger surface area. Figure 5.9 shows the adsorbed amount of charges of high molecular
PDADMAC/PSS on both SiO2 and wood fibres. In contrast to what was found when
the amount adsorbed on wood fibres was determined, the amount adsorbed on SiO2
was not higher in the first layer than in the subsequent layers. This difference may be
explained by the macroscopic structure of the wood fibre, and by the different charges
of the respective surfaces. However, the trends observed for the two surfaces are very
similar and, considering the low specific surface area assumed for the fibres, there is
also a fairly good agreement between the absolute values achieved with the different
methods.
   As has also been shown, the amount of polyelectrolytes adsorbed onto wood fibres did
not show any deviation from linearity with the number of layers as the salt concentration
was increased. The reason for the difference between wood fibres and SiO2 surfaces in
                                                       Polyelectrolyte Multilayers for Fibre Engineering                             135

                                          90                                                         30

                                          80
                                                                                                     25
   Adsorbed amount of charges on fibres




                                                                                                          Adsorbed amount of charges on SiO2
                                          70

                                          60                                                         20
               (µekv/m2)




                                                                                                                     (µekv/m2)
                                          50
                                                                                                     15
                                          40

                                          30                                                         10
                                                                            Fibre
                                          20
                                                                            Silicon oxide            5
                                          10

                                           0                                                         0
                                               0   2   4             6              8           10
                                                       Number of layers

Figure 5.9 Amount of adsorbed polyelectrolyte charges per square meter of substrate during
PEM formation from high molecular mass PDADMAC/PSS on wood fibres and on SiO2 .
The background electrolyte concentration was 0.1 M NaCl and a specific surface area of
1.37 m2 /g, for the fibres, was used to recalculate the data to equivalents/m2 (39). The
measurements were conducted without any adjustment of the pH which was determined to
be 5.5-6. Data from (29).


this respect is not obvious, and the conclusion is that the data for multilayer formation
on flat surfaces must be translated to porous wood fibres with care, especially at high
salt concentration (40) when the porous structure of the fibres has a larger importance
due to the extensive coiling of polyelectrolytes. The contributions from nonelectrostatic
interactions are different for cellulose and SiO2 , and these also have an influence on the
adsorbed amount.
   Despite these differences (Figure 5.9), it must be concluded that fully bleached chem-
ical softwood fibres and SiO2 show very similar trends in terms of PEM formation, and
that SiO2 can be used as a convenient model surface for screening measurements to
predict PEM formation on this type of wood fibres.
   The formation of the PEM on fibres was also followed by single fibre measurements
using the Cahn balance set-up shown in Figure 5.1. With this technique it is possible to
determine the wetting force created by the interaction between the fibres and the liquid
and this value can then be used, if the fibre perimeter is known to calculate a contact
angle between the liquid and the fibre. Typical results from this type of measurement are
shown in Figure 5.10, recently published by Lingstrom et al. (29), where the force-traces
for an individual fibre coated with three and four individual layers of high molecular
mass PDADMAC/PSS, treated to a depth of 0.8 mm and washed to a depth of 1.2 mm,
are shown. It can be seen that the fibre showed a certain difference in advancing wetting
136   The Nanoscience and Technology of Renewable Biomaterials

                             10

                              8

                              6




                    F (µN)
                              4

                              2

                              0                                       layer 4
                                                                      layer 3
                             −2

                              0.0   0.2   0.4   0.6    0.8      1.0   1.2
                                          Fibre distance (mm)

Figure 5.10 Force traces from the washing step of a fibre treated with 3 and 4 layers of high
molecular mass PDADMAC/PSS to a depth of 0.8 mm and then washed to a depth of 1.2 mm.
The lower curves show the force trace when the fibre was immersed (advancing) and the
upper curves show the force trace when the fibre was withdrawn (receding). The adsorption
was carried out with a background electrolyte concentration of 0.1 M NaCl without any
further pH adjustment (pH 5.5–6). Data from (29).


force depending on the polymer adsorbed in the outermost layer, with a lower force, i.e.
a lower wettability, when PDADMAC was adsorbed than when PSS was adsorbed in
the outermost layer. A similar trend can be seen in Figure 5.11 which shows the force
curve for a similar analysis of an individual fibre treated with 8 and 9 layers of PAH and
PAH (41), adsorbed at pH 5. This fibre was treated to a depth of 1.0 mm and washed
to a depth of 1.7 mm.
   In order to obtain at least a semi-quantitative estimate of the influence of the polymer
in the outer layer on the wettability of the PEM, the advancing contact angles of the
films were calculated from the Cahn balance measurements. Assuming the receding
contact angle to be 0◦ , the perimeter of each fibre was calculated over certain regions of
the treated part of the fibre. The validity of this assumption was supported by previous
Dynamic contact angle (DCA) measurements of single, nontreated, wood fibres (42–45)
showing that the receding contact angle is close to 0◦ , indicating complete wetting of the
fibres. These perimeter-values were then used when the average values of the advancing
contact angle were calculated.
   Contact angles for individual fibres treated by high molecular mass (70 k/240 k)
(35) and low molecular mass (15 k/8 k) PAH/PAA (41) using different pH strate-
gies are presented in Figure 5.12. Figure 5.13 shows corresponding results for high
molecular mass PDADMAC/PSS (29) and low molecular mass PDADMAC/PSS (46).
Figure 5.14 shows the advancing contact angle data for PEO/PAA (41), PAH/CMC (46)
and PAH/PEDOT:PSS (36). All, except PEO/PAA, demonstrated a difference in contact
angle depending on the polymer adsorbed in the outermost layer, the contact angle being
higher when the individual fibre was capped with the cationic polymer. In the case of
PAH/PAA, there was also a difference depending on the pH strategy. Comparing the
data for the LMw combination of PAH/PAA at pH 7.5/7.5 and at pH 7.5/3.5, it was found
                                                                      Polyelectrolyte Multilayers for Fibre Engineering   137

                                                  8
                                                  7
                                                  6
                                                  5
                                         F (µN)   4
                                                  3
                                                  2
                                                                                                   Layer 9 (PAH)
                                                  1                                                Layer 8 (PAA)
                                                  0
                                                   0,0    0,2   0,4       0,6   0,8   1,0   1,2   1,4   1,6
                                                                      Fibre distance (mm)

Figure 5.11 Force traces from the washing step of a fibre treated with 8 and 9 layers of low
molecular mass PAH/PAA adsorbed at pH 5, with a background electrolyte concentration of
0.01 M NaCl, and washed under the same conditions. The lower curves show the force trace
when the fibre was immersed (advancing), and the upper curves show the force trace when
the fibre was withdrawn (receding). The fibre was treated with PEM to a depth of 0.9 mm and
washed to a depth of 1.7 mm. Data from (41).



                                       110
                                       100
             Advancing contact angle




                                        90
                                        80
                                        70
                                        60
                                        50                                                              pH 7.5/3.5 HMw
                                                                                                        pH 7.5/3.5 LMw
                                        40
                                                                                                        pH 5/5 LMw
                                        30                                                              pH 7.5/7.5 LMw
                                        20
                                                      0     2         4         6      8      10        12
                                                                 Number of layers

Figure 5.12 The advancing contact angle as a function of the number of layers on an
individual fibre treated with PAH/PAA (15000/8000) (LMw ) at pH 5, 7.5/3.5, and 7.5/7.5
and with PAH/PAA (70000/240000 (HMw ) at pH 7.5/3.5. The adsorption was carried out at
0.01 M NaCl. Data from (35, 41).


that there was a significantly larger difference in advancing contact angle depending on
the polyelectrolyte in the most external layer for the pH 7.5/3.5 adsorption strategy than
for the pH 7.5/7.5 strategy.
   Since it is well known that the contact angle is influenced by the first nm of a polymer
film, the advancing contact angle can be used to estimate the difference in structure of
the PEMs formed. Small differences in contact angle may indicate very thin individual
138   The Nanoscience and Technology of Renewable Biomaterials



                                                             90
                                                             80

                                  Advancing contact angle
                                                             70
                                                             60
                                                             50
                                                             40
                                                             30
                                                             20
                                                                                                           PDADMAC/PSS, 30 k/80 k
                                                             10                                            PDADMAC/PSS, >500 k/1 ,000 k
                                                                 0
                                                                     −2       0   2       4      6    8    10    12   14       16
                                                                                              Number of layers

Figure 5.13 The advancing contact angle as a function of the number of layers on individual
fibres treated with high molecular mass PDADMAC/PSS and low molecular mass PDADMAC/
PSS. The adsorption was carried out with a background electrolyte concentration of 0.1M
NaCl without further adjustment of pH (pH 5.5–6). Data from (29, 35).




                                           120
                                           110
                                           100
        Advancing contact angle




                                                            90
                                                            80
                                                            70
                                                            60
                                                            50
                                                            40
                                                                                                                                    PEO/PAA
                                                            30
                                                                                                                                    PAH/PDOT:PSS
                                                            20
                                                                                                                                    PAH/CMC
                                                            10
                                                                          0           2             4         6            8           10
                                                                                                  Number of layers

Figure 5.14 The advancing contact angle as a function of the number of layers on an individ-
ual fibres treated with PEO/PAA, PAH/PDOT:PSS and PAH/CMC respectively. PAH/PDOT:PSS
and PAH/CMC were both adsorbed using the pH7.5/3.5 strategy. PEO/PAA was adsorbed at
pH 2.2 with a first layer of PAH adsorbed at pH 7. The experiment was conducted at 0.01 M
NaCl. Data from (36, 41, 46).
                                                              Polyelectrolyte Multilayers for Fibre Engineering                         139

layers and/or a high degree of interpenetration of polymer chains between the different
layers. The results for the PEMs formed from low molecular PAH/PAA at pH 7.5/3.5
and pH 5/5 indicate that the layer was thicker and better defined when the PEMs were
formed at pH 7.5/7.5.
   A difference in contact angle was also detected for high molecular mass PDADMAC/
PSS, PAH/PEDOT:PSS and PAH/CMC. In comparison, PAH/CMC showed the largest
difference on the wettability depending on the polyelectrolyte in the external layer. When
5–9 layers were adsorbed, the advancing contact angle was calculated to be 110–115◦
when PAH was in the outermost layer, and about 40◦ when the PEM was capped by CMC.
   The reasons behind the difference in wettability for different polyelectrolytes is still
under debate, but it has recently been shown that it is not the cationic or anionic nature
of the polyelectrolyte that determines the wettability of a polyelectrolyte, but rather the
intrinsic hydrophobicity (47) of the polymer chain.



5.5   Influence of PEM on Properties of Fibre Networks

Sheets made from cellulose fibres carrying a PEM have, for different polymer systems,
shown a great increase in paper strength compared to sheets made from nontreated
fibres (Figures 5.15 and 5.16) (27, 35). In the case of sheets made from fibres treated
with low molecular mass PAH/PAA (27) a tensile index of about 55 kNm/kg was found,
compared with a value of about 20 kNm/kg for sheets made from nontreated fibres. With
this polymer system, the increase was larger for the low molecular mass combination.
In the case of sheets made from fibres treated with highly charged PDADMAC/PSS
(29), the increase was lower, and in contrast to PAH/PAA, the improvement was most
significant for the high molecular mass combination of polyelectrolytes; about 80% when


                                                      Strain at break, PAH/PAA, 15 k/5 k
                                                      Strain at break, PAH/PAA, 70 k/240 k                    6
                                         55
                                                                                                                  Strain at break (%)




                                         50
                Tensile index (kNm/kg)




                                         45
                                         40                                                                   4

                                         35
                                         30
                                         25
                                                                                                              2
                                                                     Tensile index, PAH/PAA, 15 k/5 k
                                         20                          Tensile index, PAH/PAA, 70 k/240 k

                                         15
                                              0   1      2      3  4     5     6             7     8      9
                                                                Number of layers

Figure 5.15 Tensile index and strain at break of sheets made from fibres treated with PEMs
formed from PAH 70 k and PAA 240 k (35) and from PAH 15 k and PAA 5 k (27), presented
as functions of the number of layers in the PEM.
140   The Nanoscience and Technology of Renewable Biomaterials

                                         40           Tensile index, PDADMAC/PSS, 30 k/80 k               8
                                                      Tensile index, PDADMAC/PSS, >500 k/1000 k
                                         35                                                               7




                Tensile index (kNm/kg)
                                         30                                                               6




                                                                                                              Strain at break (%)
                                         25                                                               5

                                         20                                                               4

                                         15                                                               3

                                         10                                                               2

                                         5                  Strain at break, PDADMAC/PSS, >500 k/1000 k   1
                                                            Strain at break, PDADMAC/PSS, 30 k/80 k
                                         0                                                                0
                                              0   1     2    3    4 5 6 7 8                 9     10 11 12
                                                                  Number of layers

Figure 5.16 Tensile index and strain at break for sheets made from fibres coated with PEM of
PDADMAC 30 k and PSS 80 k, presented as a function of the number of adsorbed layers. The
data for high molecular mass PDADMAC (>500 k) and PSS (1000 k), taken from Lingstrom
et al. (29), are included in the figure for comparison.


11 layers were adsorbed, but significantly lower, 25%, for the same number of layers
of the low molecular mass combination (35). A significant result was also that both
systems showed a larger improvement in paper tensile index when the cationic polymer
was in the outermost layer.
   The difference in strength-enhancing efficiency between the different polyelectrolyte
combinations and between the conditions during preparation of the PEM can naturally
be linked to the amount of polyelectrolyte adsorbed and to the properties of the adsorbed
layers. The effect on the strength-enhancing efficiency of the charge of the polyelec-
trolyte used in the most external layer must be linked to the properties of the adsorbed
layers since the amount of polyelectrolyte is increased also when the anionic polyelec-
trolyte is added. It is not, however, possible to elucidate the molecular mechanism solely
from the evaluation of fibre network properties, and this is discussed further in the next
section.
   The adsorbed amount will naturally have a considerable influence on the strength-
enhancing efficiency of the various chemical combinations, and in Figure 5.17 different
chemical systems are compared as a function of the adsorbed amount of chemicals (48).
   Efficiency, on the basis of the adsorbed amount, is clearly highest for the combina-
tion of amylose-rich cationic starch and anionic starch and efficiency is lowest for the
combination of cationic potato starch, cationic amylopectin starch and anionic starch.
It is also obvious that the differences between the systems are rather large for a given
adsorbed amount. This indicates either that the adhesive properties of the layers are
significantly different or that the ways in which the chemicals affect the joint-forming
mechanism between the fibres in the fibre network are very different. The results show
considerable promise for the PEM technology for fibre and sheet engineering, and they
also indicate that the selection of chemicals and how they are added are of paramount
importance if the changes desired with the added chemicals are to be achieved.
                                                           Polyelectrolyte Multilayers for Fibre Engineering   141

                                                  70

                                                  60




                           Tensile index (Nm/g)
                                                  50

                                                  40

                                                  30

                                                  20

                                                  10

                                                   0
                                                       0      20     40     60       80
                                                           Adsorbed amount (mg/g)

Figure 5.17 Improvement in tensile index for sheets made from PEM-treated fibres with
different chemical combinations. Filled triangles show low molecular mass PAH/PAA with a
7.5/3.5 pH strategy, unfilled triangles show anionic potato starch and cationic potato starch
with a D.S. (Degree of Substitution) of 0.06, unfilled squares show anionic potato starch and
cationic amylose-rich starch from potato both with a D.S. = 0.06 and unfilled circles show
anionic potato starch and cationic amylopectin starch from potato both with a D.S. of 0.06.
Fully bleached, unbeaten chemical softwood fibres were used in all the experiments and all
the sheets were prepared with a background NaCl concentration of 0.01 M (48).


5.6   Influence of PEM on Adhesion between Surfaces

As was indicated earlier, based on long-term experience with fibre networks, it is almost
impossible to identify the molecular mechanisms responsible for the differences between
different chemical systems simply by testing fibre network properties. To identify
the molecular reasons for the detected differences, more clear-cut model experiments
are needed where the methods used have a sensitivity and geometrical resolution that
permit differentiation between different mechanisms. This has been done in earlier,
nonfibre-related investigations regarding the formation of adhesive joints between poly-
mer surfaces (49). In this investigation (49), it was shown that a diblock co-polymer
A-B added to the interface of blocks of A and B can dramatically improve the adhesion
between the polymers A and B via a mechanical entanglement (49) of molecular chains
on the two sides of the interface by the co-polymer. The number of chains interacting,
as well as the length of the chains, influences the fracture toughness of the joint formed
between the polymers. Translated into the interaction between PEMs formed on fibres,
this suggests that the number of chains, and the length of the chains, indirectly deter-
mined by the molecular mass (50) of the polymers at a certain adsorbed amount, may
control the adhesion between the layers. However, this hypothesis has to be tested in
model experiments before any further conclusions can be drawn.
   The QCM-D experiments, described earlier, are one type of model experiment
that can be used to test the hypothesis of molecular mobility. A PEM formed from
PAH/PAA (19) showed a higher dissipation measured by QCM than PEMs formed from
PDADMAC/PSS (46), indicating a less rigid structure with a larger number of chain
ends and loops in the PAH/PAA interacting to give a stronger adhesion. The dissipation
142   The Nanoscience and Technology of Renewable Biomaterials

for both systems was also higher when the cationic polymer was adsorbed outermost,
indicating a difference in the PEM structure depending on the polymer adsorbed in the
outermost layer. This is in agreement with the larger strength found in sheets made
from PEM-covered fibres when the PEM was capped by the cationic polyelectrolyte.
   To test this hypothesis further the adhesion between PEMs was directly measured in
model experiments with the AFM colloidal probe technique. In these experiments, the
pull-off force was directly measured for PEMs formed from low and high molecular
mass PAH/PAA (35, 51) (Figure 5.18), both adsorbed at pH 7.5. The adhesion results
show that there is an increase in adhesion between the layers with increasing number of
layers. It can also be concluded that the adhesion was higher when PAH was adsorbed
in the outermost layer and that the effect was more significant for the low molecular
mass than for the high molecular mass combination.
   Figure 5.18 shows that there was a significant increase in the adhesion when the
contact time at maximum load was increased from 0 to 5 s. This indicates that, if the
chains are given a longer time to diffuse across the interface, a stronger adhesion is
developed between the PEM covered-surfaces.
   Since it has been shown that the number of chain ends rather than the number of
loops is important to achieve a strong adhesion between surfaces (50), and since the
number of free chain ends decreases when the molecular mass of the interacting chains is
increased, higher molecular mass polymers would in general tend to give a less significant
contribution to the adhesion at least at short contact times, when the high molecular mass
polyelectrolytes will not have sufficient time to diffuse the required distance across the
PEM/PEM interface. Low molecular mass polymers are assumed to possess a higher
mobility and a higher rate of interpenetration than the high molecular fractions and it
can also be assumed that they contribute to a more significant improvement in adhesion.
For the very thin and rigid layers of PDADMAC/PSS, it is however reasonable that
the low molecular mass PDADMAC/PSS gives fewer and shorter interacting chains due
to the flat conformation of these polyelectrolytes, which explains the comparably small
improvement in paper strength compared to that obtained with the high molecular mass
combination. This process of chain mixing between two opposite surfaces carrying
PEMs is schematically shown in Figure 5.19.
   For individual fibre-fibre joints (52) of fibres treated with PAH/PAA adsorbed at pH
7.5/3.5, a study using light microscopy and specific staining of nonbonded areas showed
that the molecular degree of contact was increased from about 18% to 32% for fibres
treated with five layers compared to nontreated fibres. The data for layers 3–5 also
indicated that the degree of contact, in the fibre/fibre contact, was increased when PAH
was adsorbed in the outermost layer. This shows that a high contact area between the
fibres, in addition to a strong interaction due to entanglement between the layers, is very
important for the development of strong fibre-fibre joints. The results also clearly shows
the need for clear-cut model experiments in order to elucidate the strength-enhancing
mechanism of different additives.
   It has recently been shown that fibres having the lowest wettability, both for
PDADMAC/PSS and PAH/PAA, when the cationic polylelectrolyte is in the outermost
layer, also show the strongest adhesion. This seems contradictory to a recently
published hypothesis where it is suggested that a more hydrophilic (53) agent will
more efficiently improve the strength of papers made of treated fibres. However, the
                                                            Polyelectrolyte Multilayers for Fibre Engineering   143


                                       6

                                       5

               Pull-off force (mN/m)
                                       4

                                       3

                                       2

                                       1                                                       0s
                                                                                               1s
                                       0                                                       5s

                                           0   1   2   3   4 5 6 7 8 9 10 11 12
                                                           Number of layers
                                                                  (a)
                                       4


                                       3
               Pull-off force (nM/m)




                                       2


                                       1
                                                                                                0s
                                                                                                1s
                                       0                                                        5s

                                           0   1   2   3   4 5 6 7 8           9 10 11 12
                                                            Number of layers
                                                                (b)

Figure 5.18 Normalised pull-off force as a function of layer number and contact time at
maximum load for PEM covered silica surfaces. (a) Low molecular weight PAH and PAA
and (b) high molecular PAH/PAA were adsorbed at pH 7.5/7.5 in a background electrolyte
concentration of 0.01 M NaCl. Contact time at maximum load: 0 seconds ( ), 1 second ( ),
5 seconds ( ). Reprinted with permission from (51). Copyright (2009), American Chemical
Society.


formation of a strong fibre–fibre joint is a rather complex process, and to form strong
joints it is important that:
• efficient contacts are formed;
• the fibres are conformable (on the molecular and macroscopic levels) during water
  removal, when capillaries are created between the fibres;
• the fibres contain surface layers that allow a high degree of entanglement.
To form efficient joints between the fibres when they are totally immersed in water, the
fibres must have a high wet adhesion, and this is definitively determined by the wettability
of the fibres. A strong dry adhesion between the fibres in the sheet requires a strong
wet adhesion between the fibres when the joint is being formed. The work of adhesion
144    The Nanoscience and Technology of Renewable Biomaterials




Figure 5.19 Schematic representation of the probable mixing of polyelectrolytes across the
interface of two PEMs on adjacent surfaces. It is suggested that this intermixing is vital for the
formation of strong adhesive joints between the surfaces and that the molecular mobility and
the number of interacting chain ends are important for the development of strong adhesion at
short contact times.

between the substrates in water can be described by the following equation (54):
                                    Wsl = Wsv − 2γlv cos θ
where Wsl is the adhesion between two surfaces in water, Wsl is the adhesion between
two surfaces in vacuum, and γlv is the surface tension. This means that the adhesion
between two hydrophobic surfaces (i.e. with a contact angle >90◦ ) in water will be
greater than that between two surfaces that are more hydrophilic.
   Fibres in water are forced towards each other more strongly when the contact angle is
increased. A larger contact angle results in a better contact between the fibres, which is
important for the formation of strong, dry fibre–fibre joints. Considering the hypothesis
that the level of wettability is an important factor for creating strong adhesion between
the fibre, these results are in agreement with the results of individual fibre measurements
showing that fibres treated at pH 7.5/3.5 have a lower wettability than fibres treated
at pH 7.5/7.5. This hypothesis is also consistent with the results for high molecular
PDADMAC/PSS which showed a higher paper strength and a lower wettability when
PDADMAC was in the outermost layer.


5.7 Concluding Remarks

This chapter has focused on the formation of PEM on model surfaces and on fibres,
in order to describe how these layers can be used to improve the interaction between
                                     Polyelectrolyte Multilayers for Fibre Engineering   145

PEM-covered surfaces under wet conditions. A section has also been included which
examines the relation between the properties of the formed multilayers in model experi-
ments, the adhesion between wet PEMs studied using the AFM colloidal probe technique
and the properties of dry fibre/fibre joints, as well as the dry properties of papers pre-
pared from PEM-treated fibres. This shows that the PEM technology is a new and
versatile methodology for fibre engineering and that detailed information about the
properties of the PEM is essential in order to optimize the use of PEM for different
end purposes. Since the technique is based on treatment in aqueous solutions at neu-
tral pH, it is also a very gentle technique where the properties of the fibres will be
maintained.
   It should be stressed, however, that the work described here has so far been focused
mainly on the formation of the layers and joints between PEM-covered surfaces. Less
work has been devoted to the characterisation of the dry properties of PEM layers on
cellulose surfaces, apart from the determination of their electrical conducting properties
(30, 36). This is naturally a very important task, since it is simple to imagine that the
properties of dry PEM films will be very important for the preparation of strong, weak,
ductile etc. adhesive joints between fibre surfaces. It is therefore anticipated that a lot of
future work will be devoted to the dry characterisation of PEMs formed with different
components and under different conditions. This work is indeed currently in progress in
the laboratory of the authors.



Acknowledgements

              o
Rikard Lingstr¨ m thanks the Biofibre materials research centre (BiMaC) at KTH for
                                           a
financial support; Erik Johansson and Lars W˚ gberg thank the Biomime research centre
at KTH and Lyckeby Research Foundation for financial support.


References

1.   G. Decher, J.D. Hong, Buildup of ultrathin multilayer films by a self-assembly
     process. Makromol. Chem. Macromol. Symp. Vol. 46. 1991, Mainz. 321.
2.   R.K. Iler, Multilayers of colloidal particles, J. Colloid Interface Sci. 21 (1966) 569.
3.   G. Decher, Fuzzy Nanassemblies: Toward Layered Polymeric Multicomposites,
     Science 277 (1997) 1232.
4.   P. Bertrand, A. Jonas, A. Laschewsky, R. Legras, Ultrathin polymer coatings by
     complexation of polyelectrolytes at interfaces: suitable materials, structure and prop-
     erties, Macromol. Rapid. Commun. 21 (1999) 319.
5.                                 o
     F. Caruso, E. Donath, H. M¨ hvald, Influence of polyelectrolyte multilayer coatings
           o
     on F¨ rster resonance energy transfer between 6-carboxyfluorescin and rhodamine
     b-labeled particles in aqueous solution, J. Phys. Chem. B 102 (1998) 2011.
6.                                                     o
     F. Caruso, E. Lichtenfeld, E. Donath, H. M¨ hvald, Investigation of electrostatic
     interactions in polyelectrolyte multilayer films: binding of anionic fluorescent probes
     to layers assembled onto colloids, Macromolecules 32 (1999) 2317.
146     The Nanoscience and Technology of Renewable Biomaterials

7.    P.A. Neff, A. Naji, C. Ecker, R. Nickel, R. v Klitzing, A.R. Bausch, Electri-
      cal detection of self-assembled polyelectrolyte multilayers by a thin film resistor,
      Macromolecules 39 (2006) 463.
8.    R. v Klitzing, Internal structure of polyelectrolyte assemblies, Phys. Chem. Chem.
      Phys 8 (2006) 5012.
9.    A.V. Dobrynin, A. Deshkovski, M. Rubinstein, Adsorption of polyelectrolytes at
      oppositely charged surfaces, Macromolecules 34 (2001) 3421.
10.   M.A. Cohen Stuart, C.W. Hoogendam, A. de Keizer, Kinetics of polyelectrolyte
      adsorption, J. Phys.: Condens Matter 9 (1997) 7767.
11.   S.T. Dubas, J.B. Schlenoff, Factors controlling the growth of polyelectrolye multi-
      layers, Macromolecules 32 (1999) 8153.
12.              a
      M. Salom¨ ki, I.A. Vinikurov, J. Kankare, Effect of temperature on the buildup of
      polyelectrolyte multilayers, Langmuir 21 (2005) 11232.
13.   Z. Sui, D. Salloum, J.B. Schlenoff, Effect of molecular weight on the construction
      of polyelectrolyte multilayers: stripping versus sticking, Langmuir 19 (2003) 2491.
14.             a               a
      M. Salom¨ ki, P. Tervasm¨ ki, S. Areva, J. Kankare, The Hofmeister anion effect and
      the growth of polyelectrolyte multilayers, Langmuir 20 (2004) 3679.
15.   D. Yoo, S.S. Shiratori, M.F. Rubner, Controlling bilayer composition and surface
      wettability of sequentially adsorbed multilayers of weak polyelectrolytes, Macro-
      molecules 31 (1998) 4309.
16.   S.S. Shiratori, M.F. Rubner, pH-dependent thickness behaviour of sequentially
      adsorbed layers of weak polyelectrolytes, Macromolecules 33 (2000) 4213.
17.   J.B. Schlenoff, S.T. Dubas, Mechanism of polyelectrolyte multilayer growth: Charge
      overcompensation and distribution, Macromolecules 34 (2001) 592.
18.   K. Glinel, A. Moussa, A.M. Jonas, A. Laschewsky, Influence of polyelectrolyte
      charge density on the formation of mulilayers of strong polylelectrolytes at low
      ionic strength, Langmuir 18 (2002) 1408.
19.                                       a
      S.M. Notley, M. Eriksson, L. W˚ gberg, Visco-elastic and adhesive properties pf
      adsorbed polyelectrolyte multilayers determined in situ with QCM-D and AFM
      measuremensts, J. Colloid Interface Sci. 292 (2005) 29.
20.   Decher, G., Schlenoff, J.B. (eds), Multilayer Thin Films. John Wiley & Sons-VCH:
      New York/Weinhem, 2003, p. 524.
21.   M.L. Bruening, D.M. Dotzauer, P. Jain, L. Ouyang, G.L. Baker, Creation of func-
      tional membranes using polyelectrolyte multilayers and polymer brushes, Langmuir
      24 (2008) 7663.
22.                                   o
      F. Caruso, R.A. Caruso, H. M¨ hvald, Chem. Mater. 11 (1999).
23.   E. Donath, G.B. Sukhorukov, F. Caruso, D.A. Davies, H. Mohvald, Novel hallow
                                                                     ¨
      polymer shells by colloid-templated assembly of polyelectrolytes, Angew. Chem.
      Int. Ed. 37 (1998) 2201.
24.         a
      L. W˚ gberg, S. Forsberg, A. Johansson, P. Juntti, Engineering of fibre surface prop-
      erties by application of polyelectrolyte multilayer concept. Part I: Modification of
      paper strength, J. Pulp Paper Sci. 28 (2002) 222.
25.   Y. Lvov, G. Grozdits, Z. Zheng, L. Zonghuan, Layer-by-layer nanocoating mill
      broken fibres for improved paper, Nord. Pulp Pap Res. J. 21 (2006) 552.
                                    Polyelectrolyte Multilayers for Fibre Engineering   147

                       a
26. S. Forsberg, L. W˚ gberg (eds). Production of Particles or Fibres having a Coating of
    Polyelectrolytes Interacting with Each Other and Paper or Nonwoven Products with
    Improved Opacity Therefrom. 2000; SCA Hygiene Products AB: Sweden, p. 19 pp.
                                      a
27. M. Eriksson, S.M. Notley, L. W˚ gberg, The influence on paper strength properties
    when building multilayers on weak polyelectrolytes onto wood fibres, J. Colloid
    Interface Sci. 292 (2005) 38.
                                        a
28. M. Eriksson, G. Pettersson, L. W˚ gberg, Application of polymeric multilayers of
    starch onto wood fibres to enhance strength properties onto fibres, Nord. Pulp Pap
    Res. J. 20 (2005) 270.
                o         a
29. R. Lingstr¨ m, L. W˚ gberg, P.T. Larsson, Formation of polyelectrolyte multilayers
    on fibres: influence on wettability and fibre/fibre interaction, J. Colloid Interface
    Sci. 296 (2006) 396.
30. Z. Zheng, J. McDonald, R. Khillan, et al., J. of Nanoscience and Nanotechnology 6
    (2006) 1.
          a
31. L. W˚ gberg, I. Nygren, The use of stagnation point adsorption reflectometry to
    study molecular interactions relevant to papermaking chemistry, Colloids Surf . 159
    (1999) 3.
32. J.C. Dijt, M.A.C. Stuart, G.J. Fleer, Advances in Colloid and Interface Science 50
    (1994) 79.
         o¨
33. F. H¨ ok, B. Kasemo, C. Fant, K. Scott, H. Elwing, Anal. Chem 73 (2001) 5796.
34. J.B. Schlenoff, S.T. Dubas, Factors controlling the growth of polyelectrolyte multi-
    layers, Macromolecules 32 (1999) 8153.
                o         a
35. R. Lingstr¨ m, L. W˚ gberg, Polyelectrolyte multilayers on wood fibres: Influence
    of molecular weight on layer properties and mechanical properties of papers from
    treated fibres, J. Colloid Interface Sci. 328 (2008) 233.
                              o          a
36. I. Wistrand, R. Lingstr¨ m, L. W˚ gberg, Preparation of electrically conducting
    cellulose fibres utilizing polyelectrolyte multilayers of poly(3,4-ethylenedioxy-
    thiophene):poly(styrene sulphonate) and poly(allyl amine), European Polymer
    Journal 43 (2007) 4075.
             a                        a
37. M. Gim˚ ker, A. Horwath, L. W˚ gberg, Influence of polymeric additives on short-
    time creeo of paper, Nord. Pulp Pap Res. J. 22 (2007) 217.
                               o
38. A.E. Horvath, T. Lindstr¨ m, J. Laine, On the indirect polyelectrolyte titration of
    cellulosic fibres. conditions for charge stoichiometry and comparison with ESCA,
    Langmuir 22 (2006) 824.
         o                    o
39. G. S¨ derberg, T. Lindstr¨ m, Nord. Pulp Pap Res. J. 1 (1986) 26.
40. A.E. Horvath, in Fibre- and Polymer Technology. 2003, Royal Institute of Technol-
    ogy: Stockholm. p. 89.
                o                       a
41. R. Lingstr¨ m, S.M. Notley, L. W˚ gberg, Wettability changes in the formation of
    polymeric multilayers on cellulose fibre and their influence on wet adhesion, J. Col-
    loid Interface Sci. 314 (2007) 1.
42. J.H. Klungness, Measuring the wettwetting angle and perimeter of single wood fiber:
    A modified method, Tappi J 64 (1981) 65.
43. K.T. Hodgson, J.C. Berg, Dynamic wettability properties of single wood fibres and
    their relationship to absorbancy, Wood Fiber Sci. 20 (1988) 3.
44. J.J. Kreuger, K.T. Hodgson, Single-fiber wettability of highly sized pulp fibers,
    Tappi J 77 (1994) 83.
148   The Nanoscience and Technology of Renewable Biomaterials

45. W. a. f. s. Deng, M. Abazeri, Contact angle measurement of wood fibers in surfac-
    tant and polymer solutions, Wood Fibre Sci. 30 (1998) 155.
               o
46. R. Lingstr¨ m, in Pulp and Paper Technology. 2008, KTH: Stockholm. p. 71.
          a
47. K. H¨ nni-Ciunel, G.H. Findenegg, R. von Klitzing, Water contact angle on
    polyelectrolyte-coated surfaces: effects of film swelling and dropplet evaporation,
    Soft Materials 5 (2007) 61.
48. M. Eriksson, in Fibre and Polymer Technology. 2006, KTH: Stockholm. p. 81.
49. C. Creton, E.J. Cramer, C.-Y. Hui, B.R. Brown, Failure mechanisms of polymer
    interface reinforces with block copolymers, Macromolecules 25 (1992) 3075.
50. N. Chen, N. Maeda, M. Tirrel, J. Israelachvili, Adhesion and friction of polymer
    surfaces: The effect of chain ends, Macromolecules 38 (2005) 3491.
                                               o         a
51. E. Johansson, E. Blomberg, R. Lingstr¨ m, L. W˚ gberg, Adhesive interaction
    between polyelectrolyte multilayers of polyallylamine hydrochloride and polyacrylic
    acid studied using atomic force microscopy and surface force apparatus, Langmuir
    2009 (accepted).
                                            a
52. M. Eriksson, A.-S. Torgnysdotter, L. W˚ gberg, Surface modification of wood fibres
    using the polyelectrolyte multilayers technique: effects on fiber joint and paper
    strength properties, Ind. Eng. Chem. Res 45 (2006) 5279.
53. R.H. Pelton, Appita Journal 57 (2004) 181.
54. K. Kendall, Molecular Adhesion and its Applications – The Sticky Universe. Kluwer
    Academics/Plenum Publishers: New York, 2001. p. 108.
                                                        6
         Hemicelluloses at Interfaces:
       Some Aspects of the Interactions

                                                                ¨
                       Tekla Tammelin, Arja Paananen and Monika Osterberg



6.1     Overview

Hemicelluloses play an important role in papermaking. They can enhance paper strength
properties but on the other hand, as liberated in the white waters, they may bring harmful
side effects, e.g. interact with other papermaking chemicals and additives increasing
the consumption of these. At present there is also growing interest for the use of by
products of forest industry, for example, as a source for value added chemicals. Thus
hemicelluloses, as being an abundant plant material, can be considered as a large source
of renewable raw material for such purposes.
   The goal of this study was to investigate the interfacial behavior of the hemicelluloses
in order to enhance the understanding of the formation of hemicellulose films on cellulose
and how the film formation is affected by parameters such as ionic strength and hemi-
cellulose charge density. The adsorption of dissolved hemicellulose fractions isolated
from unbleached and peroxide bleached spruce thermomechanical pulp (TMP) as well as
pure galactoglucomannan (GGM), pure pectin and pure xylan on a Langmuir-Schaefer
cellulose film was studied using the quartz crystal microbalance with dissipation moni-
toring (QCM-D). The QCM-D data was further modeled using the Voigt-based model for
viscoelastic solids to estimate layer thicknesses and shear viscosity and shear elastic mod-
ulus of the adsorbed hemicellulose layers. These results were combined with colloidal
probe microscopy. Spruce hemicelluloses significantly adsorbed on cellulose forming
a uniform film whereas birch xylan seemed to form cluster like assemblies. Based on
the results the driving force for adsorption of different hemicelluloses on cellulose was
discussed.

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
150   The Nanoscience and Technology of Renewable Biomaterials

6.2   Introduction

Hemicelluloses are the second most abundant plant material after cellulose (Fengel and
                    e
Wegener 1989; Al´ n 2000). The amount of hemicelluloses of the dry weight of wood
                         o o
is usually 20–30% (Sj¨ str¨ m 1993). The most abundant hemicelluloses in trees are
xylans followed by glucomannans. In TMP process the hemicelluloses are dissolved
into the process waters during pulping and papermaking whereas the degradation of
hemicelluloses is one of the limiting factors in alkaline cooking. Hemicelluloses are
known to have positive effects on paper strength whereas it might harmfully consume
papermaking chemicals when dissolved in e.g. white waters.
   There is a growing interest in utilization of by-products of forest industry in order to,
for example, develop biodegradable polymers to replace synthetic oil-based ones. Hemi-
celluloses present in wood-processing industries process waters and side streams can be
considered as large sources for this renewable raw material. Although hemicelluloses
are industrially used, for example, as food additives and thickeners there are limitations
to fully exploit these resources due to the lack of isolation, purification and modification
techniques.
   The major soft wood hemicelluloses are galactoglucomannans (GGMs), arabinoglu-
curonoxylans and arabinogalactans. The dominating dissolved hemicelluloses in ther-
momechanical pulping of spruce are O-acetyl-galactoglucomannan and arabinogalactan
(Thornton et al. 1994). Arabino-4-O-methylglucuronoxylan is not notably released
during the mechanical pulping. Arabinogalactan and pectins (rhamnogalacturonans)
contribute to the anionic charge in the TMP waters. Peroxide bleaching at alkaline
conditions changes the composition of the dissolved hemicelluloses present in TMP pro-
cess waters: deacetylation of GGMs leads to their readsorption on the fiber surfaces.
Thus, the concentration of galactoglucomannans in the water phase is diminished. Con-
sequently, highly charged pectins are released which results in a higher anionic charge
of the waters (Holmbom et al. 1991; Thornton et al. 1994).
   In the papermaking process water dissolved hemicelluloses often give rise to various
problems like the growth of fungi and bacteria. Due to the anionic charge of these
polyelectrolytes they tend to interact with cationic papermaking chemicals, e.g. retention
                                                 a             ¨
aids, increasing the consumption of these (W˚ gberg and Odberg 1991). On the other
hand, hemicelluloses have been found to be beneficial for paper quality and, for example,
strength properties are claimed to be improved by adsorbing galactoglucomannans and
galactomannans on fiber surfaces (Hannuksela et al. 2004).
   The adsorption of dissolved hemicelluloses on extractive colloids originating from
spruce (Picea abies) and their stabilizing effect on colloids in TMP process waters has
been extensively studied by Johnsen et al. (2004); Sihvonen et al. (1998); Sundberg
et al. (1994a,b). The steric hindrance created around colloidal wood extractives by
dissolved hemicelluloses, especially GGMs, prevents the aggregation and accumulation
of colloids also at high ionic strengths in the presence of sodium and calcium salt
ions (Hannuksela and Holmbom 2004; Sundberg et al. 1996a). This enhanced sta-
bility of wood extractives reduces its tendency to form deposits (Otero et al. 2000).
Thus, the runnability of the papermachine is improved. GGMs have also been shown
to strongly adsorb on bleached kraft pulp fibers whose surfaces mainly consisted of
                       Hemicelluloses at Interfaces: Some Aspects of the Interactions   151

cellulose (Hannuksela et al. 2002). Adsorption increases with decreasing deacetylation
and decreasing amounts of galactose side groups.
   In our previous publications (Tammelin et al. 2007; Johnsen et al. 2006) we have
further clarified on a molecular level the interactions in the presence of hemicelluloses
using quartz crystal microbalance with dissipation monitoring (QCM-D). We were able to
show why the hemicelluloses derived from the waters of spruce consuming TMP process
can sterically stabilize extractive colloids. The hemicelluloses adsorbed extensively on
the extractives forming a layer with loops and tails pointing out to the solution phase
and thus they were able to form the steric hindrance around the colloids. The results
fully supported the conclusions drawn by Sundberg et al. (1994a; 1996a). These results
also showed that a few nanometers thin film of hemicelluloses significantly changed the
properties of cellulose and extractives surfaces.
   The major hardwood hemicellulose is glucuronoxylan. Xylan has been shown to
                                                o
have affinity towards fibers (Yllner and Enstr¨ m 1956) and at the end of pulp cooking
some dissolved xylan is readsorbed to fibers (Mitikka-Eklund 1996). Mora et al. (1986)
have concluded that xylan preferentially readsorbs to xylan rather than to cellulose, and
Henriksson and Gatenholm (2001) have suggested several types of association between
cellulose and xylan. In early studies by Marchessault et al. (1967) it has been shown that
the orientation of xylan molecules is parallel to the fiber axis and may hence affect the
mechanical properties of individual pulp fibers. That xylan molecules align themselves
on cellulose was later theoretically calculated by Kroon-Batenburg et al. (2002). Other
studies also propose that xylan on the fiber surfaces improves paper strength (Buchert
                 o                                                           ˚
et al. 1995; Sch¨ nberg et al. 2001). FTIR spectroscopy experiments by Akerholm and
      e
Salm´ n (2001) have indicated that xylan associates with mannan more closely than with
cellulose. Based on CP/MAS 13C NMR experiments Teleman et al. (2001) have con-
cluded that the supermolecular structure of xylan is highly dependent on the immediate
environment.
   Most studies dealing with cellulose-hemicellulose interactions have been bulk exper-
iments and there are only a few direct measurements of the interaction forces in the
presence of xylan. The forces between xylan-coated mica surfaces have been studied
                                                         ¨
by Neuman et al. (1993), Claesson et al. (1995) and Osterberg et al. (2001) using the
surface force apparatus (SFA, Israelachvili and Adams 1978). Mica is a highly anionic
mineral surface and the adsorption of xylan to mica was minor and mainly driven by
the low solubility of the xylan. In our previous publication (Paananen et al. 2003) the
adsorption of xylan to cellulose model surfaces and their effect on the forces between
cellulose surfaces was studied using the atomic force microscopy (AFM) colloidal probe
technique developed by Ducker et al. (1991). We found that xylan adsorbed onto
cellulose and steric forces dominated over double-layer forces (Paananen et al. 2003).
   The goal of this work was to link together our previous scattered studies (Tammelin
et al. 2006a; Paananen et al. 2003) dealing with hemicellulose interactions with differ-
ent fibrous components and to update the results conducted on cellulose with viscoelastic
modellings. The QCM-D instrument (Rodahl et al. 1995) was used to study the adsorp-
tion of dissolved hemicellulose fractions isolated from unbleached and peroxide-bleached
spruce thermomechanical pulp (TMP) as well as pure O-acetyl-galactoglucomannan,
pure pectin and pure xylan onto cellulose model surface. The AFM colloidal probe
152    The Nanoscience and Technology of Renewable Biomaterials

technique is used to study the effect of adsorbed xylan on the forces between cellulose
surfaces.
   The QCM-D data were modelled using the Voigt-based model for viscoelastic solid
(Voinova et al. 1999). In this way estimations for the layer thickness development
during the adsorption process could be attained and in addition, the shear viscosity and
shear elastic modulus of the formed hemicellulose layer could be estimated.
   By combining the results of the adsorption behavior of different hemicelluloses with
direct force measurements and a thorough investigation of the viscoelastic properties
of the adsorbed hemicellulose films the aim was to gain a deeper understanding of the
formation of the hemicellulose films on cellulose, its properties and structure as well
as how it affects the interactions present in the system. In addition, it was clarified
how parameters such as ionic strength and the source of the hemicellulose change the
properties of the adsorbed film and the interaction forces. We hope that the enhanced
understanding of the properties of different hemicellulose films facilitates the use of
hemicelluloses in novel applications.


6.3   Theoretical Basis for Interpreting QCM-D and AFM Data

6.3.1 QCM-D Data
QCM-D technique enables in-situ adsorption studies at solid/liquid interface (Rodahl
et al. 1995). Without adsorbate the piezoelectric quartz crystal oscillates at a resonant
frequency f0 which is lowered to f when material adsorbs on the surface of the crystal. If
the adsorbed mass is evenly distributed, rigidly attached, fully elastic and small compared
to the mass of the crystal, the shift in the resonant frequency is related to the adsorbed
                                     o¨
mass by the Sauerbrey equation (H¨ ok et al. 1998):
                                                  C f
                                        m=−                                             (6.1)
                                                    n
where m is the adsorbed mass per unit surface, f = f − f0 is the frequency shift, n
is the overtone number (n = 1, 3, 5, 7) and C is a constant that describes the sensitivity
of the device to changes in mass. For the crystals used, C ≈ 0.177 mg m−2 Hz−1
(Edvardsson et al. 2005).
   The resonant frequency of the crystal depends on the total oscillating mass, including
water coupled to the oscillation. By measuring several frequencies and the dissipation
it becomes possible to determine whether the adsorbed film is rigid or water-rich (soft)
which is not possible by looking only at the frequency response. If the adsorbed material
is not fully elastic, frictional losses occur that lead to a damping of the oscillation with
a decay rate of amplitude that depends on the viscoelastic properties of the material.
With the QCM-D instrument the change in the dissipation factor, D = D − Do , when
material is adsorbed can be measured. Do is the dissipation factor of the pure quartz
crystal immersed in the solvent and D is the dissipation factor when material has been
adsorbed. D is defined by
                                              Ediss
                                       D=                                               (6.2)
                                             2π Estor
                       Hemicelluloses at Interfaces: Some Aspects of the Interactions    153

where Ediss is the total dissipated energy during one oscillation cycle and Estor is the
total energy stored in the oscillation.

6.3.1.1 Interpretation of Viscoelastic Properties
Using appropriate models, the QCM-D data, f and D, can be interpreted in terms
of adsorbed mass and structural changes in the adsorbed layer. The interpretation of
the viscoelastic properties of the adsorbed layer film was based on the model presented
by Voinova et al. (1999). In this model, the adsorbed layer is represented by a single
Voigt element and it is described using a frequency dependent complex equation when
the layer is subjected to oscillated stress:
                  G = G + iG = µf + 2π if ηf = µf (1 + 2π if τf )                       (6.3)
where µf is the shear elastic (storage) modulus, ηf is the shear viscosity (loss modulus),
f is the oscillation frequency, and τf is the characteristic relaxation time of the film. The
quartz crystal is assumed to be purely elastic and the surrounding solution is assumed
to be purely viscous and Newtonian. Further, it is assumed that thickness (hf ) and
density of the adsorbed layer are uniform, that the viscoelastic properties are frequency
independent and that there is no slip between the adsorbed layer and the crystal during
shearing. For detailed equations showing correlations between frequency and dissipation
changes and shear elastic modulus (µf ), shear viscosity (ηf ) and film thickness (hf ),
see Voinova et al. (1999) and Tammelin et al. (2004).

6.3.2 Measuring Interaction Forces with AFM
A unique property of AFM, initially developed for imaging sample topography (Binnig
et al. 1986), is the possibility for measuring interaction forces between surfaces and
between molecular pairs directly (Butt et al. 2005). In a force measurement the tip and
the sample are first brought into contact (approach curve) and then withdrawn (retract or
separation curve). The interaction forces between surfaces are recorded during the force
measurement cycle. The nature of the interaction, i.e. whether repulsion or attraction is
involved, can be seen in the approach curve. Theoretical analysis of the data obtained
in different environments, for example in different electrolyte concentrations, may yield
information about whether the studied system is electrostatically or sterically stabilized
(Butt et al. 2005). The retract curve can, in addition to the adhesion data, also show
stretching and unfolding of polymers (Butt et al. 2005).
   The raw data obtained from the force measurements is a plot of cantilever deflec-
tion as a function of the sample position. In order to analyze the interaction forces
between surfaces, the raw data is converted to force-versus-distance curves, so-called
‘force curves’. The cantilever acts like a spring so the actual force can be calculated
according to Hooke’s law:
                                         F =k·x                                         (6.4)
where k is the spring constant (nN nm−1 ) and x the deflection of the cantilever (nm).
The nominal spring constants delivered by the manufacturers are often used directly,
but for obtaining more quantitative data, the cantilevers must be calibrated. There
are several different methods available for determining the spring constant (Burnham
154   The Nanoscience and Technology of Renewable Biomaterials

et al. 2003). The most widely used calibration methods determine the change in the
resonance frequency due to added mass (Cleveland et al. 1993), record the thermal
noise (Hutter and Bechhoefer 1993), or use an accurately calibrated reference cantilever
(Torii et al. 1996). The distance between the tip and the sample cannot be determined
directly. For presenting the interaction force as a function of tip-sample separation, the
distance is inferred from the raw data by adding the cantilever deflection to the sample
position. In this way the force can be plotted as a function of relative separation in a
force curve.


6.4   Experimental

6.4.1 Materials
6.4.1.1 Adsorption Experiments
QCM-D crystals. The sensor crystals used as a substrate for the model film coatings
and in the QCM-D experiments were AT-cut quartz crystals supplied by Q-sense AB,
Gothenburg, Sweden with a resonance frequency f0 ≈ 5 MHz and a sensitivity constant
C ≈ 0.177 mg m−2 Hz−1 . The crystals were spin-coated with polystyrene by the sup-
plier. The polystyrene surface was hydrophobic, the contact angle values of pure water
on the crystal surface were 95◦ ± 2◦ (Tammelin et al. 2006b).
Cellulose model film. Trimethylsilyl cellulose (TMSC) dissolved in chloroform was
deposited on the polystyrene coated QCM-D crystal using Langmuir-Schaefer technique
as described by Tammelin et al. (2006b). TMSC was hydrolyzed back to cellulose with
acid hydrolysis according to Schaub et al. (1993).
Mixture of dissolved hemicelluloses. Hemicelluloses used in the adsorption experiments
were isolated from the hexane extracted TMP by using the procedure described by
Thornton et al. (1994). Dissolved hemicellulose fractions were isolated from unbleached
and peroxide bleached TMP. The monosaccharide composition of the aqueous hemicel-
lulose fractions determined by gas chromatography is listed in Table 6.1. The fractions
from unbleached and bleached TMP were anionic with a charge density of 0.51 meq g−1
and 1.38 meq g−1 , respectively. Charge densities were determined by titration with a
cationic polyelectrolyte (1 meq l−1 pDADMAC, Mw < 300 kDa), using a particle charge
             u
detector (M¨ tek PCD 03, Germany) to indicate the end-point.
O-acetyl-galactoglucomannan. GGM (charge density 0.09 meq g−1 , weight average
Mw ≈ 50 kDa with unimodal Mw distribution analyzed by size exclusion chromatogra-
phy (SEC) was isolated from the mixture of TMP derived dissolved hemicelluloses by
                                               o
the ultrafiltration technique described by Willf¨ r et al. (2003).
Pectin. Pectin samples with a Mw comparable to pectin found in spruce were pre-
pared from commercial citrus fruit pectin (Sigma-Aldrich Chemie BmbH, Germany) by
alkaline hydrolysis. The reaction mixtures were cooled to room temperature, acidified,
concentrated in a vacuum rotor-evaporator and freeze-dried. The charge density of the
final product was 2.1 meq g−1 . The weight average Mw was ≈12 kDa with a unimodal
Mw distribution analyzed by SEC.
                      Hemicelluloses at Interfaces: Some Aspects of the Interactions   155

                 Table 6.1 Monosaccharide compositions (%) of the
                 dissolved hemicelluloses derived from unbleached and
                 peroxide bleached spruce TMP.
                                             Unbleached        Bleached
                 Arabinose                        2.7             3.7
                 Xylose                           3.0             8.9
                 Rhamnose                         3.2             9.0
                 Mannose                         47.1            14.0
                 Galactose                       14.3            10.6
                 Glucose                         14.6             5.7
                 Galacturonic acid               14.2            46.7
                 Glucuronic acid                  0.9             0.7
                 4-O-Me-glucuronic acid           0               0.6



Xylan. Xylan was commercial birch xylan from Roth (4-O-methylglucuronoxylan,
Mw ≈ 14 kDa, DP ≈ 100 as reported by the supplier) and it contained 7.8% 4-O-methyl-
α-D-glucuronic acid side groups. The same xylan was used in the cellulose-xylan force
measurements.
  All other chemicals were of p.a. grade if not otherwise specified.

6.4.1.2 Force Measurements
Cellulose beads for the force measurements. The cellulose surfaces employed in the
AFM force measurements were crosslinked cellulose beads obtained from Kanebo Co.
(Japan). The degree of crystallinity of the beads was 5–35% and they consisted mainly
of type II cellulose (Carambassis and Rutland 1999). The diameter of the cellulose
beads changed by 5% to 20% during swelling (Paananen et al. 2003) and hence, fully
swollen beads were used in the experiments. The surface roughness of wet cellulose
beads was determined from AFM images (3 × 3 µm2 with 512 × 512 pixels) and was
approximately 30 nm.

6.4.2 Methods
6.4.2.1 QCM-D Adsorption

QCM-D instrument. Adsorption of hemicelluloses (fractions isolated from unbleached
TMP and bleached TMP as well as pure GGM, pectin and xylan) on cellulose was studied
using a QCM-D instrument from Q-Sense, Gothenburg, Sweden (Rodahl et al. 1995).
Hemicellulose adsorption took place in an axial flow chamber (Q-Sense D300 system).
With QCM-D the changes in frequency and dissipation can be followed simultaneously
at the fundamental resonance frequency (5 MHz) and its overtones (15, 25 and 35 MHz).
Sample preparation. To avoid pH fluctuations in the QCM-D chamber during the
adsorption experiments, the samples of hemicelluloses except xylan samples were pre-
pared using an aqueous sodium acetate/acetic acid buffer solution with the ionic strength
of 10 mM (NaAc/HAc, pH 5.6).
156   The Nanoscience and Technology of Renewable Biomaterials

   The concentration of the dissolved hemicellulose fractions used in the adsorption
experiments was 100 mg l−1 . Samples diluted with the buffer solution are denoted as
‘low ionic strength’. Higher ionic strengths were obtained by adding NaCl. The ‘high
ionic strength’ used in the studies of hemicellulose adsorption was 110 mM. Pure GGM
and pure pectin samples were prepared in the same way as dissolved hemicellulose
samples.
   Due to the very limited solubility of the xylan, the similar procedure described above
could not be used. A stock solution of xylan (1 mg ml−1 in 0.1 M NaOH) was prepared.
Dissolution of xylan was promoted by heating. Preparation of xylan solutions was done
just prior to experiments, and included dilution of the stock solution, addition of NaCl
to desired final concentration and adjustment of pH 10 by HCl.
QCM-D experiments. Some water was bound to the cellulose coated crystals due to
swelling of the film. Therefore, prior to the adsorption experiments the QCM-D crystals
were allowed to stabilize in the appropriate buffer or electrolyte solution to ensure stable
zero baselines for the frequency and dissipation changes. The cellulose surface stabilized
within ≈7 hours. The application of the cellulose model surfaces in QCM-D experiments
has been described in detail in a recent publication (Tammelin et al. 2006).
  After replacement of a pure buffer or electrolyte solution in the QCM-D chamber
with a buffer or electrolyte solution containing 100 mg l−1 hemicelluloses, frequency
and dissipation changes were recorded as a function of time. Adsorption experiments
were performed batch wise and the adsorption process was followed until an adsorption
plateau level was attained. The plateau level for dissolved hemicelluloses isolated from
unbleached TMP and pure GGM was attained within 200 min and for dissolved hemi-
celluloses isolated from peroxide bleached TMP and pure pectin within 100 min. The
plateau level for xylan was attained within 200 min.
Voigt-based modellings. The adsorption results ( f and D) from QCM-D measure-
ments at several overtones were fitted to the Voigt-based model for viscoelastic solids
using the program Q-Tools from Q-Sense (Voinova et al. 1999). In practice, when
conducting QTools fitting the following parameters are used:
• known parameters: fluid viscosity and density (ηl and ρl ), f and D;
• assumed parameter: density of the adsorbed layer (ρf );
• fitted parameters: shear elastic modulus (µf ), shear viscosity (ηf ) and hydrodynamic
  thickness (hf ) of the adsorbed layer.
The fitting is performed by assuming the layer densities to be 1200 kg m−3 in order
to achieve comparable results between different adsorption experiments. The different
overtones are modeled all together (15, 25 and 35 MHz) and separately (15 and 25 MHz;
15 and 35 MHz; 25 and 35 MHz) with the purpose of finding the best fits and to
examine how the calculated shear viscosity, shear elastic modulus and hydrodynamic
thickness values depend on the experimental data used in the estimations. If the different
combinations of the overtone data gave similar values and fit well to the data, the results
were accepted. Usually, the best fit was obtained when all the overtones were included
and the most deficient results were obtained if only the combination of the overtone data
15 and 35 were used. The idea of the mechanical model was not to describe polymer
morphology in detail, but model the interfacial behavior in a phenomenological manner.
                       Hemicelluloses at Interfaces: Some Aspects of the Interactions   157

6.4.2.2 AFM

Sample preparation for force measurements. Single cellulose beads were glued (epoxy
glue, UHU+) to the end of a tipless cantilever (silicon nitride, Digital Instruments/Veeco,
a nominal spring constant 0.12 N m−1 ) as described in Paananen et al. (2003). For
obtaining more quantitative data, each cantilever was calibrated separately during sample
preparation. The thermal method (Hutter and Bechhoefer 1993) was used for determining
the spring constant. It was chosen for its simplicity, applicability to V-shaped cantilevers
and nondestructive nature. Cellulose beads were attached onto the glass sample support
on a thin layer of glue. Cantilevers and sample supports with cellulose beads were
dried in a desiccator and prepared prior to each force measurement. Also reference
(1 mM NaCl, pH 10) and xylan solutions (100 mg ml−1 in 1 mM NaCl, pH 10) were
freshly made.
Force measurements. The force measurements were done by the colloidal probe tech-
nique (Ducker et al. 1991) using a NanoScope IIIa Multimode AFM (Digital Instruments
(Veeco, CA) equipped with a scanner E with vertical engagement, and using an O-ring.
Swelling of the cellulose beads and behavior of the soft cellulose surfaces during force
measurements were studied separately (Paananen et al. 2003) in order to perform proper
experiments and interpret the results. Based on these results the cellulose beads were
allowed to swell in water overnight and to stabilize in reference solution (1 mM NaCl,
pH 10) for 2 h prior to the force measurements. Softness of the cellulose surfaces also
required recording force curves with different loading forces. Force curves taken with
low loading force (here ∼0.6 mN m−1 ) gave more realistic results for the ‘true’ interac-
tions than curves of high loading force (here ∼1.7 mN m−1 ), but part of the repulsion
measured on approach was due to compression of the cellulose beads and the region of
constant compliance was not reached. Hence, the cantilever sensitivity values needed
for analyzing the force curves with low loading force were taken from the ones with
high loading force, where the constant compliance was reached.
   The interaction forces between cellulose surfaces with different loading forces were
measured in a reference solution (1 mM NaCl, pH 10) and in the presence of xylan
(100 mg ml−1 in 1 mM NaCl, pH 10). The total exposure time to xylan solution was
5 h before force measurements. All measurements were performed in pH 10 to ensure
that xylan was soluble. The force curves were recorded at slightly different spots still
being on the central area of the beads. The time gap between consecutive force curves
was varied from 0.5 min to 10 min for obtaining information of the relaxation time of
the cellulose surfaces. The measurements were performed several times and the trend
observed in the results was repeatable.
   The effect of electrolyte concentration on the interaction between xylan-coated cel-
lulose surfaces (Paananen 2007) was studied by measuring interaction forces between
cellulose beads in 100 mg ml−1 xylan solution, pH 10 with varying electrolyte concentra-
tions (1, 10 and 100 mM NaCl). After changing the solution in the measurement cham-
ber, the system was allowed to stabilize for 3 h. The force curves were recorded using
different loading forces and the time gap between consecutive force curves was 5 min.
Analysis of the force curves. The raw force curve data were converted into ASCII format
using a Scanning Probe Image Processor (SPIP, Image Metrology, Denmark) and further
158    The Nanoscience and Technology of Renewable Biomaterials

handled in Excel. The cantilever sensitivity value was determined separately for each
measurement from the force curve with high loading force. The sensitivity value was
used for calculating the deflection of the cantilever from the raw data. Forces were
calculated by the Hooke’s law and normalised by the radii of the interacting beads (R1
and R2 ). The normalized force is related to the interaction free energy Wf between flat
surfaces by the Derjaguin approximation (Derjaguin 1934):
                                                R1 R2
                               F (D) = 2π ·            · Wf                             (6.5)
                                               R1 + R2
where F (D) is the force as a function of distance D. The distance between the sample
and the cantilever was calculated as the sum of the deflection and the sample position.
AFM imaging. AFM was also used to verify the xylan deposition on the cellulose
surface. The images of the cellulose coated QCM-D crystals after the xylan adsorption
experiments were scanned in tapping mode in air using silicon cantilevers (Pointprobes,
type NCH) delivered by Nanosensors, Neuchald, Switzerland with a resonance frequency
around 300 kHz. No image processing except flattening was made and several areas on
each sample were measured.



6.5   Results

6.5.1 Adsorption of Hemicelluloses on Cellulose
Figures 6.1a and 6.1b compare the adsorption of dissolved hemicelluloses isolated from
unbleached and peroxide bleached TMP as well as the adsorption of pure GGM, pectin
and xylan on cellulose at low ionic strength. QCM-D data are presented as change
in dissipation as a function of change in frequency (change in mass detected by the
quartz crystal). This procedure enables concomitant comparison of several adsorption
experiments.
   Figure 6.2 compares the adsorption data of dissolved hemicelluloses isolated from
unbleached TMP and xylan at high ionic strength.
   From these figures and on the basis of the detailed investigations reported in Tammelin
et al. (2007) and Paananen et al. (2003), the following preliminary interpretations can
be made about the adsorption of hemicelluloses from different sources on cellulose
surface, that are essential to the interpretation of the viscoelastic properties of the formed
hemicellulose layers.
1. Dissolved hemicelluloses isolated from unbleached TMP significantly adsorbed on a
   weakly anionic cellulose surface and the adsorption increased with increasing ionic
   strength since the repulsion between anionic charges within the hemicellulose chain
   and cellulose surface was screened (Figures 6.1 and 6.2), see also Tammelin et al.
   (2007).
2. Dissolved hemicelluloses isolated from peroxide bleached TMP adsorbed much less
   on cellulose since higher anionic charge of the hemicellulose molecules leads to stiffer
   and more rodlike conformation of the polyelectrolyte chains due to higher repulsion
   between anionic segments. Thus, these hemicelluloses tend to form a flatter and
                                Hemicelluloses at Interfaces: Some Aspects of the Interactions      159


                                                           DD × 10−6
DD × 10−6                                                  7
7
                                                           6
6
5                                   Ubl TMP                5
                                                                   Galactoglucomannan
4                                                          4

3                                                          3
        Bl TMP                                             2
2                                                                                   Xylan
1                                                          1
                                                                           Pectin
0                                                          0
    0    −20 −40 −60 −80 −100 −120 −140 −160                   0     −20 −40 −60 −80 −100 −120 −140 −160
                    Df (Hz)                                                     Df (Hz)
                              (a)                                                    (b)

Figure 6.1 Change in dissipation vs. change in frequency for adsorption of 100 mg l−1
hemicellulose solutions on cellulose. (a) Dissolved hemicelluloses isolated from unbleached
TMP (Ubl TMP) and peroxide bleached TMP (Bl TMP) (b) pure hemicelluloses. Hemicelluloses
were in 10 mM NaAc/HAc buffer at pH 5.6, except xylan which was in 1 mM NaCl at pH
10. f0 = 5 MHz, n = 3, t = 200 min, except t = 100 min for pectin and Bl TMP.



            DD × 10−6
            7
                                                                                Ubl TMP
            6


            5                       Xylan

            4


            3


            2


            1


            0
                0       −20           −40     −60    −80            −100     −120     −140   −160
                                                    Df (Hz)

Figure 6.2 Change in dissipation vs. change in frequency for adsorption of 100 mg l−1
solutions of dissolved hemicelluloses (10 mM NaAc/HAc buffer, pH 5.6, 100 mM NaCl)
and xylan (pH 10, 10 mM NaCl) at high ionic strength on cellulose. Ubl TMP = dissolved
hemicelluloses isolated from unbleached TMP. f0 = 5 MHz, n = 3, t = 200.
160   The Nanoscience and Technology of Renewable Biomaterials




Figure 6.3 AFM topography (left) and phase contrast (right) images of xylan on cellulose
after QCM adsorption measurement. Image size is 1 µm × 1 µm.


   more rigid layer resulting in lower adsorption (Figure 6.1); see also Tammelin et al.
   (2007).
3. Neutral GGM adsorbs on cellulose forming a dissipative layer whereas pectin with
   low Mw and high anionic charge and small pectin formed thin, flat and rigid layer
   on cellulose (Figure 6.1); see also Tammelin et al. (2007).
4. Xylan adsorbed on cellulose as notable amounts at low ionic strength. Increase in
   ionic strength leads to lower frequency change (lower mass change on the crystal)
   and higher dissipation change (Figures 6.1 and 6.2). This was not expected since an
   increase in ionic strength should facilitate adsorption of more xylan on the surface.
   Screened repulsion of anionic charges between the cellulose surface and xylan as well
   as between anionic charges within the xylan molecule should lead to more pronounced
   adsorption behavior, see also Paananen et al. (2003).
5. The large changes in dissipation for most of the systems indicate the formation of
   viscoelastic hemicellulose layer and thus, the use of the Voigt model is motivated.
Figure 6.3 shows the AFM topography and phase contrast images of cellulose coated
QCM-D crystal surface after the adsorption of xylan from 100 mg g−1 solution at pH 10
and 1 mM NaCl. The fine structure of the cellulose surface can be seen in both images
as described in Tammelin et al. (2006). Granular shapes are interpreted to be xylan
aggregates.

6.5.2 Viscoelastic Properties of the Hemicellulose Layers
To further characterize the properties of the adsorbed hemicellulose films and the effect of
substrate and ionic strength on the film properties, the formed layers were analyzed using
the Voigt-based model for viscoelastic solid. The authors would like to point out that
too much significance should not be attached to the absolute values of shear viscosity,
elastic modulus and hydrodynamic thickness. On the contrary, the modeling results can
be compared relative to each other thus giving information on how the layer properties
are affected by parameters such as ionic strength and charge of the polyelectrolyte.
   Figure 6.4 shows an example of the measured and fitted frequency and dissipation
change curves for the adsorptions of 100 mg l−1 hemicellulose solutions isolated from
            −10                                                                                                                   −12
            −11                                                                                                                   −14
                                                                                                                                                                                                                                   F3/3 (Hz)
                                                                                                                                  −16
            −12                                                                                                                                                                                                                    F5/5 (Hz)
                                                                                                                                  −18                                                                              6
            −13                                                                                                                                                                                                                    F7/7 (Hz)
                                                                                                      4                           −20
                                                                                                                                                                                                                                   f3 fit
            −14                                                                                                                   −22                                                                                              f5 fit
            −15                                                                                                                   −24                                                                                              f7 fit
                                                                                                                                                                                                                   5
                                                                                                                                  −26                                                                                              D3 (1E-6)
            −16
                                                                                                                                  −28                                                                                              D5 (1E-6)
            −17
                                                                                                                                  −30                                                                                              D7 (1E-6)
            −18                                                                                       3                           −32                                                                                              D3 fit
                                                                                                                                                                                                                   4
                                                                                                                                                                                                                                   D5 fit




                                                                                                          D3 (1E-6)
            −19                                                                                                                   −34




                                                                                                                      F3/3 (Hz)
                                                                                                                                                                                                                       D3 (1E-6)




F3/3 (Hz)
                                                                                                                                                                                                                                   D7 fit
            −20                                                                                                                   −36
                                                                                                                                  −38
            −21
                                                                                                                                  −40                                                                              3
            −22                                                                                                                   −42
                                                                                                      2
            −23                                                                                                                   −44
            −24                                                                                                                   −46
                                                                                                                                                                                                                   2
                                                                                                                                  −48
            −25
                                                                                                                                  −50
            −26                                                                                                                   −52
                  40   60   80   100   120   140   160      180   200   220   240   260   280   300                                     20   40   60   80   100   120   140   160    180   200   220   240   260
                                                   Time (min)                                                                                                           Time (min)

                                                      (a)                                                                                                                  (b)

Figure 6.4 Best fits obtained using the Voigt model to the adsorption data of hemicellulose solution at (a) low and (b) high ionic strength on
cellulose. Hemicelluloses were isolated from unbleached TMP. Lines indicate the f and D vs. time original QCM-D data at n = 3, n = 5
and n = 7 and the squares indicate fitted values. (a) I = 10 mM. (b) I = 110 mM, 10 mM NaAc/HAc buffer, pH 5.6. Assumed layer density =
1.2 g cm−3 .
                                                                                                                                                                                                                                               Hemicelluloses at Interfaces: Some Aspects of the Interactions
                                                                                                                                                                                                                                               161
162      The Nanoscience and Technology of Renewable Biomaterials


  hf (10−3)/Nm−2                                        µf (×105)/Nm−2
  2.0                                 Ubl TMP           2.5                               Ubl TMP
                                      high I                                              high I
  1.8
                                                         2
                                                                         Bl TMP
  1.6                                     Xylan
                                                                         low I                  Ubl TMP
                                          low I         1.5                                     low I
  1.4        Bl TMP
             low I          Ubl TMP                                                     Xylan
                            low I                        1                              low I
  1.2

  1.0                                                   0.5
        50       100      150       200           250         50    100       150         200       250
                       Time (min)                                          Time (min)
                          (a)                                                 (b)

Figure 6.5 Variations in (a) shear viscosity and (b) shear elastic modulus as a function of
time corresponding to the best fittings based on Voigt model for a viscoelastic solid.

unbleached TMP at low (a) and high ionic strength (b) on cellulose. Similar fittings
were carried out for dissolved hemicelluloses isolated from peroxide bleached TMP at
low ionic strength (I = 10 mM) and xylan at low and high ionic strength at pH 10
(fitting data not shown). In spite of the simplicity of the Voigt model and a very strong
approximation of its contributions to the elastic properties of an adsorbed polymer layer,
the fitting to the frequency and dissipation data was reasonable. Only the modeling of
xylan layer properties at high ionic strength failed.
   The development of the hemicellulose layers shear viscosities (ηf ) and shear modulus
(µf ) values during the adsorption process estimated by the application of the Voigt
model is shown in Figure 6.5.
   The shear viscosity of the formed hemicellulose films stayed more or less at the same
level showing a slightly decreasing trend as the film formation proceeded for all the
systems except for the dissolved hemicelluloses isolated from unbleached TMP at high
ionic strength, which stayed constant. The shear elastic modulus was highest for the
layer of dissolved hemicelluloses from unbleached TMP at high ionic strength and lowest
for xylan layer at low ionic strength. The layer formed by adsorbing peroxide bleached
TMP hemicelluloses showed relatively high shear elastic modulus values. According to
these results, the hemicelluloses isolated from unbleached TMP at high ionic strength
seemed to be most strongly bound on cellulose. Surprisingly the xylan layer at low ionic
strength appeared to be rather loosely bound on cellulose. Furthermore, it can be noted
that the Voigt model failed when the xylan film was modeled at high ionic strength. The
reasons for the low shear viscosity and low shear elastic modulus are discussed later.
   The estimated hydrodynamic thickness of the hemicellulose films as the adsorption
proceeds is plotted in Figure 6.6. At the end of the adsorption process the film thickness
of dissolved hemicelluloses at low ionic is approximately 4 nm and at high ionic strength
the final thickness is 9 nm. The hemicellulose film isolated from peroxide bleached
TMP has a final thickness of roughly 2 nm. The xylan film reaches a final thickness
value of 4 nm.
                                      Hemicelluloses at Interfaces: Some Aspects of the Interactions                        163

                            hf (nm)
                            10
                                                                                            Ubl TMP
                                                                                            high I
                             8
                                                                   Xylan
                             6                                     low I

                             4
                                                       Bl TMP
                             2                         low I

                             0
                                 50              100            150                     200           250
                                                             Time (min)

Figure 6.6 The hydrodynamic thickness of the formed hemicellulose films on cellulose as a
function of time.

                                                                                  1.0
                                             ref                                  0.8                       ref
                 1                           xyl 100 mgl−1                                                  xyl 100 mgl−1
                                                                                  0.6
                                                                  F/R (m Nm−1)
F/R (m Nm−1)




                                                                                  0.4
                0.1                                                               0.2
                                                                                  0.0
               0.01                                                              −0.2
                                                                                 −0.4
                      0   10 20 30 40 50 60 70 80                                       0   10 20 30 40 50 60 70 80
                                 Distance (nm)                                                   Distance (nm)
                                       (a)                                                            (b)

Figure 6.7 The forces measured on approach (a) and separation (b) between two cellulose
beads across a reference solution (1 mM NaCl, pH 10) and a xylan solution (100 mg ml−1
xylan in 1 mM NaCl, pH 10). (Modified from Paananen et al. 2003.)

6.5.3 Effect of Xylan Adsorption on the Interaction between Cellulose Beads
Xylan adsorbs on cellulose in weakly alkaline solutions at low ionic strength and the
adhesion between cellulose surfaces decreases. The adsorption is observed in the force
curves as a stronger and longer-ranged repulsion on approach (Figure 6.7a), the range of
repulsion changing from less than 10 nm to approximately 60 nm. Adsorption of xylan
onto cellulose was supported by the QCM-D experiments (Figure 6.1). The slight pull-off
force around 0.1–0.4 mN m−1 observed between pure cellulose spheres is replaced by
a weak repulsion upon adsorption of xylan (Figure 6.7b). Both the force measure-
ment and the corresponding QCM-D results are described in greater detail in Paananen
et al., 2003.
164   The Nanoscience and Technology of Renewable Biomaterials

6.5.4 Effect of Electrolyte on the Interaction between Xylan-coated Cellulose
      Surfaces
The electrolyte (NaCl) did not appear to have a notable effect on the interaction forces
on approach and hence data is not shown. Lack of a clear trend in the interaction
with increasing electrolyte concentration indicated that the interaction was dominated by
steric contributions rather than by double-layer repulsion due to overlapping counterion
clouds. Electrostatic interactions cannot, however, be excluded and this type of forces
are sometimes called electrosteric to emphasis the contribution of both electrostatic and
steric effects. The magnitude of the adhesion in the separation curves was low in all
electrolyte concentrations and varied between different experiments (data not shown).
The results showed, though, that when performing measurements at the highest elec-
trolyte concentration, repulsion was seen in the separation curves more often than with
the lowest electrolyte concentration.


6.6   Discussion

6.6.1 Adsorption of Dissolved Hemicelluloses on Cellulose
The dissolved hemicelluloses present in the waters of unbleached TMP of spruce mainly
contains neutral GGM, see Table 6.1. The slight anionic charge of the mixture origi-
nates from the acidic arabinogalactan, which is composed of galactose, arabinose and
glucuronic acid, and pectins. Pectins are present only in minor amounts in unbleached
fractions. The alkaline conditions during peroxide bleaching of the TMP significantly
change the chemical composition of the dissolved material. GGMs are deacetylated
which leads to lower solubility of the polymer resulting its re-adsorption on the fiber
surface. Thus, the glucose and mannose units are decreased in the water phase. Con-
comitantly highly charged demethylated pectins are dissolved from the TMP, which
results in a higher concentration of polyanions in waters. These changes can be clearly
seen in Table 6.1 and the determined charge densities of the different hemicellulose
fractions support the results. The unbleached fraction was anionic with a charge den-
sity of 0.51 meq g−1 whereas the charge density of the peroxide bleached fraction was
1.38 meq g−1 . These results are very well in accordance with the results achieved by
Holmbom et al. (1991) and Thornton et al. (1994) who have extensively studied the
properties of the TMP process waters.
   The main driving force for the adsorption is not the attraction between oppositely
charged surfaces since both, the cellulose surface and hemicellulose mixtures, are anionic.
However, the osmotic repulsion between overlapping counterion clouds is not strong
enough to prevent the hemicellulose adsorption and significant amounts of hemicel-
luloses attach to the cellulose surface; see Figure 6.1. Still the electrostatics affect
the conformation of the hemicellulose chains and, thus, cannot be excluded from the
discussion. In the following the adsorption of dissolved hemicelluloses isolated from
unbleached and peroxide bleached TMP on cellulose will partly be discussed with respect
to the effect of electrostatic interaction on the adsorption behavior.
   Increasing the ionic strength decreases the range of the double-layer force and hence
the anionic hemicellulose molecules can come closer to the anionic cellulose surface and
                       Hemicelluloses at Interfaces: Some Aspects of the Interactions   165

other short-ranged attractive forces become important. The conformation of the hemi-
cellulose chain also changes. At high ionic strength the polyelectrolyte chain becomes
more coiled and as a consequence, more polymer can fit to adsorb onto the cellulose sur-
face. Due to these facts, the hemicellulose layer seems to be very strongly bound on the
cellulose surface. Both the shear viscosity and shear elastic modulus are high indicating
the formation of a strongly bound hemicellulose film (Figure 6.5). Furthermore, it can
be assumed that the formation of relatively thick hemicellulose layer (the hydrodynamic
thickness is estimated to be ∼9 nm at high ionic strength) is due to better packing of
the molecules for two reasons: (1) the more coiled conformation of the hemicellulose
chains and (2) the lower solubility of the hemicelluloses at higher ionic strength, which
promotes adsorption.
   Hemicelluloses isolated from the peroxide bleached TMP adsorbed to a lesser extent
on cellulose compared to hemicellulose fraction isolated from the unbleached TMP.
This is as expected since the repulsion between charged carboxylic acid segments of
the hemicellulose chain is high and, thus, the polyelectrolyte molecules in solution take
a stiff and rodlike conformation (Fleer et al. 1993). When adsorbing on the surfaces
these hemicelluloses should tend to form a thin and flat layer, and according to the Voigt
based modellings, a relatively thin layer of hemicelluloses adsorbed on cellulose surface.
The final hydrodynamic thickness was approximately 2 nm, see Figure 6.6. Although
the adsorbed amount is lower compared to the adsorbed amount of the unbleached
hemicellulose fraction, the hemicellulose layer is relatively strongly bound on cellulose
surface (Figure 6.5). Especially the shear elastic modulus values are high and at the end
of the adsorption process the shear elastic modulus reaches nearly the same level as the
film formed from unbleached hemicellulose fraction at high ionic strength.
   The adsorption behavior of anionic hemicelluloses on slightly anionic cellulose surface
and the effect of ionic strength on the adsorption can largely be explained by screen-
ing of intramolecular, electrostatic repulsive interactions. Consequently, the screened
repulsion leads to a more coiled conformation of the hemicellulose chain and decreased
double layer forces which enables closer contact between the hemicellulose chains and
the cellulose surface. The closer contact may facilitate the other important attractive
forces, such as van der Waals forces, to become more predominant (Israelachvili 1992).
Adsorption behavior of pure GGM and pure pectin clarifies and supports the idea of the
effect of electrostatics on the hemicellulose adsorption. More or less neutral and rela-
tively large GGM adsorbs on cellulose forming a layer with loops and tails pointing out
to the solution phase whereas the highly charged and small pectin molecules adsorbed
on cellulose forming a very thin and flat layer (Figure 6.1b). However, the main driving
force of adsorption is unclear. The dissolved hemicelluloses may prefer the contacts with
cellulose to contacts with solvent and they probably adsorb due to the similarities in the
molecular structure with cellulose. When polymer reaches sufficiently close contact to
cellulose the formation of hydrogen bonds between cellulose and hemicellulose chains
may be promoted. However, in aqueous environment the hydrogen bonding with water
is probably dominating.
   Strongly bound hemicellulose layers, especially those forming loops and tails point-
ing out to the solution phase, can effectively sterically stabilize surfaces. Thus, the
hemicelluloses can be used as stabilizers. It is well known that hemicelluloses can sta-
bilize extractive colloids in TMP process waters preventing them from aggregating and
166   The Nanoscience and Technology of Renewable Biomaterials

accumulating on the fiber surfaces or on the process equipment surfaces even at the
addition of electrolyte (Hannuksela et al. 2004; Sundberg et al. 1996b). This way the
negative effects of wood extractives on paper properties can be diminished. Investiga-
tions dealing with wood extractive colloid interactions with TMP fine material showed
that hemicelluloses which were adsorbed on cellulose rich fibrillar fines prevented the
further adsorption of colloidal extractives (Johnsen et al. 2007; Mosbye et al. 2003).
Colloidal extractives widely adsorbed on fine material when the dissolved hemicellu-
loses were not present in the system and the adsorption increased with increasing ionic
strength. The similar behavior was observed when the adsorption experiments were
conducted with QCM-D using cellulose model surfaces as substrates for extractive col-
loid adsorptions (Tammelin et al. 2007). Extractive colloids significantly adsorbed on
pure cellulose surface but the adsorption was prevented when dissolved hemicelluloses
were allowed to sterically stabilize the cellulose surface, the colloids or both the cel-
lulose surface and the extractive colloids. The adsorption was prevented at high ionic
strengths as well.

6.6.2 Adsorption Behavior and Interaction Forces between Xylan and Cellulose
The results of the QCM-D adsorption experiments and the AFM force measurement
showed that despite the negative charges on both interacting partners, xylan adsorbed
onto cellulose in weakly alkaline solutions at low ionic strength. Addition of charges on
weakly charged cellulose surfaces by adsorption of xylan increases naturally electrostatic
repulsion, but steric repulsion is increased as well: charges also cause swelling of the
                                                    ¨
adsorbed layer. The domination of steric forces (Osterberg et al. 2001) and electrostatic
repulsion (Claesson et al. 1995) at large distances between xylan-coated surfaces has
been reported previously.
   The force measurement results also showed that the adhesion between cellulose sur-
faces is very low before and after adsorption of xylan. The QCM-D results showed
that the interaction between cellulose and xylan is weak, but according to the force
measurements the cellulose-xylan interaction is strong enough to prevent desorption
upon dilution (Paananen et al. 2003). These findings suggest that a combination of the
increase in inherent entropy increase associated with the release of solvent molecules
upon adsorption of xylan and weak van der Waals’ attraction is the driving force of
the cellulose-xylan association, rather than formation of hydrogen bonds as has repeat-
edly been cited (Mora et al. 1986). This does not exclude, though, the importance of
hydrogen bonding in dry systems. Hence, the results indicated that the role of xylan in
increasing the paper strength (interfiber bonding) is probably to increase the contact area
between fibers, and this would be associated with processes taking place during drying.
   The adsorption behavior of xylan at higher ionic strengths was not as expected; xylan
adsorption was not enhanced by increased electrolyte concentration. On the contrary,
less was adsorbed and the dissipation response was relatively high indicating soft and
water containing film of xylan on cellulose surface (Figure 6.2). Xylan reacted to the
changes in electrolyte concentration in a completely different way compared to dissolved
TMP hemicelluloses as shown in Figure 6.2.
   The expected effect of electrolyte on the interaction forces between surfaces containing
charges is reducing the electrostatic repulsion with increasing electrolyte concentration,
                       Hemicelluloses at Interfaces: Some Aspects of the Interactions   167

                  ¨
as reported by Osterberg et al. (2001) and Claesson et al. (1995) for xylan-coated
surfaces. In our force measurements (Paananen 2007) no clear trend in the interaction
with increasing electrolyte concentration was observed (data not shown). This indicates
domination of steric forces, although the presence of electrostatic interaction cannot be
excluded. This is contradictory to the reported results by Claesson et al. (1995) and
 ¨
Osterberg et al. (2001), where adsorption of xylan on mica and interaction between
xylan-coated mica surfaces has been investigated. The observed dominance of steric
or electrosteric repulsion in our results could partially be explained by the material
differences between cellulose and mica. The anionic charge of the cellulose surface
is relatively low and the the RMS roughness was approximately 30 nm whereas the
roughness of the highly charged anionic mica surface is less than 0.3 nm. As a result,
the adsorption of xylan on cellulose most likely differs from that on mica.
   Claesson et al. (1995) explained the presence of steric forces by the existence of
long dangling tails in the adsorbed xylan layers due to prolonged times at elevated pH.
On the other hand, increase in electrolyte concentration decreases the steric interac-
tion, because the charges of the polymer molecules are screened out resulting in more
                                             ¨
compact conformation of the molecules (Osterberg et al. 2001). However, this trend
was not observed by our force and adsorption measurements and it seemed that elec-
trostatics had no strong effect on detected behavior. In order to further explain the
xylan-cellulose interactions the adsorbed xylan films were modeled with the Voigt-based
model. The Voigt model estimated relatively weakly bound xylan film on cellulose at
low ionic strength (Figure 6.5) and at high ionic strength the Voigt model failed. The
reason for the behavior detected may be the moderately limited solubility of the xylan
molecules. In solution xylan molecules probably take relatively coiled conformation and
polymer-polymer contacts are presumably more preferable than polymer-solution con-
tacts. At high ionic strength when the solubility is even more impaired, the xylan chains
may form soluble clusters which adsorbs on cellulose forming patches. The formation of
xylan clusters was supported by the AFM images in Figure 6.3. In the figure the xylan
is unevenly distributed as globular structures on the cellulose. This might be the reason
why the Voigt model could not estimate the xylan layer properties. The model fails if
the adsorbing material is not evenly distributed on the crystal surface as has been found
earlier when the Voigt model was used in an attempt to estimate the viscoelastic prop-
erties of extractive colloids on cellulose surface (unpublished results). Xylan assembly
studies on cellulose fibers (Linder et al. 2003) and on model cellulose (Henriksson and
Gatenholm 2001) suggest that xylan forms particle shaped, globular structures on these
surfaces in agreement with our observations. These investigations also suggest that the
assembly process is influenced by changes in xylan solubility and the affinities between
xylan and cellulose.
   The magnitude of the adhesion between xylan coated cellulose surfaces was low in all
electrolyte concentrations (see Paananen 2007). The results showed, though, that repul-
sion was seen in the separation curves more often with the highest than with the lowest
electrolyte concentration. Adhesion between surfaces may originate from interpenetra-
tion of xylan chains from one surface to the other. When the electrolyte concentration is
increased, there are less protruding chains for interpenetration. In addition to this, when
xylan-coated surfaces are pushed together, more segments of xylan are forced to adsorb
168   The Nanoscience and Technology of Renewable Biomaterials

to the surface (Klein 1988). On the other hand, as discussed above, most probably more
xylan adsorbed from solution on the interacting surfaces during the measurements. It
would be expected that this should result in more dangling chains on the surfaces. Due
to the reasons described above, at higher electrolyte concentrations the adsorbed xylan
most probably has a rather compact conformation, possesses less dangling chains to
interpenetrate and to cause adhesion, but is still a source for steric interaction.



6.7 Conclusions

Dissolved spruce hemicelluloses isolated from the waters of TMP strongly adsorbed on
cellulose forming a thin, nanometer scale film the structure and thickness of which was
dependent on the ionic strength and on the pulp treatment, e.g., pulp peroxide bleach-
ing. Although the dependence of adsorption behavior on ionic strength can largely be
explained by electrostatics, the main driving force for adsorption seems to be nonelectro-
static in nature. Other factors such as polymer solubility and preferable polymer contacts
need to be considered when explaining the adsorption behavior. The affinity of birch
xylan towards cellulose was mainly explained by the low solubility of xylan molecules.
Xylan probably forms soluble clusters which adsorbs on cellulose and the electrosteric
rather than double-layer repulsion is dominating between xylan-coated cellulose surfaces.
The combined results from our QCM experiments and AFM force measurements bring
out information that deepens the understanding of adsorption behavior of the different
hemicelluloses and the properties of hemicellulose films on cellulose surfaces.



Acknowledgements

                                           ˚
Laboratory of Wood and Paper Chemistry, Abo Akademi University is acknowledged for
donating the galactoglucomannan and pectin samples and Prof Mark Rutland, from the
                                                                              a a
Royal Institute of Technology for donating the cellulose spheres. Mrs Marja K¨ rkk¨ inen,
                 a              a¨ o
Mrs Ritva Kivel¨ and Mr Timo P¨ akk¨ nen are warmly thanked for their skilful laboratory
assistance.


References

˚                        e
Akerholm, M. & Salm´ n, L. (2001) Interactions between wood polymers studied by
  dynamic FT-IR spectroscopy. Polymer 42(3), 963–9.
  e
Al´ n, R. (2000), Structure and chemical composition of wood. In Stenius P. (ed.)
  Forest Products Chemistry, Papermaking Science and Technology, Book 3, Fapet Oy,
      a    a
  Jyv¨ skyl¨ . Finland, 12–57.
Barnes, H.A., Hutton, J.F. & Walters, K. (1989), An Introduction to Rheology, Rheology
  Series, 3, Elsevier, Amsterdam.
Binnig, G., Quate, C.F. & Gerber, C. (1986) Atomic force microscope. Phys. Rev. Lett.
  56(9), 930–3.
                      Hemicelluloses at Interfaces: Some Aspects of the Interactions   169

                                  a¨
Buchert, J., Teleman, A., Harjunp¨ a, V., Tenkanen, M., Viikari, L. & Vuorinen, T. (1995)
  Effect of cooking and bleaching on the structure of xylan in conventional pine kraft
  pulp. Tappi J . 78(11), 125–30.
Burnham, N.A., Chen, X., Hodges, C.S., et al. (2003) Comparison of calibration meth-
  ods for atomic-force microscopy cantilevers. Nanotechnology 14(1), 1–6.
Butt, H.-J., Cappella, B. & Kappl, M. (2005) Force measurements with the atomic force
  microscope: Technique, interpretation and applications. Surf. Sci. Rep. 59(1-6),
  1–152.
Carambassis, A. & Rutland, M.W. (1999) Interactions of cellulose surfaces: Effect of
  electrolyte. Langmuir 15(17), 5584–90.
Claesson, P.M., Christenson, H.K., Berg, J.M. & Neuman, R.D. (1995) Interactions
  between mica surfaces in the presence of carbohydrates. Colloid Interface Sci . 172(2),
  415–24.
Cleveland, J.P., Manne, S., Bocek, D. & Hansma, P.K. (1993) A nondestructive method
  for determining the spring constant of cantilevers for scanning force microscopy. Rev.
  Sci. Instrum. 64(2), 403–5.
                                                                       a
Derjaguin, B.V. (1934) Untersuchungen uber die Reibung und Adh¨ sion, IV: Theorie
                                           ¨
  des Anhaftens kleiner Teilchen. Kolloid Z . 69(2), 155–64.
Ducker, W.A., Senden, T.J. & Pashley, R.M. (1991) Direct measurement of colloidal
  forces using an atomic force microscope. Nature 353(6341), 239–41.
                                               o¨
Edvardsson, M., Rodahl, M., Kasemo, B & H¨ ok, F. (2005) A dual-frequency QCM-D
  setup operating at elevated oscillation amplitudes. Anal. Chem. 77(15), 4918–26.
Fengel, D. & Wegener, G. (1989), Wood: Chemistry, Ultrastructure, Reactions, Walter
  de Gruyter, Berlin, Germany.
Fleer, G.J., Cohen Stuart, M.A., Scheutjens, J.M.H.M., Cosgrove, T. & Vincent, B.
  (1993) Polymers at Interfaces, Chapman & Hall, London.
Hannuksela, T., Tenkanen, M. & Holmbom, B. (2002) Sorption of dissolved galactoglu-
  comannans and galactomannans to bleached kraft pulps. Cellulose 9(3-4), 251–61.
Hannuksela, T., Fardim, P. & Holmbom, B. (2003) Sorption of spruce O-acetylated
  galactoglucomannans and galactomannans to bleached kraft pulp. Cellulose 10(4),
  317–24.
Hannuksela, T. & Holmbom, B. (2004) Stabilization of wood-resin emulsions by dis-
  solved galactoglucomannans and galactomannans. J. Pulp Pap. Sci ., 30(6), 159–64.
Hannuksela, T., Holmbom, B., Mortha, G. & Lachenal, D. (2004) Effect of sorbed
  galactoglucomannans and galactomannans on pulp and paper handsheet properties,
  especially strength properties. Nord. Pulp Pap. Res. J . 19(2), 237–44.
              ˚
Henriksson, A. & Gatenholm, P. (2001) Controlled assembly of glucuronoxylans onto
  cellulose fibres. Holzforschung 55(5), 494–502.
                              o
Holmbom, B., Ekman, R., Sj¨ holm, R., Eckerman, C. & Thorton, J. (1991) Chemical
  changes in peroxide bleaching of mechanical pulps. Papier 45(10A), V16–V22.
  o¨
H¨ ok, F., Rodahl, M., Brzezinski, P. & Kasemo, B. (1998) Energy dissipation kinetics
  for protein and antibody-antigen adsorption under shear oscillation on a quartz crystal
  microbalance. Langmuir 14(4), 729–34.
Hutter, J.L. & Bechhoefer, J. (1993) Calibration of atomic-force microscope tips. Rev.
  Sci. Instrum. 64(7), 1868–73.
170   The Nanoscience and Technology of Renewable Biomaterials

Israelachvili, J. (1992) Intermolecular & Surface Forces, 2nd Edn. Academic Press, San
   Diego, CA.
Israelachvili, J. & Adams, G.E. (1978) Measurement of forces between two mica surfaces
   in aqueous electrolyte solutions in the range 0–100nm. J. Chem. Soc., Faraday Trans.
   1 74(4), 975–1001.
Johnsen, I.A., Lenes, M. & Magnusson L. (2004) Stabilization of colloidal wood resin
   by dissolved material from TMP and DIP. Nord. Pulp Paper Res. J . 19(1), 22–8.
                                            ¨
Johnsen, I.A., Stenius, P. Tammelin, T., Osterberg, M., & Laine, J. (2006) The influence
   of dissolved substances on resin adsorption to TMP fine material. Nordic Pulp Paper
   Res. J . 21(5), 629–37.
Johnsen, I.A. & Stenius, P. (2007) Effects of selective wood resin adsorption on paper
   properties. Nord. Pulp. Pap. Res. J 22(4), 452–61.
Klein, J. (1988) Surface forces with adsorbed and grafted polymers. In: Molecular
   Conformation and Dynamics of Macromolecules in Condensed Systems (Studies in
   polymer science, Vol. 2). Elsevier, Amsterdam. 382 p.
Kroon-Batenburg, L.M.J., Leeflang, B.R., van Kuik, J.A. & Kroon, J. (2002) Interaction
   of mannan and xylan with cellulose microfibrils: An X-ray, NMR, and modeling
   study. In: 223rd ACS National Meeting, American Chemical Society, Washington
   D.C.
          ˚
Linder, A., Bergman, R., Bodin, A. & Gatenholm, P. (2003) Mechanism of assembly of
   xylan onto cellulose surfaces. Langmuir 19(12), 5072–5077.
Marchessault, R.H., Settineri, W. & Winter, W. (1967) Crystallization of xylan in the
   presence of cellulose. Tappi 50(2), 55–59.
Mitikka-Eklund, M. (1996) Sorption of Xylans on Cellulose Fibres. Licenciate Thesis,
                      a    a
   University of Jyv¨ skyl¨ , Department of Chemistry, Laboratory of Applied Chemistry,
       a     a
   Jyv¨ skyl¨ , Finland.
Mora, F., Ruel, K., Comtat, J. & Joseleau, J.-P. (1986) Aspect of native and redeposited
   xylans at the surface of cellulose microfibrils. Holzforschung 40(2), 85–91.
Mosbye, J., Laine, J. & Moe, S. (2003) The effects of dissolved substances on the
   adsorption of colloidal extractives to fines in mechanical pulp. Nord. Pulp Pap. Res.
   J . 18(1), 63–8.
Neuman, R.D. Berg, J.M. & Claesson, P.M. (1993) Direct measurement of surface forces
   in papermaking and paper coating systems. Nord. Pulp Paper Res. J . 8(1), 96–104.
 ¨
Osterberg, M., Laine, J., Stenius, P., Kumpulainen, A. & Claesson, P.M. (2001) Forces
   between xylan-coated surfaces: Effect of polymer charge density and background
   electrolyte. J. Colloid Interface Sci . 242(1), 59–66.
Otero, D., Sundberg, K., Blanco, A., Negro, C., Tijero, J. & Holmbom, B. (2000) Effects
   of wood polysaccharides on pitch deposition. Nord. Pulp Pap. Res. J . 15(5), 607–13.
Paananen, A. 2007. On the Interactions and Interfacial Behaviour of Biopolymers. An
   AFM study. Ph.D. Thesis. VTT Publications 637, VTT, Espoo, Finland. http://www.
   vtt.fi/inf/pdf/publications/2007/P637.pdf
                  ¨
Paananen, A., Osterberg, M., Rutland, M., et al. (2003) Interaction between cellulose
   and xylan: an atomic force microscope and quartz crystal microbalance study. In:
   Hemicelluloses: Science and Technology (eds. Gatenholm, P. and Tenkanen, M.),
   ACS Symp. Ser. 864, pp. 269–90. American Chemical Society, Washington, DC.
                      Hemicelluloses at Interfaces: Some Aspects of the Interactions   171

                o¨
Rodahl, K., H¨ ok, F., Krozer, A., Brezezinski, P. & Kasemo, B. (1995) Quartz crystal
  microbalance setup for frequency and Q-factor measurements in gaseous and liquid
  environments. Rev. Sci. Instrum. 66, 3924–30.
Schaub M., Wenz G., Wegner G., Stein A. & Klemm D. (1993) Ultrathin films of
  cellulose on silicon wafers. Adv. Mater. 5, 919–22.
    o                               a
Sch¨ nberg, C., Oksanen, T., Suurn¨ kki, A., Kettunen, H. & Buchert, J. (2001) The impor-
  tance of xylan for the strength properties of spruce kraft pulp fibres. Holzforschung
  55(6), 639–44.
Sihvonen, A.-L., Sundberg, K., Sundberg, A. & Holmbom, B. (1998) Stability and
  deposition tendency of colloidal wood resin. Nord. Pulp Paper Res. J ., 13(1), 64–7.
  o o
Sj¨ str¨ m, E. (1993) Wood Chemistry: Fundamentals and Applications. 2nd edn. Aca-
  demic Press, San Diego, CA.
Sundberg, K., Thornton, J., Ekman, R. & Holmbom, B. (1994a) Interactions between
  simple electrolytes and dissolved and colloidal substances in mechanical pulp. Nord.
  Pulp Pap. Res. J . 9(2), 125–8.
Sundberg, K., Thornton, J., Petterson, C., Holmbom, B. & Ekman, R. (1994b) Calcium-
  induced aggregation of dissolved and colloidal substances in mechanical pulp suspen-
  sions. J. Pulp Pap. Sci . 20(11), J317–J321.
Sundberg, K., Thornton, J., Holmbom, B. & Ekman, R. (1996a) Effects of wood polysac-
  charides on the stability of colloidal wood resin. J. Pulp Pap. Sci . 22(7), J226–J230.
Sundberg, A., Sundberg, K., Lillandt, C. & Holmbom, B. (1996b) Determination of
  hemicelluloses and pectins in wood and pulp fibres by acid methanolysis and gas
  chromatography. Nord. Pulp Paper. Res. J . 11(4), 216–19.
Tammelin, T., Merta, J., Johansson, L.-S. & Stenius P. (2004) Viscoelastic properties of
  cationic starch adsorbed on quartz studied by QCM-D. Langmuir 20(25), 10900–9.
Tammelin, T. (2006a) Surface Interactions in TMP Process Waters. PhD Thesis, Helsinki
  University of Technology, Laboratory of Forest Products Chemistry reports A4, Espoo,
  Finland. http://lib.tkk.fi/Diss/2006/isbn9512284839
                              ¨
Tammelin, T., Saarinen, T., Osterberg, M. & Laine J. (2006b) Preparation of Langmuir-
  Blodgett – cellulose surfaces by using horizontal dipping procedure. Application for
  polyelectrolyte adsorption studies performed with QCM-D. Cellulose 13(5), 519–35.
                                ¨
Tammelin, T., Johnsen, I.A., Osterberg, M., Stenius, P. & Laine, J. (2007) Adsorption
  of colloidal extractives and dissolved hemicelluloses on thermomechanical pulp fiber
  components studied by QCM-D. Nord. Pulp Paper. Res. J . 22(1), 93–101.
Teleman, A., Larsson, P.T. & Iversen, T. (2001) On the accessibility and structure of
  xylan in birch kraft pulp. Cellulose 8(3), 209–15.
                                               ¨ a
Thornton, J., Ekman, R., Holmbom, B. & Ors˚ , F. (1994) Polysaccharides dissolved
  from Norway spruce in thermomechanical pulping and peroxide bleaching. J. Wood
  Chem. Technol . 14, 159.
Torii, A., Sasaki, M., Hane, K. & Okuma, S. (1996) A method for determining the
  spring constant of cantilevers for scanning probe microscopy. Meas. Sci. Technol .
  7(2), 179–84.
Voinova, M., Rodahl, M., Jonson, M. & Kasemo, B. (1999) Viscoelastic acoustic
  response of layered polymer films at fluid-solid interfaces: continuum mechanics
  approach, Phys. Scr. 59(55), 391–6.
172   The Nanoscience and Technology of Renewable Biomaterials

  a               ¨
W˚ gberg, L. & Odberg, L. (1991) The action of cationic polyelectrolytes used for the
  fixation of dissolved and colloidal substances. Nord. Pulp Paper. Res. J . 6(3),
  127–35.
     o
Willf¨ r, S., Rehn, P., Sundberg, A., Sundberg, K. & Holmbom, B. (2003) Recovery of
  water-soluble acetylgalactoglucomannans from mechanical pulp of spruce. Tappi J .
  2(11), 27–32.
                     o
Yllner, S. & Enstr¨ m, B. (1956) Adsorption of xylan on cellulose fibers during the
  sulfate cook. I. Svensk Papperstidn. 59(6), 229–32.
                                                        7
    Lignin: Functional Biomaterial
  with Potential in Surface Chemistry
           and Nanoscience

                                Shannon M. Notley and Magnus Norgren



7.1     Introduction

Lignin along with the carbohydrates cellulose or hemi-cellulose and other extractive
                                                                 o o
materials, are the major components of the wood cell wall (Sj¨ str¨ m 1993). Indeed the
term lignin is derived from the Latin word for wood ‘lignum’. Lignin is considered to
be, after cellulose, the second most abundant natural polymer, found chiefly in the cell
wall of woody tree species however it is also found in all vascular plant materials includ-
ing herbs and grasses (Sarkanen and Ludwig 1971). It is estimated that approximately
30% of all carbon in the biosphere may be attributed to lignin and as such, provides
ample opportunity for the use of this material in future applications involving sustainable
resources (Guo et al. 2008). Currently, global production of lignin based materials and
chemicals, mostly as a by-product from the pulping of wood fibres, exceeds 50 million
tonnes per annum with the majority used in low technology and low value added appli-
cations such as fuel or simply discharged as waste (Gosselink et al. 2004a). Whilst
lignin is found in many sources, the pulp and paper industry provides the best oppor-
tunity for securing vast quantities of this important raw material at economically viable
rates and so the discussion in this chapter on the potential use of lignin and derivatives
in nanotechnological applications will be focussed on lignin derived mainly from wood
fibres. Furthermore, as with all aspects of nanoscience, molecular interactions are of
great importance, whether considering lignin as a polymer in solution or in the solid



The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
174   The Nanoscience and Technology of Renewable Biomaterials

state, and hence this chapter will discuss the topochemical and interfacial properties
of lignin.
   The structure of lignin is based upon the polymerised phenylpropane unit linked
together through varied covalent bonds. The functions of lignin in the wood cell wall
are numerous with the aromatic rings as well as various substituents such as ether or
hydroxyl groups providing the molecular means for imparting strength and structural
integrity to the wood fibre wall often through covalent or nonspecific interactions with
                                  o o
the polysaccharides present (Sj¨ str¨ m 1993). For an analogy from materials science, if
a wood fibre is considered as a composite, the crystalline cellulose microfibrils would
be the load-bearing component of the fibre wall whilst lignin is the matrix material. The
highest proportion of lignin can be found in the secondary cell wall however the greatest
concentration is found in the middle lamella between wood fibres. This is exploited in
the chemical pulping of wood where specific reagents such as hydrogen sulphide and
hydroxide are used to target lignin usually at the ether linkages causing a degradation
of the polymeric structure and hence a liberation of the fibres. This suggests that lignin
aids in binding the fibres together providing strength to not only individual fibres on an
ultra-structural level but also to the wood macrostructure itself by effectively acting as a
glue between fibres. Furthermore, due to its inherent hydrophobicity from its aromatic
structure and low charge in the native state, lignin acts to inhibit the swelling of wood
fibres thereby waterproofing the cell wall and providing an efficient means for the trans-
port of water and nutrients throughout the plant vascular system. Another major function
of lignin is to impart protection to the plant cell wall against microbial attack. Lignin
acts to solidify or compact the cell wall. This makes the penetration of enzymes and
proteins secreted by bacteria and fungi into the cell wall for digestion of the polysaccha-
rides extremely difficult to achieve. Perhaps the best evidence for lignin acting to inhibit
the attack of microbes is the relatively slow degradation of more heavily lignified plant
materials such as wood compared to plants such as grasses which have significantly less
lignin. Indeed, much of the humus of soil is derived from lignin polymer fragments with
components such as humic or fulvic acids present due to oxidation reactions and not the
action of microbes.


7.2   Lignin Synthesis and Structural Aspects

As mentioned above, lignin is a highly branched amorphous biomacromolecule with
variable composition dependent on the plant source. However, lignin can be sim-
ply conceived as the polymerised product of the three basic substituted phenylpropane
repeat units known collectively as ‘monolignols’: p-coumaryl alcohol, coniferyl alcohol
and sinapyl alcohol as shown in Figure 7.1 (Sarkanen and Ludwig 1971, Freudenberg
and Neish 1968). As seen in Figure 7.1, the structure of the lignin monomers varies
only in the number of substituted methoxy groups on the aromatic ring. Whilst these
three monolignols account for the overwhelming majority of the repeat units mak-
ing up the lignin polymer molecules, other lignols may be present in much smaller
quantities. Furthermore, the proportion of each of these monomers in lignin varies
considerably depending upon type of plant material under consideration as shown in
Table 7.1. It is henceforth convenient to describe lignin in terms of its source. Softwood
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience   175

                      HO             HO                      HO




                                             OMe       MeO         OMe
                       OH               OH                    OH

Figure 7.1 The common lignin monomers p-coumaryl alcohol, coniferyl alcohol and sinapyl
alcohol.


Table 7.1 Proportions of monolignols in plants.
                   p-Coumaryl alcohol              Coniferyl alcohol      Sinapyl alcohol
Softwood                     <5                          >95              Trace amounts
Hardwood                    0–8                         25–50                 46–75
Grasses                     5–33                        33–80                 20–54



lignin, found in coniferous trees, contains predominantly repeat units of coniferyl alco-
hol with very little trace of sinapyl alcohol whereas hardwood lignin, present in the
angiosperms or broad leaf trees, contains both of these monolignols in significant pro-
portions (however often the ratio of sinapyl alcohol to coniferyl alcohol may be as high
as 3:1). Neither hardwood nor softwood lignin contains high levels of p-coumaryl alco-
hol which is found in grass lignin along with both the coniferyl and sinapyl alcohol
lignin monomers.
   Now that the basic units of lignin have been defined, the way in which these monomers
are inter-connected should be discussed. Analysis of isolated lignin samples over the
past decades has led to many advances in knowledge of the prevalence of the various
covalent linkages present in the three-dimensionally branched lignin macromolecule.
However, it is very important to note that it is still beyond the realms of possibility to
directly study naturally occurring lignin in its unaltered form although some extraction
methods are more benign than others. All investigations of lignin isolated from native
sources result in some form of chemical modification of the three dimensional network
through cleavage of bonds from other lignin monomers or from the covalent attachments
to polysaccharides. In this chapter, we will limit the discussion to what is known about
types of bonding interactions and only mention in passing the biosynthetic pathways for
the native production of the lignin macromolecules as there is continuing conjecture as
to whether there is full biochemical control over the monolignol polymerisation or if the
coupling reactions proceed in a random fashion (Boerjan et al. 2003, Vanholme et al.
2008, Lewis 1999). Such a discussion is beyond the scope of this chapter focussing on the
surface properties and possibilities for use in nanotechnology from sustainable sources
so the interested reader is referred to other reviews which discuss the relative merits of
the ‘random’ and ‘directed’ synthetic pathways (Vanholme et al. 2008, Lewis 1999).
   It is commonly accepted that lignin is synthesised through the enzymatic dehydro-
genation of the monolignols, transported to the cell wall, to radical species followed
176     The Nanoscience and Technology of Renewable Biomaterials

by the subsequent radical coupling eventually resulting in polymerisation to form lignin
(Chakar and Ragauskas 2004). Whilst this basic mechanism for the formation of the
phenolic radicals can be agreed upon, the actual processes involved in the polymeri-
sation of these coupled dimers and oligomers, whether random or controlled, is the
subject of ongoing investigation and debate. The coupling of the radicals to the grow-
ing lignin chain proceeds in an essentially combinatorial fashion and, depending on the
species (and hence the prevalence of monolignol building blocks), the chemical linkages
between the monomer radicals vary significantly. In both hardwood and softwood, the
highest proportion of inter-unit bonds occur through coupling of the phenolic radicals of
the β carbon with the methoxyl group at the 4 position of the phenylpropane structure
as shown in Table 7.2. This β-O-4 bond typically accounts for approximately half of


                                                                        ¨ ¨
Table 7.2 Common linkages between monolignols in lignin. Adapted from Sjostrom (1998).
Bond                 Chemical structure               Percentage of total bond linkages

                                                     Softwood                 Hardwood
β-O-4                HO          MeO                    50                        60
                                 O




                               OMe
                       OH

α-O-4                          OH
                                     MeO               2–8                         7
                                     O




                    MeO
                            OH

β-5                                                    9–12                        6
                      HO


                                             OMe
                                     O



                                 OMe
                          OH

5-5                            OH                     10–11                        5




                    MeO                       OMe
                               OH        O



                                                                      (continued overleaf)
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience       177

Table 7.2 (continued).
Bond                 Chemical structure                 Percentage of total bond linkages

                                                      Softwood                  Hardwood
4-O-5                       OH                             4                          7




                                     O
                    MeO
                            OH       MeO

β-1                    HO                OMe               7                          7
                                             O




                                 OMe
                          OH

β-β                                      O                 2                          3
                               MeO



                            HO




                                     OMe
                               OH




all inter-unit bonds in the lignin macromolecule and this bond is particularly susceptible
to nucleophilic attack, a fact which is exploited in the kraft pulping process (Gellerstedt
and Lindfors 1984).



7.3     Isolation of Lignin from Wood, Pulp and Pulping Liquors

In 1838, the Frenchman Anselme Payen treated wood with nitric acid followed by an
alkaline solution. Payen discovered that some of the wood dissolved by this procedure
                                                                            e
had a higher carbon content than the insoluble fibrous material; ‘la mati` re incrus-
tante’. This material was nineteen years later designated lignin by the German scientist
F. Schulze. In 1890 the Swede Peter Klason found an analytical method based on treat-
ment with strong sulphuric acid to determine the lignin content in wood. In 1897 he
studied lignin’s composition in sulfite spent liquors and found that it was chemically
related to coniferyl alcohol. Somewhat later Klason also proposed that lignin was a
macromolecular substance and that the coniferyl alcohol units were coupled together
178   The Nanoscience and Technology of Renewable Biomaterials

by ether linkages. Today we know that wood, depending on the tree species, part
of the tree and growing conditions, consists of roughly between 20 and 40% lignin
   o o
(Sj¨ str¨ m).
   Perhaps the greatest difficulty in the study of naturally occurring lignin is that it is
not possible to comprehensively investigate this material in situ. As such, the majority
of advances in the understanding of this important biomacromolecule have come from
investigations into the isolated lignins from various natural sources as well as from syn-
thetically prepared phenylpropane based polymers. The elucidation of structural units
and types of bonding between monolignols has come from the analysis of lignin poly-
mer fragments using sophisticated techniques such as nuclear magnetic resonance, mass
spectrometry and spectrophotometric techniques to name a few. All of these techniques,
however, rely on some knowledge of the starting raw material (i.e. from where was the
lignin isolated) and also how the lignin fragments were isolated.
   There are numerous techniques for isolating lignin on vastly different scales. For
example, small quantities of lignin with only relatively minor chemical changes can be
prepared by the ball milling of wood samples whereas high quantities of kraft lignin
and lignosulfonates are produced as a by-product during the chemical pulping of wood
fibres.
   To characterize and make further use of lignin it is necessary to separate the lignin
from the wood material. It is generally considered that this can only be done by more
or less changing the lignin chemical structure to different degrees, often with significant
reductions in molecular weight and introduction of non-native chemical species. In
Payen’s first experiment the carbohydrate matrix as well as the lignin was partly oxidized
and degraded by the nitric acid treatment. This gave a lignin of lower molecular mass
that was richer in carboxyl and phenolic groups than in the native state, and thereby
more easily dissolved during the following alkaline extraction. From those days and up
to now several different routes and industrial processes for the removal and isolation of
lignin from wood, pulp fibres and spent liquors have been developed. Some of these
different isolation methods along with the advantages and possibility of using these
techniques in the bio-refining of lignin are presented in the subsections below. Ideally,
any method that produces a well defined lignin material which can be constantly prepared
in a reproducible fashion will lend itself to further use in value added applications in the
future bio-refinery. This perhaps is a goal that might seem difficult to achieve when the
starting material is a branched, three-dimensional amorphous macromolecular structure
that appears randomly polymerized.

7.3.1 Isolation of Lignin from Wood and Pulp Fibres
Amongst the methods used for isolation of lignin from wood, the ones referred to below
are considered to give lignin substrates essentially unchanged in their chemical struc-
tures. The first method described involves extraction with organic solvents and was
suggested by Brauns (Brauns 1939) and gave very low yields. Due to this it is thus
                                                                                o
doubtful if the Brauns lignin is fully representative for the lignin in wood. Bj¨ rkman
introduced grinding as a pretreatment before toluene extraction to yield what today is
                                            o                         o
defined as milled wood lignin (MWL) (Bj¨ rkman 1956, 1957, Bj¨ rkman and Person
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience     179

                                                                                 o
1957). Later, Lundquist et al. (1977) modified the purification steps of Bj¨ rkman’s
original protocol to reduce the risk of solvolytic reactions. The expected yield of the lat-
ter MWL method is typically about 25% of the Klason lignin content. The MWL yields
can be improved by treating the finely ground wood meal with cellulolytic enzymes
that remove the associated polysaccharides before the solvent extraction and a frac-
tion named Cellulolytic Enzyme Lignin (CEL) was first isolated by Pew (Pew 1957,
Pew and Weyna 1962). By using an enzymatic preparation with even greater cel-
lulolytic and hemicellulolytic activities, Chang et al. (1975) extracted the insoluble
residue obtained after enzymatic hydrolysis successfully with 96% and 50% aqueous
dioxane at higher total yields than MWL and CEL. The content of carbohydrates in
the lignin fraction soluble in 50% dioxane was however twice as high as with the
former methods. Recently, a novel lignin isolation procedure (EMAL) was proposed
(Argyropoulos et al. 2002, Wu and Argyropoulos 2003). Starting with an initial mild
enzymatic hydrolysis of milled wood to remove most of the carbohydrates, followed by a
mild acid hydrolysis stage to cleave the remaining lignin–carbohydrate bonds, significant
improvements in yield and purity are obtained (Guerra et al. 2006a). The molecular
mass of the EMAL is also substantially increased compared to MWL and CEL and the
weight-average molecular mass (Mw ) is in the range of 30–60 kDa dependent on the
wood-species of origin. One other way of increasing the yield is through very extensive
milling. This might improve the yield to around 50% of the Klason lignin. The risk of
introducing severe chemical modifications is however considered overwhelming (Guerra
et al. 2006b).
   The extracted MWL’s morphological origin has been discussed in the literature. Lai
and Sarkanen (1971) suggested that MWL mainly originates from regions adjacent
to the middle lamella, whereas, for example Whiting and Goring (1981) as well as
Terashima et al. (1992) and Maurer and Fengel (1992) found that MWL from spruce
mostly descents from the secondary wall of the tracheids, which is a general opinion
today.
   Regarding ways of isolating residual lignin from chemical pulp fibres methods based
on acidolysis (extraction with 1,4-dioxane under acidic conditions, Gellerstedt et al.
1994) or repeated enzymatic treatment and precipitation similar to the CEL and EMAL
procedures described earlier, are found useful. Moreover, a tuned combination of LiCl
and dimethyl acetamide (LiCl-DMAc) can be used to dissolve both hardwood and soft-
                                                                              o
wood kraft pulp fibre components (Westermark and Gustafsson 1994, Sj¨ holm et al.
1999a-b).The solvent is exceptionally good for cellulose, but lignin can also be dissolved
in lower concentrations. This protocol has mostly been applied for analytical purposes,
e.g. in determining the molecular mass distributions of wood polymers in chemical
pulps by size exclusion chromatography but also in the preparation of cellulose micro-
spheres and model films; however it can be used together with chromatography to isolate
the lignin.
   In 2003 Lu and Ralph published a paper that described a method for dissolution
of ball-milled wood in dimethylsulphoxide–tetrabutyl ammonium fluoride and dimethyl
sulphoxide–imidazole binary solvent systems (Lu and Ralph 2003). Recently, further
improvements of this procedure were suggested by Fasching et al. (2008).
   Protocols involving dissolution of wood by the use of ionic liquids have lately
                         a                                          a
been developed (Kilpel¨ inen et al. 2007, Pu et al. 2007). Kilpel¨ inen et al. found that
180    The Nanoscience and Technology of Renewable Biomaterials

softwoods (sawdust or thermomechanical pulp fibres) can be completely dissolved up to
a concentration of 8 wt% and the most efficient solvents tested were 1-butyl-3-methyl-
and 1-allyl-3-methylimidazolium chloride. These novel methods are judged as very
promising to yield relatively unmodified lignins that can be considered representative
for the overall composition in the wood sample. Ionic liquids could be expected to be
of great help in the future to increase the understanding of lignin biosynthesis, structural
characteristics and in investigations of lignins’ interactions with other materials, as well
as in the development of usage and new products from lignin.

7.3.2 Isolation of Lignin from Spent Pulping Liquors
The main source of lignin readily available for use in larger scale comes from spent
pulping liquors after chemical liberation of wood fibres. By far the most important
pulping methods are kraft and sulfite cooking and only the lignin recovered from these
processes will be discussed in this section.

7.3.2.1 Kraft Lignin
Lignin isolated from the black liquor remaining after the chemical pulping of wood using
hydroxide and hydrogen sulphide is known as kraft lignin. This material accounts for
the overwhelming majority of lignin produced worldwide with up to 50 million tonnes
produced annually. Currently, by far the greatest use of kraft lignin is in fuel applications
where the burning of lignin produces more than sufficient energy for the powering of
the pulping plant. The chemical composition of kraft lignin is influenced by a number
of factors including wood species and cooking conditions but typical molecular weights
of the lignin polymers (and oligomers) are in the range of 1–5 kDa with a relatively
high polydispersity (MW /Mn ∼ 3–4). The kraft process is relatively harsh on the native
lignin structure leading to significant depolymerisation. Furthermore, the nucleophilic
attack by the hydrogen sulphide ion on the β-O-4 ether linkages results in a significant
increase in the charge on the macromolecules with up to 13% of phenolic monomers
containing a carboxyl group (compared to MWL which is virtually uncharged). The
presence of the charged functional groups increases the solubility of lignin substantially
under alkaline conditions with the solution properties of kraft lignin to be discussed in
the next section.
   Kraft lignin is usually isolated by precipitation of the spent liquor (black liquor) that
remains after cooking through acidification, where sulphuric acid, hydrochloric acid but
                               ¨
also carbon dioxide is used (Ohman et al. 2007). During the acidification large amounts
of gaseous dihydrogen sulfide are released. Depending on the pH value to which the
black liquor is acidified, different composition and yield of the lignin is obtained. If the
liquor is brought to pH 2 in one step more or less all of the lignin, including the highly
charged low molecular fraction, is precipitated. Kraft lignin may also be separated in
relatively large scales by ultrafiltration of the black liquor and precipitated as above
(Wallberg et al. 2003). Thereafter the precipitated kraft lignin is washed and finally
filtered and dried. Further purification to remove extractives and carbohydrates is done
according to methods described elsewhere.
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience      181

7.3.2.2 Lignosulfonates
A hydrosoluble form of lignin is prepared from the by-products of the sulfite pulping
process. During the pulping which involves the use of solutions containing sulphur diox-
ide and hydrogen sulfite ions at elevated temperatures, the lignin becomes sulfonated
and as such is soluble in water and under a range of aqueous solution conditions. The
weight-averaged molecular weight of the lignosulfonate polymers is clearly above kraft
lignin and typically in the range of 5–400 kDa, and the polydispersity is at least as
high (Buchholz et al. 1992, Fredheim et al. 2002). The most widely used industrial
process is the Howard process, in which a yield of 90–95% of precipitated calcium
lignosulfonates can be obtained by adding an excess of calcium hydroxide. Ultrafiltra-
tion and treatment with a long-chained alkyl amine to form a water insoluble complex
extractable with organic solvents can also be used to separate lignosulfonates from the
spent pulping liquid.
   Lignosulfonate, the far most extensively used technical lignin polymer, is utilized in
many applications including as a binder in ceramics and in animal feed or as a dispersant
through the steric stabilisation of particulates such as clays, dyes and pigments. As
the lignosulfonate macromolecule is water soluble, this class of polymer shows great
promises in future nanotechnological and surface chemistry applications beyond those
where it is already finding use.



7.4   Solution Properties of Kraft Lignin

From studies of the reaction pattern in the three different process phases of the kraft
cook, four famous principles for selective kraft pulping was established at the Swedish
Pulp and Paper Institute (STFI) and the Royal Institute of Technology (KTH) in the
mid-1970s–mid-1980s (Hartler 1978, Johansson et al. 1984).
1. The alkali profile should be levelled out, which means that a high hydroxide concen-
   tration in the beginning of the cook should be avoided and that the hydroxide ion
   concentration during the final part of the pulping should be increased.
2. The concentration of hydrogen sulfide ions should be held as high as possible at the
   beginning of the pulping.
3. The concentration of sodium ions and dissolved lignin should be low throughout the
   pulping.
4. The temperature should be relatively low.
One principle (II) is directly attributed to the nucleophilic cleavage of aryl-ether bonds in
the lignin structure that is an organochemical matter. The other three principles describe
the influence of parameters, which usually are of physicochemical importance in many
macromolecular systems. The analogies between the impact of these principles on the
kraft cook and the kraft lignin (KL) solution behaviour at various solution conditions
are found very interesting.
   A polymer dissolves spontaneously in a solvent if the free energy of mixing ( Gmix )
is negative. The phase transition is mainly due to increased configurational entropy of
the polymer chain or, for charged polymers, an increase in counterion entropy, due to
182    The Nanoscience and Technology of Renewable Biomaterials

                                                    1.0




                           Phenolic groups per SU
                                                    0.8

                                                    0.6

                                                    0.4

                                                    0.2

                                                          1000           10,000
                                                           M (g mol−1)

Figure 7.2 Phenolic groups per structural unit (SU) versus molecular mass for softwood kraft
lignin. The full curve represents the outcome of the computer calculated Guarana model
(Jurasek 1995). The broken curve is a fit to the values calculated from conductometric
titration (◦) and the number-average molecular weight. The data given by filled circles (•), are
calculated based on the corresponding phenolic content given by the full curve but taking
into account the molecular mass distributions of the samples represented by the empty circles.
                                                        ¨
Reproduced with permission from Norgren and Lindstrom (2000a). Copyright (2000), Walter
de Gruyter.

dissociation. Since the number of counterions of a polyelectrolyte generally is large, the
latter is often most important to consider and it explains why polyelectrolytes usually
                                                 o
are more soluble than uncharged polymers (J¨ nsson et al. 1998).
   After depolymerising the lignin in the fibre wall the KL is solubilised mainly through
dissociation of phenolic groups, due to the alkaline conditions in the digester. The KL
fragments formed are widely polydisperse, both chemically and physically (see Figure 7.2).
For example, generally high molecular weight KL fragments might be considered having
their pKa’s at much higher levels than low molecular species (Norgren and Lindstr¨ m     o
2000b). For polyelectrolytes carrying weakly acidic groups, the dissociation and thus the
solubility is governed by an increase in hydroxide ion concentration. This is also the case
for KL:s. The pKa value of coniferyl alcohol, the most frequent structural unit in softwood
lignin, is 10.25 at room temperature, as calculated from Hammet equation (Perrin 1981).
   When the temperature is elevated in a system containing neutral electrolytes, the
solubility of the salt increases due to the increased entropy. This is often also valid
for polyelectrolytes in aqueous solutions. However, concerning polyelectrolytes bearing
weakly acidic groups, the explanation is not as straightforward as it may seem. In
Table 7.3, data of the dissociation behaviour at different temperatures of some phenolic
substances are presented.
   For all substances investigated the pKa values decrease as the temperature increases,
normally indicating increased dissociation. At the same time, the negative logarithm of
the ion product constant of water, pKw decreases even more, see Figure 7.3. Due to
that, the net dissociation (α) will decrease when the temperature is elevated.
   The polydispersity of KL will of course also introduce differences in the solubility and
colloidal stability characteristics within the macromolecular distribution of fragments, see
Figure 7.4.
   Numerous studies dealing with the colloidal behaviour of lignin derivatives have earlier
                                                         o
been presented in the literature (Junker 1941, Lindstr¨ m 1980, Sarkanen et al. 1982,
    Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience              183

Table 7.3 Calculated and measured pKa , pKw and apparent pKa values for simple phenolic
substances, water and KL samples.
Sample                                                                     pK values

                                           20 ◦ C              50 ◦ C        70 ◦ C    125 ◦ C   175 ◦ C
Phenol†                                    9.97                9.62           9.47       –         –
Phenol∗                                    9.92                9.59            –         –         –
2-hydroxymethylphenol∗                     9.91                9.53            –         –         –
Coniferyl alcohol                         10.2556              9.94‡          9.74‡    9.32‡     9.03‡
Indulin AT, 0.01 M NaCl†                  10.6                10.2           10.0        –         –
Indulin AT, 0.1 M NaCl†                   10.4                10.1            9.8        –         –
Indulin AT, 0.01 M NaCl‡                  10.54               10.31          10.07       –         –
Indulin AT, 0.1 M NaCl‡                   10.47               10.25          10.00       –         –
Lignin, 8 kDa‡                            10.73               10.43          10.25     9.87      9.64
∗
  Zavitsas 1967
†
  experimentally obtained
‡
  calculated




                                               kwater = d(−ln Kw)/d(1/T)
                                     30

                                               kconiferyl alc. < kwater
                            − ln K




                                     20

                                               kHAc < 0

                                     10

                                             2.4                2.8        3.2
                                                        T−1 × 103 (K −1)

Figure 7.3 Van’t Hoff plots for (•) water, (◦) coniferyl alcohol and ( ) acetic acid. As long
as the slope of the water curve (kwater = Hwater /R) is the steepest, a net decrease in the
dissociation will appear at temperature elevation. If ki < 0, the pKa value will increase by
increasing temperature.



Woerner and McCarthy 1988, Rudatin et al. 1989, Norgren et al. 2001ab, Norgren et al.
2002). For decades it has been known that in presence of high concentrations of monova-
lent metal ion salts at pH neutral conditions, KL starts to coagulate and finally precipitate
(Junker 1941). Moreover, by elevating the temperature in alkaline KL solutions the disso-
ciation of phenolic groups decreases, sometimes to levels below the threshold of solvency
                        o
(Norgren and Lindstr¨ m 2000b, Norgren et al. 2001b). From the work of Lindstr¨ m        o
(1980) it was found that colloidal KL, in conformity with other colloids, exhibits a
strongly marked critical coagulation concentration of added electrolytes (CCC). More
recent, Norgren et al. (2001b) showed that the experimentally observed phase behaviour
of colloidal KL can be described reasonably well by a theoretical approach derived from
the DLVO-theory.
184    The Nanoscience and Technology of Renewable Biomaterials


                                                               0.20 M (p)
                                                               0.50 M (s)
                                                               0.75 M (s)
                                                               1.0 M (s)




                          A.U.
                           0
                           10−11       10−10            10−9                10−8
                                           D   (m2   s−1)

Figure 7.4 Log-normal distributions of self-diffusion coefficients on some sample super-
natants (s) and one precipitates (p) obtained at different NaCl concentrations. The curve
showing the lowest mass-weighted median self-diffusion coefficient is obtained from mea-
surements on a re-dissolved KL precipitate. The KL macromolecules in the supernatants show
increasingly faster self-diffusion, indicating a decrease in molecular weight due to precipita-
tion as the ionic strength of the sample solutions increases. Reproduced with permission from
Norgren et al. (2001a).


   Self-aggregation of colloidal particles into larger clusters has been subjected to serious
scientific studies for more than a century. For aggregation due to Brownian motion, two
well-defined limiting regimes of kinetics have been identified; DLCA and RLCA (Leath
and Reich 1978, Weitz et al. 1987, Weitz et al. 1991, Julien and Botet 1987, Lin et al.
1989, Lin et al. 1990ab, Hildago-Alvarez et al. 1996). The rapid diffusion-limited
cluster(colloid)-cluster(colloid) aggregation is the result of negligible repulsive forces
between the colloidal particles, following the von Smoluchowski equations, and thus
causing particles to stick upon contact and to form loosely jointed and highly dis-
ordered structures. In case of reaction-limited cluster(colloid)-cluster(colloid) aggre-
gation, several collisions are possible before the particles finally aggregates since the
sticking probability is much lower as a result of a substantial repulsive force (electro-
static, electrosteric) between the particles. The creation of somewhat denser aggregates
is characteristic in the RLCA regime. It has further been shown that the described
processes are universal in the sense that they are independent of the detailed nature
of the colloid, if the essential physical interactions are the same (Lin et al. 1989).
The mentioned two classes of aggregation processes and their crossover behaviour are
suggested to be sufficient to describe the complete range of kinetic aggregation (Lin
et al. 1990b).
   Aggregation kinetics are often quantified in terms of stability ratios, W , defined as
the ratio of the rate constant for DLCA to the experimentally determined rate constant
                                                                                  o
for formation of doublets (Reerink and Overbeek 1954, Evans and Wennerstr¨ m 1994).
As the ionic strength in the system increases, the stability ratio approaches unity, which
is where the CCC of an electrolyte is most strictly defined. A theoretical W can be
calculated by integration of an assumed total interaction potential, which might be derived
from the DLVO-theory (Reerink and Overbeek 1954, Evans and Wennerstr¨ m 1994).  o
   Figure 7.5 shows the kinetics of KL aggregate formation and growth, as followed
by quasi-elastic light scattering (QELS). The measurements were performed at 70 ◦ C,
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience                          185

                                                    2000       1.3 M


                                                    1500




                        Eff. D (nm)
                                                                    1.2 M

                                                    1000
                                                                                 1.1 M
                                                     500
                                                                                                1.0 M

                                                       0
                                                           0           10      20         30            40
                                                                             Time (min)

Figure 7.5 Hydrodynamic diameter of KL aggregates as a function of time and sodium
chloride concentration at 3.2 10−4 M OH− (pOH 3.5) and 70 ◦ C. Reprinted with permission
from Norgren et al. (2002). Copyright (2002), American Chemical Society.


                                                    105
                            W-ratio (kfast/kslow)




                                                    104

                                                    103
                                                                                                 CCC
                                                    102

                                                    101          QELS
                                                                 Turbidity
                                                    100
                                                           0.1                            1.0
                                                                             [NaCl] (M)

Figure 7.6 Stability ratio of unfractionated KL as a function of NaCl concentration at pOH
3.5 and 70 ◦ C. The CCC is 1.3 M. Reprinted with permission from Norgren et al. (2002).
Copyright (2002), American Chemical Society.


alkaline conditions and different sodium chloride concentrations, as indicated in the
figure legend. At higher ionic strengths, very large aggregates are formed. Consequently,
some samples are found to be settling during the time of experiment. In the interval
1.3 ≤ [NaCl] < 1.5 M, the aggregation curves are overlapping. At a sodium chloride
concentration of 1.5 M, samples start to phase-separate already at room temperature.
   The results obtained from the early-time kinetic data in Figure 7.5 as well as some
supplementary results from turbidity measurements were used in the calculations of sta-
bility ratios (W ) for the Indulin AT system at 3.2 10−4 M OH− (pOH 3.5) and 70 ◦ C.
In Figure 7.6, a plot of the W -ratio as the function of sodium chloride concentration is
displayed. The onset of KL aggregation (coagulation) is found in the interval 0.2–0.7 M.
Sometimes, the electrolyte concentration at the onset of aggregation is defined as CCC
         o
(Lindstr¨ m 1980). In its strictest definition, however, the CCC is obtained at the intersec-
tion between still reaction-limited and purely diffusion-limited cluster-cluster aggregation
                                                      o
(Reerink and Overbeek 1954, Evans and Wennerstr¨ m 1994).
   As mentioned earlier, the RLCA and DLCA aggregation processes are universal, and
known to give aggregates of fractal geometry (Lin et al. 1989). This is also the case
186    The Nanoscience and Technology of Renewable Biomaterials




                                                                                500 nm
      0.2 µm


Figure 7.7 To the left; Cryo-TEM of fractal cluster of kraft lignin (reprinted with permission
from Norgren et al. 2002. Copyright (2002), American Chemical Society). To the right;
self-aggregated gold colloids (Weitz et al. 1987. Reprinted with permission from John Wiley
& Sons, Inc.).

during KL aggregation, as can be viewed in Figure 7.7 (left). The resemblance between
the lignin aggregate to the left and the fractal gold colloid cluster to the right is striking.
   In colloid science, the analysis of mass fractal dimensions of aggregates has shown to
be a good discriminator between different aggregation processes. For DLCA, the mass
fractal dimension often is found to be around df ≈ 1.8, while in the case of RLCA
aggregate df is usually situated around 2.1 (Weitz et al. 1991). Much concern has
been devoted to RLCA, due to the existence of a stability threshold in this regime.
The DLVO-theory, which divides the interaction forces into one attractive part (van
der Waals forces) and one repulsive part (the Columb forces), has been a great source
of understanding RLCA (Reerink and Overbeek 1954, Evans and Wennerstr¨ m 1994).    o
Additional stabilising effects such as steric stabilisation might however also be attributed
(Napper 1983). Electrosteric stabilisation, which is a combination of both electrostatic
and configurational entropic repulsive forces between colloidal particles, gives sometimes
an explanation of why a colloidal dispersion still is stable at high ionic strengths and
elevated temperatures.
   By fitting data from Figure 7.5 to R ∝ t −d f , where R is the cluster radius at time t,
the fractal dimensions, df , of the clusters are obtained (Hoekstra et al. 1992). Figure 7.8
shows a plot of df as a function of the W -ratio. According to Kim and Berg (2000), the
outcome suggests that it is reasonable to assume that the W -ratio can be thought as the
common denominator for fractal aggregation in KL systems.
   Due to the chemical and physical heterogeneity of KL, self-aggregation in KL systems
is complex. The presence of larger (KL) macromolecules is proposed to determine the
onset (Leubner 2000). Depending on the KL sample composition, nuclei may either be
present from the beginning or are being formed due to changes in the solution conditions.
Figure 7.9 illustrates the probable modes of KL aggregation as proposed by Norgren
et al. (2002).
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience                                                                  187




                                                       Fractal Dimension, df
                                                                               2,4



                                                                               2,2



                                                                               2,0

                                                                                           1                  10
                                                                                               Stability Ratio, W

Figure 7.8 Fractal dimensions as a function of stability ratio. The straight line in the figure is
derived from a linear regression of the data points and follows: df = 2.04 + 0.324 log10 (W ).
Reprinted with permission from Norgren et al. (2002). Copyright (2002), American Chemical
Society.




                                                                                                           KL self-associates


                  MeO       OH

                                            CH2
                                                                               OMe
       HOH2C          HC         OMe
                                              O       C C                        CH2
                 CH                                   H H
                                 OH
             O                              OH
                                                      OMe
      MeO                                    O
                                                           OH
                             CH2OH      C=O
                                                      HC
             CH2                                                                     OMe
      HOOC             OH                              CH                       OH
             MeO            HC                    O
                             CH
                 OH                     OH                               OH
       MeO                                             OH
                                                  CHOH
                                  HO              CH
              HC       CH                         CH2OH
                                      OMe
             S CH      COOH
                 CH2OH




                 macromolecular KL                                                               colloidal KL particles         fractal KL cluster

Figure 7.9 A schematic representation of the modes of aggregation in kraft lignin systems
starting from macromolecular kraft lignin and finally reaching fractal kraft lignin clusters.
Reprinted with permission from Norgren et al. (2002). Copyright (2002), American Chemical
Society.


7.5     Surface Chemistry of Solid State Lignin

Lignin, as stated previously, is difficult to describe in terms of a single macromolecular
chemical structure with a defined repeat monomer unit connected in a uniform way.
Thus, no two lignin polymer chains will be alike. Furthermore, as solid state lignin
is amorphous due to the irregular branching and molecular weight of the biomacro-
molecules, structural heterogeneity also persists. This poses a significant problem in the
study of the fundamental physicochemical properties of lignin which has part way been
overcome by the preparation of model lignin surfaces. The goal of any model surface
188    The Nanoscience and Technology of Renewable Biomaterials

is to closely mimic the naturally occurring material, however, with the reduction or
elimination of the chemical and structural heterogeneity. This may be achieved through
the use of a well characterised sample of lignin (either isolated from naturally occurring
materials or through synthetic means) and modern techniques in the preparation of thin,
smooth and continuous films. In much the same way as model films of the predominant
carbohydrate structure in wood fibres, cellulose, have been prepared and used in the
past decade for the measurement of fundament physical and chemical properties (Notley
                           a
et al. 2004, Notley and W˚ gberg 2005, Notley et al. 2006, Kontturi et al. 2006, Eriks-
son et al. 2007, Notley 2008), so too have lignin films been prepared recently by a
number of research groups using slightly different methodology, bearing in mind the
underlying goals of achieving chemical and structural uniformity.

7.5.1 Preparation and Properties of Lignin Thin Films
Many different types of model lignin surfaces have been prepared for use in the measure-
ment of fundamental properties of lignin in the solid state. Table 7.4 summarises these
recently published studies. The first model films were prepared by Lee and Luner (1972)
from a commercial sample for the use in investigating the wettability and interfacial prop-
erties of lignin. It has long been speculated that lignin provides a hydrophobising means
for the wood cell wall. The contact angle of water on these lignin model surfaces was
measured and showed a rapid decrease with time from about 60◦ to 0◦ indicating the
highly porous nature of the films and that the lignin polymer has a strong affinity for
water. Whilst providing the first experimental data of the surface properties of lignin
and in particular, a relative measure of the hydrophilicity of lignin, films prepared in this
manner are far from ideal for the calculation of surface energy. The surfaces in this study
were prepared by either evaporation of a drop of lignin solution on a microscope slide,
resulting in extensive cracking, or through heat moulding of the lignin powder under
pressure to give smoother surfaces. A key point is worth reinforcing here; a model
surface must closely mimic the naturally occurring material. Severe chemical methods
for the isolation of lignin will significantly alter the natural structure limiting the utility
of the ‘model’ surface.
   Subsequent studies have attempted to improve the quality of the lignin surfaces in
order to overcome these initial limitations. However, a number of problems have per-
sisted including high surface roughness, nonuniformity including discontinuous films and
instability, particularly in aqueous solution conditions. Constantino et al. (1996, 1998,
2000) have used the Langmuir-Blodgett technique to prepare model lignin films for over
a decade now and have extended their use to sense heavy metals (Martins et al. 2008).
The group of Micic et al. have investigated the use of model lignin macromolecules in
determining the interactions between lignin globules thus not requiring a smooth, con-
tinuous film over a large area (Micic et al. 2001a,b). Spin-coating seems particularly
promising for the reproducible preparation of lignin model films which are both smooth
and continuous. Norgren and co-workers have prepared smooth, continuous lignin films
by the spin-coating of softwood kraft lignin dissolved in ammonium hydroxide solu-
tion onto oxidised silicon wafers (Norgren et al. 2006, 2007, Notley and Norgren 2006
and 2008). These lignin surfaces have been optimised in terms of their thickness with
reproducible films prepared in the range of 30–150 nm with minimum roughness as
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience     189

Table 7.4 Previously published methods for the preparation of lignin model surfaces on
various substrates.
Film preparation           Lignin source(s)            Substrate           References
method
Evaporation, heat     Softwood (Indulin AT,         Glass             Lee and Luner 1972
  molding               dioxane lignin,
                        periodate lignin),
                        hardwood (REAX 31)
Evaporation,          ZL-DHP (synthesized)          Glass             Micic et al. 2000,
  spin-coating                                                          2001a, 2001b
                                                                        and 2004
Langmuir-Blodgett     Sugar cane bagasse            Glass, calcium    Constantino et al.
                        (acetosolv lignin), Pinus     fluoride           1996, 1998 and
                        caribaea hondurensis                            2000
                        (organosolv lignin)
Langmuir-Blodgett,    Sugar cane bagasse            Mica              Pasquini 2002 and
  evaporation           (saccharification lignin,                        2005
                        ethanol lignin, acetone-
                        oxygene lignin, soda
                        lignin)
Adsorption            Lignosulfonate                Glass, quartz     Paterno and
                        (commercial)                                    Mattoso 2001
Adsorption            Softwood (CURAN 100)          Mica              Maximova et al.
                                                                        2004
Spin coating          Softwood kraft lignin         Oxidized          Norgren et al. (2006
                                                     silicon wafer      and 2007), Notley
                                                                        et al. (2006),
                                                                        Notley and
                                                                        Norgren (2008)
Spin coating          Milled wood lignin            Polystyrene       Tammelin et al.
                                                                        (2006 and 2007)



demonstrated by the atomic force microscopy imaging shown in Figure 7.10. Typically,
the surface roughness of the lignin surfaces is less than 1 nm over a 1 µm2 image.
Macroscopically, the lignin films supported on the silica substrate are continuous over
greater than 1 mm. Furthermore, the ToF-SIMS analysis showed only a minimal amount
of solvent retained in the lignin layer and that the chemical integrity of the monolignols
was maintained.
   Importantly, lignin surfaces made using the methodology of Norgren et al. remain
intact upon exposure of the film to a range of aqueous solution conditions. This has
allowed advancement in the study of the physicochemical properties of lignin in pulp and
paper applications that has not been previously possible. To test the stability of the thin
films, kraft lignin surfaces prepared on silica wafer were subjected to various aqueous
electrolyte solutions. Low concentrations of NaCl had little effect on the thickness of
the films whilst only a minimal decrease in thickness was observed for concentrations up
to 0.1 M. Furthermore, no changes were observed for solution pH in the region of 6–9.
190                    The Nanoscience and Technology of Renewable Biomaterials




                                                   (a)                                                                                            (b)

Figure 7.10 AFM Tapping mode height images of a 55 nm thick lignin surface. (a) 5.0 ×
5.0 µm2 , RMS roughness 1.01 nm. (b) 1.0 × 1.0 µm2 RMS roughness 0.59 nm. Peak-to-peak
roughness is less than 10 nm on the 25 µm2 image. The surface roughness does not change
significantly with the thickness of the lignin film. Adapted with permission from Norgren et al.
(2006). Copyright (2006), American Chemical Society.

                                                                                                                             1600                                       pH 10           4
                     500       H2O   pH 3.5 pH 7 pH 8.5 pH 9   pH 9.5         2                                                                                 pH 10
                                                                                                                                                        pH 10




                                                                                                                                                                                             D Dissipation (x 106)
                                                                                  D Dissipation (x 106)




                                                                                                                             1400
  D Frequency (Hz)




                                                                                                          D Frequency (Hz)




                     400                                                                                                     1200
                                                                              1                                                                                                         2
                                                                                                                                                                                pH 11
                     300                                                                                                     1000
                                                                              0                                               800                                                       0
                     200
                                                                                                                              600
                     100                                                      −1                                              400
                                                                                                                                              pH 10                                     −2
                      0                                                                                                       200
                                                                              −2
                                                                                                                                0
                                                                                                                                                                                        −4
                           0     5     10    15    20    25    30   35   40                                                         0   20   40       60           80            100
                                              Time (minutes)                                                                                 Time (minutes)


Figure 7.11 The effect of alkalinity on the stability of a spin-coated lignin film as monitored
by QCM-D. The arrows indicate the given characteristics of the rinsing fluid introduced into
the measuring chamber. Adapted with permission from Norgren et al. (2007). Copyright
(2007), American Chemical Society.

A subsequent study using lignin films spin-cast on silica coated quartz crystals surfaces
used in the quartz crystal microbalance showed no effect on the uptake of solvent or
degradation of the film at pH less than 9.5 as shown in Figure 7.11. However, with
further increases to the solution pH, the films initially became swollen at pH 10 before
completely dissolving from the underlying substrate at pH 11. This study demonstrates
the limitation of these particular softwood kraft lignin films for use in other fundamental
investigations involving the interactions of solid state lignin. The stability of the kraft
lignin films closely resembles the solubility of the kraft lignin in solution, that is, at pH
greater than 10, the films delaminate from the silica supporting substrate and dissolve
into solution.
     Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience   191

7.5.2 Use of Lignin Thin Films for the Investigation of Surface
      Chemical Properties
Once lignin thin films can be prepared in a reproducible fashion, the possibility of
studying the physical and chemical properties of this material can be studied on a fun-
damental basis. Thus, a lignin surface which is continuous and smooth over a large area
lends itself to use in many surface specific analytical techniques and as such, a range of
interactions may be probed. Such studies have included the interaction of lignin with
other wood polymers such as cellulose as well as with other materials such as poly-
electrolytes, complexes and inorganic particles. As it has only been recently that such
well-defined surfaces have been prepared, the surface chemistry of lignin is still yet to
be fully investigated. Some of the recent studies are summarised below.
   A key property of lignin that has been quantified by the preparation of lignin films
is its surface energy and wettability. The surface energy of lignin is highly dependent
on a number of factors, not least the method of evaluation, the isolation of the polymer
from the wood, the tree species and the surface preparation. Lee and Luner (1972)
studied six different lignin preparations and observed no significant differences in their
wetting characteristics. Recently, Notley and Norgren (manuscript in preparation) have
undertaken a similar study investigating the differences in the surface energy components
between kraft lignin films and milled wood lignin films determined by measuring contact
angles with test liquids of varying polar and dispersive components according to the
method of Fowkes. Table 7.5 shows that while there is not a significant variation
between the samples in terms of the total surface energy, the polar contribution is much
greater for the kraft lignin films. This is expected as the kraft pulping process is known
to introduce a large amount of polar functional groups such as carboxyl groups through
cleavage of the β-O-4 ether linkages.
   Understanding the surface energy of lignin has a number of important implications,
particularly to the pulp and paper industry. A wood fibre that has a surface chemistry
rich in lignin will have a significantly different surface energy and hence wettability with
water to one that is predominantly cellulose thus having a dramatic effect on the devel-
opment of capillary forces during the drying and consolidation phases of paper-making
even though the total surface energies of lignin and cellulose are similar. The data for the
Figure 7.12 shows the contact angle that water makes with both the softwood kraft lignin
and softwood MWL films. It has long been suggested that one of the major functions
of lignin in the plant cell wall is to aid in waterproofing. However, for samples tested
in these studies, the contact angle is significantly less than 90◦ indicating that lignin is
far from being classified as hydrophobic. This fact is exploited by some plant species

Table 7.5 Surface energy, including polar and dispersive components, of model lignin
films.
Energy (mJm−2 )             Softwood kraft   Softwood MWL      Hardwood MWL         REAX 31∗
γT                                57.1           58.8                57.0             52.5
γd                                33.7           44.5                43.9             43.5
γp                                23.4           14.3                13.1              9.0
∗
    Data taken from Lee and Luner (1972).
192   The Nanoscience and Technology of Renewable Biomaterials




            (a)                            (b)                           (c)

Figure 7.12 Contact angle of water on lignin model films. (a) softwood kraft lignin film;
(b) softwood (Radiata Pine) milled wood lignin film and (c) hardwood (Eucalyptus Regnans)
milled wood lignin film.


where sourcing water has been an issue throughout evolution. Kohonen has shown that
water transport though the lumen of wood tracheids is aided by a combination of surface
chemistry, which is predominantly lignin, and surface structuring, the so-called warty
layer (Kohonen 2006). The surface structuring leads to a reduction in the contact angle
of water, which may be described theoretically according to Wenzel’s equation (Hunter
1993), increasing the wettability and hence transport of water due to the effect of the
Laplace pressure.
   The preparation of model surfaces also provides an excellent opportunity to study the
properties of single lignin macromolecules. This may be probed by stretching lignin
polymers away from the surface by some specific physisorption interaction with an
atomic force microscopy tip in a method which has been termed single molecule force
spectroscopy (Rief et al. 1997, Chattelier et al. 1998). Only a few polymer chains are
pulled away from the surface, with each chain causing an ‘event’ which can be observed
in the force-distance curve as shown in Figure 7.13 (Notley and Norgren, manuscript
in preparation). Typically, the shape of the events in the force-distance curve gives an
indication of the solvency of the polymer in the solvent into which it is being stretched
(Chattelier et al. 1998, Senden et al. 1998). A good solvent gives rise to ‘Langevin’
type interaction curves that may be described as an increasing, nonlinear adhesion force
with separation before desorption back to the baseline. The increasing adhesion is due
to the decrease in entropy as the polymer chain is stretched such that all bonds are
in the trans configuration. These curves may be fit using either a freely-jointed-chain
or wormlike-chain model to yield the contour length of the polymer as well as the
persistence length. A poor solvent interaction, as shown in Figure 7.13 for a DHP
lignin surface interacting at pH 6 and 1 mM NaCl background electrolyte concentration,
is characterised by a constant force plateau as a function of separation of the probe
from the surface. If more than one chain is pulled away from the surface, then a
plateau is observed for each chain, with the magnitude of the plateau corresponding to
an integer multiple of the lowest force. In Figure 7.13, 6 plateaus can be observed,
each a multiple of about 78 pN. If the surface energy of the solid polymer is known
(in this case 58 mJ m−2 ), then the molecular radius may be determined. For this DHP
                                                                ˚
lignin sample, the molecular radius was calculated to be 2.1 A which is of the order
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience    193

                       0   100     200         300         400          500           600
                0.00

               −0.01

               −0.02

               −0.03

               −0.04
  Force (nN)




               −0.05

               −0.06

               −0.07

               −0.08

               −0.09

               −0.10
                                         separation (nm)

Figure 7.13 Poor solvent interaction of a synthetic lignin polymer as probed using single
molecule force spectroscopy.


of what may be expected for a single linear polymer chain. Kraft lignin films were
                                                              ˚
also investigated and found to have a radius of nearly 20 A which indicates its highly
branched nature.
   The interaction of lignin with other materials has also been studied extensively. One
example is the surface forces between lignin and cellulose (Notley and Norgren 2006).
In this study, the measured potential energy of interaction was investigated as a func-
tion of the aqueous solution conditions. Softwood kraft lignin surface were used which
allowed the influence of a broad range of pH and ionic strength on the surface poten-
tial to be probed due to the stability of the films. The surface potential and forces of
interaction with cellulose were determined using the colloidal probe microscopy tech-
nique. In the pH range of 3.5 to 9 and for ionic strengths up to 0.01 M, the forces
could be well fit using the DLVO theory between the limits of constant charge and
constant potential as shown in Figure 7.14 (Deryagin and Landau 1941, Verwey and
Overbeek 1948, Chan and Horn 1985). As this theory could be applied to the experi-
mental data, the lignin films must behave similarly to solid state surface. Furthermore,
as a function of solution conditions, the surface potential of the kraft lignin films could
be determined. The surface potential increased as a function of pH, which may be
expected, through the successive ionisation of the carboxyl groups followed subse-
quently by dissociation of the phenolic functional groups on the lignin polymer. At
pH 8.5, the measured surface potential of the lignin films was −75 mV which corre-
sponds to a relative charge per area of 1 charge per 67 nm2 . At pH greater than 9.5,
where the kraft lignin becomes significantly charged resulting in reduced stability of the
194    The Nanoscience and Technology of Renewable Biomaterials

                                  0.50
                                                                                      0   20     40     60     80    100




                                                        Force/radius (mN/m)
                                  0.45                                           1

                                  0.40                                          0.1



            Force/radius (mN/m)
                                  0.35                                         0.01
                                  0.30
                                                                              0.001
                                  0.25                                                    apparent separation (nm)

                                  0.20
                                  0.15
                                  0.10
                                  0.05
                                  0.00
                                         0   20       40           60                                   80            100
                                                  apparent separation (nm)

Figure 7.14 Normalised force-distance curve for the interaction of cellulose sphere and kraft
lignin film at a pH of 8.5 in a background electrolyte of NaCl with concentration of 0.1 mM.
The data were fit to DLVO theory in the limits of constant charge (upper fit) and constant
potential (lower fit). The fitting parameters were ψcell = −3 mV, ψlig = −75 mV, κ −1 = 30 nm.
Inset shows the same data on a log-linear scale to demonstrate the exponential decay. Adapted
with permission from Notley and Norgren (2006). Copyright (2006), American Chemical
Society.


film, a short range steric force was observed consistent with the increase in solubility
of the polymer.
   The determination of the surface charge and the surface potential of the lignin films
as a function of aqueous solution conditions allow an understanding of the surface
interactions apparent in a number of industrial applications with a prominent example
being paper-making where wood fibres with lignin rich surfaces interact in aqueous
media. In the paper-furnish, there are many other components with which lignin may
interact, in particular, polymeric additives that may be used to improve strength properties
of the finished paper or to retain inorganic filler particles. Thus, knowledge of the
interaction of lignin surfaces with a variety of polymers, polyelectrolytes and soft-matter
complexes is of great importance. Lignin model films have been used to study the
adsorption of polyelectrolytes in terms of both the kinetics and surface excess under
variable solution conditions such as pH, ionic strength and polymer concentration.
   Polyelectrolytes of opposite charge to the kraft lignin films, that is, cationic poly-
electrolytes have been used in two studies. In the first, poly(allylamine hydrochloride)
(PAH), a polymer with weakly ionisable charged groups was used (Norgren et al. 2007).
There was an observed increase in adsorbed amount measured using the quartz crystal
microbalance as the pH of the solution was increased which is as expected for the
adsorption of a polyelectrolyte to an oppositely charged surface through purely ionic
interactions. Interestingly, though, in this study, a significant amount of an anionic
polyelectrolyte, poly(acrylic acid) (PAA) also adsorbed to the kraft lignin surface indi-
cating that nonionic interactions are also possible. When polyelectrolyte complexes of
PAH and PAA were prepared in different ratios which influences both size and charge,
both cationic and anionic charged complexes adsorbed to the surface as presented in
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience           195


                                1.00


                                0.75


                                0.50

                                                                     x 2.000 mm/div
                                0.25                                 z 500.000 nm/div


                                       100 nm                        2           4         6 mm
                                 0
   0     0.25   0.50   0.75   1.00
                               mm




       lignin-coated qcm crystal                PECs in solution   anionic PECs adsorbed



Figure 7.15 Schematic illustration of the apperance of (a) the lignin film on a QCM-D crystal
imaged by afm, (b) PECs in solution as caught by cryo-TEM, and (c) anionic PECs adsorbed
on a lignin-coated qcm crystal imaged by afm. Adapted with permission from Norgren et al.
(2007). Copyright (2007), American Chemical Society.

Figure 7.15. This showed that polyelectrolyte adsorption to lignin can be through both
ionic and nonionic interactions.
   A subsequent study to investigate the possibility of nonionic interaction was under-
taken with the same type of surfaces however with this time, a strong polyelectrolyte
whose charge density is effectively constant with pH, poly(diallyldimethylammonium
chloride) (PDADMAC) (Notley and Norgren 2008). In that study, it was shown that the
adsorption to the kraft lignin film agreed well with the Scheutjens-Fleer theory (Fleer
et al. 1993). The adsorbed amount decreased as the ionic strength of the adsorbing
polymer solution was increased. However, because of the problems surrounding the sta-
bility of the films at salt concentrations greater than 0.1 mM, no definitive conclusions
could be made on the likelihood of nonionic interactions.
   Tammelin et al. have used model lignin films to investigate the interaction of hemi-
cellulose and extractives with lignin (Tammelin et al. 2006 and 2007). In their studies,
a milled wood lignin from Norway Spruce was used as the raw lignin material and was
spin-coated onto polystyrene coated quartz crystals resulting in smooth films suitable for
measuring the adsorption of hemi-cellulose and extractives. The adsorbed amount of
hemi-cellulose onto the lignin surface was quite low. Furthermore, it was observed that
the layer conformation was soft and highly viscoelastic for hemi-cellulose isolated from
unbleached mechanical pulp but was rigid for material isolated from bleached pulp.
The observed difference was ascribed to the increased anionic charge density for the
hemi-cellulose from the bleached mechanical pulp which leads to flatter conformation
of the polymer at the interface in agreement with theories describing the adsorption of
highly charged polymers to solid surfaces.
196   The Nanoscience and Technology of Renewable Biomaterials

7.6 Lignin: Current and Future Uses

Starting from about the beginning of the 1920s, lignin has been prepared for use in larger
scales and as a raw material to produce other chemicals (McCarthy and Islam 2000). In
the beginning lignosulfonates from sulfite spent liquors were processed at industrial plants
to make vanillin and lignin preparations. In 1936 the Marathon Corporation began the
production of leather tanning agents and dispersants from lignosulfonates. Some years
later, in 1942, West Virginia Pulp and Paper Company (Westvaco) started to produce
kraft lignin products from softwoods and hardwoods. Some of these products (Indulin A
and C) were sold to the rubber, ceramic, and printing ink industries. In the late 1940s,
Puget Sound Pulp and Timber Co began the production of a fermented lignosulfonate
product, Lignosite, which was 1960 patented in the form of a chrome and ferrochrome
derivative as an additive to oil well drilling muds. For some sulfite mills that sold all
of their produced lignosulfonates, during the mid-seventies and some years beyond the
lignin business was actually more profitable than the pulp business.
   In 1990 the world production of lignin products was 138.5 thousand tons per year
(Lin and Zhong 1990). In the latter part of the 1990s, the worldwide amount of
lignin recovered and isolated from pulping processes and later sold totalled 1% of all
the lignin generated. Of this lignosulfonate was the main lignin derivative produced.
Larger companies involved in the production were Fraser Paper Inc., Georgia-Pacific
Inc., Nippon Paper Industries Ltd, LignoTech-USA, Borregaard Lignotech-Norway and
Tembec, Inc. In 1998, the lignosulfonate sold by Georgia-Pacific Inc. alone was 220,000
tons. Borregaard Lignotech-Norway and Westvaco also produced kraft lignin but part
of it was later sulfonated. The total sales in 1996 were estimated to around USD
600 million.
   Lignosulfonates are a very effective and economical adhesive, acting as a binding
agent in pellets or other compressed materials (Lignin Institute 2007). Used on unpaved
roads, lignosulfonate reduce airborne dust particles and stabilize the road surface. This
binding ability makes it a useful component e.g. of coal briquettes, bricks, plywood
and particle board, ceramics, animal feed pellets, carbon black, fibreglass insulation,
fertilizers and herbicides, linoleum paste, soil stabilizers etc. Due to their amphiphilic
nature lignosulfonate also stabilizes emulsions of immiscible liquids, such as oil and
water, making them highly resistant to breaking. Uses as emulsifiers in e.g. asphalt
emulsions, pesticides, pigments and dyes and wax emulsions are common. Metal ions
can be complex bound to lignosulfonates, preventing them from reacting with other
compounds and becoming insoluble. Metal ions sequestered with lignosulfonates stay
dissolved in solution, and are thus available to plants. Scaly deposits in water systems
can also be prevented through this. As a result, they are used in e.g. micronutrient
systems, cleaning compounds and water treatments for boilers and cooling systems.
This high cationic exchange capacity of lignosulfonates lends itself for use in measuring
heavy metals concentrations in sensor applications. Martins et al. (2008) have shown
that lignin films can be well utilised for exactly this application. Recently, Guo et al.
(2008) ranked the cation exchange capacity and found that the metal ion adsorption
is heavily dependent on pH, with higher metal concentrations detected at elevated pH
where some of the phenolic groups become dissociated.
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience    197

   Lignosulfonate has also been used successfully as an acid template in synthesis of
inherently conducting polymers (ICPs) through polymerisation of aniline or pyrrole
(Berry and Viswanathan 2002, Roy et al. 2002). One benefit from using lignosulfonates
is the increased dispersibility in a range of solvent including water. The final product
has been proven to possess corrosion protective ability when coated on metals, and is
also evaluated for use where electrostatic dissipative (ESD) materials are needed; e.g. in
sensitive electronic equipment, explosive materials, and when static electricity is gener-
ated in dangerous amounts. Ferromagnetic nanocomposites based on the lignosulfonic
acid-doped polyanilin have also been prepared recently.
   In 1998, Westvaco marketed a variety of specialty lignin chemicals derived from
kraft black liquor, finding uses as dyes and pigment chemicals, in mineral technology,
asphalt, agricultural, lead storage batteries and phenolic resins (McCarthy and Islam
2000). The chemical heterogeneity however limits the potential for use in phenolic
resins to an additive level of about 5–10% with a consequential increase in molecu-
lar weight of the resins (Turunen et al. 2003). Another area where there has been
strong interest is utilising lignin in epoxy resins. Simionescu et al. (1993) showed
that high lignin loads could be sustained without a significant drop in the important
mechanical properties of the epoxy resin. Lignin has also been incorporated into the
production of polyolefins such as polyethylene and polypropylene with mixed success
with results indicating a reduction in strength and poor adhesion between lignin and
the polyolefin (Gosselink et al. 2004a, 2004b, Cazacu et al. 2004). This was perhaps
due to once again the heterogeneity of the lignin samples used. However, the increased
biodegradability due to the incorporation of lignin into the matrix material presents
an interesting method for improving the environmental compatibility of this common
polymer.
   A critical factor that will enhance the potential of using polymeric lignin in nanotech-
nological applications will be the ability to produce a well-defined raw material with
reproducible properties. This may be achieved through biosynthetic control (Boudet
et al. 2003), the extended use of synthetic lignins through the polymerisation of mono-
lignols components or through the better processing of technical lignins isolated from
pulping liquors (Chakar and Ragauskas 2004). Already, there are numerous reports
detailing the use of lignin in applications as diverse as the production of carbon fibre
(Kadla et al. 2002, Kubo and Kadler 2005) and activated carbon materials (Suhas et al.
2007). The advantage in these two applications in particular is that the use of highly het-
erogeneous technical lignins is possible without the need for molecular reproducibility.
Activated carbons prepared from kraft lignin have also been used to study the adsorption
of phenol (Fierro et al. 2008) and benzene vapour (Blanco et al. 2008) in sensor type
applications.
   However, still today the available technical lignins are always by-products and the
properties of the lignins produced are thus substantially dependent on the core process
that is mainly dedicated to pulp and paper production. A change is foreseen in the future
due to emerging environmental demands in substituting oil-based sources for production
of fuels and chemicals. The huge economical efforts attributed in the ongoing worldwide
development of different biorefinery concepts, utilizing parts of the wood and other
plants for the main purpose of making other things than fibre products will possibly
198   The Nanoscience and Technology of Renewable Biomaterials

make the difference (Ragauskas et al. 2006). The shear volume of available biomass at
economically viable rates bodes well for the use of lignin as a cheap source of aromatic
monomers for subsequent polymerisation (Gandini 2008). Breaking down the larger
lignin fragments into monomeric or oligomeric precursors either through chemical or
enzymatic means overcomes the greatest difficulty in utilisation of this material: the
heterogeneous nature of the isolated lignin.


7.7 Concluding Remarks

A number of recent publications have detailed the ongoing efforts to increase the potential
of use of the underutilized resource lignin (Chakar and Ragauskas 2004, Gosselink et al.
2004a, Stewart 2008, Gandini 2008). With the cost of oil and petroleum based chemicals
at historical highs and the continually increasing demand for this scarce resource, it is
inevitable that biomass will play an important role in providing for future economic
prosperity. Lignin, with its unique structure and properties, large available volume and
surety of supply, is well placed to fill this role. However, lignin as a resource does have
some limitations that may be overcome in the future through better understanding of the
natural structure and its physical chemistry. Furthermore, if lignin could be produced
with a constant molecular structure free of both organic and inorganic contaminants will
enhance its potential in nanotechnology.


References

Argyropoulos, D.S.; Sun, Y.; Palus, E. (2002). ‘Isolation of residual kraft lignin in high
  yield and purity’, J. Pulp Paper Sci. 28: 50–4.
Berry, B.C.; Viswanathan, T. (2002). In: Chemical Modification, Properties, and Usage
  of Lignin, Hu, T.Q. Ed. Kluwer Academic/Plenum Publishers, New York.
  o
Bj¨ rkman, A. (1956). ‘Studies on finely divided wood. 1. Extraction of lignin with
  neutral solvents’, Sv. Papperstidn. 59: 477–85.
  o
Bj¨ rkman, A. (1957). ‘Lignin and lignin-carbohydrate complexes – extraction from
  wood meal with neutral solvents’, Ind. Eng. Chem. 49: 1395–8.
  o
Bj¨ rkman, A.; Person, B. (1957) ‘Studies on finely divided wood. Part II. The prop-
  erties of lignins extracted with neutral solvent from softwoods and hardwoods’, Sv.
  Papperstidn. 60: 158–69.
Blanco, F.; Vilanova, X.; Fierro, V., et al. (2008). ‘Fabrication and characterisation of
  microporous activated carbon-based pre-concentrators for benzene vapours’, Sensors
  and Actuators B-Chemical 132(1): 90–8.
Boerjan, W.; Ralph, J.; Baucher, M. (2003). ‘Lignin biosynthesis’, Annual Review of
  Plant Biology 54: 519–46.
Boudet, A.M.; Kajita, S., Grima-Pettenati, J., et al. (2003). ‘Lignins and lignocel-
  lulosics: a better control of synthesis for new and improved uses’, Trends in Plant
  Science, 8: 576–81.
Brauns, F.E. (1952). The Chemistry of Lignin, Academic Press, New York.
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience     199

Buchholz, R.F.; Neal, J.A.; McCarthy, J.L. (1992). ‘Some properties of paucidisperse
  gymnosperm lignin sulfonates of different molecular-weights’, J. Wood Chem. Technol.
  12: 447–69.
Cazacu, G.; Mihaies, M.; Pascu, C., et al. (2004). ‘Polyolefin/lignosulfonate blends’,
  Macromol. Mater. Eng 289: 880–9.
Chakar, F.S.; Ragauskas, A.J. (2004). ‘Review of current and future softwood kraft
  lignin process chemistry’, Industrial Crops and Products 20(2): 131–41.
Chan, D.Y.C.; Horn, R.G. (1985). J. Chem. Phys. 83: 5311.
Chang, H.M.; Cowling, E.B.; Brown, W.; Adler, E.; Miksche, G. (1975). ‘Comparative
  studies on cellulolytic enzyme lignin and milled wood lignin of sweetgum and spruce’,
  Holzforschung 29: 153–9.
Chatellier, X.; Senden, T.J.; Joanny, J.-F., et al. (1998). ‘Detachment of a single
  polyelectrolyte chain adsorbed on a charged surface’, Europhys. Lett. 41(3): 303–8.
Constantino, C.J.L.; Juliani, L.P.; Botaro, V.R., et al. (1996). ‘Langmuir-Blodgett films
  from lignins’, Thin Solid Films 284–5: 191–4.
Constantino, C.J.L.; Dhanabalan, A.; Curvelo, A.A.S., et al. (1998). ‘Preparation and
  characterization of composite LB films of lignin and cadmium stearate’, Thin Solid
  Films 327–9(1–2): 47–51.
Constantino, C.; Dhanabalan, A.; Cotta, M., et al. (2000). ‘Atomic force microscopy
  (AFM) investigation of Langmuir-Blodgett (LB) films of sugar cane bagasse lignin’,
  Holzforschung 54(1): 55–60.
Deryagin, B.; Landau, L. (1941). ‘Theory of stability strongly charged lyophobic soles
  and coalescence of strongly charged particles in solutions of electrolytes’, Acta. Phys.
  Chim. URSS 14: 633–62.
                                   a
Eriksson, M.; Notley, S.M.; W˚ gberg, L. (2007). ‘Cellulose thin films: Degree of
  cellulose ordering and its influence on adhesion’, Biomacromolecules 8: 912–19.
                         o
Evans, D.F.; Wennerstr¨ m, H. (1994). The Colloidal Domain, VCH Publishers Inc.,
  New York.
Fasching, M.; Schroder, P.; Wollboldt, R.P.; Weber, H.K.; Sixta, H. (2008). ‘A new and
  facile method for isolation of lignin from wood based on complete wood dissolution’,
  Holzforschung 62: 15–23.
Fierro, V.; Torne-Fernandez, V.; Montane, D., et al. (2008). ‘Adsorption of phenol onto
  activated carbons having different textural and surface properties’, Microporous and
  Mesoporous Materials 111(1–3): 276–84.
Fleer, G.J.; Cohen Stuart, M.A., Scheutjens, J.M.H.M., et al. (1993). Polymers at
  Interfaces, Chapman & Hall.
Fredheim, G.E.; Braaten, S.M.; Christensen, B.E. (2002). ‘Molecular weight determina-
  tion of lignosulfonates by size-exclusion chromatography and multi-angle laser light
  scattering’, J. Chromatography A, 942: 191–9.
Freudenberg, K. and A.C. Neish (1968). Constitution and Biosynthesis of Lignin. New
  York, Springer-Verlag.
Gandini, A. (2008). ‘Polymers from renewable resources: A challenge for the future of
  macromolecular materials’, Macromolecules In press.
Gellerstedt, G.; Lindfors, E.L. (1984). ‘Structural changes in lignin during kraft pulping’,
  Holzforschung 38: 151–8.
200   The Nanoscience and Technology of Renewable Biomaterials

Gellerstedt, G.; Pranda, J.; Lindfors, E.L. (1994). ‘Structural and molecular-properties
   of birch kraft ligins’, J. Wood Chem. Technol. 14: 467–82.
Glasser, W.G.; Barnett, C.A. (1979). ‘Structure of lignins in pulps. 2. Comparative
   evaluation of isolation methods’, Holzforschung 33: 78–86.
Gosselink, R.J.A.; DeJong, E.; Guran, B., et al. (2004a). ‘Co-ordination network for
   lignin-standardisation, production and applications adapted to market requirements
   (EUROLIGNIN)’, Ind. Crops Prod. 20: 121–9.
Gosselink, R.J.A.; Snijder, M.H.B.; Kranenbarg, A., et al. (2004b). ‘Characterisation
   and application of NovaFiber lignin’, Ind. Crops Prod. 20: 191–203.
Guerra, A.; Filpponen, I.; Lucia, L.A., et al. (2006a). ‘Comparative evaluation of
   three lignin isolation protocols for various wood species’, J. Agric. Food Chem. 54:
   9696–9705.
Guerra, A.; Filpponen, I.; Lucia, L.A., et al. (2006b). ‘Toward a better understand-
   ing of the lignin isolation process from wood’, J. Agric. Food Chem. 54: 5939–
   47.
Guo, X.; Zhang, S.; Shan, X. (2008). ‘Adsorption of metal ions on lignin’, J. Hazard.
   Mater. 151: 134–42.
Hartler, N. (1978). ‘Extended delignification in kraft cooking – New concept’, Sv. Pap-
   perstidn. 81: 483–4.
           ´                            a
Hidalgo-Alvarez, R.; Martin, A.; Fern´ ndez, A., et al. (1996). ‘Electrokinetic properties,
   colloidal stability and aggregation kinetics of polymer colloids’, Adv. Colloid Interface
   Sci. 67: 1–118.
Hoekstra, L.L.; Vreeker, R.; Agterof, W.G.M. (1992). ‘Aggregation of colloidal
   nickel hydroxycarbonate studied by light-scattering’, J. Colloid Interface Sci. 151:
   17–25.
Hunter, R.J. (1993). Introduction to Modern Colloid Science, Oxford University Press.
 a¨     a
J¨ askel¨ inen, A.S.; Sun, Y.; Argyropoulos, D.S., et al. (2003). ‘The effect of isola-
   tion method on the chemical structure of residual lignin’, Wood Sci. Technol. 37:
   91–102.
Jiang, J.-E.; Chang, H.-m.; Bhattacharjee, S.S., et al. (1987). ‘Characterization of
   residual lignins isolated from unbleached and semi-bleached softwood kraft pulps’,
   J. Wood Chem. Technol. 7: 81–96.
                     o                 o
Johansson, B.; Mj¨ berg, J.; Sandstr¨ m, P., et al. (1984). ‘Modified continuous kraft
   pulping – Now a reality’, Sv. Papperstidn. 87: 30–5.
 o
J¨ nsson, B.; Lindman, B.; Holmberg, K., et al. (1998). Surfactants and Polymers in
   Aqueous Solution, John Wiley & Sons Ltd., Chichester.
Julien, R.; Botet, R. (1987). Aggregation and Fractal Aggregation, World Scientific
   Publishing Co Pte Ltd, Singapore.
Junker, E. (1941). ‘Zur Kenntnis der kolloidchemischen Eigenschaften des Humus
                          a
   Beitrag zur Dispersit¨ tschemie des Lignins’, Kolloid-Zeitschr. 95: 213–50.
Jurasek, L.J. (1995). ‘Toward a three-dimensional model of lignin structure’, Pulp Paper
   Sci. 21, J274–J279.
Kadla, J.F.; Kubo, S.; Venditti, R.A., et al. (2002). ‘Lignin-based carbon fibers for
   composite fiber applications’, Carbon 40(15): 2913–20.
       a
Kilpel¨ inen, I.; Xie, H.; King, A., et al. (2007). ‘Dissolution of wood in ionic liquids’,
   J. Agric. Food Chem. 55: 9142–8.
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience   201

Kim, A.Y.; Berg, J.C. (2000). ‘Fractal aggregation: Scaling of fractal dimension with
  stability ratio’, Langmuir 16: 2101–4.
Kohonen, M.M. (2006). ‘Engineered wettability in tree capillaries’, Langmuir 22:
  3148–53.
                                ¨
Kontturi, E.; Tammelin, T.; Osterberg, M. (2006). ‘Cellulose – Model films and the
  fundamental approach’, Chemical Society Reviews 35(12): 1287–1304.
Kubo, S.; Kadla, J.F. (2005). ‘Lignin-based carbon fibers: Effect of synthetic poly-
  mer blending on fiber properties’, Journal of Polymers and the Environment 13(2):
  97–105.
Lai, Y.Z.; Sarkanen, K.V. (1971). In: Lignins – Occurrence, Formation Structure and
  Reactions, Sarkanen, K.V., Ludwig, C.H. (Eds). John Wiley & Sons, Inc., New York,
  pp. 165–240.
Lawoko, M.; Henriksson, G.; Gellerstedt, G. (2003). ‘New method for quantitative
  preparation of lignin-carbohydrate complex from unbleached softwood kraft pulp:
  Lignin-polysaccharide networks I’, Holzforschung 57: 69–74.
Leath, P.L.; Reich, G.R. (1978). ‘Scaling form for percolation cluster sizes and perime-
  ters’, J. Phys. C: Solid St. Phys. 11: 4017–35.
Lee, S.B.; Luner, P. (1972). ‘The wetting and interfacial properties of lignin’, Tappi J.
  55: 116.
Leubner, I.H. (2000). ‘Particle nucleation and growth models’, Curr. Opin. Colloid
  Interface Sci. 5: 151–9.
Lewis, N.G. (1999). ‘A 20th century roller coaster ride: a short account of lignification’,
  Current Opinion in Plant Biology 2(2): 153–62.
Lignin Institute 2007, http://www.lignin.org
Lin, M.Y.; Lindsay, H.M.; Weitz, D.A.; Ball, R.C.; Klein, R.; Meakin, P. (1989).
  ‘Universality in colloid aggregation’, Nature 339: 360–2.
Lin, M.Y.; Lindsay, H.M.; Weitz, D.A.; Klein, R.; Ball, R.C.; Meakin, P. (1990a).
  ‘Universal diffusion-limited colloid aggregation’, J. Phys. Condens. Matter 2: 3093–
  3113.
Lin, M.Y.; Lindsay, H.M.; Weitz, D.A.; Ball, R.C.; Klein, R.; Meakin, P. (1990b).
  ‘Universal reaction-limited colloid aggregation’, Phys. Rev. A, 41: 2005–20.
Lin, S.Y.; Zhong, X.J. (1990). China Pulp Pap. 9: 45–53.
        o
Lindstr¨ m, T. (1980). ‘The colloidal behavior of kraft lignin.2. Coagulation of kraft
  lignin sols in the presence of simple and complex metal-ions’, Colloid Polymer Sci.
  258: 168–73.
Lora, J.H.; Glasser, W.G. (2002). ‘Recent industrial applications of lignin: A sustain-
  able alternative to nonrenewable materials’, Journal of Polymers and the Environment
  10(1–2): 39–48.
Lu, F.; Ralph, J. (2003). Non-degradative dissolution and acetylation of ball-milled plant
  cell walls: high-resolution solution-state NMR’, Plant J. 35: 535–44.
McCarthy, J.L.; Islam, A. (2000). In: Lignin: Historical, Biological, and Materials
  Perspectives, Glasser, W.G., Northey, R.A., Schultz, T.P. Eds., ACS Symp. Ser.
  742: 2–99.
Martins, G.F.; Pereira, A.A.; Straccalano, B.A., et al. (2008). ‘Ultrathin films of lignins
  as a potential transducer in sensing applications involving heavy metal ions’, Sensors
  and Actuators B-Chemical 129(2): 525–30.
202   The Nanoscience and Technology of Renewable Biomaterials

Maurer, A.; Fengel, D. (1992). ‘On the origin of milled wood lignin I. The influence
 of ball-milling on the ultrastructure of wood cell walls and the solubility of lignin’,
 Holzforschung, 46: 417–23.
                 ¨
Maximova, N.; Osterberg, M.; Laine, J., et al. (2004). ‘The wetting properties and
 morphology of lignin adsorbed on cellulose fibres and mica’, Colloids and Surfaces
 A: Physicochemical and Engineering Aspects 239(1–3): 65–75.
Micic, M.; Jeremic, M.; Radotic, K., et al. (2000). ‘Visualization of artificial lignin
 supramolecular structures’, Scanning 22(5): 288–94.
Micic, M.; Benitez, I.; Ruano, M., et al. (2001a). ‘Probing the lignin nanomechanical
 properties and lignin-lignin interactions using the atomic force microscopy’, Chemical
 Physics Letters 347(1–3): 41–5.
Micic, M.; Radotic, K.; Benitez, K., et al. (2001b). ‘Topographical characterization and
 surface force spectroscopy of the photochemical lignin model compound’, Biophysical
 Chemistry 94(3): 257–63.
Micic, M.; Radotic, K.; Jeremic, M., et al. (2004). ‘Study of the lignin model compound
 supramolecular structure by combination of near-field scanning optical microscopy and
 atomic force microscopy’, Colloids and Surfaces B: Biointerfaces 34(1): 33–40.
Napper, D.H. (1983). Polymeric Stabilization of Colloidal Dispersions, Academic Press
 Inc, London.
                     o
Norgren, M.; Lindstr¨ m, B. (2000a). ‘Physico-chemical characterization of a fractionated
 kraft lignin’, Holzforschung 54: 528–34.
                     o
Norgren, M.; Lindstr¨ m, B. (2000b). ‘Dissociation of phenolic groups in kraft lignin at
 elevated temperatures’, Holzforschung 54: 519–27.
Norgren, M.; Edlund, H.; Nilvebrant, N.-O. (2001a). ‘Physicochemical differences
 between dissolved and precipitated kraft lignin fragments as determined by PFG
 NMR, CZE and Quantitative UV Spectrophotometry’, J. Pulp Paper Sci. 27:
 359–63.
                             a                  o
Norgren, M.; Edlund, H.; W˚ gberg, L.; Lindstr¨ m, B.; Annergren, G. (2001b). ‘Aggre-
 gation of kraft lignin derivatives under conditions relevant to the process. Part I. Phase
 behaviour’, Colloids Surf. A, 194: 85–96.
                                a
Norgren, M.; Edlund, H.; W˚ gberg, L. (2002). ‘Aggregation of lignin derivatives
 under alkaline conditions.        Kinetics and aggregate structure’, Langmuir 18:
 2859–65.
Norgren, M.; Notley, S.M.; Majtnerova, A., et al. (2006). ‘Smooth model surfaces
 from lignin derivatives. I. Preparation and characterization’, Langmuir 22(3):
 1209–14.
Norgren, M.; Gardlund, L.; Notley, S.M., et al. (2007). ‘Smooth model surfaces
 from lignin derivatives. II. Adsorption of polyelectrolytes and PEC’s monitored by
 QCM-D’, Langmuir 23: 3737–43.
Notley, S.M. (2008). ‘Effect of introduced charge in cellulose gels on surface interactions
 and the adsorption of cationic polyelectrolytes’, Physical Chemistry Chemical Physics
 10: 1819–25.
                                 a
Notley, S.M.; Pettersson, B.; W˚ gberg, L. (2004). ‘Direct measurement of attractive van
 der Waals’ forces between regenerated cellulose surfaces in an aqueous environment’,
 J. Am. Chem. Soc. 126(43): 13930–31.
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience   203

                  a
Notley, S.M.; W˚ gberg, L. (2005). ‘Morphology of modified regenerated model cellulose
  II surfaces studied by atomic force microscopy: Effect of carboxymethylation and heat
  treatment’, Biomacromolecules 6(3): 1586–91.
                                    a
Notley, S.M.; Eriksson, M.; W˚ gberg, L., et al. (2006). ‘Surface forces measure-
  ments of spin-coated cellulose thin films with different crystallinity’, Langmuir 22(7):
  3154–60.
Notley, S.M.; Norgren, M. (2006). ‘Measurement of interaction forces between lignin
  and cellulose as a function of aqueous electrolyte solution conditions’, Langmuir
  22(26): 11199–11204.
Notley, S.M.; Norgren, M. (2008). ‘The adsorption of a strong polyelectrolyte to model
  lignin surfaces’, Biomacromolecules 9: 2081–6.
¨
Ohman, F.; Wallmo, H.; Theliander, H. (2007). ‘Precipitation and filtration of lignin
  from black liquor of different origin’, Nordic Pulp Paper Res. J. 22: 188–93.
Pasquini, D.; Balogh, D.T.; Antunes, P.A., et al. (2002). ‘Surface morphology and
  molecular organization of lignins in Langmuir-Blodgett films’, Langmuir 18(17):
  6593–6.
Pasquini, D.; Balogh, D.T.; Oliveira, O.N.J., et al. (2005). ‘Lignin molecular arrange-
  ments in Langmuir and Langmuir-Blodgett films: The influence of extraction pro-
  cesses’, Colloids and Surfaces A: Physicochemical and Engineering Aspects 252(2–3):
  193–200.
Paterno, L.G.; Mattoso, L.H.C. (2001).           ‘Effect of pH on the preparation of
  self-assembled films of poly(o-ethoxyaniline) and sulfonated lignin’, Polymer 42(12):
  5239–45.
Perrin, D.D. (1981). pKa Prediction for Organic Acids and Bases’ , Chapman & Hall,
  London.
Pew, J.C. (1957). ‘Properties of powdered wood and isolation of lignin by cellulytic
  enzymes’, Tappi 40: 53–8.
Pew, J.C.; Weyna, P. (1962). ‘Fine grinding, enzyme digestion, and the lignin-cellulose
  bond in wood’, Tappi 45: 247–56.
Pu, Y.Q.; Jiang, N.; Ragauskas, A.J. (2007). ‘Ionic liquid as a green solvent for lignin’,
  J. Wood Chem. Technol. 27: 23–33.
Ragauskas, A.J.; Williams, C.K.; Davison, B.H., et al. (2006). ‘The path forward for
  biofuels and biomaterials’, Science 311: 484–9.
Reerink, H.; Overbeek, J.Th.G. (1954). ‘The rate of coagulation as a measure of the
  stability of silver iodide sols’, Disc. Faraday Soc. 18: 74–84.
Rief, M.; Oesterhelt, F.; Heymann, B., et al. (1997). ‘Single molecule force spec-
  troscopy on polysaccharides by atomic force microscopy’, Science 275: 1295.
Ringena, O.; Saake, B.; Lehnen, R. (2005). ‘Isolation and fractionation of lignosul-
  fonates by amine extraction and ultrafiltration: A comparative study’, Holzforschung
  59: 405–12.
Roy, S.; Fortier, J.M.; Nagarajan, R.; et al. (2002). ‘Blomimetic synthesis of a water
  soluble conducting molecular complex of polyaniline and lignosulfonate’, Biomacro-
  molecules 3: 937–41.
Rudatin, S.; Sen, Y.L.; Woerner, D.L. (1989). In: Lignin: Properties and Materials,
  Glasser, W.G., Sarkanen, S., Eds., ACS Symp. Ser. 397: 144–54.
204   The Nanoscience and Technology of Renewable Biomaterials

Sarkanen, K.V.; Ludwig, C.H. (1971). Lignin, Occurence, Formation, Structure and
  Reactions. New York, Wiley/Interscience.
Sarkanen, S.; Teller, D.C.; Abramowski, E.; McCarthy, J.L. (1982). ‘Lignin.19. Kraft
  lignin component conformation and associated complex configuration in aqueous alka-
  line solution’, Macromolecules 1982, 15: 1098–1104.
Senden, T.J.; Di Meglio, J.-M., Auroy, P. (1998). ‘Anomalous adhesion in adsorbed
  polymer layers’, Eur. Phys. J. B 3: 211–16.
Simionescu, C.I.; Rusan, V.; Macoveanu, M.M., et al. (1993). ‘Lignin/epoxy compos-
  ites’, Compos. Sci. Technol. 48: 317–23.
  o                                    o
Sj¨ holm, E.; Gustafsson, K.; Colmsj¨ , A. (1999a). ‘Size exclusion chromatography of
  lignins using lithium chloride/N,N-dimethylacetamide as mobile phase. I. Dissolved
  and residual birch kraft lignins’, J. Liquid Chrom. Rel. Technol. 22: 1663–85.
  o                                    o
Sj¨ holm, E.; Gustafsson, K.; Colmsj¨ , A. (1999b). ‘Size exclusion chromatography of
  lignins using lithium chloride/N,N-dimethylacetamide as mobile phase. II. Dissolved
  and residual pine kraft lignins’, J. Liquid Chrom. Rel. Technol. 22: 2837–54.
  o o
Sj¨ str¨ m, E. (1993). Wood Chemistry. Fundamentals and Applications 2nd ed. Aca-
  demic Press Inc., San Diego.
Stewart, D. (2008). ‘Lignin as a base material for materials applications: Chemistry,
  application and economics’, Industrial Crops and Products 27(2): 202–7.
Suhas, Carrott, P.J.M.; Carrott, M. (2007). ‘Lignin – from natural adsorbent to activated
  carbon: A review’, Bioresource Technology 98(12): 2301–12.
Sun, R.C.; Lawther, J.M.; Banks, W.B. (1997). ‘Fractional isolation and physico-
  chemical characterization of alkali-soluble lignins from wheat straw’, Holzforschung
  51: 244–50.
Tammelin, T.; Osterberg, M.; Johansson, L.S., et al. (2006). ‘Preparation of lignin and
  extractive model surfaces by using spin-coating technique – Application for QCM-D
  studies’, Nordic Pulp Paper Res. J. 21: 444–50.
Tammelin, T.; Jonhsen, I.A.; Osterberg, M., et al. (2007). ‘Adsorption of colloidal
  extractives and dissolved hemi-celluloses on thermomechanical pulp fiber components
  studied by QCM-D’, Nordic Pulp Paper Res. J. 22: 93–101.
Terashima, N.; Fukushima, K.; Imai, T. (1992). ‘Morphological origin of milled wood
  lignin studied by radiotracer method’, Holzforschung 46: 271–5.
Turunen, M.; Alvila, L.; Pakkanen, T.T., et al.              (2003).    ‘Modification of
  phenol-formaldehyde resol resins by lignin, starch and urea’, J. Appl. Polym. Sci. 88:
  582–8.
Vanholme, R.; Morreel, K.; Ralph, J., et al. (2008). ‘Lignin engineering’, Current
  Opinion in Plant Biology 11(3): 278–85.
Verwey, E.G.W.; Overbeek, J.T.G. (1948). Theory of the Stability of Lyophobic Colloids.
  New York, Elsevier.
                 o
Wallberg, O.; J¨ nsson, A.S.; Wimmerstedt, R. (2003). ‘Fractionation and concentration
  of kraft black liquor lignin with ultrafiltration’, Desalination 154: 187–99.
Weitz, D.A.; Lin, M.Y.; Huang, J.S. (1987). In: Physics of Complex and Supramolecular
  Fluids, Safran, S.A., Clark, N.A., Eds., Wiley, New York.
Weitz, D.A.; Lin, M.Y.; Lindsay, H.M. (1991). ‘Universality laws in coagulation’,
  Chemometr. Intell. Lab. 10: 133–40.
 Lignin: Functional Biomaterial with Potential in Surface Chemistry and Nanoscience   205

Westermark, U.; Gustafsson, K. (1994). ‘Molecular-size distribution of wood polymers
  in birch kraft pulps’, Holzforschung 48: 146–50.
Whiting, P.; Goring, D.A.I. (1981). ‘The morphological origin of milled wood lignin’,
  Sv. Papperstidn. 84: R120–R122.
Woerner, D.L.; McCarthy, J.L. (1998). ‘Lignin. 24. Ultrafiltration and light-scattering
  evidence for association of kraft lignins in aqueous solutions’, Macromolecules 21:
  2160–6.
Wu, S.; Argyropoulos, D.S. (2003). ‘An improved method for isolating lignin in high
  yield and purity’, J. Pulp Paper Sci. 29: 235–240.
Zavitsas, A.A. (1967). ‘Acid ionization constants of phenol and some hydroxy-
  methylphenols between 20.degree. and 60.degree’, J. Chem. Eng. Data 12: 94–7.
                                                         8
   Cellulose and Chitin as Nanoscopic
              Biomaterials

               Jacob D. Goodrich, Deepanjan Bhattacharya and William T. Winter



8.1     Overview

Cellulose and chitin nanoparticles were isolated from bagasse, and shrimp shells, respec-
tively. The nanoparticles were characterized by optical, electron, and atomic force
microscopy, solid-state NMR spectroscopy, and X-ray powder diffraction methods. The
nanoparticles were then topochemically modified with maleate ester groups in the case
of cellulose, and medium- to long-chain aliphatic esters in the case of chitin. The
derivatized nanoparticles were further characterized with spectroscopic techniques and
subsequently melt processed with elastomeric thermoplastics to create nanocompos-
ites having a significant improvement in the mechanical properties relative to the neat
thermoplastics.


8.2     Introduction

Since the start of the 20th century, polymeric materials have begun replacing conven-
tional materials such as wood and metals in a diverse array of industries. Today, polymers
have a ubiquitous presence in our society. They are often mixed with fillers or fibers as a
versatile route to fabricating advanced materials with improved thermal and mechanical
properties, to form what are called polymer composites. Typically, polymer composites
have greater mechanical strength and stiffness than any of their individual components,
and are used in a variety of applications. Historically, polymer composites consisted of
synthetic thermoset resins reinforced with inorganic filler materials like glass fiber. The

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
The contribution of Dr Goodrich and Dr Bhattacharya has been written in the course of their employment with Eastman
Chemical Company c Eastman Chemical Company
208   The Nanoscience and Technology of Renewable Biomaterials

greatest advantage with thermoset resins is the low viscosity of the materials prior to
cross-linking, allowing for low-pressure processing conditions and good particle disper-
sions. The main disadvantages with such materials are that: (1) the thermoset resins are
very intractable and inherently brittle, (2) they cannot be recycled or biodegraded after
their useful lifetime, (3) almost all thermosetting resins are derived from diminishing
petroleum resources, and (4) the incorporated inorganic filler phases are very dense,
adding to the overall weight and shipping costs of the materials. Recently, improved
polymer processing technology and the demand for materials with more diverse ther-
mal and mechanical performance properties have placed thermoplastic composites in
strong competition with conventional thermoset materials. The high melt-viscosity of
thermoplastics has made it difficult to create adequate particle dispersions in thermo-
plastic matrixes. High shear rates and elongational flow patterns generated in modern
melt-processing equipment (i.e. compounders and extruders) have facilitated the sepa-
ration of particle aggregates in high viscosity systems. Generally, once aggregates are
disrupted, the high viscosity of the thermoplastic melt prevents reaggregation, producing
fairly uniform, high performance composites.
   Most reinforcement phases in conventional composites are macroscopic particles, such
as glass or aramid fibers. Developments in analytical techniques, such as microscopy,
and increased understanding of composite behavior, have stimulated massive amounts
of research, which recently has moved into the nanostructure of composite materials.
The term nanotechnology is pervasive in modern scientific research and, as defined by
the National Nanotechnology Initiative (1), implies three essential characteristics. At
least one of the objects in the system must be on a nanoscale, meaning that it must
have at least one dimension less than 100 nm. The second trait is that the work must
involve the creation or use of structures, devices, or systems having novel properties
consequential to their size. The third trait is the element of control or manipulation at
the nanoscale (i.e. the molecular-to-atomic scale). Polymer nanocomposites embody
all three of these traits. It is generally recognized that the gains in surface area as
a consequence of reducing particle sizes to nanometer dimensions (1 nm = 10−9 m)
can lead to outstanding properties in composites reinforced with these particles. Such
materials have been heavily researched in the last few decades. Since the seminal
work published by researchers at Toyota on Nylon 6,6 reinforced with montmorillonite
nanoclay, the considerable potential of polymer nanocomposites has been realized (2).
The materials showed remarkable improvements in tensile modulus, tensile strength, and
heat resistance at very low filler loadings (2). In much of the subsequent nanocomposite
research, the filler materials studied have been inorganic materials such as nanoparticles
of silica, boron, clay, calcium carbonate, or carbon nanotubes, to name a few. Several
research groups, particularly those in Europe and Japan, as well as a few in North
America, including our own laboratories, have investigated natural polysaccharide based
nanoparticle composites containing cellulose and chitin (3–5).
   Cellulose and chitin are the two most abundant biopolymers. It is estimated that the
combined worldwide annual production of cellulose and chitin by nature is nearly 2
× 1012 metric tons (6, 7). The seafood industry alone generates annually some 105
metric tons of chitin waste for industrial use (8, 9), with the availability of cellulose
far exceeding this number. Both polymers represent underutilized, readily available,
sustainable feedstock alternatives to petroleum-based materials. Cellulose and chitin
                                    Cellulose and Chitin as Nanoscopic Biomaterials     209

are chemically similar, and both serve a natural role in structural reinforcement for
many plant and animal organisms. By mimicking nature, the materials industry might
benefit from enhanced utilization of these feedstocks. Useful properties of chitin and
cellulose are their high stiffness and strength, low density (∼1.35 g/cm3 for chitin
and ∼1.6 g/cm3 for cellulose compared to 2.6 g/cm3 for glass fiber), biodegradability,
renewability, inherent nanoscale architecture, coupled to an abundance of opportunities
for chemical functionalization. The elastic modulus of cellulose whiskers is reported
to approach 145 GPa (10) while that of glass fibers averages around 70 GPa (11),
suggesting that composites reinforced with cellulose nanoparticles may be superior to
conventional glass fiber reinforced composites. The issue of filler density also becomes
very important when trying to maximize the strength-to-weight ratio for lower ship-
ping costs of materials and reduce fuel consumption when the materials are used in
transportation.
   For thousands of years cellulose fibers in the form of straw have been used by people
to mechanically reinforce mud or clay to create adobe bricks for the construction of their
dwellings. These early composites were the first precursors of the present day thermo-
plastic cellulose-based nanocomposites. Thermoplastic nanocomposites reinforced with
cellulose or chitin whiskers derived from sources such as wood pulp, straws, bacteria,
and bagasse, for cellulose, and shrimp, crab, or lobster shells, for chitin, have shown
promising results. In a few particular cases, a 2- to 3-order of magnitude improvement
in modulus of the composites was observed at low filler loadings of cellulose or chitin
(3, 4). Much of the work in nanocomposites reinforced with highly crystalline polysac-
charide nanoparticles has made use of synthetic, petroleum-based matrix polymers that
are largely nonbiodegradable after their usable lifetimes.
   The design and utilization of green processes and sustainable materials is relevant
to the urgent need to develop technologies that minimize our dependence on petroleum
feedstocks and to concerns regarding the management of excessive amounts of municipal
solid-waste generated by an expanding human population. Animal, plant, and microbial-
based biopolymers, and derived materials, such as biocomposites, are promising alter-
natives to currently employed petroleum-based plastics. To effectively compete with
existing products, this class of materials still faces challenges to their widespread uti-
lization, including their high cost, and poorer performance relative to petroleum-based
plastics. Bioplastics such as polyhydroxyalkanoates, poly(lactic acid), and cellulose
esters reinforced with cellulose or chitin nanoparticles have recently attracted attention
as renewable and sustainable alternatives to current plastic materials (12, 13). For such
composites to replace traditional materials, further work in characterizing the materials,
enhancing compatibility, achieving stable dispersions, and developing faster and cheaper
‘green’ processing is essential. The ultimate goal is to create nanocomposites that are
entirely bio-based, controllably biodegradable, and match or exceed the performance
properties of synthetic composites reinforced with inorganic fillers. As such, the cre-
ation and study of cellulose or chitin as the structural reinforcing phase is the main focus
of this report.
   Many thermoplastics, whether petroleum or bio-based, are hydrophobic in charac-
ter. Conversely, cellulose and chitin are hydrophilic materials, with their abundance of
hydroxyl groups, and thus exhibit poor compatibility with hydrophobic plastics. Due to
the high surface area of cellulose and chitin nanoparticles, which creates a very large
210    The Nanoscience and Technology of Renewable Biomaterials

interfacial area, the interactions between the particle and matrix phases become critically
important to composite performance. Typical methods to improve compatibility between
the polysaccharide filler and the thermoplastic matrix involve either creating dispersive
coatings around the particles or covalently modifying the particles with either hydropho-
bic molecules, or coupling agents that covalently link the two phases. The approach used
in this study involves the topochemical modification of cellulose and chitin nanoparti-
cles with different hydrophobic moieties to achieve better phase compatibility. The
mechanical effects that cellulose and chitin nanoparticles, and their derivatives, have
on different bio-based or biodegradable thermoplastic nanocomposites are also briefly
described.



8.3   Preparation and Microscopic Characterization of Cellulose
      and Chitin Nanoparticles

Processes for the purification of cellulose and chitin are well established, and involve
the removal of the organic and/or inorganic materials naturally associated with them,
generally through enzymatic, acidic or basic treatments. When purified, both cellulose
and chitin are semicrystalline materials. Both materials are susceptible to degradation in
strongly acidic media. Acid hydrolysis of cellulose and chitin is essentially the same,
since the occurrence of an acetamido group at C(2) in the latter is the only primary struc-
ture difference between them. The first step in acid hydrolysis involves the protonation
of the acetal oxygen of the glycosidic linkage. An intermediate carbocation is formed at
the anomeric carbon through heterolysis, causing a destruction of the glycosidic bond (6).
The carbocation then reacts with water, forming a hydroxyl group, and a proton (6). The
reaction is first-order, with the speed of reaction highly dependent on both the cellu-
lose and acid concentrations (6). The amorphous regions of the cellulose and chitin are
digested first due to their higher accessibility relative to the crystalline regions. Hetero-
geneous acid hydrolysis was performed in this work, where the hydrolysis occurs first
in the amorphous regions, then decreases considerably when the amorphous cellulose or
chitin is digested. The acid hydrolysis conditions are controlled to digest the amorphous
cellulose and chitin segments connecting crystallites in the elementary and microfibrils,
to create smaller, highly crystalline segments, called nanocrystals or nanoparticles. The
nanocrystals are desirable for their retention of the native crystalline properties, and their
high stiffness and specific surface areas. A process of high-energy mechanical dispersion
via homogenization is often used as an additional route to the production of individ-
ualized nanocrystals. After processes of acid hydrolysis and homogenization, well-
dispersed aqueous suspensions of the nanoparticles are obtained.
   As demonstrated through the preparation of cellulose nanocrystals from bagasse, a
change in the morphological structure of the whole cellulose fibers occurs upon acid
hydrolysis and can be observed using SEM. Figure 8.1 is an optical micrograph of the
cellulose fibers prior to acid hydrolysis.
   The average particle length is approximately one to two millimeters. However, the
SEM micrographs of the fibers post hydrolysis indicate that the majority of the microfib-
rils are in the submicron range having high aspect ratios between 50 and 120. The larger
                                    Cellulose and Chitin as Nanoscopic Biomaterials    211




Figure 8.1 Optical micrograph of whole cellulose fibers isolated from bagasse after pulping.
Reprinted from (14). Copyright (2008), with permission from Elsevier.




Figure 8.2 Scanning electron micrograph showing the presence of the individual cellulose
microfibers obtained from bagasse. Reprinted from (14). Copyright (2008), with permission
from Elsevier.

bundles from which the microfibers were released after hydrolysis, ultrasonication and
homogenization, can be seen in Figure 8.2.
   A broad distribution of fiber lateral dimensions is evident, owing to the fact that
some of the microfibrillar bundles were not completely dispersed and/or re-aggregated
during the preparation of samples for scanning electron microscopy (SEM) and atomic
force microscopy (AFM) studies. Depending on their origin, cellulose microfibrils may
have transverse dimensions that range from 20–200 nm but these particles are often
aggregates, and the individual microfibrils are usually in the range of 3–20 nm (15).
212   The Nanoscience and Technology of Renewable Biomaterials

Compared to the cellulose whiskers obtained from other sources like ramie (16), cotton
(16), filter paper (17), or bleached kraft wood pulp (18), the microfibril dimensions for
particles isolated from bagasse appeared to be less uniform. The distribution of the
particle lengths for the cellulose whiskers has also been reported (15).
   There have been some recent reports on the crystal morphology of cellulose using
AFM. Sugiyama and coworkers have used it to study the crystalline order in the bulk as
well as on the surface for microcrystalline cellulose isolated from Valonia ventricosa (19).
Images generated from our studies using AFM illustrated the fiber bundle morphology
in cellulose microfibers (MFs) isolated from bagasse (20). In Figure 8.3, we see whole
microfibrillar bundles as well as individual nanofibers (14).
   In this figure, the left-hand image is a height image that represents surface topog-
raphy, while the right-hand image is a phase image whose contrast differentiates soft
(amorphous) and hard (crystalline) polymer segments. Both height and phase images are
recorded. These images agree well with the scanning electron micrographs. Increased
magnification of microfibrillar bundles reveals nanometer-scale (30 nm) structures as
shown in Figure 8.4.
   These dimensions are comparable to those proposed by Hess et al., for their schematic
representation of cellulose fiber structure (20). The banding apparent in these images
(see Figure 8.5) is consistent with the density fluctuations in the Hess model for the
microfibrillar assembly.
   The presence of periods from 60 to 100 nm is representative of crystalline (bright
regions) and amorphous (dark) regions in the direction of the fiber axis. For semicrys-
talline polymers, lighter areas in phase images have been interpreted as crystalline




 0                                     10.0 µm 0                                      10.0 µm
           Data type          Height                     Data type           Phase
           2 range            400 nm                     2 range            30.0 de

Figure 8.3 AFM images showing the fiber bundle morphology in cellulose isolated from
bagasse. Reprinted from (14). Copyright (2008), with permission from Elsevier.
                                    Cellulose and Chitin as Nanoscopic Biomaterials       213




                                                                        800



                                                                        600



                                                                        400



                                                                        200



                                                                        0
                 0        200       400       600        800            nm

Figure 8.4 Microfibrillar bundles are also observed to be composed of nanometer-sized
(∼30 nm) nanofibers. Reprinted from (14). Copyright (2008), with permission from Elsevier.




                                       Crystalline

                                       Amorphous




                                          ~150 Å



                                                               ~260 Å
                                                                          ~60 Å


                     30 nm                                                        ~65 Å


Figure 8.5 AFM phase images support Hess et al. model for the presence of periods from 10
to 20 nm for the presence of crystalline and amorphous regions in the direction of the fiber
axis. Reprinted from (14). Copyright (2008), with permission from Elsevier. Reproduced with
permission from (20). Copyright (1941), Oldenbourg.
214   The Nanoscience and Technology of Renewable Biomaterials




Figure 8.6 SEM micrograph of chitin fibers isolated from shrimp shells. Actual magnification
shown is (×1200). Reprinted with permission from (23). Copyright (2007), American
Chemical Society.

domains, while darker regions are considered to be amorphous (21). Although the
banding is longer than that described in the original Hess model, it is consistent with
numerous reports of a leveling off degree of polymerization, LODP, for cellulose of
approximately 150–200 glucose residues (22). In a crystalline domain, each glucose
residue subtends 0.5 nm along the major axis so a LODP of 150–200 nm corresponds
to a crystallite length of 75–100 nm, as seen in Figure 8.5.
   Purified chitin was produced after treatments with acid and base. This material was
studied by SEM and TEM and demonstrates a hierarchal breakdown, similar to that
observed with cellulose. Figure 8.6 shows an SEM micrograph of purified shrimp shell
chitin prior to hydrolysis and homogenization.
   The particle dimensions for the isolated fibers are observed to be between 5 µm and
10 µm in width, and several hundred microns in length. Figure 8.7 is a TEM micrograph
of chitin nanocrystals cast from a dilute aqueous suspension. From this micrograph it is
apparent that individual chitin nanoparticles range from 200 to 500 nm in length, and 8
to 12 nm in width.


8.4   NMR Characterization of Cellulose and Chitin Nanoparticles

Cross Polarization/Magic Angle Spinning (CP/MAS) NMR studies proved to be a very
useful technique in monitoring the morphological changes taking place in cellulose dur-
ing the course of hydrolysis. The 13 C NMR spectra of (a) intact cellulose fibers and
(b) hydrolyzed microfibers (MFs) are illustrated in Figure 8.8a and b respectively.
   The six carbon atoms that are assigned to the cellulose molecule dominate the spectrum
in both cases. The chemical shift values range from 105 ppm to 60 ppm. The anomeric
carbon (C1) appears furthest downfield at around 105 ppm. This is followed by the
signal from the C4 atom between 82 and 89 ppm, a range arising from the different
                                    Cellulose and Chitin as Nanoscopic Biomaterials    215




Figure 8.7 TEM micrograph of chitin nanocrystals from shrimp shells formed after hydrolysis
and mechanical dispersion. Magnification shown is (×10,000). Reprinted with permission
from (23). Copyright (2007), American Chemical Society.


structural domains within the microfiber. Peaks arising due to C2, C3 and C5 atoms
occur between 72 and 79 ppm and finally the C6 peak has a chemical shift value of
65 ppm. The absence of any aromatic signals between 110 and 140 ppm clearly indicated
that the lignin component present in bagasse had been successfully eliminated as a result
of the alkali treatment and the subsequent bleaching process.
   The earliest published work on the solid-state NMR spectra of cellulose showed two
peaks in the chemical shift range 80–92 ppm and these have been assigned to the C4
carbon atom (24, 25). A relatively sharp peak was assigned to crystalline regions, and
a relatively broad peak was attributed to the crystallite surfaces and the amorphous/
disordered domains.
   Previous assignments by Newman (26, 27), Horii (28), and Iversen (29) have corre-
lated the solid-state NMR spectra of cellulose with its structure and morphology. A weak
shoulder on the C6 peak, between 63 and 65 ppm, has been attributed to the amorphous
and disordered component in cellulose. This includes the surface of crystal domains
since they need not participate in the symmetry of the crystal. The 13 C NMR spectra of
the cellulose fibers before and after hydrolysis have some very distinct differences:
1. The signal assigned to the C4 peak changed dramatically. There was a significant
   difference between the peak profiles of the crystalline and the amorphous compo-
   nents attributed to signals in this region, (80–92 ppm) before and after hydrolysis and
   mechanical shearing. The unhydrolysed cellulose fibers exhibited roughly equal con-
   tributions from the crystalline and the amorphous domains (Figure 8.9a). Newman
   and his co-workers have used curve-fitting specific assignments to the signals arising
   from crystallite interiors, crystallite surfaces as well as the amorphous regions. No
   such attempt was made in our case because of the large number of scans required
216   The Nanoscience and Technology of Renewable Biomaterials




                         140               100              60
                                          (ppm)
                                           (a)




                      140                100                 60
                                         (ppm)
                                           (b)

Figure 8.8 13 C CP/MAS NMR spectrum of cellulose (a) fiber (whole cells) and (b) microfibers
after hydrolysis.


   to obtain a sufficiently large signal to noise ratio that would justify such curve fit-
   ting. However, after hydrolysis of the cellulose fibers, we observe that the signal
   attributed to the crystalline regions at around 89 ppm is sharper and far better defined
   (Figure 8.9b). Moreover, the ratio of the peak intensities for the crystalline region
   to that of the amorphous region also increases significantly with hydrolysis. This
   observation agrees with our model of cellulose hydrolysis in which the accessible
   amorphous and surface regions react before the crystalline interiors. The increase in
   both intensity and sharpness of the C4 crystalline component peak clearly indicates
   that we were successfully able to eliminate the amorphous and disordered domains
   leaving behind collections of well-defined crystalline microfibrils.
2. The C6 signal appearing at 63 ppm had a well-defined shoulder that has normally
   been attributed to the amorphous component in cellulose. We find that the hydrolytic
   treatment of the whole cellulose fibers resulted in the main C6 peak becoming sharper.
                                     Cellulose and Chitin as Nanoscopic Biomaterials      217




                                             (a)




                                     100              80
                                            (ppm)
                                             (b)

Figure 8.9 Expansion of the C4 region, 80–100 ppm, in cellulose 13 C CP/MAS NMR spectra,
for (a) intact cellulose fibers and (b) cellulose MFs obtained after hydrolysis and dispersion.


   The shoulder in the C6 signal that was associated with the disordered regions also
   decreased considerably. This provides further support for preferential degradation of
   cellulose as a result of hydrolysis and mechanical dispersion.
  13
      C CP-MAS NMR spectroscopy was also used to estimate an average degree of
N-acetylation (DSacetyl ) for the chitin nanocrystals, and to assess changes in acetylation
after hydrolysis. The DSacetyl was determined by the ratio of the integration values of
the methyl carbon to the anomeric carbon signal. The methyl signal is preferred over
the carbonyl signal due to the attached protons, which allow for better magnetization
transfer in the cross-polarization experiment.
    The DSacetyl from CP-MAS NMR was calculated to be 0.90 for the chitin nanocrystals
after hydrolysis and 0.89 prior to hydrolysis, indicating that the hydrolysis treatment had
little effect on the DSacetyl . The NMR spectrum for the shrimp shell chitin nanocrystals is
displayed in Figure 8.10. The signal assignments shown are based upon a paper published
by Tanner and Chanzy et al. (30). Furthermore, from the spectrum in Figure 8.10 we
can confirm that the chitin nanocrystals are essentially free from residual protein. The
solid-state NMR spectrum for a pure, unhydrolyzed α-chitin sample is not noticeably
different from that of the nanocrystals, and is not shown. Surface and amorphous
regions of chitin are not detectable by 13 C CP MAS NMR, thus an analysis similar to
that performed with cellulose, as described above, was not possible.
    X-ray powder diffraction (XRD) was used to monitor changes in crystallinity and
morphology of the chitin nanocrystals upon acid hydrolysis, and to estimate crystallite
sizes in the chitin samples. Figure 8.11 shows the diffraction profiles of the chitin
nanocrystals, and the purified shrimp shell chitin prior to hydrolysis. The diffraction
patterns for both materials exhibit Bragg reflections typical of pure α-chitin, indicating
that the chitin crystal structure is maintained after hydrolysis.
    Analysis of the XRD data from the native chitin and the chitin nanocrystal
samples indicated that the crystallinity of the material was observed to increase after
218                      The Nanoscience and Technology of Renewable Biomaterials

                                                                                C-3
                                                                    C-5

                                                           C-1
                                                                                      C-2
                                                                                                       CH3
                                                                   C-4


                                                                                C-6



                               C=O




                        190 180 170 160 150 140 130 120 110 100 90 80 70           60   50   40   30    20   PPII
                                                13C Chemical shift (ppm)

Figure 8.10 13 C CP MAS NMR spectrum of chitin nanocrystals. Reprinted with permission
from (23). Copyright (2007), American Chemical Society.

                        3000
                                                                  110
                                     020
                        2500               Chitin NC                                          013

                                                                          120
                                                                                  130
                        2000                         021
  Arbitrary Intensity




                        1500
                                           Pure Chitin

                        1000



                         500



                           0
                               7                12         17                22                   27
                                                           2 Theta (Degrees)

Figure 8.11 X-ray diffraction curves of chitin nanocrystals (top), and shrimp shell chitin
before hydrolysis (bottom). Assigned Miller indices for α -chitin are noted above the upper
curve. Reprinted with permission from (23). Copyright (2007), American Chemical Society.
                                                 Cellulose and Chitin as Nanoscopic Biomaterials            219

hydrolysis. To determine sample crystallinity, the diffraction data was smoothed with a
Savitzky-Golay filter using a second-order polynomial regression with 10 points, decon-
voluted, and fit with Gaussian-Lorentzian line shapes, all performed in the Peak Fit V.4
software package. This method of peak fitting provided a reasonable overall curve fit
with a correlation coefficient of 0.95 for all samples. The six most intense crystalline
diffraction peaks were observed for 5◦ ≤ 2θ ≤ 30◦ , and were indexed as the 020, 021,
110, 120, 130 , and 013 reflections according to the unit cell of α-chitin as reported by
Minke and Blackwell (31). The deconvolution of the XRD data for the chitin nanocrys-
tals sample is presented in Figure 8.12 and assumed the 2θ values predicted by the
reported Minke and Blackwell unit cell as noted in Table 8.1.
   The crystallinity was measured as the ratio of the sum of the areas under the crystalline
diffraction peaks to the total area under the curve for 5◦ ≤ 2θ ≤ 30◦ , based on a method
proposed for cellulose and used here for chitin (32). From this analysis we found the
percentage crystallinity of the pure shrimp shell chitin to be approximately 76%, and


                       1250                                                                          1250
                       1000                                                                          1000
           Intensity




                        750                                                                          750
                        500                                                                          500
                        250                                                                          250

                                                               19.4
                       1000         9.368                                                            1000
                        750                                              23.72 26.648                750
           Intensity




                       500                                       20.888                              500
                                                          17.864      22.472          29.672
                       250                  12.824                                                   250
                                                                                    28.232
                         0                                                                           0
                              5                      15                        25               35
                                                      2 Theta (Degrees)

Figure 8.12 Deconvolution of chitin nanocrystal X-ray diffraction data. Top: dotted line
represents smoothed data, and the solid line represents the overall fit. Bottom: individual
peak fits. Reprinted with permission from (23). Copyright (2007), American Chemical Society.


                                  Table 8.1 Observed and literature reported
                                  2θ values for the 6 most intense crystalline
                                  diffraction peaks observed in α -chitin (31).
                                  hkl                     2θliterature                  2θObs
                                  020                       9.39                         9.37
                                  021                      12.72                        12.82
                                  110                      19.30                        19.40
                                  120                      20.95                        20.88
                                  130                      23.54                        22.47
                                  013                      26.37                        26.65
220   The Nanoscience and Technology of Renewable Biomaterials

increasing to 84% for the nanocrystals after acid hydrolysis and homogenization. This
signifies a reduction in the amorphous contribution to the diffraction data, and suggests
that some of the amorphous regions in the cellulose and chitin are digested during the
hydrolysis process. The intensity of the Bragg peaks in the nanocrystal sample is slightly
less than those in the pure chitin sample, indicating that some of crystalline domains
are mildly affected by the hydrolysis treatment. However, with the increasing percent
crystallinity upon hydrolysis, the losses due to the digestion of the amorphous material
are greater than the losses to the crystalline material.
   Information regarding the average crystallite sizes for the chitin materials was also
obtained from the X-ray diffraction data. Curve deconvolution permitted measuring the
peak widths at half-maximum intensity so that crystallite size could be calculated using
the Scherrer relation (33):
                        Dhkl = (0.9)(λCuKα )/(FWHM)hkl (cos θ )hkl
where FWHM is the full width at half maximum intensity, in radians, for a single Bragg
peak, λCuKα is the wavelength of the X-ray radiation (0.15418 nm), and θ represents
the location of the maximum intensity of a particular Bragg reflection. The FWHM is
used to measure the line broadening, which arises primarily from the finite size of the
crystallites. Paracrystallinity and instrumental broadening factors can contribute to line
broadening as well but were not evaluated in this work.
   Based on crystallographic studies of α-chitin from various sources it is apparent that
the molecular repeat axis and the microfibrillar axis are parallel, so that the 001 set
of planes correspond to the repeat period along the major axis of the microfibrils (31,
34–36). Also, it has been previously shown that the 100 set of planes correspond to the
growth plane for α-chitin (34). Given the orthorhombic geometry of the crystals, if the
normal to the 001 planes is parallel to the major axis of the microfibrils, then the 100 and
010 sets of planes, which are perpendicular to the 001 planes, represent periodicities on
the transverse axes of the microfibrils Line broadening from Bragg peaks for these planes
or multiplicities thereof can be used to measure crystallite widths. Crystallite dimensions
for the 020 and 110 reflections appearing in the data were used to derive the width along
the 100 set of planes, which is coincident with the crystallite width. Due to the high
crystallinity of the particles and the apparent small transverse dimensions of the particles
visualized from TEM, it is apparent that the crystallite width, in this system, measures
the width of the microfibrils. Further justification for this assumption comes from a
previous study on α-chitin from lobster tendons, where a strong resemblance between
α-chitin microfibrils and cellulose microfibrils is observed, in that with both materials,
the microfibrils are elongated single crystallites (34). From the line broadening data,
crystallite dimensions for the chitin nanoparticles were calculated at 8.33 nm for the 100
planes, and 6.65 nm for the 010 planes. This corresponds to crystallites approximately
6.65 nm × 8.33 nm in cross-section.


8.5   Chemical Modification of Cellulose and Chitin Nanoparticles

A common drawback, which has impeded the use of cellulosic and chitin fibers as
reinforcing agents in thermoplastic composites, has been the lack of compatibility at
                                                     Cellulose and Chitin as Nanoscopic Biomaterials                     221

            OH                                                                                 O(CO)CHCHCO2H
                                       O        O             NaHPO4
                    O                                   O
                                  +3                                                  O                O
     O HO                                                     xylene
                                                                D           HO(CO)CHCH(CO)O
                        OH                                                                              O(CO)CHCHCO2H
                              n                                                                                          n
      Cellulose MF Surface                 Maleic Anhydride                          Surface Modified Microfibers




      OH
                   OH                                                       OH
                                                                                      OCOCHCHCOOH
                     OH
HO
                                                                                          OH
                                                                       HO
     HO                      OH                 OH
                                                                            HO                 OH           OCOCHCHCOOH
              HO                                        OH
                                                                       HOOCHCHCOCO                                  OH
                                              OH
                             OH
                                                                                                           OCOCHCHCOOH
                                                                                               OH


Scheme 8.1 Molecular modification shown at the top. A representation of nanocrystal
surface modification is shown at the bottom.


the fiber matrix interface. Hydrophilicity of the cellulose particle surface can result
in poor interfacial adhesion of thermoplastics. This can be controlled by hydrophobic
modification of the particle surface. It is crucial to note that any chemical modification
of the hydroxyl groups in cellulose must be restricted to the surface, if the particles are
to be used as a reinforcing element in composites, since interior changes would decrease
crystallinity, particle size and thickness.
   The surface of cellulose nanocrystals isolated from bagasse were decorated with
maleate ester groups, which can serve as branch points for subsequent grafting of
aliphatic chains onto the crystal surface, resulting in the formation of brushlike struc-
tures. The presence of an unsaturated alkene group on the surface of these microfibers
provides opportunity for further topochemical modification. A scheme for the maleate
derivatization of cellulose is provided in Scheme 8.1.
   Attenuated total reflectance (ATR), FTIR was used to confirm chemical modification.
From the absorbance peaks at 1095 cm−1 (C–O–C stretching), 3400 cm−1 (–OH stretch-
ing), and 1725 cm−1 (C=O stretching), it can be concluded that the spectra are consistent
with both cellulose (in the case of the starting material) and partially substituted cellulose
(Figure 8.13).
   Several bands consistent with the maleated ester of cellulose are present in the spectra.
The most notable feature is a rather broad carbonyl band ‘a’ that increases with reaction
time. The carboxyl band initially has a maximum near 1725 cm−1 in the spectra of the 3
and 5 hour reaction time products, and then shifts to 1712 cm−1 for the 9-hour reaction
time product. This suggests that it is a composite carbonyl with contributions from both
the conjugated ester and the carboxylic acid functionality. Both of these functionalities
would be expected to exist in the maleated product spectrum, and the increasing intensity
of the carbonyl is consistent with an enhanced degree of substitution in the product.
   Another notable feature in the spectra of maleated cellulose is a very broad –OH band,
from 2100 cm−1 to 3200 cm−1 , consistent with acidic –OH functionality (band ‘b’). This
222             The Nanoscience and Technology of Renewable Biomaterials


                                 Cellulose Maleate, 9 hrs
              0.30
                                 Cellulose Maleate, 5 hrs                  d
              0.25               Cellulose Maleate, 3 hrs
                                 Cellulose
              0.20                                                  a

              0.15
Absorbance




                                 c
                                                   b
              0.10

              0.05

              0.00

             −0.05

             −0.10

                4000                 3000                   2000               1000
                                                Wavenumber (cm−1)

Figure 8.13 ATR-FTIR spectra of surface modified cellulose MFs at different reaction times.
Important spectral features are labeled ‘a’ through ‘d’. Band ‘a’ denotes the carbonyl
stretching frequency, band ‘b’ is the acid –OH stretching frequency, band ‘c’ marks the
cellulose hydroxyl functionality, and band ‘d’ marks both ester and acid C–O stretching
frequencies. All labels indicate the esterification of cellulose with the maleate functional
group.


band further supports the presence of carboxylic acid functionality from the maleated
product, with the absorption increasing with increasing reaction time and presumed
degree of substitution. A comparison of the spectra of the 3- and 5-hour products
with native cellulose suggests a progressive decrease of the cellulosic –OH contribution,
as marked by the decrease in the broad 3340 cm−1 –OH band (band ‘c’). However,
the 9-hour spectrum does not continue this pattern, possibly because of the overlapping
acidic –OH absorption band, or interactions between the acidic and cellulosic –OH units.
Finally, there were bands, consistent with both ester and acid carbonyl C–O stretches,
observed in the 1300–1100 cm−1 regions (band ‘d’) whose intensity increased with
increasing reaction time and presumed degree of substitution. No attempt was made to
further investigate these or other bands.
   X-ray diffraction studies were carried out in order to determine whether there was any
change in crystallinity during the course of the reaction. Cellulose I has a characteristic
intense peak at a 2θ value of 22.5 degrees (37, 38). This corresponds to the 002 plane
using the convention of the b axis as the chain direction. We find that with an increase
in esterification time there is a measurable broadening of the peaks associated with the
101 , 110 , and 002 planes of reflection. (These indices use the convention of the fiber
axis parallel to the b axis.) This broadening implies shrinkage in crystallite size along
the normal to each indexed set of planes. Since there were no significant changes in the
position of maximum intensity during the course of the reaction, we conclude that the
crystalline core of these microfibers remained in the cellulose I allomorph. Since this
                                      Cellulose and Chitin as Nanoscopic Biomaterials       223

                   4500


                   4000
                                                          Pure Cellulose

                   3500


                   3000                                    5 h reaction


                   2500
       Intensity




                                                           9 h reaction
                   2000


                   1500


                   1000


                    500


                      0
                          0   10            20             30              40          50
                                                   2q

Figure 8.14 X-ray diffraction studies showing the relative decrease in crystallinity at different
reaction times.


crystal structure cannot accommodate the bulky maleate groups, we conclude that the
esterification reaction was restricted to the surface at low reaction times.
   We did find a decrease in the intensity of the 002 peak with an increase in reaction
time. The cellulose MFs that were esterified up to nine hours showed a significant
decrease in peak intensity at 22.7◦ indicating the possibility that the reaction had pro-
gressed from the amorphous regions to the crystalline domains at long esterification
times (Figure 8.14).
   Chitin nanoparticles were derivatized with medium-chain aliphatic esters to improve
their surface hydrophobicity. These materials were characterized by FTIR. The IR spec-
tra for the chitin nanocrystals, chitin hexanoate, chitin nonanoate, and chitin stearate
are shown in Figure 8.15. Arrows on the top spectrum representing the chitin stearate
denote the most significant new features associated with esterification. After reaction,
with any of the organic acids, a signal arises at 1750 cm−1 that can be assigned as
the ester carbonyl stretching frequency by analogy with cellulose esters, where the ester
carbonyl signal appears at a very similar frequency (39). FTIR spectra of the corre-
sponding pure organic acids shows a higher frequency for the acid carbonyl stretching
signals, at 1760 to 1780 cm−1 (40). The presence of a carbonyl signal in the chitin
esters, its moderate shift to lower energy relative to the typical free acid carbonyl signal,
and its stability to repeated washing with various solvents lead us to conclude that the
long aliphatic side-chains are attached to chitin through an ester linkage, as opposed
224    The Nanoscience and Technology of Renewable Biomaterials

                                                                                   1.00

                                                                                   0.90
                          C-H stretching    C=O ester stretch
                                                                                   0.80

                                                                                   0.70
                                        d
                                                                                   0.60




                                                                                          Absorbance
                                                                                   0.50
                                        c
                                                                                   0.40

                                                                                   0.30
                                        b
                                                                                   0.20

                                                                                   0.10
                                        a
                                                                                   0.00
      3900      3400      2900       2400       1900        1400      900       400
                                 Wavenumbers (cm−1)

Figure 8.15 Stacked FTIR spectra of chitin and chitin esters. (a) chitin nanocrystals, (b) chitin
hexanoate, (c) chitin nonanoate, (d) chitin stearate.


Table 8.2 Contact angle measurements for the chitin nanoparticles.
Sample              Contact angle (deg)             Polar          Dispersive    Surface energy
                                                  (dyne/cm)        (dyne/cm)       (dyne/cm)
                Water      Methylene iodide
Chitin NC         27               41                  41             25               66
Chitin C6         95               30                   1             45               46
Chitin C9        121               38                   4             48               53
Chitin C18       119               41                   3             46               49



to existing as free unattached acids perhaps stabilized by hydrogen bonds to the chitin
molecule.
   Also, after modification, we observe increasing signal intensities in the region between
2930 and 2860 cm−1 , corresponding to asymmetric and symmetric stretches of the methy-
lene groups, respectively. Chitin has broad bands in this region with signals arising
from C–H stretching in the pyranose ring. After esterification, the signals in this region
sharpen and increase in intensity as longer aliphatic esters are attached and the population
of methylene groups increases.
   The results from static contact angle experiments using the sessile drop method are
provided in Table 8.2. They demonstrate the increase in hydrophobic character at the
particle surface as a consequence of derivatization. Other analyses were performed, but
are not shown, which further confirm the chemical derivatization of the surface of chitin
nanoparticles.
                                       Cellulose and Chitin as Nanoscopic Biomaterials   225

8.6   Nanocomposite Properties

Surface derivatized cellulose and chitin nanoparticles were melt processed with differ-
ent biobased plastic matrices to yield nanocomposites and the effects of the surface
modification on thermal and mechanical properties were measured. Maleated cellulose
nanoparticles were blended with a biodegradable co-polyester of adipic acid, terephalic
acid, and 1,4 butanediol. The neat plastic is elastomeric at room temperature with a
Tg = −32 ◦ C, and has a relatively low melt processing temperature of 140 ◦ C. The low
melt temperature of the matrix plastic reduces the risk of thermal decomposition of
the cellulose nanoparticles during processing. Nanocomposites prepared with varying
weight % filler showed improved mechanical properties. Table 8.3 provides the stor-
age modulus data for the nanocomposites. Modulus values were taken from dynamic
mechanical analysis profiles. The results indicate a large reinforcing effect imparted to
the materials by the nanoparticles. This reinforcing effect is more marked with sur-
face derivatization, presumably due to the greater surface compatibility. Tan δ plots
of the nanocomposites at different filler loadings indicate an improvement in damping
properties with reinforcement as shown in Figure 8.16.
   Surface derivatized chitin nanoparticles were melt processed with a C-9 long chain
cellulose ester matrix plastic in a similar manner to the cellulose nanocomposites. The
neat C-9 cellulose ester was an experimental type provided by Eastman Chemical Com-
pany, and is also a low Tg elastomeric thermoplastic with poor mechanical strength in
neat form. When reinforced with chitin nanoparticles surface derivatized with medium
to long chain aliphatic esters, a marked improvement in modulus of the composite
material is recognized. Mechanical data for the composite materials is provided in
Table 8.4. Tensile testing of the composites reveals that both the yield stress and
Young’s modulus of the materials are increased nearly 2-fold with nanoparticle rein-
forcement. Figure 8.17 illustrates this effect, where the slope of the linear part of the
curve equates to the Young’s modulus, and the point of deviation from linearity or elastic


                Table 8.3 Comparative values of storage modulus of the
                thermoplastic filled with unmodified and surface modified
                cellulose nanoparticles at different filler loadings.
                Weight fraction (wf)             Storage modulus (MPa)

                                             −25◦ C       25◦ C        75◦ C
                0.0                           9100         5500       1400
                Unmodified MFs
                0.05                         11200         6200       1800
                0.10                         14450         7400       1900
                0.20                         17300         8500       2400
                0.30                         20400        10500       2900
                Maleated MFs
                0.05                         10700         6800       2600
                0.10                         16200        10500       3900
                0.20                         25100        16200       5700
                0.30                         39800        24500       8300
226               The Nanoscience and Technology of Renewable Biomaterials



                   0.5
                                                                                Pure Matrix
                  0.45                                                          10% filler
                   0.4                                                          30% filler
                                                                                20% filler
                  0.35
                   0.3
      tan d




                  0.25
                   0.2
                  0.15
                   0.1
                  0.05
                        0
                         −80          −60         −40          −20                 0                     20
                                                        Temp °C


                   Figure 8.16 Tan δ plots for the neat plastic and the reinforced composites.

                  6.0


                  5.0


                  4.0
   Stress (MPa)




                  3.0


                  2.0                                                      Cellulose Nonanoate

                                                                           Cellulose Nonanoate + 5% C9

                  1.0                                                      Cellulose Nonanoate + 10% C9

                                                                           Cellulose Nonanoate + 15% C9


                  0.0
                        0        20          40           60          80                100                   120
                                                        % Strain

Figure 8.17 Stress-strain curves for the C-9 cellulose ester reinforced with C-9 surface
derivatized chitin nanoparticles.

behavior marks the yield stress. It is also observed that much of the mechanical rein-
forcement gained is achievable with only 5 wt% filler loadings. The mechanical data
provided for both cellulose and chitin nanoparticles demonstrates that these biomaterials
have great potential for providing mechanical reinforcement to elastomeric thermoplastics
when used as filler materials in composite applications.
                                    Cellulose and Chitin as Nanoscopic Biomaterials    227

Table 8.4 Storage modulus values for C-9 cellulose ester reinforced with chitin and surface
derivatized chitin nanoparticles. The particles are observed to provide similar or better
reinforcement relative to nanoclay fillers.
Weight fraction                                         Storage modulus (MPa)

                                            −100 ◦ C              30 ◦ C              70 ◦ C
0.00                                          1906                 112                  4
Unmodified chitin nanocrystals
0.05                                          2006                 169                 28
0.10                                          2096                 190                 32
0.15                                          2070                 174                 27
C6 esterified nanocrystals
0.05                                          1994                 157                 24
0.10                                          2260                 217                 33
0.15                                          2145                 203                 30
C9 esterified nanocrystals
0.05                                          2392                 175                 29
0.10                                          2325                 180                 34
0.15                                          2475                 222                 38
C18 esterified nanocrystals
0.05                                          1973                 146                 24
0.10                                          2440                 168                 31
0.15                                          2072                 205                 33
Nanoclay
0.05                                          1906                  94                  9
0.10                                          2240                 150                 13
0.15                                          2190                 190                 17



8.7 Conclusions

Cellulose and chitin nanoparticles were effectively isolated from bagasse, and shrimp
shells, respectively, both waste products of the food industry. Microscopic character-
ization of these materials has shown them to be nanoparticle and highly crystalline in
nature. The abundance of hydroxyl groups available on the surface of these materials
facilitated their topochemical modification. The confinement of the reactions to the par-
ticle surface, and the preservation of the crystal integrity were confirmed by XRD, and
FTIR techniques. Contact angle analysis has shown that the hydrophobicity of the chitin
particles was greatly enhanced after modification with medium- to long-chain aliphatic
esters. The nanoparticles were melt processed with a biodegradable co-polyester and
biobased cellulose ester matrix phase. Mechanical results indicate that these materials
impart substantial strength in composite applications. The data provided within this
chapter is a mere subset of the potential that nanoscopic biomaterials such as cellulose
and chitin possess in the field of composite materials. There are suggestions from sev-
eral laboratories that transcrystallization at the nanoparticle surface may contribute to
the increased modulus of nanocomposites. If so then facilitating interfacial crystalliza-
tion through surface modification may prove to be a robust route to the formation of
228     The Nanoscience and Technology of Renewable Biomaterials

strong nanocomposites. There are endless combinations cellulose and chitin nanoparticles
derivatives and thermoplastic composite blends that can be explored.


Acknowledgements

The authors would like to thank Dr Lou Germinario at Eastman Chemical for the AFM
and SEM analyses of the cellulose microfibers. Much of this work was also supported
by a grant from the EPA to William T. Winter.


References

1.    National Nanotechnology Initiative, URL: http://www.nano.gov/html/facts/
      whatIsNano.html
2.    Okada, A.; Kawasumi, M.; Usuki, A.; Kojima, Y.; Kurauch, T.; Kamigaito, O.
      Mater. Res. Symp. Proc. 1990, 171, 45.
3.    Favier, V.; Canova, G.R.; Cavaille, J.Y.; Chanzy, H.; Dufresne, A.; Gauthier, C.
      Polym. Adv. Technol. 1995, 6, 351.
4.    Nair, K.G.; Dufresne, A. Biomacromolecules. 2003, 4, 1835.
5.    Grunert, M.; Winter, W.T. J. Polym. Envir. 2002, 10, 27–30.
6.    Krassig, H.; Schurz, J.; Steadman, R.G.; Schliefer, K.; Albrecht, W. Cellulose, in
      Ullman’s Encyclopedia of Industrial Chemistry. 5th ed. Vol. A., New York, VCH
      Publishers, 1986, 375.
7.    Muzzarelli, R.A.A. Chitin, Permagon Press, Oxford, 1977.
8.    Roberts, G.A.F. In Chitin Chemistry; Roberts, G.A.F., ed.; Macmillan Press Ltd:
      London 1992.
9.    Knorr, D. Food Technol. 1991, 45, 114.
10.   Sturcova, A.; Davies, G.R.; Eichhorn, S.J. Biomacromolecules. 2005, 6, 1055.
11.   Kotaro, G. Zairyo. 1993, 42(475), 355.
12.   Bhardwaj, R.; Mohanty, A.K.; Drzal, L.T.; Pourboghrat, F.; Misra, M. Biomacro-
      molecules. 2006, 7, 2004.
13.   Huda, M.S.; Drzal, L.T.; Mohanty, A.K.; Misra, M. Comp. Sci. Technol. 2006, 66,
      1813.
14.   Bhattacharya, D.; Germinario, L.T.; Winter, W.T. Carbohydr. Polym. 2008, 73(3),
      371.
15.   Elazzouzi-Hafraoui, S.; Nishiyama, Y.; Putaux, J.; Heux, L.; Dubreuil, F.;
      Rochas, C. Biomacromolecules. 2008, 9, 57.
16.   Frey-Wyssling, A. Science. 1954, 119, 80.
17.   Dong, X.; Revol, J.; Gray, D. Cellulose. 1998, 5, 19.
18.   Revol, J.; Bradford, H.; Giasson, J.; Marchessault, R.; Gray, D. Int. J. Biol. Macro-
      mol., 1992, 14, 170.
19.   Baker, A.; Helbert, W.; Sugiyama, J.; Miles, M. Biophys. J. 2000, 79, 1139.
20.   Hess, K.; Kiessig, H.; Gundermann, R. Z. Phys. Chem., 1941, Vol. B 49, 64.
21.   Boyd, R.; Badyal, J. Adv. Mat. 1997, 9, 895.
22.     a
      H˚ kansson, H; Ahlgren, P. Cellulose, 2005, 12, 177.
                                Cellulose and Chitin as Nanoscopic Biomaterials   229

23. Goodrich, J,D.; Winter, W. T. Biomacromolecules. 2007, 8(1), 252.
24. Earl, W.; Vanderhart, D. J. Amer. Chem. Soc. 1980, 102, 3251.
25. Attala, R.; Gast, J.; Sindorf, D.; Bartuska, V.; Maciel, G. J. Amer. Chem. Soc.
    1980, 3249, 102.
26. Newman, R.; Hemmingson, J. Cellulose, 1995, 2, 95.
27. Newman, R. Holzforschung, 1998, 52, 157.
28. Horii, F.; Hiraii, A.; Kitamaru, R. Macromolecules, 1987, 20, 2117.
29. Lennholm, H.; Larson, T.; Iversen, T. Carbohydr. Res. 1994, 261, 119.
30. Tanner, S.F.; Chanzy, H.; Vincendon, M.; Roux, J.C.; Gaill, F. Macromolecules.
    1990, 23, 3576.
31. Minke, R.; Blackwell, J. J. Mol. Biol. 1978, 120, 167.
32. Hermans, P.H.; Weidinger, A. J. Appl. Phys. 1948, 19, 491.
33. Patterson, A.L. Phys. Rev. 1939, 56, 978.
34. Revol, J.F.; Int. J. Biol. Macromol. 1989, 11, 233.
35. Persson, J.E.; Domard, A.; Chanzy, H. Int. J. Biol. Macromol. 1992, 14, 221.
36. Saito, Y.; Okana, T.; Chanzy, H.; Sugiyama, J. J. Struct. Biol. 1995, 114, 218.
37. Mukherjee, S.; Wood, H. Biochim. Biophys. Acta. 1953, 10, 499.
38. Isogai, A.; Usuda, M.; Attala, R. Macromolecules. 1989, 22, 3172.
39. Jandura, P.; Kokta, B.V.; Riedel, B. J. Appl. Polym. Sci. 2000, 78, 1354.
40. SDBSWeb: http://www.aist.go.jp/RIODB/SDBS/ (National Institute of Advanced
    Industrial Science and Technology, 10/25/06).
                                                        9
Bacterial Cellulose and Its Polymeric
           Nanocomposites

                                           Marie-Pierre G. Laborie



9.1     Introduction

In the last decade, a great deal of materials development has involved bacterial cellulose
(BC) as reinforcement for both thermosetting and thermoplastic polymers, synthetic or
bio-based polymers. Due to the network and nanoscale structure of bacterial cellulose
fibers, the reinforcement of polymers with BC leads to nanocomposites with unique
morphologies and properties.
   A common challenge in developing nanocomposites is to finely disperse the nanoscale
reinforcement into the matrix such that homogeneous nanocomposites with optimum per-
formance can be achieved. To circumvent this challenge with BC nanocomposites, sev-
eral manufacturing approaches have been taken. In a first approach, reactive monomers
or oligomers are polymerized in situ, i.e. within a BC mat, resulting in an interpenetrat-
ing network or at least a well dispersed nanocomposite. Such an approach has been taken
with phenolic, acrylic and epoxy resins yielding applications as optoelectronic devices
as well as biomaterials and membranes. A second approach that has been most com-
monly used with thermoplastic polymers consists of solvent casting BC and a polymer
solution into solid shapes. With this second approach, BC has been incorporated into
various bio-based polymers including cellulose acetate butyrate (CAB), xylans, starch
and proteins but also with synthetic polymers such as poly(vinyl alcohol) (PVA). Finally,
biomimetic approaches, in which the growth medium of BC is augmented with host poly-
mers allowing for the in vitro assembly of the components, have paved the way for the
development of nanocomposites with unique morphologies and properties. In particular
a wide range of BC/hemicelluloses or BC/lignin nanocomposites have been developed
with this approach with a view to shedding light on the biosynthesis, morphology and

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
232   The Nanoscience and Technology of Renewable Biomaterials

properties of the primary and secondary cell walls of plants and trees. Synthetic polymers
have also been used in biomimetic approaches to develop well dispersed nanocomposites.
In this review, research and developments on BC nanocomposites obtained from these
three manufacturing approaches are successively reviewed, placing a special emphasis
on the nanocomposite morphology and performance. A few studies have used bacterial
cellulose nanocrystals to reinforce polymeric matrices and these advances are reviewed
next. Prospect for the future developments in BC nanocomposites are finally proposed.


9.2   Bacterial Cellulose: Biosynthesis and Basic Physical
      and Mechanical Properties

The biosynthesis and properties of BC have been extensively reviewed (1–3) and are
therefore only briefly examined in this review. Special attention is given to the mechan-
ical properties of BC sheets as they constitute a good reference from which to examine
the performance on BC nanocomposites.

9.2.1 Synthesis and Properties of BC
Cellulose is a semi-crystalline high molecular weight homopolymer of β-1,4 linked
anhydroglucose (Figure 9.1). Many living organisms synthesize cellulose affording a
wide range of supramolecular structures, morphologies and properties. Among all cel-
lulosic materials, bacterial cellulose displays the highest mechanical properties. In fact,
it has a tensile strength and Young modulus comparable to Kevlar (4). The outstanding
performance of bacterial cellulose stems from its high purity, high crystallinity (75%)
and ultra-fine network structure (4). Bacterial cellulose also has the highest DPn (up
to 8000), aspect ratio and strength to weight ratio, making it an ideal reinforcement for
natural composites. Besides it can hold ca 100% of its weight of water, defining it as
an hydrogel.
   Cellulose produced by bacteria is most commonly of the cellulose I type, although
one bacteria also produces cellulose II (2). In cellulose I the chains are oriented parallel
with a spacing of 0.53 nm between the glucan chains. Native cellulose I has two
suballomorphs, Iα and Iβ . The former exists as a single chain triclinic unit cell while
the latter exists as a two-chain monoclinic unit cell (Figure 9.1). Many organisms
produce cellulose, including plants, eukaryotic bacteria, procariotic organisms and fungi.
However Gluconacetobacter xylinum, a rod shape aerobic gram negative bacteria, is
most commonly used to produce BC nanocomposites.
   When cellulose is produced from Acetobacter Xylinum, 12 to 15 cellulose chains are
extruded from the enzymatic terminal complexes into the culture medium as subelemen-
tary fibrils that have a lateral width of 1.5 nm and are amorphous (5). The subelementary
fibrils aggregate and crystallize into 3–6 nm wide microfibrils that comprise cellulose
Iα and Iβ allomorphs (Figure 9.1) (4–6).

9.2.2 Performance of BC Mats
Researchers in Japan first evaluated the mechanical performance of air dried and hot-
pressed bacterial cellulose mats (7–9). Tensile measurements of BC sheets hotpressed
                                     Bacterial Cellulose and Its Polymeric Nanocomposites                      233

                      Glucan chain aggregate        Microfibril      Cellulose ribbon
                      (6–8 chains)                  (3–4 nm wide)    (40–60 nm wide)




                                     Acetobacter Xylinum
   (a)




   (b)
                                                    b

                                                           a


                      Cellulose 1α                                     Cellulose 1β



                                                                      OH              CH2OH              OH
                               OH               CH2OH
         CH2OH
                                                           O   HO            O                O   HO
                 O   HO                O
   (c)                                                         O        O    HO                    O       O
                     O           O     HO

                                                     HO             CH2OH               HO             CH2OH
            HO             CH2OH




Figure 9.1 Biosynthesis (a), crystalline structure (b) and chemical structure (c) of cellulose
produced by Acetobacter Xylinum (Partially reproduced, with permission of Taylor & Francis
Informa UK, Ltd – Journals, from: The biosynthesis of cellulose. Brown, R.M.J., Journal of
Macromolecular Science – Pure and Applied Chemistry A33(10), 1996; permission conveyed
through Copyright Clearance Center, Inc.) (1).


under various conditions of pressure (49 to 1960 Mpa) and temperature (120, 150,
200 ◦ C) for 5 min revealed that tensile strength and elongation were sensitive to the
hotpressing conditions whereas Young modulus was little affected (7). Young modulus
was rather constant, in the 16–18 GPa range, while the tensile strength and elongation
varied from the 102–260 MPa range and the 0.8–2.1% range, respectively. Interestingly
air-dried BC sheets were found to perform as well as hotpressed sheets since their tensile
modulus, strength and elongation reached 16.9 GPa, 256 MPa and 1.7% respectively,
falling within the range of those measured for the hotpressed samples. The tensile
strength and elongation decrease with higher pressure was ascribed to the introduction
of defects (9). Addition of disintegrated BC to cotton lint pulp also improved the Young
modulus and tensile strength of paper sheets while elongation was little affected by BC
addition (7). The linear increase of Young modulus and tensile strength with BC con-
tent was ascribed to the finer BC fibers that were able to H-bond and give much better
strength (9).
234   The Nanoscience and Technology of Renewable Biomaterials

   The washing procedure of the BC pellicle, was also found to impact the performance
of BC sheets (8). When BC was washed with varying concentrations of NaOH or
NaClO (0–1% wt/wt, as active Cl), the tensile modulus varied with the treating solution
concentration and passed through a maximum suggesting competition of two factors. The
maximum Young modulus was measured (24 GPa) at 0.5% NaClO% and at 5% NaOH.
In the case of the NaClO treatment, the noncellulosic components, i.e. proteinaceous
cell debris, were effectively removed with increasing concentration thus allowing better
interfiber bonding. At the same time NaClO caused a disintegration of the cellulose
chains thus lowering the overall stiffness. These two competing effects converged to
provide the highest modulus at 5% concentration. Similarly NaOH which also removed
proteinaceous components thus enhancing interfacial bonding also induced the fibers to
curl, an effect that was more pronounced at high concentrations and impaired the stiffness
of the BC fibers and sheets. Again these two competing effects converged to define
5% NaOH as the optimum concentration for mechanical performance. Interestingly an
even higher modulus was obtained when the BC sheets were consecutively treated with
NaOH and NaClO. In this case a modulus of 30 GPa was obtained. Further rheological
and morphological characterization of BC sheets showed that two transitions occured,
one at 50 ◦ C and 230 ◦ C that were ascribed to water desorption and cellulose degradation
respectively. The BC sheets also exhibited no preferential orientation of fibers (9).
   In order to enhance various properties of BC mats, many researchers have combined
BC with synthetic polymers, thermoplastic or thermosetting, utilizing various manufac-
turing approaches for these fiber reinforced nanocomposites.


9.3   BC Nanocomposites by in situ Polymerization

9.3.1 BC Nanocomposites with Thermosetting Phenolic and Epoxy Resins
The first report on BC reinforced thermosetting composites used phenol-formaldehyde
(PF) resins (10) and clearly demonstrated that the nanoscalar structure of BC con-
tributes significantly to the fiber reinforcing potential. In this study, BC sheets dried
between metal plates at 70 ◦ C were impregnated with PF resins and cured under pres-
sure (15–150 MPa) and temperature to deliver composites with resin concentrations of
2.7%, 12.4% and 21.9%. Comparison of the bending and tensile properties of the BC/PF
nanocomposites with those of microfibrillated cellulose, MFC/PF composites (11) with
similar density (1.5 g/cm3 ) indicated that BC was a better reinforcement than MFC.
For example, when pressed at 100 MPa, the BC/PF nanocomposites reached a Young
modulus of 28 GPa well above that of the MFC/PF composite at 15 Gpa (10). Bending
strength was also higher for the BC/PF nanocomposite compared to the MFC/PF com-
posite although differences were not as marked as for the Young modulus. The high
modulus of the BC nanocomposites was ascribed to the straight and planar orientation of
the BC fibers and to the fact that the fibers are continuous, uniform and intertwined, not
the case of MFC. In contrast PF resin content and pressure did not affect the mechanical
properties of the nanocomposites. In fact, the Young moduli reached by adding the
phenolic resin (28 GPA) were not as high as those reached in neat BC sheets (30 GPa)
after proper washing and hot-pressing as previously reported (8).
                               Bacterial Cellulose and Its Polymeric Nanocomposites     235

   Further research concentrated on the manufacture of transparent BC nanocomposites
providing great strides for developing BC materials for optoelectronic devices (12–16).
Yano (16) first reported that when dried and purified BC pellicles are filled with trans-
parent thermosets such as epoxy, PF and acrylic resins under vacuum and then cured
in situ, BC nanocomposites with as much as 60–70% fiber content retain a very good
transparency. For example the loss in transmittance due to the fiber in a BC/epoxy
nanocomposite was less than 10% compared to the neat epoxy. Similarly, good optical
properties were reported for the BC composites made of PF and acrylic resins. With
a BC fiber diameter less than 1/10 of the visible light wavelength, the BC network
does not scatter light, thus constituting an excellent reinforcement for transparent resins
without altering their optical property. Added benefits of reinforcing such resins with
a BC network included a lower coefficient of thermal expansion (CTE) and unique
mechanical properties of the nanocomposites. For PF and epoxy nanocomposites the
CTE ranged from 3 to 6 10−6 /◦ C, well below that of epoxy alone at 1.2 10−4 /◦ C when
measured between 50 and 150 ◦ C. At the same time, the epoxy/BC nanocomposites had a
Young modulus around 20 GPa and a tensile strength around 325 MPa while elongation
remained at ca 2% affording great flexibility and good mechanical performance for use
as flexible displays in optical devices (Figure 9.2).

9.3.2 BC Nanocomposites with Acrylic Resins
Subsequent research by Nogi’s group focused on BC/acrylic resins attempting to assess
and optimize many properties of these nanocomposites for use as optoelectronic devices
(12–15). Close matching of the refractive index (RI) of the reinforcement and the matrix




Figure 9.2 Flexibility of a 65 mm thick BC sheet with an acrylic resin and 60% fiber content.
(Hiroyuki Yano, Junji Sugiyama, Antonio Norio Nakagaito, Masaya Nogi, Tohru Matsuura,
Makoto Hikita, Keishin Handa, Optically transparent composites reinforced with networks
of bacterial nanofibers, Advanced Materials, 2005, 17, 2,153–5. Copyright John Wiley &
Sons-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)
236    The Nanoscience and Technology of Renewable Biomaterials

is normally required for composites to maintain transparency. Besides a resin RI varies
with temperature. The impacts of temperature and RI on the nanocomposites trans-
parency were thus first evaluated for high fiber content (60% BC) nanocomposites (14).
A series of acrylic resins with RI ranging from 1.492 to 1.636, close to the cellulose
RI, 1.544 in the transverse direction and 1.618 in the longitudinal direction, was used
to manufacture various BC/acrylic nanocomposites. Again less than 10% degradation
in luminous transmittance was observed in the nanocomposites compared to the neat
resins. At 20 ◦ C, the total transmittance of the nanocomposites only slightly varied with
RI while remaining above 85% (Figure 9.3). The regular transmittance showed a more
pronounced variation while also remaining above 75% (Figure 9.3). Both the total and
regular transmittances appeared to peak at a RI around 1.56–1.60 which was consistent
with the average RI of BC (Figure 9.3). Although the RI of an acrylic decreased signif-
icantly upon heating to 80 ◦ C, no loss in the regular transmittance of the nanocomposite
was observed (Figure 9.4). The relative insensitivity of the nanocomposite transparency
on resin RI and temperature are additional positive attributes of high fiber content BC
nanocomposites as electronic displays.
   Nogi’s group also expected the nanocomposites optical transparency and the CTE
to depend on fiber content and therefore embarked on evaluating BC nanocomposites
having a much wider variation of fiber content (15). In that objective, the manufacturing
process of the dry BC sheet was altered such that the interstitial space or void in the mat
could be tailored. Two manufacturing procedures were used. In the first procedure, BC
pellicles were first pressed to remove excess water yielding mats with 1/3 void content


                                        90

                                        88

                                        86

                                        84
                    Transmittance (%)




                                        82

                                        80

                                        78

                                        76

                                        74

                                        72

                                        70
                                         1.46   1.50       1.54     1.58        1.62   1.66
                                                       Resin refractive index

Figure 9.3 Total transmittance (filled circles) and regular transmittance (open circles) of BC
nanocomposites at 20 ◦ C and 590 nm versus the refractive index of their resins at 20 ◦ C
and 589.3 nm. (Reused with permission from Masaya Nogi, Keishin Handa, Antonio Norio
Nakagaito, and Hiroyuki Yano, Applied Physics Letters, 2005, 87, 243110, Copyright 2005,
American Institute of Physics.)
                                                                              Bacterial Cellulose and Its Polymeric Nanocomposites                                   237

                                              1.535                                                                               85.0
                                                                                                            TCDDMA




                                                                                                                                         Regular transmittance (%)
                     Resin refractive index
                                              1.530                                                                               84.5


                                              1.525                                                                               84.0


                                              1.520                                                                               83.5


                                              1.515                                                                               83.0
                                                                              0            20         40        60        80    100
                                                                                                Temperature (°C)

Figure 9.4 Temperature dependence of the refractive index of the neat acrylic sheet at
589.30 nm (filled circles) and the regular transmittance of BC nanocomposites (open circles)
at 590 nm for the TCDDMA neat acrylic sheet and BC nanocomposite. (Reused with
permission from Masaya Nogi, Keishin Handa, Antonio Norio Nakagaito, and Hiroyuki Yano,
Applied Physics Letters, 2005. 87, 243110 Copyright 2005, American Institute of Physics.)


after which the mats were soaked into binary solvents of water/acetone of different
compositions to prevent the fibers from adhering to each other and to leave more void
for resin impregnation. In the second manufacturing procedure, ethanol was used as
solvent. The resulting nanocomposites had a fiber content varying from 7.4 to 66.1 wt%.
In the 500–800 nm range, the regular admittance remained above 75% regardless of fiber
content and the transmittance at 590 nm, normalized to the nanocomposite thickness,
showed an expected decreasing trend with higher fiber content (Figure 9.5). The loss


                                                                              100
                                               Normalized transmittance (%)




                                                                                                            : Acrylic resin
                                                                                                            : Heat drying
                                                                                  97                        : Acetone exchange
                                                                                                            : Ethanol exchange

                                                                                  94

                                                                                  91

                                                                                  88

                                                                                  85
                                                                                       0         20        40        60    80    100
                                                                                            Fiber weight ratio (wt %)

Figure 9.5 Normalized regular transmittance at 590 nm and 100 mm thickness of BC
nanocomposite against fiber content. (Reused with permission from Nogi, M.; Ifuku, S.;
Abe, K.; Handa, K.; Nakagaito, A.N.; Yano, H., Fiber-content dependency of the optical
transparency and thermal expansion of bacterial nanofiber reinforced composites. Applied
Physics Letters, 2006, 88(13), 133124, Copyright 2006, American Institute of Physics.)
238    The Nanoscience and Technology of Renewable Biomaterials

in transparency with higher fiber content was rather small with for instance a 3.3%
transmittance loss at 11.7 wt% fiber and a 13.7% transparency loss at 66.1wt% fiber.
While low fiber contents appeared slightly advantageous to maintain transparency, high
fiber content nanocomposites had a lower CTE and therefore were less sensitive to tem-
perature variations. In fact, the incorporation of only 7.4 wt% BC fiber in the resin
contributed to a large reduction in the materials CTE from 86 × 10−6 to 38 × 10−6 K−1
while minimally deteriorating the light transmittance. It thus appeared from this report
that minimal addition of BC fibers allowed maintaining a high transparency while dras-
tically reducing the CTE (15).
   Another limitation of BC nanocomposites lies in the hygroscopicity of BC which
imparts poor dimensional stability to the nanocomposites. To circumvent the low
dimensional stability of BC nanocomposites acetylated BC fibers were utilized in BC
nanocomposites in conjunction with acrylic resins (12, 13). At low degree of fiber
acetylation with acetic anhydride (degree of substitution DS up to 0.17) acetylated BC
fibers had similar dimensions than untreated BC fibers but appeared better separated
from each other in a scanning electron microscope (SEM), a likely result of lower inter-
fiber bonding between acetylated fibers (Figure 9.6). As a result of lower interfiber
bonding, acetylated BC sheets had a lower modulus (17.3 GPa) than an untreated BC
sheet (23.1 GPa). As expected acetylation was effective at decreasing the hygroscopy
of the BC/acrylic nanocomposites with two acrylic resins, one based on dimethacrylate
(TCDDMA) and another commercial resin. Nanocomposites with 33% acetylated fiber
content reached similarly low moisture content than the neat acrylic (0.8%) when equili-
brated at 20 ◦ C and 55% relative humidity. Slight changes in optical transparency were
observed in the nanocomposites after acetylation and could go either direction depend-
ing on whether it improved the match in resin and fiber RI (Figure 9.7). Interestingly,
while sheets of acetylated BC had a lower CTE (0.8 ppm/◦ K) compared to untreated
BC sheets (3 ppm/◦ K), this improvement in CTE was not observed in the corresponding
nanocomposites, at least within this small range of DS. As acetylation also increases the
thermal stability of cellulose, it was no surprise to see that after heat treatment at 200 ◦ C
for various time periods, the acetylated BC nanocomposites did not experience as high
of a loss in optical transparency than the control nanocomposites. The overall positive
effect of acetylation on the properties of the nanocomposites prompted this group to
examine further the impact of a broader DS range from 0.17 to 1.8 on the nanocom-
posite properties (12). Thus 40% fiber content nanocomposites were manufactured with
acetylated fibers and the TCDDMA based resin. With higher DS, wider nanofibers were
observed (Figure 9.8) along with a change in crystalline structure indicating that acety-
lation occurred from the surface to the core of semicrystalline fibers. Also within that
broader DS range, both the nanocomposite transparency and the equilibrium moisture
content (20 ◦ C, 55% RH) passed through an optimum as a function of DS (Figure 9.9).
Maximum transmittance (87.8%) was measured for a DS of 0.74 above which transmit-
tance decreased down to ca. 70% and minimum equilibrium moisture content (EMC)
of the nanocomposite (20 ◦ C and 55% RH) was found (0.5% MC) for a DS = 0.5. At
this substitution level the nanocomposite hygroscopicity was reduced by 1/3 compared
to nanocomposites using untreated BC (MC of 1.5%) but was still higher than that
of the acrylic resins alone (0.35%). Up to a DS of approximately 0.6, the CTE of the
acetylated nanocomposites was similar to that of the control nanocomposite, after which it
                              Bacterial Cellulose and Its Polymeric Nanocomposites    239




                                                             1 µm

                                           (a)




                                                             1 µm

                                           (b)

Figure 9.6 SEM image of (a) untreated BC sheet and (b) acetylated BC one. (Reprinted with
permission from Nogi, M.; Abe, K.; Handa, K.; Nakatsubo, F.; Ifuku, S.; Yano, H., Property
enhancement of optically transparent bionanofiber composites by acetylation. Applied
Physics Letters, 2006, 89(23), 233123, Copyright 2006, American Institute of Physics.)


rapidly increased to 25.10−6 /K. The fact that all the properties monitored passed through
an optimum was ascribed to a positive effect of BC surface acetylation up to the point
where the crystalline structure of the BC was disrupted.
   These advances in BC utilization for optoelectronic devices brought about by Nogi’s
group undoubtedly constitute a very significant development of BC application in mate-
rials of high commercial potential. As a result, these developments have been largely
patented (17–23). Two other research groups have reported on the development of
acrylic/BC nanocomposites for potential use as cation exchange membranes (24) or
as biomaterials for soft tissue replacement or repair (25). In the first approach, dried
240    The Nanoscience and Technology of Renewable Biomaterials

                                        100                                                          100




                                                                         Regular transmittance (%)
            Regular transmittance (%)
                                        80                                                            80

                                        60                                                            60

                                        40                                                            40

                                        20                                                            20
                                                            TCDDMA                                                      ABPE300
                                         0                                                             0
                                          200   400   600     800 1000                                  200   400     600   800 1000
                                                Wavelength (nm)                                               Wavelength (nm)
                                                      (a)                                                             (b)

Figure 9.7 Regular transmittance spectra of acetylated (solid line) and untreated (broken
line) BC nanocomposites (a) with the TCDDMA resin matrix and (b) with the ABPE300 resin
matrix. (Reprinted with permission from Nogi, M.; Abe, K.; Handa, K.; Nakatsubo, F.;
Ifuku, S.; Yano, H., Property enhancement of optically transparent bionanofiber composites
by acetylation. Applied Physics Letters, 2006, 89(23), 233123, Copyright 2006, American
Institute of Physics.)

      (a)                                                    (b)                                                (c)




Figure 9.8 SEM images of (a) untreated, (b) DS 0.45 and (c) DS 1.55 BC samples. For all
photographs, the length of the scale bar is 500 nm (12). (Reprinted with permission from
Ifuku, S.; Nogi, M.; Abe, K.; Handa, K.; Nakatsubo, F.; Yano, H., Surface modification of
bacterial cellulose nanofibers for property enhancement of optically transparent composites:
Dependence on acetyl-group DS. Biomacromolecules, 2007, 8(6), 1973–1978. Copyright
2007. American Chemical Society.)


bacterial cellulose sheets were first activated by UV irradiation after which they were
immersed in an acrylic acid solution and once again irradiated with UV (24). Thorough
washing of the membrane allowed removing unreacted chemicals. Weight gain percent-
age showed that the acrylic content increased linearly with the UV irradiation time up
to 25%. The acrylic acid apparently grafted to the cellulose as suggested from Fourier
transform infrared spectroscopy (FTIR) and filled in the pores of the bacterial cellulose
membrane such that the pore structure was no longer distinguishable after 20 min of
UV irradiation (Figure 9.10). Tensile strength and elongation at break of the dry mem-
branes also appeared to increase with the UV irradiation time up to 180 MPa and ca
6% elongation respectively (Figure 9.11). Besides the electrochemical properties of the
BC/acrylic composite were comparable to those of a commercial membrane (24).
   In another publication a slightly different approach was used to manufacture BC/
acrylic interpenetrating networks (25). Namely a purified and never dried bacterial
                                                        Bacterial Cellulose and Its Polymeric Nanocomposites                       241

                               90                                                              2.5

                               88
   Regular transmittance (%)



                               86                                                               2




                                                                        Moisture content (%)
                               84
                               82                                                              1.5
                               80
                               78                                                               1
                               76
                               74                                                              0.5
                               72
                               70                                                               0
                                    0   0.5       1         1.5     2                                0   0.5       1         1.5   2
                                        Degree of substitution                                           Degree of substitution

Figure 9.9 Regular transmittance of a series of acetylated BC nanocomposites at 580 nm (left)
and moisture content vs degree of substitution of acetylated BC/resin nanocomposites (right)
(12). (Reprinted with permission from Ifuku, S.; Nogi, M.; Abe, K.; Handa, K.; Nakatsubo, F.;
Yano, H., Surface modification of bacterial cellulose nanofibers for property enhancement
of optically transparent composites: Dependence on acetyl-group DS. Biomacromolecules,
2007, 8(6), 1973–1978. Copyright 2007. American Chemical Society.)




                                               (a)                                                            (b)




                                               (c)                                                            (d)

Figure 9.10 SEM of BC (a), BC/acrylic acid composite obtained after 5 min of irradiation
(b), 10 minutes of irradiation (c) and 20 minutes of irradiation (d) (24). (Preparation and
characterization of acrylic acid-treated bacterial cellulose cation-exchange membrane, Y.J.
Choi, Y.H. Ahn, M.S. Kang, H.K. Jun, I.S. Kim and S.H. Moon, Journal of Chemical Technology
and Biotechnology, 2004, 79(1), 79–84, Copyright 2004. Copyright John Wiley & Sons Ltd.
Reproduced with permission.)
242   The Nanoscience and Technology of Renewable Biomaterials

                                         200
                                         180                                                    14




                                                                                                     Elongation at break [%]
                Tensile strength [MPa]
                                         160                                                    12
                                                                                   CMX
                                         140                                  (Tokuyama co.)    10
                                         120
                                         100                                                    8
                                          80                                                    6
                                          60                 CMX                                4
                                          40            (Tokuyama co.)

                                          20                                                    2
                                           0                                                    0
                                               0   5         10          15       20       25
                                                       UV-irradiation time [min.]

Figure 9.11 Mechanical properties of BC/acrylic acid membranes compared to a commercial
(CMX) membrane (24). (Preparation and characterization of acrylic acid-treated bacterial
cellulose cation-exchange membrane, Y.J. Choi, Y.H. Ahn, M.S. Kang, H.K. Jun, I.S. Kim
and S.H. Moon, Journal of Chemical Technology and Biotechnology, 2004, 79(1), 79–84,
Copyright 2004, John Wiley & Sons Ltd. Reproduced with permission.)

cellulose membrane was impregnated with various acrylate and methacrylate monomers
and crosslinker solutions and irradiated with UV light for photopolymerization
(wavelength between 320–520 nm). A good balance of strength and flexibility was
obtained when using a 30–60 wt% acrylate/crosslinker concentration (25). By photo-
polymerizing the acrylates and methacrylates into the porous structure of never dried
BC, flexible and stable materials were obtained that retained the original BC shape
without discharging monomers. As expected, the morphology and properties of the
nanocomposites varied with the choice and content of the acrylate monomers and
crosslinker. With 60% monomers the pores of BC were completely filled; with 30%
monomers the surface of the fibers were coated as demonstrated in the SEM pictures.
The resulting composites could then be washed to remove any residual monomers to
be suitable as a biomaterial. Note that the procedure is different from that of Nogi’s
group where BC sheets are dried before impregnation and cure of the acrylic resins.
In addition, in situ polymerized BC/acrylate were used as hydrogels (25) although
their water adsorption capacity appeared to decrease significantly with the in situ
polymerization of acrylics. Young modulus measured at room temperature within a
frequency range from 1 to 50 Hz revealed that with enough irradiation cycles, the
modulus of the composites can fall in the range of that of hyaline cartilage (25).


9.4   BC Nanocomposites by Polymer Impregnation and Solution Casting

Mixing of a polymer solution and a BC suspension followed by solvent casting is another
common method to prepare BC nanocomposites. Due to the synthesis and nature of
thermoplastic polymers, this approach has been more common for preparing BC/thermo-
plastic nanocomposites and has involved both natural and synthetic polymers (26–29).
Due to the lower modulus of thermoplastics compared to thermosets one might expect
                              Bacterial Cellulose and Its Polymeric Nanocomposites   243

the former polymers to be more efficiently reinforced by stiff nanometer scale fillers
such as BC fibers.

9.4.1 BC/Biopolymer Nanocomposites
In an attempt to understand the secondary plant cell wall, Gatenholm’s group embarked
in preparing and characterizing BC/xylans nanocomposites (30–32). To do so, solutions
of 4-O-Methyl glucuronoxylans that were alkali extracted from birch wood were mixed
with BC in different ways. One way consisted of depositing the xylans on BC surfaces
while incubating at 90 or 170 ◦ C for various time periods (32). Atomic force micro-
scopic examination of the cellulose surfaces demonstrated that the xylans aggregated
onto the cellulose fibers as globule in amounts and sizes that depended on the autoclave
treatments. In fact, the nanocomposite morphology and its dependence on the xylan
content was ascribed to the colloidal state of xylan solutions induced by the autoclaving
conditions. Further analysis of the xylan adsorbed by the BC showed that preferential
interactions occurred with the less substituted xylans. Less substituted xylans have a
greater tendency to self associate thus explaining their greater colloidal size after the
170 ◦ C autoclaving that removes the 4-O-methyl glucuronic acid (32). Another way to
prepare the BC/xylans nanocomposites consisted of first homogeneizing the bacterial
cellulose in a blender after which the xylan solutions (alkali extracted from aspen wood)
were added in different ratios and the blends were allowed to interact for 30 min and
solution casted (30). Again, xylans were found to aggregate on BC to produce a lam-
inated structure (Figure 9.12). Besides, the incorporation of xylans in the composites
reduced the tensile strength of homogeneized bacterial cellulose from ca 110 MPa to
65 MPa in nanocomposites containing 50% xylans. However Young modulus increased
from ca 4 GP to 6.5 Gpa when the xylan content increased to 30% after which the
modulus was found to decrease again. The authors proposed that the optimum mod-
ulus observed in the 30% xylan composite might represent the best formulation for
optimal interactions between the two components, which incidentally also corresponded
to the composition of wood secondary cell wall (30). The homogeneized BC/xylan
nanocomposites also displayed very small extensibility (<6%), lower than that of the
homogeneized BC. When the BC/xylan nanocomposites (Aspen glucuronoxylans with
MW or 10,530 g/mol, DS acetyl = 0.52) were characterized by humidity scans in a
DMA at 30 ◦ C, a significant modulus drop assigned to the hemicelluloses softening was
observed at 85% RH for the pure glucuronoxylan and for 50:50 BC/xylan nanocomposite
but not for the pure BC (31). With dynamic FTIR during uniaxial loading, strong elastic
signals from both the cellulose and the glucoronoxylan components of the nanocompos-
ites were detected, indicating that both polymers participated in load transfer. While
this observation suggested strong interactions between xylans and bacterial cellulose, no
chemical interactions could be demonstrated in this work (31). According to the authors,
these BC/xylan nanocomposites are good models of the secondary wood cell wall. Fur-
thermore the nanocomposites are transparent thus opening opportunities for applications
besides that of the secondary cell wall model.
   Chitosan, has also been used in several instances to create BC/chitosan nanocom-
posites (33–36). While it has been shown that incorporation of chitosan in the culture
medium of bacterial cellulose leads to the incorporation of N-acetyl glucosamide into
244   The Nanoscience and Technology of Renewable Biomaterials


                                         5 µm




                   Bacterial Cellulose               Bacterial Cellulose
                                         5 µm




             BC/Glucuronoxylan Composite        BC/Glucuronoxylan Composite

Figure 9.12 Micrographs of bacterial cellulose and bacterial cellulose/Glucoronoxylan
                                                                        e
nanocomposites. (Reprinted from Polymer, 46, Dammstrom, S.; Salm´ n, L.; Gaten-
holm, P., The effect of moisture on the dynamical mechanical properties of bacterial
cellulose/glucuronoxylan nanocomposites, 10364–10371, Copyright (2005), with permission
from Elsevier.)

the cellulose chain and therefore the production of copolymers fibers rather than BC
nanocomposites (33, 35, 37), other studies have shown that true nanocomposites can be
formed by immersing a purified BC pellicle in a chitosan solution followed by solvent
casting (34, 36). The resulting nanocomposites were considered ideal for wound dress-
ing applications (34). The potential of BC/chitosan nanocomposite as membranes has
also been demonstrated (36). Upon impregnation of BC with chitosan, the porous struc-
ture of the BC was masked as chitosan filled the pores of the BC fleece (36). Besides
chitosan lowered the mechanical properties of the membrane from a tensile strength of
74 MPa to 54 MPa while elongation at break increased from 6.8% to 7.4%. No clear
chemical bonding between cellulose and chitosan moieties was detected by FTIR. How-
ever, pervorative experiments demonstrated that the nanocomposite had good potential
to separate ethanol/water azeotrope (36).
   Proteins, including gelatin and silk proteins, have also been used in combination with
bacterial cellulose to develop high mechanical strength double network hydrogels or
dry nanocomposites (38–40). While bacterial cellulose in the wet state has high tensile
properties, its compressive properties are mediocre limiting its utilization in various
                               Bacterial Cellulose and Its Polymeric Nanocomposites     245




Figure 9.13 Layered structure of BC (left) and BC/gelatin composite (right) (39). (Nakayama,
A.; Kakugo, A.; Gong, J.P. et al., High mechanical strength double-network hydrogel with
bacterial cellulose. Advanced Functional Materials, 2004, 14(11), 1124–8. Copyright John
Wiley & Sons-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)


biomedical applications. To remedy this disadvantage, Nakayama et al. have immersed
a purified BC pellicle in a gelatin solution after which it was chemically crosslinked
to various crosslinking densities (39). The resulting double networks (DN) combined
the outstanding tensile properties of BC with very high compressive strength. The DN
showed a stratified structure similar to that of BC, consisting of dense and sparse layers
of cellulose fibers (Figure 9.13). The gelatin did not affect the crystalline structure of
BC and was proposed to fill in the cellulose-poor layers. The improvement in properties
imparted by the combination of these two polymers was outstanding. In the direct
ion perpendicular to the stratified structure of the DN, the compressive modulus was
1.7 MPa, that is over 200 times higher than that of BC alone (Figure 9.14). The fracture
strength of the BC/gelatin network was also very largely improved compared to that of
the gelatin gel alone, being 30 times higher. Furthermore the BC/gelatin hydrogel was
capable of recovery after a first compressive deformation up to 30% strain which was
not the case of neat BC. The authors proposed that in the DN gel, mechanical properties
are mainly dominated by the degree of crosslinking of the gelatin and by the unit ratio
of the second network to the first network (R = amino acid/glucose) (Figure 9.14). As
a result the DN could achieve properties similar to that of articular cartilage.
   A recent study also evaluated composites of silk fibroin and BC in the dry and in the
wet state (40). Silk protein, a mixture of glycine, alanine and serine, can take various
conformations, amorphous and crystalline, and has generated interest in the biomedical
field as membranes due to its biocompatibility, permeability in the wet state and high
mechanical properties. The authors therefore prepared BC/silk fibroin nanocomposites
by immersing a purified and never dried bacterial cellulose pellicle into aqueous silk
fibroin solution (8 wt%) produced from Bombix mori cocoons and by then drying the
nanocomposites into films. Field emission scanning electron microscopy (FESEM) of
fracture surfaces suggested that the silk protein was finely dispersed between the cel-
lulose fibers resulting in well dispersed nanocomposites. Concomitantly, the fibroin
morphology changed from a mostly amorphous coil structure to a crystalline structure
as a β-sheet structure as demonstrated by X-ray diffraction and FTIR spectroscopy. As
a result the mechanical properties of the nanocomposites differed from those of the pure
fibroin or bacterial cellulose films. In the dry state, the tensile modulus of the composite
film (465 ± 57 MPa) was higher than that of the neat BC film (118 ± 9 MPa) or silk
246                        The Nanoscience and Technology of Renewable Biomaterials

                           101                                                                        6.0                                            6.0
                                                          BC-Gelatin
                                                                                                      5.0                                            5.0
   Elastic Modulus [MPa]




                                                                              Elastic Modulus [MPa]




                                                                                                                                                           Fracture Stress [MPa]
                           100
                                                                                                      4.0                                            4.0
                                                            Gelatin
                                                                                                                                             3.1
                           10−1                                                                       3.0                                            3.0
                                                                                                                                       3.6
                                                                                                      2.0                                            2.0
                           10−2                                                                                                  5.8
                                    BC
                                                                                                      1.0                   11                       1.0

                           10−3                                                                       0.0                                            0.0
                                  0.0   0.2   0.4   0.6     0.8   1.0   1.2                                 0     5      10            15          20
                                         EDC concentration [M]                                              Amino Acid/Glucose [unit/unit]
                                                    (a)                                                                  (b)

Figure 9.14 Compressive modulus of a BC/gelatin composite as a function of crosslinker
density (a) and of network unit ratio (b) (39). (Nakayama, A.; Kakugo, A.; Gong, J.P. et al., High
mechanical strength double-network hydrogel with bacterial cellulose. Advanced Functional
Materials, 2004, 14(11), 1124–8. Copyright John Wiley & Sons-VCH Verlag GmbH & Co.
KGaA. Reproduced with permission.)


fibroin film (355 ± 56 MPa). However the composite films were more brittle than the
silk fibroin film with an elongation at break of only 2% and a tensile strength (9.4 ± 1.2
MPa) slightly lower than that of the pure silk fibroin film (11.5 ± 3.3 MPa). On the
other hand, in the hydrated state, the composite was tougher and more flexible with an
elongation at break reaching 13.5 ± 2.9% but exhibited very low tensile strength and
modulus. Besides the crystallization of the silk into the cellulose fiber mats appeared to
impart a high water resistance to the composite (40).
   Solvent casted BC network/thermoplastic nanocomposites have also been developed
using cellulose acetate butyrate (CAB) as the matrix (26). BC/CAB nanocomposites were
prepared by impregnating an unpurified BC sheet into various CAB acetone solutions
and solvent casting such that a 10.3% (Composite A) and a 32% (composite B) volume
fraction BC nanocomposites were obtained. The tensile properties of the composites were
evaluated in static and cyclic modes and the fracture surfaces examined with SEM. The
stress-strain curves revealed an initial linear behavior followed by a yielding at a strain
of 0.6–0.8% and second linear region for strains higher than 2% (Figure 9.15). While the
strain at the maximum load was similar for the two composites at ca 3.5%, the Young
modulus and the tensile strength of BC/CAB nanocomposites improved significantly
with higher BC content. Tensile stiffness and strength more than doubled with ca 10%
addition of BC from 25.9 MPa and 1.2 GPa to 52.6 MPa and 3.2 GPa, respectively.
With 32% addition of BC, tensile stiffness and strength increased approximately 5 folds
to 128.9 MPa and 5.8 GPa. However the nanocomposites never reached the tensile
stiffness and strength of control BC films at 260 MPa and 15–18 GPa, respectively (9).
The cyclic test showed that in a loading cycle the unloading modulus was systematically
higher than the loading modulus and that the elastic moduli systematically increased
with each successive loading/unloading cycle, suggesting reorientation and slippage of
                                           Bacterial Cellulose and Its Polymeric Nanocomposites   247

                                 140
                                                               Composite B
                                 120




                  Stress (MPa)
                                 100
                                  80
                                  60                          Composite A
                                  40
                                  20                                                CAB

                                  0
                                       0    1    2     3       4        5   6   7         8
                                                           Strain (%)

Figure 9.15 Stress-strain curves of tensile tests. (Reprinted from Composites Science and
Technology, 64, W. Gindl and J. Keckes, Tensile properties of cellulose acetate butyrate
composites reinforced with bacterial cellulose, 2407–13. Copyright (2004), with per-
mission from Elsevier.)


the BC fibers beyond the linear elastic range and along the loading direction (26). It was
proposed that yielding occurred in interfacial shear between fibers and matrix allowing
a reorientation of the fibers in the composites and thereby causing a modulus increase
with repeated loading. BC was therefore established as a good reinforcement for CAB,
especially under cyclic loading.

9.4.2 BC/Synthetic Polymer Nanocomposites
Stiffening upon cyclic loading was also observed in physically crosslinked BC/PVA
nanocomposites that were developed for biomedical applications (28, 41). For appli-
cations such as stents and heart valve leaflets, the substitute material must match the
tissue strength and stiffness but also its stress relaxation behavior in order to abide to
the cardiac cycle. After demonstrating the suitability of PVA as hydrogels for such
applications (27), Millon and Wan embarked on finetuning the hydrogel properties by
combining BC and PVA in various proportions and by physically crosslinking PVA
through heat/freeze cycles (28). Note that in this case, the BC nanocomposites were
intended for use in the wet state, i.e. as hydrogels. Cyclic tensile tests up to 75%
strain and tensile stress relaxation experiments indicated that the BC/PVA nanocompos-
ites could be tailored to match the behavior of a specific cardiovascular tissue. Under
tensile loading the PVA/BC nanocomposites displayed a different stress-strain behavior
than that of PVA alone with notably a significant increase in modulus at ca 40% strain
for the nanocomposite and a trend for increasing stiffness with number of cycles. It was
also evident that the higher the BC content in the nanocomposite, the higher the tensile
modulus (Figure 9.16). Increasing the PVA content in the hydrogel nanocomposite also
induced a higher tensile modulus. Consequently by varying composition and crosslink-
ing extent the authors were able to tailor the stress strain properties of the BC/PVA
nanocomposites to match the behavior of specific cardiovascular tissue. For example
the tensile properties of porcine aorta in both the circumferential and axial directions
could be well matched by the nanocomposites (Figure 9.17) whereas the stress relaxation
248                The Nanoscience and Technology of Renewable Biomaterials

                   2.0



                                            P10
                   1.5                      P10 BC 0.15
                                            P10 BC 0.23
                                            P10 BC 0.31
   Modulus (MPa)




                                            P10 BC 0.61
                   1.0




                   0.5




                   0.0

                           0.0       0.1       0.2         0.3     0.4        0.5   0.6
                                                          Strain

Figure 9.16 Tensile moduli of 4 composites with 10% PVA and various BC contents (0.15
to 0.61%) for cycle 6 (28). ( Journal of Biomedical Materials Research, Part B: Applied Bio-
materials, 79B (2), 2006, 245–53. Copyright 2006 John Wiley & Sons, Inc. Reprinted with
permission of John Wiley & Sons, Inc.)

rate for these nanocomposites was significantly faster than that of porcine aorta. Such
BC/PVA nanocomposites therefore matched the tensile stiffness of living tissues while
allowing for a faster recovery during a cardiac cycle.


9.5                BC Nanocomposites via Biomimetic Approaches

Both Van der Waals and H-bond play a major role in glucan chain association and cel-
lulose crystallization. Cousins and Brown (42) proposed that the glucan chains exuded
from the enzymatic complex at the bacterial membrane spontaneously form minisheets
via Van der Waals forces and then associate into minicrystals via H-bond to finally con-
verge into cellulose I microfibrils. In acetobacter, the TC consists of a single row of
subunits, thus facilitating the spontaneous formation of sheets by van der Waals forces
in an aqueous medium. On the other hand, if mini-sheet formation occurs via H-bond,
then it suffers competition of water molecules and could further be manipulated with
appropriate H-bonding molecules. In fact, it has been repeatedly shown that when a
strong H-bonding polymer such as carboxymethyl cellulose (CMC) is added to the incu-
bation medium of BC, cellulose crystallization is altered (5, 43). As CMC concentration
increases, the resulting bacterial cellulose exhibits smaller diameter microfibrils and also
has higher Iβ crystalline allomorph content. For CMC derivative, it was empirically
                                   Bacterial Cellulose and Its Polymeric Nanocomposites       249



                0.25


                                      Aorta Circ
                0.20                  Aorta Axial
                                      P15 BC0.3-Cycle 2
                                      P10 BCE0.15-Cycle 3
 Stress (MPa)




                0.15



                0.10



                0.05



                0.00


                       0.0   0.1      0.2      0.3            0.4   0.5      0.6        0.7
                                                     Strain

Figure 9.17 Close match of stress-strain curve of various nanocomposites with porcine aorta
in circumferential and axial directions (28). ( Journal of Biomedical Materials Research, Part B:
Applied Biomaterials, 79B (2), 2006, 245–53. Copyright 2006 John Wiley & Sons, Inc. Re-
printed with permission of John Wiley & Sons, Inc.)


determined that a degree of substitution of 7 groups on 10 glucan (DS of 0.7) and a DP
of 80 was the most effective in disrupting glucan chain association to produce separate,
smaller size microfibrils (44). In this case the average size of the cellulose microfibrils
is reduced to 10 nm with 1% CMC concentration.
   Since the first observation that additives in the BC incubation medium may alter the
aggregation and crystallization pattern of bacterial cellulose (43), a wealth of studies
have reported on the effect of polymers on bacterial cellulose crystallization with a view
to understanding the biosysnthesis of the cell wall (29, 35, 37, 44–55). For example
hemicelluloses have been shown to alter the aggregation of BC chains into microfibrils
and ribbons resulting in smaller width fibers and a crystalline structure that is enhanced
in the Iβ form to the detriment of the Iα form (47–52, 54, 55). In contrast pectin
has been found to have little effect on the morphology and microstructure of BC (51).
Although such studies are principally aimed at shedding light on cellulose biosynthe-
sis, they have also paved the way for the development of BC nanocomposites using
biomimetic approaches. In a biomimetic approach, the BC is grown in a media that is
rich in another polymer such that BC and the host polymer assemble in nanocomposites.
   As pointed out by Teeri et al. (56) biomimetics can largely contribute to the develop-
ment of novel environmentally friendly biomaterials owing to the energy efficient and
mild conditions nature uses for the biosynthesis of biomaterials. The work by Gidley’s
250   The Nanoscience and Technology of Renewable Biomaterials

group (57–61) and by a few other researchers (30–32, 62, 63) clearly illustrates the
potential of biomimecry for the development of novel biomaterials that are completely
based on natural resources. Indeed Gidley group’s has extensively used the in vitro
assembly of BC with other cell wall polymers in order to mimic the biosynthesis of the
primary cell wall (57–61). In so doing a wide range of BC/hemicellulose nanocomposites
have been developed and characterized for morphology, physical and mechanical prop-
erties (57–61). In particular BC nanocomposites have been produced with mannans (64),
xyloglucans (57–59, 61, 65–67), pectins (58–61), xyloglucans in conjugation with pectins
(59) and xylans (30–32). Lignin precursors have also been polymerized in situ in order to
produce BC/lignin nanocomposites (62, 63). The structural, morphological and mechan-
ical features of these bio-based nanocomposites manufactured in vitro are reviewed next.
A similar biomimetic approach has been taken to develop a few nanocomposites with
synthetic polymers and these developments are also reviewed (68–70).

9.5.1 BC/Xyloglucan Nanocomposites
Whitney et al. first demonstrated that the in vitro assembly of cellulose/xyloglucan
networks could result in the formation of model composite materials that closely resem-
bled the primary cell wall of plants (65, 66). Xyloglucans are β-1,4-linked glucose
residues to which α-1-6 linked xylose residues are attached. Some of the xylose residues
are substituted with galactose and focusylgalactose (71). Composites were thus pro-
duced by modifying the incubation medium of acetobacter xylinum with 0.5% tamarind
xyloglucan and growing BC under agitated conditions. The deep-etch, freeze frac-
ture transmission electron microscopy of washed composite materials revealed that the
xyloglucans imparted some degree of lateral order in the BC ribbons which contrasted
with the randomness of ribbon orientation in control BC (Figure 9.18). Besides thin
strands of xyloglucans, 20–70 nm long, bridged the cellulose ribbons, whose widths
(36 nm) were similar to those of BC. Comparison of the cellulose:xyloglucan ratio of
1:0.38 with the measured level of cross-bridges further suggested that the majority of
the xyloglucans (33% of the total 38%) present in the nanocomposites might be inti-
mately associated with the cellulose ribbons, either surface bound or interwoven in
the fibers and ribbons. Furthermore, the combination of cross-polarization (CP/MAS)
and single pulse excitation with dipolar decoupling (SP/MAS) NMR techniques con-
firmed the existence of two types of xyloglucans in the nanocomposites with the evidence
of both rigid and mobile xyloglucan domains. The former domain was ascribed to the
xyloglucans that are intimately associated and oriented parallel to the cellulose microfib-
rils and the latter domain corresponded to the xyloglucans in the cross-bridges (65). In
presence of xyloglucans, BC developed with less crystallinity and with a significantly
lower Iα content and higher Iβ content compared to the control BC, confirming molecu-
lar association of xyloglucans with cellulose (65), a phenomenon that was not observed
in another study (55). Additional analysis of the crystal structures in BC/xyloglucan
nanocomposites by small-angle X-ray diffraction (XRD) and environmental scanning
electron microscopy (ESEM) concluded that during in vitro synthesis of the nanocom-
posites, xyloglucans get entrapped in the less dense shell of the ribbons (57). In contrast,
composites created by association of xyloglucans and bacterial cellulose as an abiotic
                               Bacterial Cellulose and Its Polymeric Nanocomposites         251




                                     0.5 µm                                        0.5 µm


Figure 9.18 Micrographs of cellulose ribbons obtained in standard growth
conditions – control BC (left) and obtained in presence of Xyloglucans-BC/xyloglucans
nanocomposites (right). (Whitney, S.E.C.; Brigham, J.E.; Darke, A.H.; Reid, J.S.G.; Gid-
ley, M.J., In-vitro assembly of cellulose/xyloglucan networks – ultrastructural and molecular
aspects. Plant Journal, 1995, 8(4), 491–504. Copyright 1995, Reprinted with permission from
Wiley-Blackwell.)


system only exhibited the mobile xyloglucan cross-bridges, clearly illustrating the poten-
tial of biomimetic approaches to develop nanocomposites with unique morphologies (65).
   When the small deformation rheological properties of the intact and pulverized
BC/xyloglucan nanocomposites were compared to those of tomatoes’ primary cell
wall and of BC alone, striking similarities were observed (66). Namely, the dynamic
viscosity, storage (G’) and loss (G”) shear moduli followed similar trends in the tomato
cell wall, pulverized bacterial cellulose and in BC/xyloglucan nanocomposites (66) with
a noteworthy solid-like behavior. Hence it was proposed that under small deformation
the entangled cellulosic rods rather than the xyloglucan crosslinks dominated the
shear moduli of these nanocomposites. Cellulose was therefore the key mechanical
component that contributed to the linear viscoelastic behavior of BC/xyloglucan
nanocomposites. In contrast xyloglucans appeared to play a significant role in the
nonlinear uniaxial tensile properties of BC/xyloglucan nanocomposites (66). When
tested under hydrated conditions, BC sheets were stiffer, stronger but less extensible
than BC/xyloglucan sheets (all with less than 10% cellulose content) (Figure 9.20). This
mechanical response was rationalized in terms of the different entanglements in BC
and in the BC/xyloglucan nanocomposite (66) and in a later publication modeled based
on small-angle X-ray scattering experiments performed during uniaxial loading tests
(58). Namely, it was proposed that in the randomly oriented fibers of neat BC, looped
links required more force and allowed low stretching before failure (Figure 9.19).
Alternatively, alignment of cellulose ribbons in the BC/xyloglucans nanocomposites
generated entanglements between somewhat parallel fibrils. Upon uniaxial loading the
ribbons or crosslinked domains could orient with the loading direction until the breaking
point at a large extension (Figure 9.19). These concepts would explain the lower
tensile strength and stiffness and higher extensibility of BC/xyloglucan nanocomposites
compared to neat BC in uniaxial tensile loading.
252    The Nanoscience and Technology of Renewable Biomaterials

                 A                         B




                                    Direction of extension

Figure 9.19 Conceptualized entanglements in pure BC (A) and in BC/xyloglucans nanocom-
posites (B) explaining the difference in tensile properties. (Reprinted from Whitney, S.E.C.;
Gothard, M.G.E.; Mitchell, J.T.; Gidley, M.J., Roles of cellulose and xyloglucan in deter-
mining the mechanical properties of primary plant cell walls. Plant Physiology, 1999,
121, 657–63, Copyright (1999), with permission from the American Society of Plant
Biologists.)


   Dynamic FTIR spectroscopy further shed light on the morphology and molecular
orientation that can take place during linear stretching of these nanocomposites (61).
During stretching of neat BC, cellulose fibers molecularly reoriented along the loading
direction, a reorientation that was eased by the lubricating power of water (61). In the
BC/xyloglucan nanocomposites prepared in vitro, where cellulose fibers were already
oriented, xyloglucan bridges experienced less reorientation upon stretching. With 2D
FTIR spectroscopy the authors confirmed that the xyloglucans and the cellulose moved
collectively.
   Bi-axial tensile loading revealed yet another difference between neat BC and in vitro
BC/xyloglucan nanocomposites. Under equi-biaxial tensile loading, the BC alone was
found to behave as a linear elastic material (59). The BC/xyloglucan nanocompos-
ites showed a markedly different behavior with much greater extensibility (threefold
higher than that of BC), a low stiffness, and most noticeably a nonlinear elastic behavior
(Figure 9.20). These drastic differences in behavior were confirmed by creep mea-
surements where neat BC did not exhibit any time dependent behavior whereas the
BC/xyloglucan nanocomposites displayed a large time-dependent and viscoelastic behav-
ior (Figure 9.20) (59).
   While these initial studies (57–59, 65, 66) were all conducted on Tamarind xyloglu-
cans, there exists a large diversity of xyloglucans that could translate in a wide range
of interactions with BC and thus nanocomposite performance. To test that possibility,
Whitney et al. also examined BC/xyloglucan nanocomposites prepared in vitro with
eight different xyloglucans that systematically varied in terms of molecular weight and
substitution pattern including galactose content, fucose content and degree of acetyla-
tion (67). For all nanocomposites, significant amounts of xyloglucans remained in the
composites after washing demonstrating that they all interacted strongly with the cellulose
(19–98%). Morphological characterization with SEM and CP/MAS NMR showed that
for all the xyloglucans except the one with a 60% galactose depletion of the Iα/Iβ ratio
was significantly reduced hence revealing that most xyloglucans perturbed the molec-
ular assembly of BC. The deep etch freeze fracture images also revealed a different
ultrastructure based on the characteristics of the xyloglucan. For high molecular mass
                                                    Bacterial Cellulose and Its Polymeric Nanocomposites          253


                                          1.2

                                          1.0           Cellulose


                  Stress (MPa)            0.8
                                                                                Pectin/cellulose
                                          0.6

                                          0.4
                                                                                      3 components

                                          0.2                             Xyloglucan/cellulose


                                          0.0
            (a)                                 0       20               40             60               80
                                                                     Strain (%)


                                          500


                                          400                         Cellulose
                  Pressure (kPa)




                                          300


                                          200

                                                             Pectin/cellulose
                                          100
                                                                                           Xyloglucan/cellulose
                                                                  3 components
                                            0
            (b)                                 0   1         2          3      4            5       6
                                                                  Displacement (mm)

                                          1.4
                                                                                           XG/cellulose
                  Relative displacement




                                          1.3



                                          1.2
                                                                                       Pectin/XG/cellulose

                                          1.1


                                                                          Cellulose       Pectin/cellulose
                                          1.0
            (c)                                 0       100              200              300             400
                                                                       Time (s)

Figure 9.20 Performance of bacterial cellulose based composites under (a) uniaxial loading,
(b) biaxial loading and (c) creep (59). (With kind permission from Springer Science + Business
Media: Planta, Mechanical properties of primary plant cell wall analogues, 215(6), 2002,
989–96, Chanliaud, E.; Burrows, K.M.; Jeronimidis, G.; Gidley, M.J., Figs. 3, 6 and 7.)
254    The Nanoscience and Technology of Renewable Biomaterials




                          1 µm                             1 µm                             1 µm



              (a)                              (b)                              (c)

Figure 9.21 Deep etch freeze fracture TEM of xyloglucan/BC composites showing from left
to right (a) absence of crosslinks, (b) intimate mix of thick-stranded cellulose fibrils in a
meshwork of thin stranded xyloglucan and (c) showing areas of thin stranded xyloglucans
between fibrils (67). (Reproduced from Whitney, S.E.C.; Wilson, E.; Webster, J.; Bacic, A.;
Reid, J.S.G.; Gidley, M.J., Effects of structural variation in xyloglucan polymers on interactions
with bacterial cellulose. American Journal of Botany, 2006, 93, 1402–14, with permission
from the Botanical Society of America, Copyright (2006).)

tamarind xyloglucans the typical crosslinks were observed in all nanocomposites but the
one using 60% galactose depleted xyloglucan. Local alignment of the cellulose fibers was
also observed in all nanocomposites especially with the high molecular weight tamarind
xyloglucans. Besides this aligned and tethered structure, three additional structures were
observed (Figure 9.21).
   In one structure, purely cellulosic regions devoid of cross-bridges were observed. This
structure was most prominent in the highly galactose depleted xyloglucan. In another
structure, thin xyloglucan strands appeared interdispersed between thicker and somewhat
aligned cellulose fibers, indicating intimate mixing between the two polymeric phases.
This structure was not seen in nanocomposites using high galactose depleted and/or high
molecular weight tamarind xyloglucans. A last type of structure reminiscent of a weak
gel was observed, indicating large self-association of xyloglucans. This structure was
mainly observed in nanocomposites with xyloglucans having a high degree of galactose
depletion in combination with a pure cellulose domain, demonstrating a high degree of
segregation (two phases) in those nanocomposites. Mechanical properties under uniaxial
loading also showed that the more galactose depleted BC/xyloglucan nanocomposites
had lower strength and extensibility compared to the BC/native xyloglucans. It was
proposed that direct molecular binding between the xyloglucans and cellulose occurred
mainly due to the ability of the xyloglucan backbone to bind to the nascent cellulose
and to adopt the twofold conformation of crystalline cellulose (67). Another important
parameter to molecular binding was the galactose content since high depletion of galac-
tose (60%) led to the development of two phases, a cellulose phase and a xyloglucan
gel phase, a morphology consistent with the low extension and strength observed in
                              Bacterial Cellulose and Its Polymeric Nanocomposites    255

galactose depleted BC/xyloglucan nanocomposites while their stiffness was similar to
that of cellulose alone. Thus galactose substitution appeared to be a major determinant
of the nanocomposite structure and performance. Besides the lower molecular mass led
to lower xyloglucan incorporation and absence of cross-bridges in the composites. A
secondary role of the degree of fucolysation was also noted where fucosylation appeared
to promote the formation of a crosslinked BC/xyloglucan network. While the effect of
fucose substitution was not as important on the composite formation as the galactose
content, a modulating role was proposed (67). These studies thus clearly demonstrate
the power of biomimetic approaches for tailoring the morphology and performance of
BC nanocomposites when a host polymer capable of interacting closely with cellulose
is selected.

9.5.2 BC/Mannan Nanocomposites
Only a few studies have evaluated the effect of galactomannans and glucomannans on
the in vitro synthesis of BC/mannans nanocomposites, mainly focusing on their mor-
phology rather than their performance (50, 64). A wide range of morphologies has been
observed with mannans, glucomannans leading to more pronounced changes on BC
than galactomannans (64). With glucomannans highly heterogeneous nanocomposites
have been observed comprising (1) regions where cellulose and the glucomannans could
not be distinguished and (2) regions where cellulose ribbons were easily discerned and
crosslinked by thin strands of glucomannan chains and regions where a glucomannan
network formed within a cellulose network (Figure 9.22). Besides the presence of gluco-
mannans in the growth medium was found to drastically decrease cellulose crystallinity
while slightly increasing the relative abundance of Iβ form suggesting that glucoman-
nans may be present in the cellulose microfibrils. Detection of the glucomannans by
CP/MAS NMR supported the hypothesis that glucomannans exist in a rigid form or in an
extended ‘cellulosic’ conformation. Similarly to xyloglucans, glucomannans can asso-
ciate with cellulose microfibrils leading to loose microfibrils bundles having a lower Iα
content (50).
   With galactomannans a wide range of structures and morphologies has also been
observed depending on the galactose–mannose ratio; lower galactose content resulting in
higher incorporation of mannans into the nanocomposites. Crosslinking, self-associations
into gellike structures, cellulose ribbon alignment and self-association of galactomannans
close to cellulose structures and combinations of these different morphologies have been
observed (Figure 9.23). In contrast to glucomannans, galactomannans were not found
to disrupt cellulose crystallinity. Besides when added in large amounts to BC, galac-
tomannans were found to be rather mobile from SP/MAS and therefore considered to be
simply trapped rather than intimately associated in the cellulose fibers as glucomannans.
The lower galactose content galactomannans resulted in more self-association of galac-
tomannans. Cellulose fibers coalesced and the galactomannans formed a gel. Compared
to the xyloglucans the crosslinks were also more random in size thus explaining that
the long-range alignment of BC fibrils was not pronounced in these nanocomposites
compared to BC/xyloglucan nanocomposites. Altogether mannans led to less ordered
network structures compared to xyloglucans.
256    The Nanoscience and Technology of Renewable Biomaterials




                       (a)               0.5 µm                     (b)              0.5 µm

Figure 9.22 Micrographs of Konjac Glucomannans composites showing heterogeneous
morphology comprising regions where cellulose and glucomannan cannot be distinguished
(a), and regions where glucomannans network is formed (b). (Reprinted from Carbohydrate
Research, 307, Whitney, S.E.C.; Brigham, J.E.; Darke, A.H.; Reid, J.S.G.; Gidley, M.J.,
Structural aspects of the interaction of mannan-based polysaccharides with bacterial cellulose,
299–309, Copyright (1998), with permission from Elsevier.)




              (a)       0.5 µm               (b)       0.5 µm              (c)        0.5 µm

Figure 9.23 Galactomannans composites showing heterogeneous morphology (a) cellulose
ribbon crosslinking (b) mannans gel formation and (c) cellulose ribbon alignment. (Reprinted
from Carbohydrate Research, 307, Whitney, S.E.C.; Brigham, J.E.; Darke, A.H.; Reid, J.S.G.;
Gidley, M.J., Structural aspects of the interaction of mannan-based polysaccharides with
bacterial cellulose, 299–309, Copyright (1998), with permission from Elsevier.)
                               Bacterial Cellulose and Its Polymeric Nanocomposites      257

9.5.3 BC/Pectin Nanocomposites
BC/pectin nanocomposites were also prepared in vitro and extensively characterized
(58–61). Pectin consists of galacturonans, galactans and arabinans. For pectin to be
incorporated in the BC network, calcium binding was required, and pectins having a
lower degree of methyl substitution were more easily incorporated (60). The morphology
of the BC/pectin nanocomposites differed greatly from the crosslinked and intimately
mixed morphology observed with glucans and mannans. Interpenetrating networks of
pectin bacterial cellulose formed when a slightly crosslinked pectin network existed in
the growth medium of BC (60). Pectin was found to aggregate next to cellulose fibrils
leaving voids of intermediate size between that of the cellulose and pectin networks
alone (Figure 9.24). However, pectin did not alter the width of the cellulose fibers
nor their crystallinity (60). In the absence of preexisting slightly crosslinked pectin
network, the BC/pectin nanocomposites developed as a phase separated material of BC
and pectin. With solid state NMR the absence of molecular interactions between pectin




                      (a)                                           (b)




                      (c)                                          (d)

Figure 9.24 Micrographs of (a) pectin/Ca2+ gel (b) Acetobacter cellulose (c) Ace-
tobacter cellulose/pectin/Ca2+ composite (BC:Pectin in 60:40) and (d) Acetobacter
cellulose/pectin/Ca2+ composite (BC:Pectin in 71:29). Scale bar = 0.5 µm. (Chanliaud, E.;
Gidley, M. J., In Vitro synthesis and properties of pectin/Acetobacter xylinus cellulose com-
posites. Plant Journal, 1999, 20, 25–35. Copyright 1999. Reprinted with permission from
Wiley-Blackwell.)
258    The Nanoscience and Technology of Renewable Biomaterials

and cellulose was further demonstrated. Yet pectin appeared to be rigidified in presence
of the cellulose network as it aggregated around the rigid cellulose microfibrils (60).
   The small deformation rheology of BC/pectin nanocomposites was very similar to
that previously observed with BC/xyloglucan nanocomposites. That is, the cellulose
network dominated the small deformation properties of the nanocomposite. For uniaxial
tensile loading the pectin had a much more dramatic effect than xyloglucans, increasing
the extensibility very significantly while decreasing the composite stiffness and strength
(Figure 9.20). As a result the nanocomposite toughness was much higher than that of the
pure components. While the performance of BC/pectin nanocomposites is reminiscent of
that of xyloglucan/BC nanocomposites, the change in the property was entirely ascribed
to the different cellulose structure that developed in presence of the pectin network
rather than to the pectin itself or to the cellulose/pectin interactions (60). Local isotropic
arrangement of the cellulose fibers into the pectin network caused cellulose fibers to have
less contact between each other’s allowing for greater slippage and alignment of fibers
upon tensile loading and therefore higher extension. In fact, in composites comprising
20% pectin and conditioned at different relative humidity, 2D FTIR spectroscopy con-
firmed that the cellulose and pectin networks did not display any connected motion and
behaved independently as interpenetrating networks (58).
   A 39% pectin content composite was further characterized for biaxial tensile load-
ing and creep. In biaxial tensile loading, the BC/pectin nanocomposite although it was
weaker and also had a slightly lower maximum elongation than the cellulose alone,
was found to behave as a linear elastic material similarly to cellulose (59). Again the
nanocomposite weakening was ascribed to the modified cellulose network that resulted
from cellulose deposition within a pectin gel. Similarly to cellulose, the BC/pectin
nanocomposites exhibited no creep i.e. time-dependent behavior. This was in sharp
contrast to the behavior of BC/xyloglucan nanocomposites which were prone to creep
(Figure 9.20). Differences in biaxial tensile behavior and in creep reflected the mor-
phological differences between the BC/pectin and BC/xyloglucan nanocomposites. That
is, the cellulose fibers deposited in a pectin gel possibly experienced less interfibrillar
contacts and were more able to align resulting in higher extensibility (58, 59). In these
nanocomposites cellulose was clearly the load bearing component while deposition of
cellulose within a network of pectin could help improve the extensibility and toughness
of the nanocomposites.

9.5.4 BC/Xyoglucan/Pectin Nanocomposites
To further shed light on the morphology and behavior of the primary cell wall, Gidley’s
group also manufactured ternary nanocomposites comprising 63% BC, 22% pectin and
15% xyloglucan (59). The mechanical behavior of the BC/pectin/xyloglucan nanocom-
posites was characterized in uni-axial and bi-axial tensile loading and in creep (59).
The ternary nanocomposite combined the structural and morphological properties of the
BC/xyloglucan and BC/pectin nanocomposites. That is, the typical BC/xyloglucan mor-
phology was observed with intimately bound xyloglucans and crosslinks and a pectin
layer appeared to cover the tethered BC/xyloglucan structures (Figure 9.25). As in other
BC nanocomposites, BC was found to be the loadbearing component while xyloglucans
and pectin increased the material compliance, inducing higher extensibility and lower
                               Bacterial Cellulose and Its Polymeric Nanocomposites      259




                        (a)                                       (b)




                        (c)                                      (d)      0.25 µm


Figure 9.25 TEM micrographs of bacterial cellulose (a), cellulose/pectin (b), cellulose/
xyloglucan (c) and cellulose/pectin/xyloglucan (d) (59) (With kind permission from Springer
Science + Business Media: Planta, Mechanical properties of primary plant cell wall analogues,
215, 2002, 989–96, Chanliaud, E.; Burrows, K.M.; Jeronimidis, G.; Gidley, M.J, Figures 3, 6
and 7.)


strength and stiffness to the nanocomposite compared to the neat BC (Figure 9.20). The
BC/pectin/xyloglucan nanocomposites were also prone to creep, thus exhibiting a time
dependent behavior that was not present in the BC and in the BC/pectin nanocomposites
(Figure 9.20). In fact, regardless of the loading mode, the ternary nanocomposites exhib-
ited a behavior intermediate to that of the BC/pectin and BC/xyloglucan nanocomposites,
with a closer resemblance with the later. Besides, as for the BC/xyloglucan nanocom-
posites, the ternary nanocomposites could not be modeled as linear elastic materials.

9.5.5 BC/Lignin Nanocomposites
Other model nanocomposites based on BC have also been prepared to study the lignifi-
cation of the cell wall (62, 63). Using a diffusion cell, coniferyl alcohol, a precursor of
lignin, was polymerized into a BC mat and a BC/pectin mat. Observation of the compos-
ites by SEM showed that the dehydrogenation polymers (DHP or synthetic lignin) formed
ovoid aggregates, a few microns in diameter, on the cellulose microfibrils, demonstrating
phase separation between cellulose and lignin (Figure 9.26) (62). In contrast, DHP syn-
thesized in the BC/pectin mat appeared to be more homogeneously dispersed, forming
smaller aggregates on the cellulose fibrils (62). Pectin was thus proposed to act as a
260   The Nanoscience and Technology of Renewable Biomaterials




                  100 µm                                      100 µm
                    (a)                                         (d)




                   20 µm                                       20 µm
                    (b)                                          (e)




                   10 µm                                       10 µm
                    (c)                                          (f)

Figure 9.26 Micrographs of bacterial cellulose/DHP blends and bacterial cellulose/
pectin/DHP blends at different magnification (62). (Reprinted with permission from Touzel,
J.P.; Chabbert, B.; Monties, B.; Debeire, P.; Cathala, B., Synthesis and characterization
of dehydrogenation polymers in Gluconacetobacter xylinus cellulose and cellulose/pectin
composite. Journal of Agricultural and Food Chemistry, 2003, 51, 981–6. Copyright (2003),
American Chemical Society.)
                              Bacterial Cellulose and Its Polymeric Nanocomposites   261

compatibilizer between cellulose and DHP. However with solid-state NMR no covalent
bond between lignin and pectin could be determined (63) suggesting that noncovalent
bonds might govern the association of DHP with pectin. At the nanometer scale however
the pectin and the lignin were found to phase separate (63). With thiadoacydolysis, it
was shown that presence of pectin influenced the polymerization of coniferyl alcohol
since lignin had higher amounts of β-O-4 linkages (62).

9.5.6 BC/Synthetic Polymer Nanocomposites
A couple of studies have examined the reinforcing potential of BC for polyethylene
oxide (PEO) in nanocomposites prepared in vitro (68, 70). Takai first reported on incor-
porating PEO into BC fleeces, among other water soluble polymers including cellulose
derivatives, to form nanocomposites for application as membranes (70). Incorporation of
a 50,000 g/mol PEO into the BC fleeces caused a 10% wt gain, lower than that observed
with other cellulose derivatives. The BC/PEO nanocomposite displayed a similar modu-
lus than the neat BC at around 25 GPa, which was not the case of other nanocomposites
with modulus all below 10 GPa. A similar bioengineering approach was taken in another
study in which BC was grown into media enriched with various PEO contents with a view
to tailoring nanocomposite composition, morphology and properties (68). The unpurified
nanocomposites were freeze dried and compression molded into films for characteriza-
tion by thermogravimetric analysis (TGA), atomic force microscopy (AFM), infrared
spectroscopy (FTIR), differential scanning calorimetry (DSC) and dynamic mechani-
cal analysis (DMA). As expected, increasing PEO concentration in the growth medium
resulted in a systematic change in composition with the BC: PEO ratio varying from
59:41 to 15:85. The nanocomposite morphology also changed with addition of PEO. In
compression molded samples, the BC fibers appeared to aggregate in larger bundles as
the PEO content increased (Figure 9.27).
   Fine dispersion of the cellulose fibers into the PEO matrix was further demonstrated
by the significant drop in the melting temperature (Tm) and crystallinity of PEO. In
presence of BC fiber, less stable PEO crystals might have nucleated on the fiber surface
and crystal growth might have been impinged by the finely dispersed BC fibers resulting
in a low Tm and crystallinity matrix. This fine mixing of cellulose fibers in PEO was
also consistent with the indication of hydrogen bonding between cellulose hydroxyls
and the PEO oxy group. Finally, selected physical, thermal and mechanical properties
could also be tailored from the nanocomposite composition. The thermal decomposition
temperature of PEO increased by 15 ◦ C in the nanocomposites, and this increase was
ascribed to mutual thermal stabilization with cellulose. DMA in tension mode showed
that BC effectively reinforced the PEO matrix in the glassy and rubbery states and most
significantly above the melting temperature of PEO (Figure 9.28). Surface roughness
also decreased with higher PEO content. Integrating the BC synthesis with the mixing
step with a thermoplastic polymer was thus proposed to be a potent manner to manip-
ulate the properties of BC/PEO nanocomposites. A similar approach was taken with
poly(vinyl alcohol) (PVA), demonstrating the potential to manipulate composition and
selected properties of BC/PVA nanocomposites as well (69).
   In an attempt to control the water absorption potential of BC materials for biomedical
applications, Seifert et al. (29) also used in vitro synthesis of BC in the presence of
262    The Nanoscience and Technology of Renewable Biomaterials




Figure 9.27 Atomic Force Microscopy Images of BC grown in (a) HS medium, (b) HS
medium with 1% PEO, (c) HS medium with 3% PEO, and (d) HS medium with 5% PEO
(68). (Reprinted with permission from Brown, E.E.; Laborie, M.-P.G., Bioengineering bacterial
cellulose/poly(ethylene oxide) nanocomposites. Biomacromolecules, 2007, 8, 3074–81.
Copyright (2007), American Chemical Society.)

                             109

                             108

                             107
                                            59:41
                   E’ (Pa)




                             106            53:47
                                            33:67
                                            23:77
                             105
                                            15:85
                                            0:100
                             104

                             103
                                −60   −40   −20      0     20   40     60   80   100
                                                    Temperature (°C)

Figure 9.28 Storage tensile modulus E’ versus temperature at 1 Hz for nanocomposites
of varying BC/PEO ratios (68). (Reprinted with permission from Brown, E.E.; Laborie,
M.-P.G., Bioengineering bacterial cellulose/poly(ethylene oxide) nanocomposites. Biomacro-
molecules, 2007, 8, 3074–81. Copyright (2007), American Chemical Society.)

various water-soluble polymers including poly(vinyl alcohol). It appeared that PVA was
easily washed from the BC/PVA nanocomposite during the purification step although
some PVA could be detected when 2% PVA was added in the growth medium of BC. In
this case, the BC network structure and crystalline structure appeared to be unaffected by
the presence of PVA. This contrasted with the morphology of the BC/methyl cellulose
                               Bacterial Cellulose and Its Polymeric Nanocomposites     263

(CMC and MC) nanocomposites which shifted from a fine fibrillar structure to a porous
structure with increasing MC or CMC content. As a result, the BC/CMC nanocom-
posites had superior absorption properties than neat BC and the PC/PVA nanocompos-
ites (29).


9.6 BC/Polymer Nanocomposites Based on Bacterial
    Cellulose Nanocrystals

Other attempts at preparing nanocomposites based on bacterial cellulose have utilized
hydrolyzed cellulose nanocrystals (CNXL) (72–75). Since the discovery of their rein-
forcing potential in 1995, CNXLs have been shown to reinforce a wide variety of
thermoplastic polymers via a percolation effect (76–78). That is, above a percolation
threshold the stiff rodlike nanocrystals are able to interact via hydrogen bonding to form
a percolation network which induces a very significant stiffening of thermoplastics (78).
With the large Young modulus of cellulose I at around 140 Gpa (79), an outstand-
ing reinforcing potential of BC nanocrystals in polymer matrices might be expected.
Cellulose nanowhiskers are obtained by acid treatment of cellulose which hydrolyses the
amorphous region of cellulose and liberates nanocrystals, typically 10–20 nm in width
(Figure 9.29).
   Grunert and Winter first reported on the use of bacterial CNXLs in polydimethyl-
siloxane (PDMS) polymers (72). In this work, CNXL from BC were derivatized by
trimethylsilylation with a view to improving the interfacial adhesion with PDMS. Con-
trol and surface modified CNXLs were thus blended with dimethylsiloxane and the
appropriate curing agents to develop films. Dynamic mechanical analysis (DMA) in
the – 60 ◦ C to 240 ◦ C range revealed a twofold increase in storage modulus of PDMS




                                300 nm




Figure 9.29 Transmission electron micrograph of bacterial cellulose nanocrystals. (With kind
permission from Springer Science+Business Media: Journal of Polymers and the Environment,
Nanocomposites of cellulose acetate butyrate reinforced with cellulose nanocrystals, 10,
2002, 27–30, M. Grunert and Winter, W.T., Figure 1.)
264   The Nanoscience and Technology of Renewable Biomaterials

from the 0.5–0.75 MPa range to the 1.25–1.5 MPa range when untreated CNXLs were
added in PDMS. Surprisingly, trimethylsilylation of the CNXLs lowered the reinforcing
effect of CNXLs on PDMS, a phenomenon that was ascribed to surface hydrophobicity
(72). A similar approach was taken with cellulose acetate butyrate (CAB) matrix (73).
Partially silylated (DS of 0.49) and untreated CNXLs were incorporated into CAB at var-
ious loadings from 2% to 10 wt%. Although a small increase in the melting temperature
of the CAB was reported with higher CNXL content, its crystallinity slightly decreased
in presence of silylated or untreated CNXLs. CNXLS were proposed to impinge on
CAB crystallization. As expected, the nanocomposite storage modulus increased with
higher CNXLs content and above the glass transition temperature of CAB, the sily-
lated CNXLs provided a less efficient reinforcement than the native CNXLs (73), a
phenomenon similar to that observed with PDMS (72) (Figure 9.30). Damping was
also reduced in presence of CNXLs. It was thus proposed that on a weight basis the

                        10,000


                            1000
             E’ (MPa)




                                100
                                                  10 wt.% Silyalted crystals
                                                  10 wt.% Native crystals
                                10                5 wt.% Silylated crystals
                                                  5 wt.% Native crystals
                                                  0 wt.% crystals
                                 1
                                      25   35   45   55    65     75     85    95 105 115 125 135 145
                                                             Temperature (°C)

                                1.6
                                                10 wt.% Silylated crystals
                                1.4
                                                10 wt.% Native crystals
                                1.2             5 wt.% Silylated crystals
                                                5 wt.% Native crystals
                                 1
                                                0 wt.% crystals
                        tan d




                                0.8
                                0.6
                                0.4
                                0.2
                                 0
                                      25   35   45   55    65 75 85 95 105 115 125 135 145
                                                             Temperature (°C)

Figure 9.30 Temperature dependence of the storage modulus (E’) and loss tangent (tan
d) of the composites at a frequency of 10 Hz. (With kind permission from Springer
Science + Business Media: Journal of Polymers and the Environment, Nanocomposites
of cellulose acetate butyrate reinforced with cellulose nanocrystals, 10, 2002, 27–30,
M. Grunert and Winter, W.T., Figure 4.)
                                Bacterial Cellulose and Its Polymeric Nanocomposites      265

native CNXLs had a better reinforcing effect than the silylated CNXLs and immobilized
the CAB matrix in the vicinity.
   A more recent study compared the effect of CNXLs from various sources including BC
to reinforce thermoplastic starch (74). Using different processing methods, thermoplastic
starch and pectin were blended with CNXLs. Namely films were produced by solution
casting with 3% CNXLs while monofilaments were produced by mixing in a Hobart
mixer, followed by extrusion. After equilibration at 50% RH the tensile properties of
the films were determined and the thermal properties of samples with and without CNXLs
were compared. The Young modulus of thermoplastic starch increased from 1.39 GPA in
the neat state to more than 6 GPa with the incorporation of bacterial CNXLs. Elongation
at break also increased from 2.7% to 4% with bacterial CNXLs. While bacterial CNXLs
clearly improved the performance of thermoplastic starch, they were not as effective as
CNXLs from softwood or cotton. For example the elongation at break was 8% with
softwood or cotton CNXLs, that is twice that observed with bacterial CNXLs. Blends of
starch and pectin (50/50) were also produced and exhibited better strength and stiffness
than the pure thermoplastic polymers. However, regardless of the CNXLs origin, their
incorporation in the thermoplastic blend decreased its strength, elongation and tensile
modulus altogether.
   In another study, bacterial CNXLs were incorporated in polyethylene(oxide) and effec-
tively electrospun in a 1.5 mm capillary tube (75). The BC CNXLs had dimensions
of 420 ± 190 nm X 11 ± 4 nm X 10 ± 2 nm as measured by TEM and AFM and
when added to PEO as a suspension with some additional water, the PEO/cellulose
whiskers mix was amenable to electrospinning in terms of viscosity and surface tension.
Interestingly the nanofiber diameter increased with the incorporation of the cellulose
nanowhiskers. For neat PEO the nanofibers had a diameter of 140 ± 20 nm which
increased to the 250–350 nm range with incorporation of 0.2 wt% and 0.4 wt% cellulose
whiskers. The nanofiber size distribution was also increased with the incorporation of
the cellulose whiskers (Figure 9.31). The cellulose whiskers were well incorporated into
the nanofibers, although some whiskers were observed to protrude out of the nanofiber
(Figure 9.32). Further morphological investigation of the cellulose whisker dispersion




                      0.5 µm                         0.5 µm                          0.5 µm

             (a)                             (b)                             (c)

Figure 9.31 Field emission scanning electron microscopic (FESEM) image of electrospun
PEO/cellulose whiskers having (a) 0 wt%, (b) 0.2 wt% and (c) 0.4 wt% of whiskers (Park,
W.-I.; Kang, M.; Kim, H.-S.; Jin, H.-K., Electrospinning of poly(ethylene oxide) with bacterial
cellulose whiskers, Macromolecular Symposia, 2007, 249–50, 289–94. Copyright Wiley-VCH
Verlag GmbH & Co.KGaA. Reproduced with permission.)
266   The Nanoscience and Technology of Renewable Biomaterials




            100 nm                            200 nm

                           (a)                               (b)

Figure 9.32 TEM images of PEO/bacterial cellulose whiskers (0.4 wt%) showing (a) alignment
of cellulose whiskers and (b) entanglement of cellulose whiskers in electrospun PEO fibers
Park, W.-I.; Kang, M.; Kim, H.-S.; Jin, H.-K., Electrospinning of poly(ethylene oxide) with
bacterial cellulose whiskers, Macromolecular Symposia, 2007, 249–50, 289–94. Copyright
Wiley–VCH Verlag GmbH & Co.KGaA. Reproduced with permission.)

into the PEO nanofiber by TEM indicated that in some cases the nanowhiskers aligned
along the fiber direction whereas aggregation problems were also identified in other cases
(Figure 9.32). As a result of the incorporation of BC whiskers into the PEO matrix, the
mechanical properties of the nanofibers were very significantly enhanced. Young mod-
ulus and extension at break more than doubled with incorporation of BC nanowhiskers.
The nanofiber strength was also significantly improved. Enhancement in mechanical
properties appeared to be proportional to the amount of whiskers incorporated.


9.7   Conclusions and Prospects

This review illustrates the intensification of research and development in the last decade
involving bacterial cellulose nanocomposites. There is no doubt that the growing interest
in bacterial cellulose arises from its unique properties that confer potential in high-tech
applications that are currently in high demand such as that of biomaterials. Besides,
bacterial cellulose is particularly suited for the development of complex materials using
biomimecry approach, another highly active research field. While most of the research
reviewed in this paper has used bacterial cellulose as is, bacterial cellulose has abundant
hydroxyl groups on its surface, conferring the potential for surface derivatization and
functionalization. Compatibilization of bacterial cellulose with a wider range of poly-
mers may be afforded by surface grafting of appropriate functional groups or chains to
yield new nanocomposites. Besides, more complex nanostructured materials might be
developed from bacterial cellulose by functionalizing its surface with molecules such
as proteins or other compounds that display interesting self-assembly properties. In
any case, it is clear from recent developments in the field of BC nanomaterials, that
research and developments will further intensify and new high-value applications for
BC nanocomposites will be developed.
                                Bacterial Cellulose and Its Polymeric Nanocomposites   267

References

1.    Brown, R.M.J., The biosynthesis of cellulose. Journal of Macromolecular Science-
      Pure and Applied Chemistry 1996, A33(10), 1345–73.
2.    Brown, R.M., Cellulose structure and biosynthesis: What is in store for the 21st cen-
      tury? Journal of Polymer Science Part a-Polymer Chemistry 2004, 42(3), 487–95.
3.    Klemm, D.; Schumann, D.; Kramer, F.; Hessler, N.; Hornung, M.; Schmauder,
      H.P.; Marsch, S., Nanocelluloses as innovative polymers in research and application.
      Polysaccharides Ii 2006, 205, 49–96.
4.    Klemm, D.; Schumann, D.; Udhardt, U.; Marsch, S., Bacterial synthesized cellu-
      lose – artificial blood vessels for microsurgery. Progress in Polymer Science 2001,
      26(9), 1561–1603.
5.    Haigler, C.H.; White, A.R.; Brown, R.M.J.; Cooper, K.M., Alteration of in vivo
      cellulose ribbon assembly by carboxymethylcellulose and other cellulose derivatives.
      The Journal of Cell Biology 1982, 94(1), 64–9.
6.    Haigler, C.H.; Brown, R.M.J.; Benziman, M., Calcofluor white ST Alters the in
      vivo assembly of cellulose microfibrils. Science 1980, 210(4472), 903–6.
7.    Yamanaka, S.; Watanabe, K.; Kitamura, N.; Iguchi, M.; Mitsuhashi, S.; Nishi, Y.;
      Uryu, M., The structure and mechanical properties of sheets prepared from bacterial
      cellulose. Journal of Materials Science 1989, 24, 3141–5.
8.    Nishi, Y.; Uryu, M.; Yamanaka, S.; Watanabe, K.; Kitamura, N.; Iguchi, M.;
      Mitsuhashi, S., The structure and mechanical-properties of sheets prepared from
      bacterial cellulose.2. Improvement of the mechanical-properties of sheets and their
      applicability to diaphragms of electroacoustic transducers. Journal of Materials
      Science 1990, 25(6), 2997–3001.
9.    Iguchi, M.; Yamanaka, S.; Budhiono, A., Bacterial cellulose – a masterpiece of
      nature’s arts. Journal of Materials Science 2000, 35(2), 261–70.
10.   Nakagaito, A.N.; Iwamoto, S.; Yano, H., Bacterial cellulose: the ultimate nano-
      scalar cellulose morphology for the production of high-strength composites. Applied
      Physics A-Materials Science & Processing 2005, 80(1), 93–7.
11.   Nakagaito, A.N.; Yano, H., Novel high-strength biocomposites based on microfibril-
      lated cellulose having nano-order-unit web-like network structure. Applied Physics
      A-Materials Science & Processing 2005, 80(1), 155–9.
12.   Ifuku, S.; Nogi, M.; Abe, K.; Handa, K.; Nakatsubo, F.; Yano, H., Surface modifi-
      cation of bacterial cellulose nanofibers for property enhancement of optically trans-
      parent composites: Dependence on acetyl-group DS. Biomacromolecules 2007, 8(6),
      1973–8.
13.   Nogi, M.; Abe, K.; Handa, K.; Nakatsubo, F.; Ifuku, S.; Yano, H., Property enhance-
      ment of optically transparent bionanofiber composites by acetylation. Applied
      Physics Letters 2006, 89(23), 233123: 1–3.
14.   Nogi, M.; Handa, K.; Nakagaito, A.N.; Yano, H., Optically transparent bionanofiber
      composites with low sensitivity to refractive index of polymer matrix. Applied
      Physics Letters 2005, 87(24), 3110.
15.   Nogi, M.; Ifuku, S.; Abe, K.; Handa, K.; Nakagaito, A.N.; Yano, H., Fiber-content
      dependency of the optical transparency and thermal expansion of bacterial nanofiber
      reinforced composites. Applied Physics Letters 2006, 88(13) 133124: 1–3.
268   The Nanoscience and Technology of Renewable Biomaterials

16. Yano, H.; Sugiyama, J.; Nakagaito, A.N.; Nogi, M.; Matsuura, T.; Hikita, M.;
    Handa, K., Optically transparent composites reinforced with networks of bacterial
    nanofibers. Advanced Materials 2005, 17(2), 153–5.
17. Yano, H.; Nogi, M. Fiber-reinforced composite material and method for production
    thereof, and precursor for producing fiber-reinforced composite material. WO Patent
    2006082964 A1 2006, 2006.
18. Yano, H.; Nogi, M.; Abe, K. et al., Fiber composite material and process for pro-
    ducing the same. WO Patent 2008010449 A1 2008, 2008.
19. Yano, H.; Nogi, M.; Ifuku, S.; Abe, K.; Handa, K. Nanofiber sheet, process for pro-
    ducing the same, and fiber-reinforced composite material. WO Patent 2008010462
    A1 2008, 2008.
20. Yano, H.; Nogi, M.; Ifuku, S.; Abe, K.; Takezawa, Y.; Handa, K. Fiber-reinforced
    composite resin compositions with good isotropic thermal conductivity, trans-
    parency, and strength, and low thermal expansion for adhesives and electronic
    packaging materials. WO Patent 2007049666 A1 2007, 2007.
21. Yano, H.; Nogi, M.; Nakagaito, A. N. Fiber-reinforced composite material and
    method for production thereof. WO Patent 2006087931 A1 2006, 2006.
22. Yano, H.; Nogi, M.; Nakatsubo, F.; Ifuku, S. Manufacture of transparent fiber-
    reinforced composite materials. WO Patent 2006082803 A1 2006, 2006.
23. Yano, H.; Sugiyama, J.; Nogi, M.; Iwamoto, S.; Handa, K.; Nagai, A.; Miwa, T.;
    Takezawa, Y.; Miyadera, T.; Kurihara, T.; Matsuura, T.; Koshoubu, N.; Maruno, T.,
    Preparation of fiber-reinforced composite material with good transparency for lam-
    inate, printed circuit board, and optical waveguide. WO Patent 2005012404 A1
    2005, 2005.
24. Choi, Y.J.; Ahn, Y.H.; Kang, M.S.; Jun, H.K.; Kim, I.S.; Moon, S.H., Prepara-
    tion and characterization of acrylic acid-treated bacterial cellulose cation-exchange
    membrane. Journal of Chemical Technology and Biotechnology 2004, 79(1), 79–84.
25. Kramer, F.; Klemm, D.; Schumann, D.; Hessler, N.; Wesarg, F.; Fried, W.; Stader-
    mann, D., Nanocelluloses polymer composites as innovative pool for (bio)material
    development. Macromolecular Symposia 2006, 244, 136–48.
26. Gindl, W.; Keckes, J., Tensile properties of cellulose acetate butyrate compos-
    ites reinforced with bacterial cellulose. Composites Science and Technology 2004,
    64(15), 2407–13.
27. Millon, L.E.; Mohammadi, H.; Wan, W.K., Anisotropic polyvinyl alcohol hydro-
    gel for cardiovascular applications. Journal of Biomedical Materials Research Part
    B-Applied Biomaterials 2006, 79B(2), 305–11.
28. Millon, L. E.; Wan, W. K., The polyvinyl alcohol-bacterial cellulose system as a
    new nanocomposite for biomedical applications. Journal of Biomedical Materials
    Research, Part B: Applied Biomaterials 2006, 79B (2), 245–53.
29. Seifert, M.; Hesse, S.; Kabrelian, V.; Klemm, D., Controlling the water content of
    never dried and reswollen bacterial cellulose by the addition of water-soluble poly-
    mers to the culture medium. Journal of Polymer Science, Part A: Polymer Chemistry
    2004, 42(3), 463–70.
30. Dammstrom, S.; Gatenholm, P., Preparation and properties of cellulose/xylan nano-
    composites. In Characterization of the Cellulosic Cell Wall , Stokke, D. and Groom,
    L.H., eds. Blackwell Publishing: 2006; pp. 53–66.
                              Bacterial Cellulose and Its Polymeric Nanocomposites    269

                            e
31. Dammstrom, S.; Salm´ n, L.; Gatenholm, P., The effect of moisture on the dynamical
    mechanical properties of bacterial cellulose/glucuronoxylan nanocomposites. Poly-
    mer 2005, 46(23), 10364–71.
32. Linder, A.; Bergman, R.; Bodin, A.; Gatenholm, P., Mechanism of assembly of
    xylan onto cellulose surfaces. Langmuir 2003, 19(12), 5072–7.
33. Shirai, A.; Sakairi, N.; Nishi, N.; Tokura, S., Preparation of a novel (1->4)-beta-
    D-glycan by Acetobacter xylinum – A proposed mechanism for incorporation of a
    N-acetylglucosamine residue into bacterial cellulose. Carbohydrate Polymers 1997,
    32(3–4), 223–7.
34. Hamlyn, P.F.; Crighton, J.; Dobb, M.G.; Tasker, A. Cellulose product. GB Patent
    2314856, 14.01.1998, 1998.
35. Ciechanska, D., Multifunctional bacterial cellulose/chitosan composite materials for
    medical applications. Fibres & Textiles in Eastern Europe 2004, 12(4), 69–72.
36. Dubey, V.; Pandey, L.K.; Saxena, C., Pervaporative separation of ethanol/water
    azeotrope using a novel chitosan-impregnated bacterial cellulose membrane and
    chitosan-poly(vinyl alcohol) blends. Journal of Membrane Science 2005, 251(1–2),
    131–6.
37. Ciechanska, D.; Struszczyk, H.; Guzinska, K., Modification of bacterial cellulose.
    Fibres & Textiles in Eastern Europe 1998, 6(4), 61–5.
38. Gong, J.P.; Katsuyama, Y.; Kurokawa, T.; Osada, Y., Double-network hydro-
    gels with extremely high mechanical strength. Advanced Materials 2003, 15(14),
    1155–8.
39. Nakayama, A.; Kakugo, A.; Gong, J.P.; Osada, Y.; Takai, M.; Erata, T.; Kawano,
    S., High mechanical strength double-network hydrogel with bacterial cellulose.
    Advanced Functional Materials 2004, 14(11), 1124–8.
40. Jung, R.; Jin, H.-J., Preparations of silk fibroin/bacterial cellulose composite films
    and their mechanical properties. Key Engineering Materials 2007, 342-343, 741–4.
41. Wan, W.K.; Millon, L. Poly(vinyl alcohol)-bacterial cellulose nanocomposite. US
    Patent 20050037082, 2005.
42. Cousins, S.K.; Brown, R.M., X-ray diffraction and ultrastructural analyses of dye-
    altered celluloses support van der Waals forces as the initial step in cellulose crys-
    tallization. Polymer 1997, 38(4), 897–902.
43. Ben-Hayyim, G.; Ohad, I., Synthesis of cellulose by Acetobacter xylinum. VIII.
    Formation and orientation of bacterial cellulose fibrils in the presence of acidic
    polysaccharides. Journal of Cell Biology 1965, 25(2), 191–207.
44. Hirai, A.; Tsuji, M.; Yamamoto, H.; Horii, F., In Situ crystallization of bacte-
    rial cellulose – III. Influences of different polymeric additives on the formation of
    microfibrils as revealed by transmission electron microscopy. Cellulose 1998, 5(3),
    201–13.
45. Atalla, R.H.; Hackney, J.M.; Uhlin, I.; Thompson, N.S., Hemicelluloses as structure
    regulators in the aggregation of native cellulose. International Journal of Biological
    Macromolecules 1993, 15(2), 109–12.
46. Horii, F.; Yamamoto, H.; Hirai, A., Microstructural analysis of microfibrils of bac-
    terial cellulose. Macromolecular Symposia 1997, 120, 197–205.
47. Uhlin, K.I.; Atalla, R.H.; Thompson, N.S., Influence of hemicelluloses on the aggre-
    gation patterns of bacterial cellulose. Cellulose 1995, 2(2), 129–44.
270    The Nanoscience and Technology of Renewable Biomaterials

48. Yamamoto, H.; Horii, F., In-situ crystallization of bacterial cellulose 1. Influences
    of polymeric additives, stirring and temperature on the formation celluloses I-Alpha
    and I-Beta as revealed by cross-polarization magic-angle-spinning (Cp/Mas) C-13
    Nmr-spectroscopy. Cellulose 1994, 1(1), 57–66.
49. Yamamoto, H.; Horii, F.; Hirai, A., In situ crystallization of bacterial cellulose II.
    Influences of different polymeric additives on the formation of celluloses Ia and Ib
    at the early stage of incubation. Cellulose 1996, 3(4), 229–42.
50. Tokoh, C.; Takabe, K.; Fujita, M.; Saiki, H., Cellulose synthesized by Acetobacter
    xylinum in the presence of acetyl glucomannan. Cellulose 1998, 5(4), 249–61.
51. Tokoh, C.; Takabe, K.; Sugiyama, J.; Fujita, M., CP/MAS C-13 NMR and electron
    diffraction study of bacterial cellulose structure affected by cell wall polysaccharides.
    Cellulose 2002, 9(3–4), 351–60.
52. Tokoh, C.; Takabe, K.; Sugiyama, J.; Fujita, M., Cellulose synthesized by Aceto-
    bacter xylinum in the presence of plant cell wall polysaccharides. Cellulose 2002,
    9(1), 65–74.
53. Iwata, T.; Indrarti, L.; Azuma, J. I., Affinity of hemicellulose for cellulose produced
    by Acetobacter xylinum. Cellulose 1998, 5(3), 215–28.
54. Hackney, J.M.; Atalla, R.H.; Vanderhart, D.L., Modification of crystallinity and
    crystalline-structure of acetobacter-xylinum cellulose in the presence of water-
    soluble Beta-1,4-linked polysaccharides-C-13-NMR Evidence. International Journal
    of Biological Macromolecules 1994, 16(4), 215–18.
55. Bootten, T.J.; Harris, P.J.; Melton, L.D.; Newman, R.H., WAXS and C-13 NMR
    study of Gluconoacetobacter xylinus cellulose in composites with tamarind xyloglu-
    can. Carbohydrate Research 2008, 343(2), 221–9.
56. Teeri, T.T.; Brumer, H.; Daniel, G.; Gatenholm, P., Biomimetic engineering of
    cellulose-based materials. Trends in Biotechnology 2007, 25(7), 299–306.
57. Astley, O.M.; Chanliaud, E.; Donald, A.M.; Gidley, M.J., Structure of acetobac-
    ter cellulose composites in the hydrated state. International Journal of Biological
    Macromolecules 2001, 29(3), 193–202.
58. Astley, O.M.; Chanliaud, E.; Donald, A.M.; Gidley, M.J., Tensile deformation of
    bacterial cellulose composites. International Journal of Biological Macromolecules
    2003, 32(1–2), 28–35.
59. Chanliaud, E.; Burrows, K.M.; Jeronimidis, G.; Gidley, M.J., Mechanical properties
    of primary plant cell wall analogues. Planta 2002, 215(6), 989–96.
60. Chanliaud, E.; Gidley, M.J., In Vitro synthesis and properties of pectin/Acetobacter
    xylinus cellulose composites. Plant Journal 1999, 20(1), 25–35.
61. Kacurakova, M.; Smith, A.C.; Gidley, M.J.; Wilson, R.H., Molecular interactions in
    bacterial cellulose composites studied by 1D FT-IR and dynamic 2D FT-IR spec-
    troscopy. Carbohydrate Research 2002, 337(12), 1145–53.
62. Touzel, J.P.; Chabbert, B.; Monties, B.; Debeire, P.; Cathala, B., Synthesis and
    characterization of dehydrogenation polymers in Gluconacetobacter xylinus cellulose
    and cellulose/pectin composite. Journal of Agricultural and Food Chemistry 2003,
    51(4), 981–6.
63. Cathala, B.; Rondeau-Mouro, C.; Lairez, D., et al., Model systems for the under-
    standing of lignified plant cell wall formation. Plant Biosystems 2005, 139(1),
    93–7.
                              Bacterial Cellulose and Its Polymeric Nanocomposites   271

64. Whitney, S.E.C.; Brigham, J.E.; Darke, A.H.; Reid, J.S.G.; Gidley, M.J., Structural
    aspects of the interaction of mannan-based polysaccharides with bacterial cellulose.
    Carbohydrate Research 1998, 307(3–4), 299–309.
65. Whitney, S.E.C.; Brigham, J.E.; Darke, A.H.; Reid, J.S.G.; Gidley, M.J., In-vitro
    assembly of cellulose/xyloglucan networks-ultrastructural and molecular aspects.
    Plant Journal 1995, 8(4), 491–504.
66. Whitney, S.E.C.; Gothard, M.G.E.; Mitchell, J.T.; Gidley, M.J., Roles of cellulose
    and xyloglucan in determining the mechanical properties of primary plant cell walls.
    Plant Physiology 1999, 121(2), 657–63.
67. Whitney, S.E.C.; Wilson, E.; Webster, J.; Bacic, A.; Reid, J.S.G.; Gidley, M.J.,
    Effects of structural variation in xyloglucan polymers on interactions with bacterial
    cellulose. American Journal of Botany 2006, 93(10), 1402–14.
68. Brown, E.E.; Laborie, M.-P.G., Bioengineering bacterial cellulose/poly(ethylene
    oxide) nanocomposites. Biomacromolecules 2007, 8(10), 3074–81.
69. Laborie, M.-P.G.; Brown, E.E., Bacterial cellulose/polyvinyl alcohol nanocompos-
    ites. In 235th ACS National Meeting, American Chemical Society: New Orleans,
    LA, 2008.
70. Takai, M., Bacterial cellulose composites. In Cellulosic polymers, blends and com-
    posites, Gilbert., R. D., Ed. Hanser Publishers: Munich, 1994; pp 233–40.
71. Fengel, D.; Wegener, G., Wood: Chemistry, Ultrastructure, Reactions. W. de
    Gruyter: Berlin; New York, 1984; p xiii, 613 p.
72. Grunert, M.; Winter, W.T., Progress in the development of cellulose-reinforced
    nanocomposites. Abstracts of Papers of the American Chemical Society 2000, 219,
    U484–U484.
73. Grunert, M.; Winter, W.T., Nanocomposites of cellulose acetate butyrate rein-
    forced with cellulose nanocrystals. Journal of Polymers and the Environment 2002,
    10(1–2), 27–30.
74. Orts, W.J.; Shey, J.; Imam, S.H.; Glenn, G.M.; Guttman, M.E.; Revol, J.F., Appli-
    cation of cellulose microfibrils in polymer nanocomposites. Journal of Polymers
    and the Environment 2005, 13(4), 301–6.
75. Park, W.-I.; Kang, M.; Kim, H.-S.; Jin, H.-K., Electrospinning of poly(ethylene
    oxide) with bacterial cellulose whiskers. Macromolecular Symposia 2007, 249–50,
    289–94.
                                         e
76. Favier, V.; Canova, G.R.; Cavaill´ , J.Y.; Chanzy, H.; Dufresne, A.; Gauthier, C.,
    Nanocomposite materials from latex and cellulose whiskers. Polymers for Advanced
    Technologies 1995, 6(5), 351–5.
                                        e
77. Favier, V.; Chanzy, H.; Cavaill´ , J. Y., Polymer nanocomposites reinforced by
    cellulose whiskers. Macromolecules 1995, 28(18), 6365–7.
78. Azizi Samir, M.A.S.; Alloin, F.; Dufresne, A., Review of recent research into
    cellulosic whiskers, their properties and their application in nanocomposite field.
    Biomacromolecules 2005, 6(2), 612–26.
79. Nishino, T.; Takano, K.; Nakamae, K., Elastic-modulus of the crystalline regions of
    cellulose polymorphs. Journal of Polymer Science Part B-Polymer Physics 1995,
    33(11), 1647–51.
                                                      10
     Cellulose Nanocrystals in Polymer
                 Matrices

                                    John Simonsen and Youssef Habibi



10.1     Introduction

Polymer composites have been with us since Leo Baekeland put wood flour in phenol-
formaldehyde resin and invented ‘Bakelite’ in 1907 (Rosen 1993). Since then we have
seen many improvements in composite technology, and the latest in this string of develop-
ments is polymer nanocomposites. While polymer nanocomposites are not new, e.g. car-
bon black in automobile tires is a nanocomposite and has been with us for approximately
100 years (Kohjiya et al. 2005), they have received increasing attention from researchers
over the past ∼20 years and many new developments have resulted (Koo 2006).
   While the use of cellulose nanocrystals (CNXLs) has been studied by a number of
researchers, mostly from the natural products community, for the past several years,
CNXL-filled polymer composites still represent only a tiny fraction of the research effort
being expended in the general area of polymer nanocomposites. This research has been
recently reviewed by several authors (Azizi Samir et al. 2005a; Dufresne 2006; Kamel
2007; Dufresne 2008). In this chapter, we will focus primarily on summarizing the
results obtained to date on a few polymer systems, carboxymethyl cellulose, poly(vinyl
alcohol) and polysulfone, with an added look at transport properties.


10.2     Background on CNXL Material Science

Cellulose is the largest volume polymer on earth. It is contained in virtually all plants and
is produced by certain bacteria and small sea animals. Regardless of its source, cellulose
is a semi-crystalline high molecular weight homopolymer of β-1,4 linked anhydroglucose

The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
274        The Nanoscience and Technology of Renewable Biomaterials

                   OH                                                                 OH

               4                                                                  4                                      3
                        5       O
                                                      3           OH 1                     5       O                             2   OH 1
                                                              2                                                                             O
      O                                      HO                          O                     2                HO
                            2                             5                                                                  5
                    3                1            4                                    3                1            4
          HO                             O                                   HO                             O
                                                                  O                                                                  O
                                OH                                                                 OH
                                                                  OH                                                                 OH



                                         Figure 10.1 Chemical structure of cellulose.

(Figure 10.1) (Fengel and Wegener 1983). The polymer chains are arranged in an
hierarchical order from elementary fibrils of cross dimension 2–5 nm in plant celluloses
(Hon and Shiraishi 1991; Ding and Himmel 2006).
   In the plant cell walls, the cellulose microfibrils result from the combined action
of biopolymerization spinning and crystallization. All these events are orchestrated by
specific enzymatic terminal complexes (TC) that act as biological spinnerets. If the TCs
are not perturbed, they can generate endless microfibrils having only a limited number
of defects or amorphous regions (Brown 1996; Brown 2004). These regions are located
on segments of the elementary fibril which are distorted by internal strain in the fiber to
undergo tilt and twist (Figure 10.2) (Rowland and Roberts 1972).
   After an acid treatment that hydrolyzes the cellulose and consequently cuts the
microfibrils at each defect, true cellulose rod-like nanocrystals are obtained that have a
morphology and crystallinity similar to the original cellulose fibers. Acids preferentially




                                                                                                            C




                                                                              C




                                                                  B                                                                  C




                                     A

Figure 10.2 Schematic representation of the elementary fibril illustrating the microstructure
of the elementary fibril and strain-distorted tilt and twist regions (defects) (Rowland and
Roberts 1972, reprinted with permission of John Wiley & Sons, Inc.).
                                          Cellulose Nanocrystals in Polymer Matrices      275


    (a)                                         (b)




    (c)                                         (d)




   (e)                                          (f)




Figure 10.3 TEM images of dried dispersion of CNXLs derived from (a) tunicate
(Elazzouzi-Hafraoui et al. 2008, Copyright (2008), American Chemical Society), (b) bac-
terial (Grunnert and Winter 2002, Copyright (2004), American Chemical Society), (c) ramie
(Habibi et al. 2008, reproduced by permission of The Royal Society of Chemistry), (d) cotton
linter, (e) sisal (Garcia de Rodriguez et al. 2006, with kind permission from Springer Science
+ Business Media), and (f) microcrystalline cellulose (MCC) (Brown 2004, Copyright (2007),
with permission from Elsevier).

hydrolyze the amorphous regions (Ruiz et al. 2000; Angles and Dufresne 2001).
Examples of such elements are given in Figure 10.3.
  The acid hydrolysis of native cellulose induces a rapid decrease in its degree of
polymerization (DP) to the so-called level-off DP (LODP). The DP then decreases much
276   The Nanoscience and Technology of Renewable Biomaterials

more slowly, even during prolonged hydrolysis. The value of LODP has been shown
to depend on the cellulose origin: Typical DP values of 250 have been recorded for
hydrolyzed cotton (Battista 1950) 300 for ramie fibers (Nishiyama et al. 2003) 140–200
for bleached wood pulp (Battista et al. 1956) and up to 6000 for the highly crystalline
Valonia cellulose (Kai 1976). However, a wide distribution in DP is typically seen for
all cellulose sources, even at the LODP.
   Sulfuric acid, hydrochloric acid and phosphoric acid have been used for CNXL prepa-
ration. If the CNXLs are by hydrochloric acid hydrolysis, the resulting dispersability is
limited and their aqueous suspensions tend to flocculate (Araki et al. 1998). On the other
hand, when sulfuric acid is used as hydrolyzing agent, it reacts with the surface hydroxyl
groups of cellulose to yield charged surface sulfate esters that promote a spontaneous
dispersion of the whiskers in water (Revol et al. 1992). However, the introduction of
charged sulfate groups compromises the thermostability of the nanocrystals (Roman and
Winter 2004).
   Recently Habibi et al. performed TEMPO-mediated oxidation of CNXLs that were
obtained from HCl hydrolysis of CNXLs from tunicin to introduce negative charges on
their surface (Habibi et al. 2006). They showed that after hydrolysis and TEMPO-
mediated oxidation, the CNXLs kept their initial morphological integrity and native
crystallinity, but at their surface the hydroxymethyl groups were selectively converted to
carboxylic groups, thus imparting a negative surface charge to the whiskers. When dis-
persed in water these oxidized CNXLs did not flocculate, and their suspensions appeared
birefringent.
   The geometrical dimensions (width l and length L) of the CNXLs are found to vary
with the cellulosic source material, the conditions of hydrolysis and with the degree of
agglomeration. Moreover, a wide distribution of CNXL size, especially the length, is
inevitable owing to the diffusion-controlled nature of the acid hydrolysis. This hetero-
geneity can be reduced by adding filtration and/or ultracentrifugation (using a saccharose
gradient) steps to the preparation process (de Souza Lima and Borsali 2002). The typi-
cal geometrical characteristics for CNXLs from various cellulose sources are shown in
Table 10.1. The width is generally a few nanometers, but the length ranges from tenths
of nanometers to several micrometers. In fact, there is a direct correspondence between
the length of the CNXLs and the LODP of the corresponding material, as it is generally
accepted that these whiskers consist of fully extended cellulose chain segments in an
almost perfect crystalline arrangement.
   Nanocrystalline cellulose from wood is 3–5 nm in width and 100–200 nm long; from
Valonia, a sea plant, 20 nm in width and 1000–2000 nm long; from cotton, 5–10 nm in
width and 100–300 nm long; from tunicata (or Urochordata), a sea animal, ∼10–20 nm
in width and 500–2000 nm long (Angles and Dufresne 2001). The aspect ratio, defined
as the ratio of the length to the width (L/l), varies between 10 and 30 for cotton and up
to ∼70 for tunicin.
   The morphology of the CNXL cross-section also depends on the origin of the cellulose.
Based on TEM observations, Revol reported that the cross-section of cellulose crystallites
in Valonia ventricosa was almost square, with an average side of 18 nm (Revol 1982).
CNXLs from tunicin were studied by small angle light scattering and found to have a
rectangular 8.8 × 18.2 nm2 cross-sectional shape (Terech et al. 1999). Ding et al. have
recently suggested a hexagonal shape (Ding and Himmel 2006).
                                             Cellulose Nanocrystals in Polymer Matrices         277

Table 10.1 Length (L) and cross section (l) of cellulose nanocrystals from various sources.
Source               L (nm)                l (nm)           Reference
Bacterial           100–1000s         5–10 × 30–50          Araki and Kuga 2001; Grunnert and
                                                              Winter 2002; Roman and Winter
                                                              2004
Cotton              100–300                5–10             Dong et al. 1998; Podsiadlo et al.
                                                             2005
Cotton linter       100–200              15–35              Elazzouzi-Hafraoui et al. 2008
MCC                 150–300                 3–7             Bondeson et al. 2006a, 2006b;
                                                              Elazzouzi-Hafraoui et al.
                                                              2008
Ramie               200–400                8–10             Hanley et al. 1992; Habibi et al.
                                                             2007
Sisal               100–500                 3–5             Garcia de Rodriguez et al. 2006
Tunicin             100–1000s            10–20              Habibi et al. 2007;
                                                             Elazzouzi-Hafraoui et al. 2008
Valonia              >1000               10–20              Revol 1982; Hanley et al. 1992
Wood                100–300                3–15             Araki et al. 1998, 1999;
                                                              Beck-Candanedo et al. 2005




10.3 Polymer Nanocomposite Systems

Cellulose nanocrystals have received attention as reinforcing material in nanocomposites
due to their low cost, high availability, natural renewability, nanoscale dimensions, high
surface area, unique morphology, ease of chemical modification and low density, as well
as their good mechanical response to stress.
  The density of CNXLs calculated from X-ray diffraction data is 1.566 g/cc (Battista
1975) while the density of pure crystalline cellulose Iβ is 1.61 g/cc (Nishiyama et al.
2002). The mechanical performance of CNXLs compare well with other materials (see
Table 10.2).

Table 10.2 Strength and stiffness of CNXLs compared to other materials. (Jones 1975).
Material                Tensile strength          Modulus      Reference
                             (GPa)                 (GPa)
Cellulose crystal              7.5                  145        Marks 1967; Eichhorn et al. 2005
Glass fiber                     4.8                   86        -
Steel wire                     4.1                  207        -
Graphite whisker               21                   410        -
Carbon nanotubes              11–63               270–970      Yu et al. 2000
278    The Nanoscience and Technology of Renewable Biomaterials

   CNXLs have been incorporated into many polymers, including siloxanes (Grunnert
and Winter 2000), poly(caprolactone) (Morin and Dufresne 2002; Habibi and Dufresne
2008; Habibi et al. 2008), glycerol-plasticized starch (Angles and Dufresne 2001),
styrene-butyl acrylate latex (Paillet and Dufresne 2001), poly-(styrene-co-butyl acrylate)
(poly(S-co-BuA)) (Favier et al. 1995b), cellulose acetate butyrate (Grunnert and
Winter 2002), poly(vinyl acetate) (Roohani et al. 2008; Shanmuganathan et al. 2008),
poly(vinyl alcohol)/carboxymethyl cellulose blends (He et al. 2008), epoxies (Ruiz
et al. 2000) phenol-formaldehyde (Hong et al. 2008), polypropylene (Bonini 2000),
poly(vinyl chloride) (Chazeau et al. 1999a; Chazeau et al. 1999b), and thermoplastic
starch (Orts et al. 2004). Note that most of the matrices listed above are thermoplastics,
which can compensate for the lack of CNXLs ductility with only 2% extension at break
(Marks 1967). In none of these cases were the very properties of neat CNXLs obtained
(Chazeau et al. 1999b).
   CNXLs have not yet been used extensively in the common thermoplastics, e.g.
polyethylene and polypropylene, as they are thermally sensitive at the temperatures
commonly used to extrude them.


10.4   Thermal Properties

From most of the studies in this field, surprisingly, the addition of CNXLs into poly-
mers matrices seems not to affect the values of the glass-rubber transition temperature
Tg , regardless of the nature of the host polymer, or the origin of the CNXLs or the
processing conditions (Azizi Samir et al. 2005a; Dufresne 2008). This observation is
unexpected if one considers the high specific area of CNXLs. There are a few cases
reported in the literature where the addition of CNXL as a filler in composite materi-
als affects the Tg but only slightly. This unusual effect, which showed especially in a
moisture sensitive system (Roohani et al. 2008) was related to the plasticization effect
of water and is also linked to the strong interaction between CNXLs and the respective
matrix.
   In the case of semi-crystalline polymers, it was reported that the addition of unmodi-
fied CNXLs had no influence on the melting temperature (Tm ) of the nanocomposites in
plasticized starch (Angles and Dufresne 2000; Mathew and Dufresne 2002), PEO (Azizi
Samir et al. 2004a, 2005b), CAB (Grunnert and Winter 2002), and PCL-reinforced
polymers (Habibi and Dufresne 2008; Habibi et al. 2008). However, when chemi-
cally modified CNXLs were used in nanocomposites, a change of Tm was observed.
Strong interactions between chemically modified CNXLs and matrices were reported
to be the origin of this Tm change. Moreover, CNXLs can act as a nucleating agent
in the semi-crystalline polymers which significantly increases the crystallinity of such
nanocomposites (Ljungberg et al. 2006; Habibi and Dufresne 2008). This effect is
mainly governed by CNXL-matrix compatibility which depends on surface chemistry
considerations. Finally, transcrystallization phenomenon has also been reported in a
CNXL-filled polypropylene nanocomposite (Gray 2008).
                                           Cellulose Nanocrystals in Polymer Matrices      279

10.5 Mechanical Properties

According to the estimated modulus of the native cellulose perfect crystal which is around
150 GPa (Eichhorn et al. 2005; Sturcova et al. 2005), the effect of nanoparticles on the
nanocomposite mechanical properties exceeds conventional predictions. For example,
Favier et al. showed the reinforcing effect of CNXLs for poly(S-co-BuA) loaded with
CNXLs from tunicate (Favier et al. 1995a; Favier et al. 1995b). Their system showed
a spectacular improvement in the storage modulus, measured by DMA, above the
glass-rubber transition temperature range, even at low CNXL content (Figure 10.4).
   The authors demonstrated that this unusual effect is due, in part, to the formation of
a rigid percolating filler network. The percolation phenomenon, which is well known
in the case of electrical conductivity (Stauffer and Aharony 1992), may also alter the

                                  10

                                   9
                                                                 14 %
                                   8                              6%
                      Log (G)




                                                                  3%
                                   7
                                                                  1%
                                   6

                                   5                              0%

                                   4
                                   200   250              300         350
                                               Temp (K)
                                                 (a)

                                  10

                                   9

                                   8                             6%
                        Log (G)




                                   7

                                   6

                                   5
                                                 0%
                                   4
                                   200   300           400            500
                                               Temp (K)
                                                 (b)

Figure 10.4 Logarithm of storage shear modulus vs temperature for poly(S-co-BuA) nanocom-
posite reinforced by weight fractions of tunicin CNXLs from 0 to 14% for temperature range
from 200 to 350 K (a) or up to 500 K (b). Reprinted with permission from Favier et al. (1995b),
Copyright (1995), American Chemical Society.
280   The Nanoscience and Technology of Renewable Biomaterials

mechanical and transport properties of composites (Surve et al. 2006b). The affect of
a percolating network on the mechanical properties depends on a number of variables,
primarily the size (and aspect ratio) of the particles making the dispersed phase (Garboczi
et al. 1995; Surve et al. 2006b), the interaction energy between particles relative to the
matrix (and relative to kT) (Prasad et al. 2003), and the volume fraction (Surve et al.
2006a).
   The high reinforcing effect of CNXL-reinforced nanocomposites has been well
predicted by applying the percolation concept to the classical phenomenological
series-parallel model of Takayanagi et al. (1964). In this model, all the interactions,
including matrix-matrix, matrix-filler and filler-filler interactions, that hold the perco-
lating CNXLs network are considered. The use of this model to CNXLs-containing
composites and details of the calculation are reported by Favier et al. (1997). In this
approach, the elastic tensile modulus Ec of the composite is given by the following
equation:
                            (1 − 2ψ + ψvR )ES ER + (1 − vR )ψER      2
                      Ec =
                                    (1 − vR )ER + (vR − ψ)ES

where the subscripts S and R refer to the soft and rigid phase, respectively, i.e. polymeric
matrix and filler. ψ and ER correspond to the volume fraction and modulus of the stiff
percolating network, respectively. ψ can be written as:

                 ψ =0                                      for VR < VRc
                                        b
                            VR − VRc
                 ψ = VR                                    for VR ≥ VRc
                             1 − VRc

where vR and vRc correspond to the volume fraction of the filler and the critical volume
fraction at the percolation threshold, respectively and b is the corresponding critical
exponent which is 0.4 in a three-dimensional network.
   The model assumes the formation of an infinite network of cellulose whiskers and
this gives rise to unexpectedly large composite stiffness.
   The effect of surface chemistry can also be important. For example, as the size of the
filler becomes similar to that of the polymer molecule in the matrix, polymer bridging can
occur. These bridges, depending upon the attractive forces between matrix and filler, can
have a dramatic effect on the ability of the percolating structure to withstand imposed
stresses (Surve et al. 2006b; Surve et al. 2006a). This effect is especially apparent
in the rubbery phase of the polymer matrix where the modulus can rise by orders of
magnitude at filler volumes less than 10% (Azizi Samir et al. 2004b). However, these
large effects have generally only been reported in very soft polymer matrices (Azizi Samir
et al. 2004b). The percolation effect is thus dependent upon the interaction between
matrix and filler, and can therefore be modified by altering the surface chemistry of the
nanoparticles (Zhang and Archer 2002).
   Due to the hydrophilic character of CNXLs, the simplest polymer systems to incor-
porate CNXLs are water-borne systems. In this case films can be formed via solution
casting, i.e. simply allowing the water to evaporate. However, these systems suffer
from limited utility and are only appropriate for niche markets where susceptibility to
moisture is not an issue.
                                                                           Cellulose Nanocrystals in Polymer Matrices                                 281

   An example of this type of system is carboxymethyl cellulose (CMC) (Choi and
Simonsen 2006), in which glycerine is added as a plasticizer at a concentration of 10%.
The addition of 5% CNXLs results in an improvement in mechanical properties, including
elongation at break (Figure 10.5). The resulting nanocomposite is not only stronger and
stiffer, but tougher. It is known that the addition of fillers at low filler volumes can
increase the toughness, as has been reported, for example, for silica-filled polypropylene
(Wu et al. 2002).
   To overcome the issue of water susceptibility, crosslinking methods have been
employed. For CMC, in the salt form (which is the form produced industrially) the
preparation of the composite starts by converting the matrix to the acid form, by passing
a solution of sodium CMC, NaCMC, through a cation exchange column. The resulting
acid form can be confirmed by potentiometric titration (Figure 10.6).
   Conversion of the NaCMC to the acid form and subsequently heat treating the resulting
composite increases the tensile strength of the 5% filled material by a factor of 2.6
compared to the NaCMC sample. The elongation to break was reduced from 5.5% to
3.8% (Figure 10.7). This result is explained by the increased cross-linking in, which
produced an increased brittleness. The nanocomposite thus shows significantly improved
mechanical properties compared to the unfilled composite at the 5% filler level. However,
loadings greater than 5% showed no improvement or even a decrease in properties. This
is postulated to be the result of agglomeration.

                                                                                                         4


                            6
  Elongation at break (%)




                                                                                 Tensile Modulus (GPa)




                            4
                                                                                                         2




                            2
                                0   5   10     15               20   25   30                                 0    5    10     15    20      25   30
                                        CNXL Content (wt%)                                                             CNXL Content (wt%)
                                                           38

                                                           36

                                                           34
                                               UTS (MPa)




                                                           32

                                                           30

                                                           28

                                                           26
                                                                0    5    10     15                          20   25    30
                                                                          CNXL Content (wt%)

Figure 10.5 Mechanical properties of CNXL-filled NaCMC with no heat treatment. The error
bars represent ± one standard deviation of the data (Choi and Simonsen 2006).
282                                The Nanoscience and Technology of Renewable Biomaterials

                                                       70

                                                       60
                                                                                                                                              immersed 3 days
                                                       50                                                                                     immersed 1 day
                                    Dissolution (%)


                                                       40

                                                       30

                                                       20

                                                       10

                                                           0
                                                               20                        40        60                                80              100          120
                                                                                                     Temperature (°C)

Figure 10.6 The effect of heat treatment on 5% CNXL-filled CMC composites (Choi and
Simonsen 2006).

                          6                                                                                                        3.0
Elongation at break (%)




                                                                                                           Tensile Modulus (GPa)




                          5                                                                                                        2.5



                          4                                                                                                        2.0



                          3                                                                                                        1.5


                              20                      40             60     80       100           120                                   20     40      60     80       100   120
                                                                    Temperature (°C)                                                                   Temperature (°C)

                                                                                    95


                                                                                    90
                                                                        UTS (MPa)




                                                                                    85


                                                                                    80


                                                                                    75
                                                                                         20   40   60      80      100                           120
                                                                                                   Temperature (°C)

Figure 10.7 Mechanical properties of 5% CNXL-filled heat treated CMC. The error bars
represent ± one standard deviation of the data (Choi and Simonsen 2006).
                                                       Cellulose Nanocrystals in Polymer Matrices   283

                                       2.5


                                       2.0
               Tensile modulus (GPa)
                                       1.5


                                       1.0


                                       0.5


                                       0.0
                                             0   3            6           9           12
                                                     CNXL Content (wt%)

Figure 10.8 Tensile modulus of polysulfone film filled with CNXLs formed via phase
inversion. The films were ∼ 20 µm thick (Noorani et al. 2006).


   It has been observed an increase in modulus in the system polysulfone (PSf)-CNXL
(Figure 10.8) up to ∼7% CNXL weight %. At higher concentrations the modulus
decreased. It was concluded that this was due to agglomeration of the CNXLs, as
confirmed by SEM images of the fracture surfaces of the films. Since these PSf/CNXL
films were prepared by phase inversion, i.e. taking a PSf solution dissolved in N-methyl
pyrrolidone (NMP) and submerging it in warm water, the films are porous. However,
the pores were small since the resulting films were clear and as confimed by SEM
images. This explains the observed moduli being lower than that reported for pure PSf
(2.48 GPa) (Solvay 2006). However, the addition of CNXLs increased the modulus
significantly, with the modulus of the 7%-load sample 2.8 times that of the unfilled
PSf. The large variability in the data is typical for phase inversion samples fabricated
in a laboratory. SEM indicated that the size of the pores did not change greatly with
CNXL content, although this may also be a source of variability in the samples.


10.6   Transport Properties

In addition to the reinforcing effect, it is well known that CNXLs affect the performance
of nanocomposites, from example, by improving the barrier properties and thus the mem-
brane performance, depending upon the filler particle type and its interfacial interactions.
Indeed, the permeability has been seen to increase without sacrificing selectivity (Aerts
et al. 2000a; Aerts et al. 2000b; Merkel et al. 2002; Naidu et al. 2005; Sacca et al.
2005; Yan et al. 2005). The effect can be especially pronounced in the case of nanopar-
ticle fillers (Aerts et al. 2000a; Aerts et al. 2000b; Aminabhavi and Mallikarjuna 2004).
The mechanism of this effect of nano-sized filler particles is perhaps not entirely under-
stood, but it is generally accepted that the effect of the interphase between filler and
284    The Nanoscience and Technology of Renewable Biomaterials

matrix is important in this regard (Merkel et al. 2002; Liu and Kee 2005). A widely
held view, especially in gas separation membranes, is that the presence of the nano-sized
filler alters the free volume of the matrix polymer (the polymer depletion effect) and also
introduces nano-sized cracks which serve to increase permeability without sacrificing,
and sometimes enhancing, selectivity (Zhong et al. 2005). For example, the addition
of nano-sized silica (Aerosil) to polysulfone (PSf) has been shown to improve the per-
meability without sacrificing the selectivity (Aerts et al. 2000a; Aerts et al. 2000b).
However, of the most studied nano-fillers, namely, Aerosil (Aerts et al. 2000a; Aerts
et al. 2000b; Merkel et al. 2002), zeolite (Hennepe et al. 1987), Al2 O3 (Wara et al.
1995) or ZrO2 (Genne et al. 1996) none has a large aspect ratio. The nanoclays, e.g.
montmorrillonite, have been investigated for gas separation membranes. These materials
do have a high aspect ratio (when exfoliated) and it has been reported that increasing
the aspect ratio dramatically improved the gas transport properties (Aminabhavi and
Mallikarjuna 2004). Further, Karthikeyan et al. have shown that surface modification of
the nano-filler has an important effect on both permeability and selectivity (Karthikeyan
et al. 2005). However, the clay systems have generally been difficult to exfoliate, and it
has been hard to measure the extent of exfoliation and relate it to composite properties.
   In general, increasing the attractive forces between polymer and dispersed particles
improves the dispersion stability and increases polymer bridging. On the other hand,
decreasing the attractive force tends to create agglomeration of the particles. The force
between particles at their contact points (as opposed to that for the dispersed ones in the
matrix), can also serve to increase the strength of the percolating network, this may be
enhanced by a lowered attractive force between the matrix and filler (Capadona et al.
2007). This effect has been used to create a ‘switch’ for altering mechanical properties in
a CNXL-filled poly(ethylene oxide)-based copolymer nanocomposite system (Capadona
et al. 2008).
   When the forces between polymer and filler particle are not attractive, the polymer
depletion effect can increase permeability due to an increase in free volume in the
neighborhood of the nanoparticle. This can consequently cause reverse size selectivity
in gas separation membranes (Hill 2006a; Hill 2006b). The effect can be quite dramatic
(Figure 10.9). This is not a ‘nano’ but a quantum effect, but it only arises as the filler
particle size approaches that of the matrix polymer molecule.
   Thus, this is a complex system where the filler geometry and volume, mechanical prop-
erties of matrix and filler particles and the surface chemistries of both polymer/particle
and particle/particle interactions interplay to create the properties of the final composite.
This complex system may be further complicated by grafting specific compounds on the
CNXL surface. In this way the mechanical properties of the CNXL, which are much
greater than those of the polymer matrix, may be preserved while using the grafted
compound to control particle/particle and particle/polymer matrix interactions.
   The use of nanoparticles results in fillers with very large surface areas. The effect of
this surface area on the polymer matrix can be substantial. In fact, if we make some
simple assumptions (e.g. infinite perfect cylinders, uniformly distributed and aligned) and
calculate the amount of the matrix polymer which resides in the interphase, assuming the
interphase is 20 nm thick, a typical radius of gyration for the types of polymers of interest
here, we see that at relatively low filler volumes, substantial amounts of the polymer
matrix will be in the interphase (Figure 10.10). Matrix polymers in the interphase may
                                                                                                     Cellulose Nanocrystals in Polymer Matrices   285

                                                                                         2.5




                                                 Scaled effective diffusivity, D e/D ∞
                                                                                         2.0



                                                                                         1.5



                                                                                         1.0



                                                                                         0.5
                                                                                           100          101              102
                                                                                                 Scaled inclusion radius, a/x

Figure 10.9 The scaled effective diffusivity D e /D ∞ versus the scaled inclusion radius a/ξ
where D e = the composite diffusivity, D ∞ = the bulk diffusivity, a = the radius of the
nanoparticle inclusion, ξ = 0.8 nm, the thickness of the depleted zone around the inclusion.
The circles are experimental measurements of the permeability enhancement from Merkel
et al. (Merkel et al. 2002). Graph from Hill (Hill 2006b).


                                       100
                                       90
                                       80                                                                                        50 m2/g
                                                                                                                                 100 m2/g
              % matrix in interphase




                                       70                                                                                        200 m2/g
                                                                                                                                 300 m2/g
                                       60
                                       50
                                       40
                                       30
                                       20
                                       10
                                        0
                                             0                                             5          10           15           20          25
                                                                                                      % filler content

Figure 10.10 Simple estimate of the amount of the polymer matrix residing in the interphase
as a function of filler surface area and content.

have more or less free volume than the bulk matrix and may therefore exhibit different
transport properties.
   We have observed the dependence of water vapor transport on CNXL content in
different polymer systems: CMC (not shown here), PSf, and poly(vinyl alcohol) (PVOH)
(Figure 10.11). The CMC, filled with 5% CNXLs, showed very high water vapor
286    The Nanoscience and Technology of Renewable Biomaterials

                       1000                                         PVOH, 0 % PAA
                                                                    PVOH, 10 % PAA
                                                                    PSf


                        750
       WVTR, g/m2 dy




                        500




                        250




                         0
                              0   5            10              15            20
                                          CNXL content, wt %

Figure 10.11 Water vapor transport rate (WVTR) of various polymer systems filled with
CNXLs.


transport rate (WVTR), around 1500 (g/m2 dy), which can be explained by the presence
of 10% glycerin as a plasticizer.
   The PVOH system was prepared by mixing aqueous solutions of PVOH, and
poly(acrylic acid) (PAA) with aqueous dispersions of CNXLs. A subsequent heat
treatment of 170 ◦ C for 45 minutes was shown to crosslink the polymers and probably
the CNXLs, resulting in greatly reduced swelling and weight loss upon submersion.
The crosslinking mechanism was shown to be ester bond formation between the
hydroxyls of PVOH, and perhaps CNXLs, and the carboxylic acid groups of PAA.
SEM and optical microscope images suggested that the CNXLs were agglomerated at
filler contents of 15% and above in this system.
   It is interesting to note that CNXLs in a hydrophilic system (PVOH) reduce the
WVTR up to the agglomeration region while in a hydrophobic matrix, PSf, the WVTR
increases with increasing CNXL content. In the PVOH system, the CNXLs probably
act as barriers, creating a tortuous path for water vapor transport. This is a typical effect
and a standard technique for creating barrier films. The variability of the PSf WVTR at
11% CNXL was attributed to agglomeration. In the agglomerated state there might be
voids which would increase the WVTR.
   These data underscore the nanoscale effect in the transport properties of polymer films
and the importance of the surface chemistry and matrix/filler interactions. While these
data are not conclusive, they are highly suggestive. They indicate that close attention
must be given to the chemistry of a particular system. The PSf results in particular
suggest that we may be seeing surface diffusion. Here the water may be adsorbing to
                                        Cellulose Nanocrystals in Polymer Matrices    287

the surface of the CNXLs and diffusing along their length. With a percolating network
of CNXLs, this allows for a new conduit for moisture transport in the composite mem-
brane. If this effect can be confirmed, it will open up a new technology for improving
transport properties in polymer membranes. Grafting carefully designed constituents
on the CNXL surface could allow for the control of diffusing species and consequent
membrane performance.
   Nanoparticles in general and CNXLs in particular, due to their ease of manufacture
and straightforward chemical manipulation, hold the potential to radically transform
membrane transport properties, affecting a great many areas of technology and everyday
life.


References

Aerts, P., I. Genne, S. Kuypers, R. Leysen, I.F.J. Vankelecom and P.A. Jacobs (2000a).
  ‘Polysulfone-aerosil composite membranes Part 2. The influence of the addition of
  aerosil on the skin characteristics and membrane properties.’ Journal of Membrane
  Science 178(1–2): 1–11.
Aerts, P., E. Van Hoof, R. Leysen, I.F.J. Vankelecom and P.A. Jacobs (2000b).
  ‘Polysulfone-Aerosil composite membranes Part 1. The influence of the addition of
  Aerosil on the formation process and membrane morphology.’ Journal of Membrane
  Science 176(1): 63–73.
Aminabhavi, T.M. and N.N. Mallikarjuna (2004). ‘Polymeric membranes: Polymeric
  nanocomposites: Barrier properties and membrane applications.’ Polymer News 29(6):
  193–5.
Angles, M.N. and A. Dufresne (2000). ‘Plasticized starch/tunicin whiskers nanocom-
  posites. 1. Structural analysis.’ Macromolecules 33(22): 8344–53.
Angles, M.N. and A. Dufresne (2001). ‘Plasticized starch/tunicin whiskers nanocom-
  posite materials. 2. Mechanical behavior.’ Macromolecules 34: 2921–31.
Araki, J. and S. Kuga (2001). ‘Effect of trace electrolyte on liquid crystal type of
  cellulose microcrystals.’ Langmuir 17: 4493–6.
Araki, J., M. Wada, S. Kuga and T. Okano (1998). ‘Flow properties of microcrystalline
  cellulose suspension prepared by acid treatment of native cellulose.’ Colloids and
  Surfaces, A: Physicochemical and Engineering Aspects 142(1): 75–82.
Araki, J., M. Wada, S. Kuga and T. Okano (1999). ‘Influence of surface charge on
  viscosity behavior of cellulose microcrystal suspension.’ Journal of Wood Science
  45(3): 258–61.
Azizi Samir, M.A.S., F. Alloin and A. Dufresne (2005a). ‘Review of recent research
  into cellulosic whiskers, their properties and their application in nanocomposite field.’
  Biomacromolecules 6: 612–26.
Azizi Samir, M.A.S., F. Alloin, W. Gorecki, J.-Y. Sanchez and A. Dufresne (2004a).
  ‘Nanocomposite polymer electrolytes based on poly(oxyethylene) and cellulose
  nanocrystals.’ Journal of Physical Chemistry B 108(30): 10845–52.
Azizi Samir, M.A.S., F. Alloin, J.-Y. Sanchez and A. Dufresne (2005b). ‘Nanocomposite
  polymer electrolytes based on poly(oxyethylene) and cellulose whiskers.’ Polimeros:
  Ciencia e Tecnologia 15(2): 109–13.
288   The Nanoscience and Technology of Renewable Biomaterials

Azizi Samir, M.A.S., F. Alloin, J.-Y. Sanchez, N. El Kissi and A. Dufresne (2004b).
  ‘Preparation of cellulose whiskers reinforced nanocomposites from an organic medium
  suspension.’ Macromolecules 37(4): 1386–93.
Battista, O.A. (1950). ‘Hydrolysis and crystallization of cellulose.’ Industrial and Engi-
  neering Chemistry 42: 502–7.
Battista, O.A. (1975). Microcrystal Polymer Science. New York, NY, McGraw-Hill.
Battista, O.A., S. Coppick, J.A. Howsmon, F.F. Morehead and W.A. Sisson (1956).
  ‘Level-off degree of polymerization. Relation to polyphase structure of cellulose
  fibers.’ Industrial and Engineering Chemistry 48: 333–5.
Beck-Candanedo, S., M. Roman and D.G. Gray (2005). ‘Effect of reaction conditions
  on the properties and behavior of wood cellulose nanocrystal suspensions.’ Biomacro-
  molecules 6(2): 1048–54.
Bondeson, D., I. Kvien and K. Oksman (2006a). Strategies for preparation of cellulose
  whiskers from microcrystalline cellulose as reinforcement in nanocomposites. Cellu-
  lose Nanocomposites: Processing, Characterization, and Properties. 938: 10–25.
Bondeson, D., A. Mathew and K. Oksman (2006b). ‘Optimization of the isolation of
  nanocrystals from microcrystalline cellulose by acid hydrolysis.’ Cellulose 13(2):
  171–80.
                             ´             o
Bonini, C. (2000). Mise en evidence du rˆ le des interactions fibre/fibre et fibre/matrice
                             `
  dans des nanocomposites a renfort cellulosique et matrice apolaire (atactique et iso-
  tactique). Grenoble, France, Joseph Fourier University.
Brown, R.M.J. (1996). ‘The biosynthesis of cellulose.’ Journal of Macromolecular
  Science Pure and Applied Chemistry A33(10): 1345–73.
Brown, R.M.J. (2004). ‘Cellulose structure and biosynthesis: what is in store for the
  21st century?’ Journal Polymer Science, Part A: Polymer Chemistry 42(3): 487–95.
Capadona, J.R., K. Shanmuganathan, D.J. Tyler, S.J. Rowan and C. Weder (2008).
  ‘Stimuli-responsive polymer nanocomposites inspired by the sea cucumber dermis.’
  Science 319: 1370–4.
Capadona, J.R., O. Van Den Berg, L.A. Capadona, et al. (2007). ‘A versatile approach
  for the processing of polymer nanocomposites with self-assembled nanofibre tem-
  plates.’ Nature Nanotechnology 2(12): 765–9.
                          e
Chazeau, L., J.Y. Cavaill´ , G. Canova, R. Dendievel and B. Boutherin (1999a). ‘Vis-
  coelastic properties of plasticized PVC reinforced with cellulose whiskers.’ Journal
  of Applied Polymer Science 71(11): 1797–1808.
Chazeau, L., J.Y. Cavaille and P. Terech (1999b). ‘Mechanical behaviour above Tg
  of a plasticised PVC reinforced with cellulose whiskers; a SANS structural study.’
  Polymer 40: 5333–44.
Choi, Y. and J. Simonsen (2006). ‘Cellulose nanocrystal-filled carboxymethyl cellulose
  nanocomposites.’ Journal of Nanoscience and Nanotechnology 6(3): 633–9.
de Souza Lima, M.M. and R. Borsali (2002). ‘Static and dynamic light scattering from
  polyelectrolyte microcrystal cellulose.’ Langmuir 18: 992–6.
Ding, S.-Y. and M.E. Himmel (2006). ‘The maize primary cell wall microfibril: A new
  model derived from direct visualization.’ Journal of Agricultural and Food Chemistry
  54: 597–606.
                                        Cellulose Nanocrystals in Polymer Matrices    289

Dong, X.M., J.F. Revol and D.G. Gray (1998). ‘Effect of microcrystallite prepara-
  tion conditions on the formation of colloid crystals of cellulose.’ Cellulose 5(1):
  19–32.
Dufresne, A. (2006). ‘Comparing the mechanical properties of high performance polymer
  nanocomposites from biologicial sources.’ Journal of Nanoscience and Nanotechno-
  logy 6: 322–30.
Dufresne, A. (2008). ‘Polysaccharide nano crystal reinforced nanocomposites.’ Cana-
  dian Journal of Chemistry 86(6): 484–94.
Eichhorn, S.J., R.J. Young and G.R. Davies (2005). ‘Modeling crystal and molecular
  deformation in regenerated cellulose fibers.’ Biomacromolecules 6: 507–13.
Elazzouzi-Hafraoui, S., Y. Nishiyama, J.-L. Putaux, L. Heux, F. Dubreuil and C. Rochas
  (2008). ‘The shape and size distribution of crystalline nanoparticles prepared by acid
  hydrolysis of native cellulose.’ Biomacromolecules 9(1): 57–65.
Favier, V., G.R. Canova, J.Y. Cavaille, H. Chanzy, A. Dufreshne and C. Gauthier
  (1995a). ‘Nanocomposite materials from latex and cellulose whiskers.’ Polymers for
  Advanced Technologies 6(5): 351–5.
                                                           e
Favier, V., G.R. Canova, C. Shrivastavas and J.Y. Cavaill´ (1997). ‘Mechanical percola-
  tion in cellulose whisker nanocomposites.’ Polymer Engineering and Science 37(10):
  1732–9.
Favier, V., H. Chanzy and J.Y. Cavaille (1995b). ‘Polymer nanocomposites reinforced
  by cellulose whiskers.’ Macromolecules 28(18): 6365–7.
Fengel, D. and G. Wegener (1983). Wood, Chemistry, Ultrastucture, Reactions. New
  York, Walter de Gruyter.
Garboczi, E.J., K.A. Snyder and J.F. Douglas (1995). ‘Geometrical percolation threshold
  of overlapping ellipsoids.’ Physical Review E: Statistical Physics, Plasmas, Fluids, and
  Related Interdisciplinary Topics 52(1): 819–28.
Garcia de Rodriguez, N.L., W. Thielemans and A. Dufresne (2006). ‘Sisal cellulose
  whiskers reinforced polyvinyl acetate nanocomposites.’ Cellulose 13(3): 261–70.
Genne, I., S. Kuypers and R. Leysen (1996). ‘Effect of addition of ZrO2 to polysulfone
  based UF membranes.’ Journal of Membrane Science 113: 343–50.
Gray, D.G. (2008). ‘Transcrystallization of polypropylene at cellulose nanocrystal sur-
  faces.’ Cellulose 15: 297–301.
Grunnert, M. and W.T. Winter (2000). ‘Progress in the development of cellulose rein-
  forced nanocomposites.’ Polymeric Materials Science and Engineering 82: 232.
Grunnert, M. and W.T. Winter (2002). ‘Nanocomposites of cellulose acetate butyrate
  reinforced with cellulose nanocrystals.’ Journal of Polymers and the Environment
  10(1/2): 27–30.
Habibi, Y., H. Chanzy and M.R. Vignon (2006). ‘TEMPO-mediated surface oxidation
  of cellulose whiskers.’ Cellulose 13: 679–87.
Habibi, Y. and A. Dufresne (2008). ‘Highly filled bionanocomposites from functional-
  ized polysaccharide nanocrystals.’ Biomacromolecules 9(7): 974–80.
                                    e e
Habibi, Y., L. Foulon, V. Agui´ -B´ ghin, M. Molinari and R. Douillard (2007).
  ‘Langmuir–Blodgett films of cellulose nanocrystals: Preparation and characterization.’
  Journal of Colloid and Interface Science 316: 388–97.
290   The Nanoscience and Technology of Renewable Biomaterials

Habibi, Y., A.-L. Goffin, N. Schiltz, E. Duquesne, P. Dubois and A. Dufresne (2008).
  ‘Bionanocomposites based on poly(ε-caprolactone)-grafted cellulose nanocrystals by
  ring opening polymerization.’ Journal of Materials Chemistry 18(41): 5002–10.
Hanley, S.J., J. Giasson, J.F. Revol and D.G. Gray (1992). ‘Atomic force microscopy of
  cellulose microfibrils – comparison with transmission electron-microscopy.’ Polymer
  33(21): 4639–42.
He, G., H. Zheng, F. Xiong and R. Zhao (2008). Preparation and characterization
  of physically crosslinked poly(vinyl alcohol)/carboxymethyl cellulose hydrogels.
  Abstracts of Papers, 235th ACS National Meeting, New Orleans, LA.
Hennepe, H.J.C., D. Bargeman, M.H.V. Mulder and C.A. Smolders (1987).
  ‘Zeolite-filled silicone rubber membranes, Part 1. Membrane preparation and
  pervaporation results.’ Journal of Membrane Science 35(1): 39–55.
Hill, R.J. (2006a). ‘Diffusive permeability and selectivity of nanocomposite membranes.’
  Industrial & Engineering Chemistry Research 45(21): 6890–8.
Hill, R.J. (2006b). ‘Reverse-selective diffusion in nanocomposite membranes.’ Physical
  Review Letters 96(21): 216001/1–216001/4.
Hon, D.N.-S. and N. Shiraishi (1991). Wood and Cellulosic Chemistry. New York,
  Marcel Dekker, Inc.
Hong, J.K., C.E. Frazier and M. Roman (2008). Cellulose nanocrystals as additives for
  wood adhesives. Abstracts of Papers, 235th ACS National Meeting, New Orleans,
  LA, American Chemical Society.
Jones, R.M. (1975). Mechanics of Composite Materials. New York, NY, Hemisphere
  Publishing Corp.
Kai, A. (1976). ‘The fine structure of Valonia microfibril. Gel permeation chromato-
  graphic studies of Valonia cellulose.’ Sen-i Gakkaishi 32: T326 – 34.
Kamel, S. (2007). ‘Nanotechnology and its applications in lignocellulosic composites, a
  mini review.’ eXPRESS Polymer Letters 1(9): 546–75.
Karthikeyan, C.S., S.P. Nunes, L.A.S.A. Prado, et al. (2005). ‘Polymer nanocomposite
  membranes for DMFC application.’ Journal of Membrane Science 254(1–2): 139–46.
Kohjiya, S., A. Katoh, J. Shimanuki, T. Hasegawa and Y. Ikeda (2005). ‘Nano-structural
  observation of carbon black dispersion in natural rubber matrix by three-dimensional
  transmission electron microscopy.’ Journal of Materials Science 40: 2553–5.
Koo, J.H. (2006). Polymer Nanocomposites, Processing, Characterization and Applica-
  tions. New York, NY, McGraw-Hill.
Liu, Q. and D.D. Kee (2005). ‘Modeling of diffusion through nanocomposite mem-
  branes.’ Journal of Non-Newtonian Fluid Mechanics 131: 32–43.
                                e
Ljungberg, N., J.-Y. Cavaill´ and L. Heux (2006). ‘Nanocomposites of isotactic
  polypropylene reinforced with rod-like cellulose whiskers.’           Polymer 47(18):
  6285–92.
Marks, R.E. (1967). Cell Wall Mechanics of Tracheids. New Haven, CT, Yale Univ.
  Press.
Mathew, A.P. and A. Dufresne (2002). ‘Morphological investigation of nanocompos-
  ites from sorbitol plasticized starch and tunicin whiskers.’ Biomacromolecules 3(3):
  609–17.
Merkel, T.C., B.D. Freeman, R.J. Spontak, et al.                (2002).  ‘Ultrapermeable
  reverse-selective nanocomposite membranes.’ Science 296: 519–22.
                                        Cellulose Nanocrystals in Polymer Matrices    291

Morin, A. and A. Dufresne (2002). ‘Nanocomposites of chitin whiskers from Riftia
  tubes and poly(caprolactone).’ Macromolecules 35: 2190–9.
Naidu, B.V.K., M. Sairam, K.V.S.N. Raju and T.M. Aminabhavi (2005). ‘Pervaporation
  separation of water + isopropanol mixtures using novel nanocomposite membranes
  of poly(vinyl alcohol) and polyaniline.’ Journal of Membrane Science 260(1-2):
  142–55.
Nishiyama, Y., U.J. Kim, D.Y. Kim, K.S. Katsumata, R.P. May and P. Langan (2003).
  ‘Periodic disorder along ramie cellulose microfibrils.’ Biomacromolecules 4(4):
  1013–17.
Nishiyama, Y., P. Langan and H. Chanzy (2002). ‘Crystal structure and hydrogen-
  bonding system in cellulose ib from synchrotron x-ray and neutron fiber diffraction.’
  Journal of the American Chemical Society 124: 9074–82.
Noorani, S., J. Simonsen and S. Atre (2006). Polysulfone-cellulose nanocomposites.
  ACS Symposium Series. Cellulose Nanocomposites: Processing, Characterization
  and Properties. K. Oksman and M. Sain. Washington, D.C., American Chemical
  Society. 938.
Orts, W.J., S.H. Imam, J. Shey, et al. (2004). ‘Effect of fiber source on cellulose rein-
  forced polymer nanocomposites.’ Annual Technical Conference - Society of Plastics
  Engineers 62nd(Vol. 2): 2427–31.
Paillet, M. and A. Dufresne (2001). ‘Chitin whisker reinforced thermoplastic nanocom-
  posites.’ Macromolecules 34(19): 6527–30.
Podsiadlo, P., S.-Y. Choi, B. Shim, J. Lee, M. Cuddihy and N.A. Kotov (2005). ‘Molec-
  ularly engineered nanocomposites: Layer-by-layer assembly of cellulose nanocrystals.’
  Biomacromolecules 6(6): 2914–18.
Prasad, V., V. Trappe, A.D. Dinsmore, P.N. Segre, L. Cipellettie and D.A. Weitz (2003).
  ‘Universal features of the fluid to solid transition for attractive colloidal particles.’
  Faraday Discussions 123: 1–12.
Revol, J.F. (1982). ‘On the cross-sectional shape of cellulose crystallites in Valonia
  ventricosa.’ Carbohydrate Polymers 2(2): 123–34.
Revol, J.F., H. Bradford, J. Giasson, R.H. Marchessault and D.G. Gray (1992). ‘Heli-
  coidal self-ordering of cellulose microfibrils in aqueous suspension.’ International
  Journal of Biological Macromolecules 14: 170–2.
Roman, M. and W.T. Winter (2004). ‘Effect of sulfate groups from sulfuric acid hydrol-
  ysis on the thermal degradation behavior of bacterial cellulose.’ Biomacromolecules
  5(5): 1671–7.
Roohani, M., Y. Habibi, N.M. Belgacem, G. Ebrahim, A.N. Karimi and A. Dufresne
  (2008). ‘Cellulose whiskers reinforced polyvinyl alcohol copolymers nanocomposites.’
  European Polymer Journal 44(8): 2489–98.
Rosen, S.L. (1993). Fundamental Principles of Polymeric Materials. New York, NY,
  John Wiley & Sons, Inc.
Rowland, S.P. and E.J. Roberts (1972). ‘Nature of accessible surfaces in the microstruc-
  ture of cotton cellulose.’ Journal of Polymer Science, Part A-1: Polymer Chemistry
  10(8): 2447–61.
Ruiz, M.M., J.Y. Cavaille, A. Dufresne, J.F. Gerard and C. Graillat (2000). ‘Processing
  and characterization of new thermoset nanocomposites based on cellulose whiskers.’
  Composite Interfaces 7(2): 117–31.
292   The Nanoscience and Technology of Renewable Biomaterials

Sacca, A., A. Carbone, E. Passalacqua, et al. (2005). ‘Nafion-TiO2 hybrid membranes
  for medium temperature polymer electrolyte fuel cells (PEFCs).’ Journal of Power
  Sources 152: 16–21.
Shanmuganathan, K., J.R. Capadona, S. Rowan, D.J. Tyler and C. Weder (2008).
  Stimuli-responsive mechanically dynamic polymer nanocomposites. Abstracts of
  Papers, 235th ACS National Meeting, New Orleans, LA,, American Chemical
  Society.
Solvay (2006). ‘Udel polysulfone product data.’ 2006, from http://www.solvaymem
  branes.com/static/wma/pdf/1/2/6/P1700nt.pdf.
Stauffer, D. and A. Aharony (1992). Introduction to Percolation Theory London, Taylor
  & Francis.
Sturcova, A., G.R. Davies and S.J. Eichhorn (2005). ‘Elastic modulus and stress-transfer
  properties of tunicate cellulose whiskers.’ Biomacromolecules 6: 1055–61.
Surve, M., V. Pryamitsyn and V. Ganesan (2006a). ‘Polymer-bridged gels of nanopar-
  ticles in solutions of adsorbing polymers.’ Journal of Chemical Physics 125(6):
  064903/1–064903/12.
Surve, M., V. Pryamitsyn and V. Ganesan (2006b). ‘Universality in structure
  and elasticity of polymer-nanoparticle gels.’ Physical Review Letters 96(17):
  177805/1–177805/4.
Takayanagi, M., S. Uemura and S. Minami (1964). ‘Application of equivalent model
  method to dynamic rheo-optical properties of a crystalline polymer.’ Journal of Poly-
  mer Science No. 5(Pt. C): 113–22.
Terech, P., L. Chazeau and J.Y. Cavaille (1999). ‘A small-angle scattering study of
  cellulose whiskers in aqueous suspensions.’ Macromolecules 32(6): 1872–5.
Wara, N.M., L. Falter Francis and B.V. Velamakanni (1995). ‘Addition of alumina to
  cellulose acetate membranes.’ Journal of Membrane Science 104: 43–9.
Wu, C.L., M.Q. Zhang, M.Z. Rong and K. Friedrich (2002). ‘Tensile performance
  improvement of low nanoparticles filled-polypropylene composites.’ Composites Sci-
  ence and Technology 62: 1327–40.
Yan, L., Y.S. Li and C.B. Xiang (2005). ‘Preparation of poly(vinylidene fluoride)(pvdf)
  ultrafiltration membrane modified by nano-sized alumina (Al2 O3 ) and its antifouling
  research.’ Polymer 46(18): 7701–6.
Yu, M.F., O. Lourie, M.J. Dyer, K. Moloni, T.F. Kelly and R.S. Ruoff (2000). ‘Strength
  and breaking mechanism of multiwalled carbon nanotubes under tensile load.’ Science
  287: 637–40.
Zhang, Q. and L.A. Archer (2002). ‘Poly(ethylene oxide)/silica nanocomposites: struc-
  ture and rheology.’ Langmuir 18: 10435–42.
Zhong, J., G. Lin, W.-Y. Wen, A.A. Jones, S. Kelman and B.D. Freeman (2005).
  ‘Translation and rotation of penetrants in ultrapermeable nanocomposite membrane
  of poly(2,2-bis(trifluoromethyl)-4,5-difluoro-1,3-dioxole-co-tetrafluoroethylene) and
  fumed silica.’ Macromolecules 38(9): 3754–64.
                                                      11
        Development and Application
       of Naturally Renewable Scaffold
          Materials for Bone Tissue
                 Engineering

        Seth D. McCullen, Ariel D. Hanson, Lucian A. Lucia and Elizabeth G. Loboa



11.1     Introduction

The progression of regenerative medicine has largely been catapulted by the implemen-
tation of tissue engineering based therapies with the hope of providing a replacement
for organ transplantation. Tissue engineering therapies can be defined where a popula-
tion of progenitor or stem cells are directed by their surrounding milieu to differentiate
into a desired tissue. This differentiation process is regulated by both their physical
(e.g. scaffolds, mechanical loads) and chemical (e.g. inductive soluble factors) envi-
ronments. In practice, tissue engineering approaches include the assembly of cells on
a temporary scaffold resembling the tissue’s natural extracellular matrix in vivo. The
function of this scaffold is to provide the appropriate template for cellular organiza-
tion while maintaining necessary physical and mechanical properties for the seeded
cells to differentiate/mature. The scaffold also achieves the desired physical integrity
for a specific defect site by promoting the cells to deposit their own natural extracel-
lular matrix within the scaffold prior to implantation of the cell-seeded construct at
a tissue defect site. Of the four tissue types in the human body, connective tissue
has seen the most prolific advances in tissue engineering due to its primary function
being rooted in mechanical stability and ambulatory function (1, 2). Within the realm



The Nanoscience and Technology of Renewable Biomaterials Edited by Lucian A. Lucia and Orlando J. Rojas
c 2009 Blackwell Publishing, Ltd
294    The Nanoscience and Technology of Renewable Biomaterials

of connective tissue, bone tissue engineering has emerged with the most clinical suc-
cess (3, 4, 4–7, 7–9). Currently, bone is the most transplanted tissue, second only to
blood transfusions, with approximately 500,000 cases occurring annually in the United
States (10).
   Bone tissue engineering is aimed at developing implantable substitutes to replace the
use of autograft and allograft treatments. At present, autografts and allografts are most
commonly used for bone grafting. Autografts are ideal based on their high acceptance
rate within the body and ability to become integrated into the skeletal system by being
osteoinductive, osteoconductive, and having osteogenic properties (11). Osteoinductive
refers to the graft’s ability to attract surrounding mesenchymal stem cells into the area
of repair that can then become a source of osteoblasts, while osteoconductive refers to
the facilitation of vascularization and the orientation of haversian canal systems (10).
Osteogenic potential implies that osteoprogenitor cells are present in the graft itself (10).
However, autografts have also been associated with multiple problems including donor
site morbidity (12–14), chronic pain, nerve damage, infection, fracture, pelvic instabil-
ity, hematoma, and tumor transplantation (15). Allografts negate these concerns but
have their own limitations such as carrying the risk of causing an immune response
in the host, transferring diseases to the host (16), storing and transplanting of the allo-
graft, and/or weakening of the allograft’s biological and mechanical properties during
the storage and transplant process that would have made it an ideal replacement for bone
constructs (17). The limitations of autografts and allografts have led to the use of tissue
engineered constructs for bone grafts. Scaffolds are a key component for developing
a tissue engineered bone construct for implantation into a critical bone defect. Ideally,
a scaffold should have the following characteristics for successful implantation: (1) be
biocompatible and bioresorbable with a controlled degradation rate to match cell/tissue
growth in vivo; (2) have mechanical properties capable of withstanding the mechani-
cal loads experienced in the physiological environment during cell matrix maturation;
(3) be three-dimensional and allow for adequate diffusion for cell growth, nutrient deliv-
ery, and waste removal; and (4) have suitable surface chemistry for cell attachment,
proliferation, and differentiation (18). In order to accommodate all of these qualities,
a diverse portfolio of materials, fabrication techniques, and modifications have been
implemented over the years to achieve successful skeletal integration for use in clinical
applications.
   Recently, investigators have focused on the use of natural or renewable materials as
a scaffolding choice over synthetically derived options. The main driving force for the
use of naturally renewable materials is that these materials are highly biocompatible,
biodegradable, offer chemical functionality (which is desirable for cell processes such
as attachment, migration, and differentiation), and provide a cheap and replenishing
source of material. Thus, renewable scaffolding materials can be defined as materials
that can be obtained from natural resources including plant, fungal, animal, or bacterial
derivation. Typically, these materials are some form of secondary product and must
undergo some chemical treatment and sterilization process before end use. Two reviews
by Mano et al. and Malafaya et al. have addressed the overall status of these types of
materials in tissue engineering, hence, this chapter will specifically focus on their use in
bone tissue engineering applications with in vitro or in vivo examples (19, 20).
    Development and Application of Naturally Renewable Scaffold Materials for BTE     295

11.2   Natural Renewable Materials for Bone Tissue Engineering (BTE)

The primary role of the extracellular matrix (ECM) is to endow tissues with their spe-
cific mechanical and physicochemical properties while providing a platform for cell
attachment and migration. The ECM exerts a regulatory role in promoting or maintain-
ing cellular differentiation and phenotype expressions through its composition, structure,
and morphology. For bone tissue engineering, three groups of naturally renewable mate-
rials are commonly used: polysaccharides, fibrous proteins, inorganic materials, and any
combination thereof. Polysaccharides are composed of repeating sugar rings linked by
oxygen bonds, and in their natural state they function as membranes, participate in
cellular communication, and can act as sequestering agents. On the molecular level,
the tailoring of polysaccharides and their function can be controlled by their molecular
weight, stereochemistry, primary sequence, and chemical reactivity. Polysaccharides can
be derived from a number of resources with the most common forms including cellulose,
hyaluronan, chitosan, dextran, alginates, and starches, to name a few. The main differ-
ence between these materials is the location of the linking glucosidic bonds between
rings, the relative position of this linkage either being equatorial (β) or axial (α), and
the presence of different pendant groups on each ring. Polysaccharides can be classified
into four broad categories and include the ribbon, hollow helix, crumpled, and loosely
jointed families (21).
   Fibrous proteins are materials that are formed by repeating amino acid sequences and
possess four levels of organization. These materials are the major structural compo-
nents of tissues by providing high mechanical strength and resiliency. The mechanical
integrity of proteins is preserved by the various levels of organization of its molecular
and macroscopic arrangement which include its: (1) primary structure or the sequence
of amino acids, (2) secondary structure or conformation of the chain, (3) tertiary struc-
ture or polypeptide chain arrangement, and (4) quaternary structure or configuration
of multiple polypeptide chains. Fibrous proteins display one of the following confor-
mations or secondary structures: α-helix, β-sheet, triple helix, and random coil. The
most popular fibrous proteins used as scaffolding materials include collagen, silk, ker-
atin, and fibrin. Collagen is usually derived from mammalian sources, primarily from
bovine and human origin, and its functional unit is arranged in a triple helix where
three collagen molecules are intertwined. These molecules are known as tropocolla-
gen and are approximately 300 nm in length and 1.5 nm in diameter (19, 22). Type
I collagen is largely used in bone tissue engineering due to its natural occurrence and
large quantity in bone; thus numerous collagen based systems have been developed as
a starting point for bone tissue scaffolds. Silk is another fibrous protein that is pro-
duced by spiders and silkworms. This protein is composed of β-sheet structures that
allow the tight packing of stacked sheets of hydrogen bonded anti-parallel chains and
account for its high tensile modulus and elasticity (23, 24). Keratin is another protein
that displays either an α-helix or β-sheet structure (depending on source) along with
cysteine residues to create disulfide bridges for enhanced stability and strength (25).
Fibrin is the polymerized form of fibrinogen after it has been crosslinked with thrombin,
and is known for both its coagulation effects in blood and as an extracellular matrix
substitute (26).
296    The Nanoscience and Technology of Renewable Biomaterials

   Inorganic materials include demineralized bone matrix (DBM), hydroxyapaptite, nacre,
and coral. DBM can come from allograft and xenograft sources and is also known
as decalcified cortical bone. To reduce host rejection and foreign body response, it is
processed until it only retains a highly porous collagenous structure (10). Hydroxyapatite
is the natural inorganic component of bone and is typically incorporated as a filler
material in composite systems. Nacre is a calcified structure that forms the inner layer
of some sea shells. Coral is a marine invertebrate that consists of CaCO3 . Coral has
a porous structure with an interconnected network of pores. To be used in vivo it
undergoes a partial hydrothermal exchange process that converts carbonate to phosphate
(27). These materials resemble the natural architecture and porosity of bone and are
adequate scaffold materials based on this striking similarity.
   The remainder of this chapter will present a brief background on the anatomy and
function of bone, highlighting the extracellular matrix components, physical properties,
overall architecture, and the osteodifferentiation of progenitor cell populations featuring
mesenchymal stem cells. This will be followed by specific examples of investigators
implementing naturally renewable materials for bone tissue engineering, discussing the
successes and limitations with each example.


11.3   Bone Background

Bone is involved in many diverse roles within the body such as: (1) the protection of
vital organs, (2) support and attachment to muscles for locomotion, (3) the generation
of red and white blood cells for immunoprotection and oxygenation or other tissues,
and, (4) mineral storage and ion homeostasis (28–31). The architecture of bone is
representative of the many functions it serves in the body. There are two types of bone
that make up the adult skeleton, cortical bone (80%) and cancellous (or trabecular) bone
(20%) (30). Cortical bone provides mechanical stability and protection to vital organs
and is therefore almost completely solid, having a very low porosity (10%) (28, 30). In
comparison, trabecular bone is loosely organized and very porous (50–90%) in order to
provide a proper environment for metabolic activity (28, 31). In bone, entire collagen
triple helices lie in a parallel, staggered array. 40 nm gaps between the ends of the
tropocollagen subunits probably serve as nucleation sites for the deposition of long,
hard, fine crystals of the mineral component, which is (approximately) hydroxyapatite,
Ca5 (PO4 )3 (OH), with some phosphate (mineralization during endochondral ossification
of articular cartilage occurs in a similar fashion). Collagen gives bone its elasticity and
contributes to fracture resistance.
   Bone replacement has become an important area in tissue engineering. The previously
described limitations of autografts and allografts have led researchers to investigate the
use of natural scaffold materials, in combination with human mesenchymal stem cells
(hMSCs), to provide biocompatible, biodegradable, and mechanically stable bone grafts
for critical size defects. The success of bone grafts relies heavily on the architecture of
the scaffold, specifically the pore size and porosity, and should be designed to mimic
the physical properties of native bone (32, 33). For design purposes, the porosity of
native trabecular bone is estimated at >75% (34) and typical pore sizes are approxi-
mately 1 mm in diameter (35). Investigators have recently shown a direct correlation
    Development and Application of Naturally Renewable Scaffold Materials for BTE    297

between pore size, vascularization, and bone formation (36). Klenke et al. found that
scaffolds containing pores ≥140 mm had significantly higher ingrowth and bone for-
mation as compared to scaffolds with smaller pores. These findings confirmed results
from a separate study demonstrating a relationship between increasing pore size and
bone ingrowth, with optimal pore sizes for bone formation ≤350 mm (33). Scaffold
porosity also plays a key role in bone formation, as demonstrated by Takahashi et al.
who reported higher proliferation of mesenchymal stem cells (MSCs) when grown on
polyethylene terephthalate fabrics with higher porosities compared to those of lower
porosities (37). The increased cell proliferation was attributed to the increased volume
allowing for both greater cell migration and increased nutrient and oxygen delivery and
exchange.
   While large pore sizes and higher porosities have been shown to be beneficial for vas-
cularization and bone formation, they can also result in decreased compressive strength
of the scaffold which may then fail under physiological loading (4, 38). Trabecu-
lar bone is reported to have a compressive strength of 4–12 MPa and a modulus of
0.02–0.5 GPa (39). For successful repair of critical size bone defects, it is desirable
for the bioresorbable scaffold to have similar mechanical properties to the host tissue
and retain its physical properties for at least six months (four months in vitro during
cell culture and two months in vivo) (18). The strength and stiffness of the scaffold
should match that of the host tissue until new tissue has replaced the degrading scaffold
matrix.

11.3.1 Progenitor Cells for Tissue Engineering Bone
Mesenchymal stem cells (MSCs) are defined as progenitor cells that have the ability
to differentiate into tissues of a mesenchymal lineage such as bone, cartilage, adi-
pose, tendon, muscle, ligament and stroma (40). Investigators claim to have iso-
lated these cells from multiple sites including bone marrow (40, 41), umbilical cord
blood (42), peripheral blood (43), amniotic fluid (44, 45), and adipose tissue (46–49),
although recent studies have found that bone-marrow derived MSCs behave differently
than adipose-derived stem cells with respect to growth kinetics and differentiation effi-
ciency (50). Adipose-derived stem cells are known to contain a heterogeneous population
of cells. Research has indicated clear biologic distinctions between mesenchymal stem
cells derived from multiple tissues and noted site specific differences (51). Nonetheless,
MSCs offer tissue engineering a means to fully evaluate biomaterial interactions and
afford a cell line capable of undergoing osteogenesis. Typical characterization of mes-
enchymal stem cells consists of their expression of specific protein markers such as, but
not limited to, CD44, CD71, CD90, CD105, CD106, and CD166 (40, 52, 53), and the
ability of the MSCs to differentiate down osteogenic (41, 48, 54–62), adipogenic (41,
48, 63, 64), chondrogenic (3, 41, 48, 65–68), fibrogenic (69–71) myogenic (48, 57, 72),
and neuronal (41, 46, 57, 73) lineages. The majority of applications using stem cells for
tissue or organ replacement typically use bone marrow-derived mesenchymal stem cells
(BMMSCs) (55, 66, 70–72, 74–76). Researchers have also begun to extensively investi-
gate other sources of mesenchymal stem cells, including from adipose tissue. In contrast
to bone marrow, adipose tissue provides an abundant and easily obtainable source of
cells (77) adipose derived adult stem cells (ASCs) exhibit somewhat similar capacity
298   The Nanoscience and Technology of Renewable Biomaterials

for expansion, growth kinetics, and differentiation as BMMSCs depending on the source
and site of tissue (41).

11.3.2 Natural Renewable Materials Used for Bone Tissue Engineering
Tissue regeneration schemes have revolved around the use of progenitor cell populations
and bioactive molecules to catalyze neo-bone formation. Previously stated, this type of
therapy requires a scaffold material for cellular organization and to impart the correct
physiological matrix for success. The following section will address the use of naturally
renewable materials used for bone tissue engineering and focus on specific examples of
such scaffolding with progenitor and mesenchymal stem cell populations.

11.3.3 Naturally Occurring Polysaccharide Materials in BTE
Polysaccharide materials offer many benefits over other scaffolding materials, mainly
due to their abundance in nature. Chitosan, the deacetylated derivative of chitin, is
similar in structure to glycosaminoglycans within the mammalian extracellular matrix.
Chitosan is the treated form of chitin, the second most abundant polysaccharide obtained
from crustaceans’ exoskeletons, after is has been demineralized by HCl, deproteinized by
NaOH, and deacetylated by 50% or more. Besides being biocompatible, biodegradable,
and bioresorbable, chitosan exhibits a cationic nature and is hydrophilic, aiding in cellu-
lar processes. Chitosan can be formed into numerous structures based on its method of
preparation and includes porous spheres, films, fibers, or injectable solutions (78). Due
to its high molecular weight and electrolytic properties, chitosan is highly insoluble at
neutral pH and is usually dissolved in weak acids such as acetic acid. Based on chi-
tosan’s ability to form many different structures and possess a wide range of porosities,
much research has focused on its use in composite scaffold applications including com-
bination with natural and synthetic polymers, and inorganic materials (79–81). When
chitosan has been coupled with inorganic materials such as hydroxyapatite, investigators
have reported significant increases in osteogenic markers such as calcium deposition,
alkaline phosphatase (ALP) activity, and increased gene expression of bone sialoprotein,
osteopontin, and osteocalcin (82). Ge et al. investigated large weight percentages (25,
50, 75%) of hydroxyapatite in chitin films via lyophilization and were able to show
good biocompatibility with tissue ingrowth in a rabbit femur model after 2 months (83).
Histological analysis showed that scaffolds seeded with mesenchymal stem cells greatly
influenced tissue ingrowth compared to cell-free controls, and minimized inflammatory
response noted by the paucity of inflammatory cells (83). Work by Malafaya et al.
developed a unique approach to assemble micron sized chitosan particles into a macro-
scopic scaffold capable of filling a large bone defect void (84). Chitosan particles were
precipitated by drop-casting in a 1 M NaOH bath and thermally-pressing into the desired
mold shape. They reported that the scaffold was highly biocompatible and allowed cell
ingrowth into the porous structure (84). Gravel et al. experimented with a combination
of chitosan and coralline and determined that the composite allowed distinct cell mor-
phology displaying osteoblastic phenotypes for mesenchymal stem cells at higher ratios
of coralline:chitosan compared to pure chitosan alone (85). This was attributed to the
crystalline component and calcium content of the coralline.
    Development and Application of Naturally Renewable Scaffold Materials for BTE        299

   Hyaluronan is another polysaccharide that has made a large impact on bone tissue
engineering. Hyaluronan, also known as hyaluronic acid, is an anionic, nonsulfated,
high molecular weight glycosaminoglycan and was first isolated from the vitreous body
of the eye (22). It is a natural mucopolysaccharide that consists of alternating residues
of D-glucuronic acid and N-acetyl-D-glucosamine. Commercially, hyaluronan can be
derived from bacterium such as streptococcus zooepidium and extraneous bovine mate-
rials. Hyaluronan’s main function is to provide tissue hydration based on its hygroscopic
nature (22). It functions as the backbone of the proteoglycan aggregates necessary for
the integrity of articulating cartilage such as found in joints. Hyaluronan is preferentially
expressed by cells during wound healing to aid in cell migration and proliferation. Due
to the correlation of hyaluronan expression during wound healing, researchers have long
been supplementing bone defects with exogenous hyaluronan in combination with bone
graft materials. Hyaluronan and its derivatives have been used as topical, injectable, and
wholly implantable biomaterials for the delivery of bioactive compounds (86). In fact,
it is the orginal lastoviscous biomaterial for applications in eye surgery, bone surgery,
otology, plastic surgery, and rheumatology (87). From a biomechanical perspective,
by itself or with fibropectin, it may be a potentially optimal bioimplant for vocal fold
defects and scarring (88). Interestingly, like chitosan, it too demonstrates antibacterial
activity, especially when applied for guided tissue regeneration surgery (89). Addition-
ally, it has shown enhanced activity for the treatment of noninfected, mechanical corneal
lesions where the time for epithelial defect closure was significantly reduced compared
to nontreated corneas (90).
   The versatility of hyaluronic acid for biomedical applications is seemingly limitless,
especially with respect to its use for prodrugs, delivery vehicle, and tissue scaffold. In
the area of prodrugs, hyaluronic acid can be modified chemically to develop polymeric
structures for simple drug applications such as analgesics (91). When grafted with
poly(ethylene glycol) (PEG), it was possible to incorporate insulin preferentially into the
PEG phase of this copolymer to provide ‘leakage’ (92). Indeed, the release was solely
dependent upon the PEG content.
   In terms of delivery vehicle, it has been used as an osteogenic or chondrogenic delivery
vehicle upon a similar modification (to PEG) with glucuronic acid (93). These types of
carrier have been shown by this work to be superior in terms of their delivery volume and
osteo- or chondrogenic ability relative to traditional porous calcium phosphate ceramics.
   Finally, and most relevant to this chapter, hyaluronan’s tolerability and biocompati-
bility as a three-dimensional tissue scaffolding matrix is very acceptable. For example,
studies done with rabbit autologous mesenchymal progenitor cells showed that the cells
adhered and proliferated on hyaluronan (94). In Vivo, when the cell-seeded hyaluro-
nan was implanted, there was no inflammatory response and the scaffold completely
degraded after four months of implantation.
   Surgical applications have also benefited from composite structures that include
hyaluronan, specifically hyaluronan in combination with alginates (95). Alginates are
linear polysaccharides, derived from seaweed, and are composed of D-mannuronic and
L-guluronic acid residues. When in the presence of divalent cations, notably calcium,
a semisolid gel can be formed through the ionic crosslinking between the carboxylic
acid groups located along the polymer chain. This entrapment is known as the eggbox
model where two chains are ionically bound to the calcium ion (96). This system
300   The Nanoscience and Technology of Renewable Biomaterials

is highly biocompatible and has been used as a means to create a three dimensional
environment for culturing cells for bone defects (97). This system is very successful at
encapsulating cells and growth factors due to its crosslinking occurring in physiological
saline with divalent ion concentrations ∼100 mM or less. Past research has investigated
ectopic formation of bone in alginate beads by crosslinking cells within the alginate
as they undergo crosslinking (98, 99). Cai et al. loaded alginate beads with MSCs
into a mouse model and was able to distinguish osteogenic markers including increased
gene expression of osteopontin, osteocalcin, and collagen type I eight weeks post
implantation (98). Perrot et al. also performed ectopic experimentation in a rat model,
and determined that alginate beads require MSCs to fully undergo calcification when
implanted, as blank alginate bead controls only had peripheral calcification as deter-
mined via histological analysis (99). In addition, work by Evangelista et al. analyzed
the effect of functionalized alginate chains by decorating with peptide sequences (100).
Specifically, arginine-glycine-aspartic acid (RGD) was conjugated to alginate, seeded
with MC3T3 cells, ionically crosslinked and underwent osteogenic differentiation. The
authors were able to show a pronounced increase in alkaline phosphatase, von Kossa,
and calcium deposition via staining. Another notable point was that cytoskeleton
organization was pronounced with filopodia spreading and extension as early as day 6
in RGD modified alginate beads compared to a rounded balled up cell morphology in
unmodified controls (100). As with other gelatinous systems, combination of hyaluronan
and/or alginates with other materials has rapidly increased, as investigators expand
on the portfolio of material combinations for bone tissue engineering For instance,
research by Park et al. investigated a unique combination of chitosan/alginate/MSCs/
bone morphogenetic protein-2 (BMP-2) (101). When compared to controls without
MSCs or BMP-2, the gels quickly dissolved signifying a strong interdependence
between the correct growth factor and cell population. Future work is focused on
analyzing a more robust system.
   Another polysaccharide that has great potential in bone tissue engineering is cellulose.
Cellulose is the most abundant biomaterial (polysaccharide) on earth. Its application,
however, for advanced biomedical applications such as tissue engineering has surpris-
ingly not been explored extensively until the last several years. Specifically, microbial
cellulose (MC) synthesized by the microbe Acetobacter xylinum has already been used
in wound healing applications (102). In point of fact, cellulose as derived from plants
has seen extensive applications for generic gauze dressings, dialysis membranes, bone
cell attachment, and connective tissue formation (103). However, MC, albeit chemically
similar to plant-based cellulose, is characterized by a very fibrillar nanostructure that is
very appealing for used as a potential tissue enginnering construct. The advantages of
MC (or cellulose, for that matter) are very apparent for biomedical technologies: high
biocompatibility, high hydrophilicity, high micro- and nanoporosity, and high bioab-
sorbability. Obviously, one of the main criteria for successful use of a biomaterial is
biocompatibility; cellulose has been shown to be able to remain in contact with bone
tisse and hepatocytes without any adverse side effects (104). In fact, a study by Klemm
et al. has shown that cellulose can be implanted as hollow tubular interpositions in
the carotid arteries of rats that elicited a definitive vascular angiogenic response after
12 weeks (105).
    Development and Application of Naturally Renewable Scaffold Materials for BTE        301

   The hydrophilicity of cellulose is amongst the highest of any natural biomaterial.
Its ability to bond hydrogen is what allows nanofibrils to eventually form macroscopic
fibers and also retain significant amounts of water relative to the native mass of the
cellulose (5–10 times more water). In our tissue engineering work, we have hypothe-
sized, for example, that the overall hydrophilicity of a modified carboxy methylcellulose
(CMC) can be too high for normal cellular adhesion. We found that ASCs refrained
from symmetrically occupying the interstitial pore spaces of a CMC sample that had a
high carboxylation loading level. Although the cells appeared to be viable, a unique
cluster formation was observed that was anchored from a small number of attached cells
(unpublished data).
   Another powerful feature of cellulose that abets successful tissue engineering is its
porosity and ability to allow cells to penetrate. For example, permanently implanted
MC can be penetrated by skin cells that are then able to migrate deep into the cellulose
net (106). This is a remarkable finding for the treatment of very deep burns because
the fibroblasts and keratinocytes can penetrate the porous net of cellulose, synthesize an
extracellular matrix, and form dermal tissue over time.
   The aspect of bioabsorbability merits attention. In the last example, the tissue engi-
neering of skin is a very likely event, but in the final analysis MC will not degrade in the
short term. Of course, the time dependence of degradation/absorption is clearly impor-
tant: a biomaterial cannot be too labile or else it will fail to perform its primary function
of behaving as a perfusible medium to allow cellular adhesion, proliferation, and tissue
organization. Yet, if it is not absorbed, it may eventually inhibit or severely attenuate
the final desired prospect of tissue in-growth. Interestingly, cellulose does eventually
become methodically resorbed within a time frame that is compatible with most tissue
engineering programs (90 days), but its retention for longer periods of time (in fact, for
the entire life of the host) has not been known to cause any adverse inflammatory or
allergic reactions.
   In general, cellulose is becoming a very versatile biomaterial for tissue engineering.
Cellulose can be used in a variety of applications in which it is often superior to its
synthetic counterparts due to its durability and biocompatibility.

11.3.4 Naturally Occurring Fibrous Protein Materials in BTE
Collagen, the most abundant protein in the body, has been extensively investigated for
biomedical applications. Collagen is a biocompatible, biodegradable, osteoinductive
material (6, 107). In addition, it has properties, such as amino acid sequences, that make
it an ideal material for cell attachment, proliferation, and differentiation (108). Kakudo
et al. was successful in using a three-dimensional (3D) human adipose-derived stem
cell (hASC) seeded collagen scaffold for a bone construct (109). After being cultured
in vitro for 14 days, the scaffolds were able to induce cell ingrowth and osteogenic
differentiation with the addition of osteogenic supplements in the culture media. Once
grown in vitro, the scaffolds were implanted into nude mice and new bone formation
occurred. In another application, Shih et al. showed that osteogenic differentiation of
bone marrow-derived hMSCs was significantly higher for cells grown on type I col-
lagen nanofibers compared to those seeded on polystyrene tissue culture plates (110).
302   The Nanoscience and Technology of Renewable Biomaterials

Although there have been numerous studies concluding that collagen is a suitable mate-
rial for tissue engineered bone scaffolds, collagen alone typically does not provide the
mechanical strength needed for an effective bone replacement. For this reason, col-
lagen has been modified from its original form by combining it with other materials.
Collagen-hydroxyapatite composites have been investigated in order to utilize the bio-
compatible, biodegradable, osteoinductive material properties of collagen, while also
providing a more rigid and mechanically stable structure (111–116). Rodrigues et al.
formed a hybrid scaffold made from collagen and hydroxyapatite to create a human
osteoblast-seeded scaffold for bone engineering applications (115). They observed that
osteoblasts exhibited a high degree of proliferation and were securely attached to the sur-
face. In addition, cells migrated through the composite and began covering the surface
of the material 11 days post seeding (115). In order to test the enhanced mechanical
properties of a porous collagen/hydroxyapatite composite, Yunoki et al. showed that
during compression tests at 30% strain, the shape of the specimens were well recovered
(116). They also reported that the composite was able to withstand higher compres-
sive stress, attributed to the reinforcement of hydroxyapatite nanocrystals in the collagen
matrix, than other porous materials with biopolymers.
   The emergence of silk as a scaffold material for bone has been extensively developed
by Kaplan and his colleagues (23, 24, 63, 117–124). Studies by Kaplan and others were
initiated due to silk’s unique mechanical properties, formability, biocompatibility, and
ability to undergo proteolytic degradation. Initial work was aimed at the extraction of
sericin proteins to limit immunogenic responses and the behavior of human bone marrow
stromal cells on silk fibroin mats (124). Meinel et al. investigated the use of fibroin
films conjugated with RGD peptide sequences for the promotion of integrin adhesion
and subsequent osteogenesis (122). For that work, neat silk scaffolds and collagen gels
were used as controls versus RGD sequenced silk, and all were seeded with human
MSCs. Bone differentiation was comparable on all materials as determined by alkaline
phosphatase levels, scaffold calcification, and expression bone-specific mRNA transcripts
of bone sialoprotein, osteopontin, and BMP-2 (120, 121). Both silk scaffolds expressed
significant increases in calcium content and alkaline phosphatase activity compared to
collagen. The authors attributed the fast biodegradation of collagen to the inhibition
of these markers. Jin et al. electrospun composite silk fibroin mats by blending the
poly(ethylene oxide) (124). They were able to create fibers with diameters ∼700 ± 50
that come close to mimicking the natural architecture of the extracellular matrix, and
those matrices maintained cell viability up to 14 days. Work by Li et al. followed up
on that research by producing electrospun nanocomposites that encapsulated BMP-2 and
hydroxyapatite inside the electrospun matrices (123). That nanocomposite displayed the
highest calcium deposition and upregulation of BMP-2. Accordingly, it was a pivotal
moment that began to exemplify the tremendous impact that scaffolds providing the
appropriate morphology, chemical composition, and physical properties have on bone
generation in vitro (123).
   To illustrate silk’s potential in vivo, Meinel and Kirker-Head investigated silk in both
nonloadbearing (calvarial) and loadbearing (femoral) defects in rodent models (118–120).
Meinel et al. seeded silk fibroin scaffolds with hMSCs and osteogenically differentiated
them in vitro to yield tissue engineered bone prior to implantation in a 4 mm cranial
defect. Microcomputerized tomography, x-ray, and histological analysis were performed
    Development and Application of Naturally Renewable Scaffold Materials for BTE       303

5 weeks post-implantation and cranial defects that were filled with the tissue engineered
bone appeared to be completely integrated into the skull when compared to (1) hMSC
seeded scaffolds that did not undergo differentiation prior to implantation, (2) scaffold
material alone, and (3) unfilled defects (120). The expression of osteocalcin was only
observed in the center of the defect when cellular scaffolds were used, reflecting that
the contribution of host cells to the defect was negligible with this scaffolding. Follow
up work by Meinel demonstrated that silk based scaffolds could be used in loadbearing
defects with those results displaying a similar trend in that only bone tissue engineered
in vitro a priori to implantation created significantly greater bone volumes compared
to the control groups (119). Mechanical testing of the implanted grafts displayed a
significant maximal load, torque, and torsional stiffness for the tissue engineered bone
constructs compared to undifferentiated hMSC/silk scaffolds and silk scaffolds alone,
clearly displaying its functionality in its ability for skeletal restoration (119). However,
very recent work by Kirker-Head et al. gave controversial evidence that with the sim-
ple induction of a bioactive molecule such as BMP-2, similar results can be obtained
between implanted scaffolding materials with or without a progenitor cell population
(118). These findings signify that in their case, a differentiation period was not needed
to warrant skeletal integration as determined by microcomputed tomography (118). The
work by Kirker-Head et al. exemplifies how the appropriate matrix (silk) and bioactive
molecules (BMP-2) can act as effective osteoinductive mediators of local pluripotent
osteoprogenitor and osteoblastic cell populations.
   Studies have slowly been moving towards keratin-based scaffolds for bone tissue engi-
neering. Keratins are fibrous proteins derived from wool, hair, and nail materials. These
materials contain protein sequences including RGD that are known to facilitate cell adhe-
sion via distinct integrin binding sites. Tachibana et al. began modifying keratin sponges
with calcium-phosphate absorption versus hydroxyapatite loading (125). Osteoblast dif-
ferentiation was assessed for hydroxyapatite, calcium phosphate, and control keratin
sponges and all materials maintained similar cell densities throughout the six day exper-
iment. Alkaline phosphatase activity was measured with the highest activity occurring in
the following order: Hydroxyapatite > Calcium phosphate > unmodified keratin sponge
(125). Tachibana et al. followed up their initial work by binding the bioactive protein
BMP-2 to carboxymethylated keratin sponges (126). The authors were able to show good
retention of BMP-2 and demonstrated confined differentiation to areas of the scaffold
that had bound BMP-2. Alkaline phosphatase activity for osteoblasts increased 2-fold
throughout the seven day culture period for BMP-2 modified keratin sponges, whereas
control keratin sponges displayed no change in ALP activity (126).
   In addition to these materials some work has focused on the use of fibrin as a scaf-
fold and carrier of BMP-2. Gurevich et al. encapsulated bone marrow-derived MSCs
in fibrin microbeads and examined cell proliferation and ectopic bone formation with
rodent models (127). Karp et al. investigated whether fibrin sealants with different
thrombin concentrations would provide an adequate scaffold for bony wound healing
in rat models. Histological analysis revealed a significant decrease in bone infiltration
for both high and low concentrations of thrombin compared to defect controls (128).
Xu et al. experienced similar results with a mouse model when fibrin was investigated
at a delivery vehicle for BMP-2 (129). The fibrin scaffold combined with BMP-2 was
injected subcutaneously and compared to collagen, alginate, hyaluronan, agarose, and
304   The Nanoscience and Technology of Renewable Biomaterials

pluronics. No bone formation occurred for the fibrin group determined by histology and
electron micrographs (129). These two studies illustrate how vital the scaffold material
is, and that not all materials are appropriate for bone tissue engineering.

11.3.5 Naturally Occurring Inorganic Matrices in Bone Tissue Engineering
As mentioned earlier, demineralized bone matrix is another widely used natural scaffold
material. Demineralized bone matrices are created by obtaining bone from a subject
(either from patient or another donor), dissolving the mineral, and then partially defat-
ting it (130). Once the matrix is prepared, the demineralized bone matrix is seeded with
MSCs or osteoblasts. Demineralized bone is thought to contain properties that cause
MSCs to differentiate. Urist et al. proposed that a low molecular weight oligosaccha-
ride glycoprotein exists in the intercellular matrix and perilacunar walls that is exposed
when bone is demineralized and this glycoprotein causes differentiation when it comes
into contact with surrounding cells (131). Einhorn et al. tested this hypothesis by
implanting a demineralized bone matrix, obtained from male Sprague-Dawley rats, into
a fracture site (130). After 12 weeks’ post implantation, five of the seven animals treated
with a demineralized bone matrix were found to have a bridging callus and union across
the fracture. In contrast, those animals not treated with demineralized bone matrices
demonstrated nonunion, were grossly unstable, and were unable to undergo mechanical
testing. When limbs of animals treated with demineralized bone matrices were mechan-
ically tested, they showed improved resistance to fracture and increased strength, values
comparable to early fracture repair, as compared to animals treated only with pins. The
success of this type of bone replacement has led to commercially available demineralized
bone matrices such as Allomatrix Injectable Putty (Wright Medical Technology, Inc.,
Arlington, TN, USA), demineralized bone matrix plus sodium hyaluronate (DBX), DBX
with poly(DL-lactide) mesh, Dynagraft II (Isotis OrthoBiologics, Inc., Irvine, CA,
USA), Grafton DBM line (Osteotech, Inc., Eatontown, NJ, USA), and Regenafil
Injectable Allograft Paste (Exactech Dental Biologics, Gainesville, FL, USA) (132).
One main concern with these commercial products, however, is the lack of regulation
by the FDA. Therefore, methods of sterilization vary, resulting in the creation of prod-
ucts with unreliable properties that may make the implant inferior or invoke an immune
response (133).
   An early study by Bruder et al. demonstrated that bone marrow-derived MSCs loaded
into scaffolds consisting of hydroxyapatite and β-tricalcium phosphate ceramic could be
used to treat a large defect in the femora of adult female dogs (76). Their study con-
sisted of three groups: Group A contained dogs treated with MSC-loaded scaffolds,
dogs in Group B were given scaffolds with no cells loaded into it, and the defects in
the dogs of Group C were not treated at all. Results of this study showed that union
occurred more quickly in defects treated with hMSC-seeded scaffolds than all other
conditions, with a large osseous callus developing around five of the six implants and
the adjacent host bone. In addition, more bone filled the pores of the hMSC-seeded
scaffolds compared to other groups. Kon et al. was able to use the same method of
filling a porous hydroxyapatite ceramic scaffold with autologous BMMSCs to repair
a critical-size bone defect in sheep (7). Upon retrieval of cell-seeded constructs and
unseeded constructs 2 months post implantation, bone formation around and throughout
    Development and Application of Naturally Renewable Scaffold Materials for BTE     305

the porous scaffold was higher in BMMSC-seeded scaffolds as compared to controls
(54.2% and 8.6%, respectively). Their investigation also tested the mechanical proper-
ties of seeded and unseeded scaffolds and found that cell-seeded specimens had a higher
stiffness compared with cell-free scaffolds. The success of using BMMSC-seeded scaf-
folds in animal models led to the first human clinical trial to repair critical-size bone
defects by Quarto et al. (134). Three patients were treated with bone grafts com-
prised of BMMSCs seeded on HA scaffolds, representative in size and shape to their
injury. Using radiographs and computed tomography scans, callus formation along the
implants and integration at the interfaces with native bone was observed 2 months after
surgery. By 13 months post surgery, all external fixations at the site of injury, origi-
nally supplied for mechanical stability, were removed and patients had not reported any
problems.
   Coral scaffolds have many strikingly similar characteristics to trabecular bone. Appro-
priately coral has been used as a bond scaffold material for over 20 years. With its
interconnective 3D porous structure, coral has great osteoconductive activity. Hou et al.
examined the synergistic effects of BMP-2 with coral in a rabbit critical-sized (15 mm)
cranial defect (135). Radiopacity observations were made at 16 weeks showing pref-
erence for a MSC/BMP-2/coral system over BMP-2/coral or coral alone. Fluorescently
labeled MSCs were examined histologically and confirmed a faster rate for osteogenesis
compared to coral alone, and were able to conclude that this treatment was as effective
as an autologous bone graft. A study by Foo et al. used a coral matrix and studied
the gene expression of osteoblasts. The authors analyzed gene expression of RUNX2,
osteopontin, alkaline phosphatase, and osteocalcin. Their results showed similar expres-
sion levels to osteoblast controls signifying that coral matrix did not change the genetic
expression of osteoblasts. Hou et al. analyzed the synergistic effects of coral scaffolds
with MSCs and BMP-2 in a rabbit model (135). oral scaffolds without MSCs showed
inferior results compared to when MSCs were present, as integration into the cranium
was not complete after 16 weeks. Histological sections revealed densely organized tissue
compared to a void randomness seen in the coral/BMP-2 control. The authors specu-
late that in light of their study, a one-time dose of 200 µg BMP-2 could be applied
by clinicians to initiate bone regeneration (135). When compared to Hofmann’s work,
there seems to be a carrier specific effect (136). A recent study by Cui et al. analyzed
ASCs on coral in a cranial defect for a canine model (137). The authors analyzed cel-
lularity alkaline phosphatase activity, and osteocalcin. Cells were precultured in either
growth or osteogenic medium conditions. Cell density was similar for both conditions
yet osteocalcin and was highest for the osteogenically induced group. Opacity volume
was significant for cell-seeded coral scaffolds where neat coral scaffolds underwent rapid
degradation. The authors concluded that when coral is seeded with progenitor cells, it
has a major advantage of matching its degradation rate to the kinetics of new bone
growth (137).
   Nacre is another inorganic material that has been experimented with as a scaffold
for bone tissue engineering. Rousseau et al. showed that nacre stimulated osteoblast
differentiation and mineralization after only 6 days in culture compared to the soluble
factors such as dexamethasone, which takes periods of up to 14 days in vitro to trigger
mineralization (138).
306    The Nanoscience and Technology of Renewable Biomaterials

11.4 Conclusions and Future Directions

Though much innovation continues to drive the development of naturally renewable
materials for bone tissue engineering, much controversy exists as to which materials
exhibit paramount in vivo evidence for direct clinical significance. One of the principle
issues remaining to be addressed is the understanding why some systems allow and/or
promote calcification and skeletal integration while others do not. The combination of
MSCs and growth factors, notably BMP-2, seem to be the current status quo for promot-
ing bone formation in vivo. Work has largely focused on varying the material or carrier
in use. Overall, when choosing which natural material should be used for bone tissue
engineering, one must keep in mind the anatomical location of the defect, i.e. whether
it is a loadbearing or nonloadbearing site, as this will greatly limit the application of
most gelatinous systems that are unable to withstand physiological loading forces for
that defect site. Composite systems of inorganic and organic components have become
popular and a recent review discusses some in use (139). In Vivo work has focused
on examining implanted materials via histology and quantifying radio-opacity compared
to bone controls. These techniques offer merely a glance at what has been achieved
and require more detailed investigatory work before worldwide clinical acceptance. For
instance, few implanted tissue constructs have been examined for the osteogenic proper-
ties of bone such as the organization of osteons or formation of haversian canal systems.
Though much progress has been made, bone tissue engineering is still in its infancy, as
specialized materials are being developed and analyzed for skeletal regeneration.


References

1.    Fung YC. Biomechanics: Motion, Flow, Stress, and Growth. New York: Springer-
      Verlag; 1990.
2.    Mow VC, Hayes WC, editors. Basic Orthopaedic Biomechanics. 2nd ed. Philadel-
      phia: Lippincott-Raven; 1997.
3.    Afizah H, Yang Z, Hui JH, Ouyang HW, Lee EH. A comparison between the
      chondrogenic potential of human bone marrow stem cells (BMSCs) and adipose-
      derived stem cells (ADSCs) taken from the same donors. Tissue Eng. 2007 Apr;
      13(4):659–66.
4.    Bignon A, Chouteau J, Chevalier J et al. Effect of micro- and macroporosity of
      bone substitutes on their mechanical properties and cellular response. J Mater Sci
      Mater Med . 2003 Dec;14(12):1089–97.
5.    Burg KJ, Porter S, Kellam JF. Biomaterial developments for bone tissue engineer-
      ing. Biomaterials. 2000 Dec;21(23):2347–59.
6.    Cornell CN. Osteoconductive materials and their role as substitutes for autogenous
      bone grafts. Orthop Clin North Am. 1999 Oct;30(4):591–8.
7.    Kon E, Muraglia A, Corsi A et al. Autologous bone marrow stromal cells loaded
      onto porous hydroxyapatite ceramic accelerate bone repair in critical-size defects
      of sheep long bones. J Biomed Mater Res. 2000 Mar 5;49(3):328–37.
      Development and Application of Naturally Renewable Scaffold Materials for BTE      307

8.    Shea LD, Wang D, Franceschi RT, Mooney DJ. Engineered bone development
      from a pre-osteoblast cell line on three-dimensional scaffolds. Tissue Eng. 2000
      Dec;6(6):605–17.
9.    Salgado AJ, Coutinho OP, Reis RL. Bone tissue engineering: State of the art and
      future trends. Macromol Biosci . 2004 Aug 9;4(8):743–65.
10.   Giannoudis PV, Dinopoulos H, Tsiridis E. Bone substitutes: An update. Injury.
      2005;365: S20–7.
11.   Cypher TJ, Grossman JP. Biological principles of bone graft healing. J Foot Ankle
      Surg. 1996 Sep–Oct;35(5):413–17.
12.   Ahlmann E, Patzakis M, Roidis N, Shepherd L, Holtom P. Comparison of anterior
      and posterior iliac crest bone grafts in terms of harvest-site morbidity and functional
      outcomes. J Bone Joint Surg Am. 2002 May;84-A(5):716–20.
13.   Sasso RC, LeHuec JC, Shaffrey C. Iliac crest bone graft donor site pain after
      anterior lumbar interbody fusion: A prospective patient satisfaction outcome assess-
      ment. J Spinal Disord Tech. 2005 Feb;18 Suppl:S77–81.
14.   Younger EM, Chapman MW. Morbidity at bone graft donor sites. J Orthop
      Trauma. 1989;3(3):192–5.
15.   Seiler JG, 3rd, Johnson J. Iliac crest autogenous bone grafting: Donor site compli-
      cations. J South Orthop Assoc. 2000 Summer;9(2):91–7.
16.   Friedlaender GE, Strong DM, Tomford WW, Mankin HJ. Long-term follow-up
      of patients with osteochondral allografts. A correlation between immunologic
      responses and clinical outcome. Orthop Clin North Am. 1999 Oct;30(4):583–8.
17.   Pelker RR, Friedlaender GE. Biomechanical aspects of bone autografts and allo-
      grafts. Orthop Clin North Am. 1987 Apr;18(2):235–9.
18.   Hutmacher DW. Scaffolds in tissue engineering bone and cartilage. Biomaterials.
      2000 Dec;21(24):2529–43.
19.   Mano JF, Silva GA, Azevedo HS et al. Natural origin biodegradable systems in
      tissue engineering and regenerative medicine: Present status and some moving
      trends. J R Soc Interface. 2007 Apr 3.
20.   Malafaya PB, Silva GA, Reis RL. Natural-origin polymers as carriers and scaffolds
      for biomolecules and cell delivery in tissue engineering applications. Advanced
      Drug Delivery Reviews. 2007;59(4–5):207–33.
21.   Rees DA. Polysaccharide shapes and their interactions – some recent advances.
      Pure & Applied Chemistry. 1981;53:1–14.
22.   Comper WD, ed. Extracellular Matrix: Molecular Components and Interactions.
      Australia: Harwood Academic Publishers; 1996.
23.   Vepari C, Kaplan DL. Silk as a biomaterial. Progress in Polymer Science. 2007;32:
      991–1007.
24.   Altman GH, Diaz F, Jakuba C et al. Silk-based biomaterials. Biomaterials. 2003;
      24:401–16.
25.   Furth ME, Atala A, Van Dyke ME. Smart biomaterials design for tissue engineering
      and regenerative medicine. Biomaterials. 2007;28(34):5068–73.
26.   Grinnell F. Fibronectin and wound healing. Journal of Cellular Biochemistry.
      1984;26:107–16.
308   The Nanoscience and Technology of Renewable Biomaterials

27. Demers C, Hamdy CR, Corsi K, Chellat F, Tabrizian M, Yahia L. Natural coral
    exoskeleton as a bone graft substitute: A review. Bio-Medical Materials and
    Engineering. 2002;12:15–35.
28. Marks S, Hermey. The structure and development of bone. In: Bilezikian, J.,
    Raisz, LG, Rodan, GA, eds. Principles of Bone Biology. San Diego: Academic
    Press; 1996. pp. 3–14.
29. Guilak F, Butler DL, Goldstein SA, Mooney DJ. Functional Tissue Engineering.
    New York: Springer-Verlag; 2003.
30. Sikavitsas VI, Temenoff JS, Mikos AG. Biomaterials and bone mechanotransduc-
    tion. Biomaterials. 2001 Oct;22(19):2581–93.
31. Buckwalter JA, Glimcher MJ, Cooper RR, Recker R. Bone biology. I: Struc-
    ture, blood supply, cells, matrix, and mineralization. Instr Course Lect. 1996;45:
    371–86.
32. Karageorgiou V, Kaplan D. Porosity of 3D biomaterial scaffolds and osteogenesis.
    Biomaterials. 2005 Sep;26(27):5474–91.
33. Robinson BP, Hollinger JO, Szachowicz EH, Brekke J. Calvarial bone repair with
    porous D,L-polylactide. Otolaryngol Head Neck Surg. 1995 Jun;112(6):707–13.
34. Athanasiou KA, Zhu C, Lanctot DR, Agrawal CM, Wang X. Fundamentals of
    biomechanics in tissue engineering of bone. Tissue Eng. 2000 Aug;6(4):361–81.
35. Keaveny TM, Morgan EF, Niebur GL, Yeh OC. Biomechanics of trabecular bone.
    Annu Rev Biomed Eng. 2001;3:307–33.
36. Klenke FM, Liu Y, Yuan H, Hunziker EB, Siebenrock KA, Hofstetter W. Impact
    of pore size on the vascularization and osseointegration of ceramic bone substitutes
    in vivo. J Biomed Mater Res A. 2007 Sep 26.
37. Takahashi Y, Tabata Y. Effect of the fiber diameter and porosity of non-woven PET
    fabrics on the osteogenic differentiation of mesenchymal stem cells. J Biomater
    Sci Polym Ed . 2004;15(1):41–57.
38. Borden M, El-Amin SF, Attawia M, Laurencin CT. Structural and human cellular
    assessment of a novel microsphere-based tissue engineered scaffold for bone repair.
    Biomaterials. 2003 Feb;24(4):597–609.
39. Yang S, Leong KF, Du Z, Chua CK. The design of scaffolds for use in tissue
    engineering. part I. traditional factors. Tissue Eng. 2001 Dec;7(6):679–89.
40. Pittenger MF, Mackay AM, Beck SC et al. Multilineage potential of adult human
    mesenchymal stem cells. Science. 1999 Apr 2;284(5411):143–7.
41. De Ugarte DA, Morizono K, Elbarbary A et al. Comparison of multi-lineage cells
    from human adipose tissue and bone marrow. Cells Tissues Organs. 2003;174(3):
    101–9.
42. Huang GP, Pan ZJ, Jia BB et al. Ex vivo expansion and transplantation of
    hematopoietic stem/progenitor cells supported by mesenchymal stem cells from
    human umbilical cord blood. Cell Transplant. 2007;16(6):579–85.
43. Zvaifler NJ, Marinova-Mutafchieva L, Adams G et al. Mesenchymal precursor
    cells in the blood of normal individuals. Arthritis Res. 2000;2(6):477–88.
44. De Coppi P, Bartsch G, Jr., Siddiqui MM et al. Isolation of amniotic stem cell
    lines with potential for therapy. Nat Biotechnol . 2007 Jan;25(1):100–6.
45. Kim J, Lee Y, Kim H et al. Human amniotic fluid-derived stem cells have char-
    acteristics of multipotent stem cells. Cell Prolif . 2007 Feb;40(1):75–90.
      Development and Application of Naturally Renewable Scaffold Materials for BTE   309

46.   Guilak F, Lott KE, Awad HA et al. Clonal analysis of the differentiation potential
      of human adipose-derived adult stem cells. J Cell Physiol . 2006 Jan;206(1):
      229–37.
47.   Zuk PA, Zhu M, Mizuno H et al. Multilineage cells from human adipose tissue:
      Implications for cell-based therapies. Tissue Eng. 2001 Apr;7(2):211–28.
48.   Zuk PA, Zhu M, Ashjian P et al. Human adipose tissue is a source of multipotent
      stem cells. Mol Biol Cell . 2002 Dec;13(12):4279–95.
49.   Strem BM, Hicok KC, Zhu M et al. Multipotential differentiation of adipose
      tissue-derived stem cells. Keio J Med . 2005 Sep;54(3):132–41.
50.   Izadpanah R, Trygg C, Patel B et al. Biologic properties of mesenchymal stem cell
      derived from bone marrow and adipose tissue. Journal of Cellular Biochemistry.
      2006;99:1285–97.
51.   Prunet-Marcassus B, Cousin B, Caton D, Andre M, Penicaud L, Casteilla L. From
      heterogeneity to plasticity in adipose tissues: Site-specific differences. Experimen-
      tal Cell Research. 2006;312:727–36.
52.   De Ugarte DA, Alfonso Z, Zuk PA et al. Differential expression of stem cell
      mobilization-associated molecules on multi-lineage cells from adipose tissue and
      bone marrow. Immunol Lett. 2003 Oct 31;89(2-3):267–70.
53.   Wall ME, Bernacki SH, Loboa EG. Effects of serial passaging on the adipogenic
      and osteogenic differentiation potential of adipose-derived human mesenchymal
      stem cells. Tissue Eng. 2007 Jun;13(6):1291–8.
54.   Sumanasinghe RD, Bernacki SH, Loboa EG. Osteogenic differentiation of human
      mesenchymal stem cells in collagen matrices: Effect of uniaxial cyclic tensile
      strain on bone morphogenetic protein (BMP-2) mRNA expression. Tissue Eng.
      2006 Dec;12(12):3459–65.
55.   Endres M, Hutmacher DW, Salgado AJ et al. Osteogenic induction of human bone
      marrow-derived mesenchymal progenitor cells in novel synthetic polymer-hydrogel
      matrices. Tissue Eng. 2003 Aug;9(4):689–702.
56.   Weinzierl K, Hemprich A, Frerich B. Bone engineering with adipose tissue derived
      stromal cells. J Craniomaxillofac Surg. 2006 Dec;34(8):466–71.
57.   Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix elasticity directs stem cell
      lineage specification. Cell . 2006 Aug 25;126(4):677–89.
58.   Mauney JR, Sjostorm S, Blumberg J et al. Mechanical stimulation promotes
      osteogenic differentiation of human bone marrow stromal cells on 3-D partially
      demineralized bone scaffolds in vitro. Calcif Tissue Int. 2004 May;74(5):458–68.
59.   Ward DF, Jr., Salasznyk RM, Klees RF et al. Mechanical strain enhances extra-
      cellular matrix-induced gene focusing and promotes osteogenic differentiation of
      human mesenchymal stem cells through an extracellular-related kinase-dependent
      pathway. Stem Cells Dev . 2007 Jun;16(3):467–80.
60.   Halvorsen YD, Franklin D, Bond AL et al. Extracellular matrix mineralization and
      osteoblast gene expression by human adipose tissue-derived stromal cells. Tissue
      Eng. 2001 Dec;7(6):729–41.
61.   Bancroft GN, Sikavitsas VI, van den Dolder, J. et al. Fluid flow increases miner-
      alized matrix deposition in 3D perfusion culture of marrow stromal osteoblasts in
      a dose-dependent manner. Proc Natl Acad Sci USA. 2002 Oct 1;99(20):12600–5.
310    The Nanoscience and Technology of Renewable Biomaterials

62.   Jager M, Feser T, Denck H, Krauspe R. Proliferation and osteogenic differentiation
      of mesenchymal stem cells cultured onto three different polymers in vitro. Ann
      Biomed Eng. 2005 Oct;33(10):1319–32.
63.   Mauney JR, Nguyen T, Gillen K, Kirker-Head C, Gimble JM, Kaplan DL. Engi-
      neering adipose-like tissue in vitro and in vivo utilizing human bone marrow and
      adipose-derived mesenchymal stem cells with silk fibroin 3D scaffolds. Biomate-
      rials. 2007 Aug 30.
64.   Sen A, Lea-Currie YR, Sujkowska D et al. Adipogenic potential of human adipose
      derived stromal cells from multiple donors is heterogeneous. J Cell Biochem. 2001
      Mar 26;81(2):312–19.
65.   Im GI, Shin YW, Lee KB. Do adipose tissue-derived mesenchymal stem cells have
      the same osteogenic and chondrogenic potential as bone marrow-derived cells?
      Osteoarthr Cartil . 2005 Oct;13(10):845–53.
66.   Wakitani S, Goto T, Pineda SJ et al. Mesenchymal cell-based repair of large,
      full-thickness defects of articular cartilage. J Bone Joint Surg Am. 1994 Apr;76(4):
      579–92.
67.   Saitoh S, Takahashi I, Mizoguchi I, Sasano Y, Kagayama M, Mitani H. Compres-
      sive force promotes chondrogenic differentiation and hypertrophy in midpalatal
      suture cartilage in growing rats. Anat Rec. 2000 Dec 1;260(4):392–401.
68.   Mehlhorn AT, Niemeyer P, Kaiser S et al. Differential expression pattern of extra-
      cellular matrix molecules during chondrogenesis of mesenchymal stem cells from
      bone marrow and adipose tissue. Tissue Eng. 2006 Oct;12(10):2853–62.
69.   Yang G, Crawford RC, Wang JH. Proliferation and collagen production of human
      patellar tendon fibroblasts in response to cyclic uniaxial stretching in serum-free
      conditions. J Biomech. 2004 Oct;37(10):1543–50.
70.   Young RG, Butler DL, Weber W, Caplan AI, Gordon SL, Fink DJ. Use of mes-
      enchymal stem cells in a collagen matrix for achilles tendon repair. J Orthop Res.
      1998 Jul;16(4):406–13.
71.   Lee IC, Wang JH, Lee YT, Young TH. The differentiation of mesenchymal stem
      cells by mechanical stress or/and co-culture system. Biochem Biophys Res Com-
      mun. 2007 Jan 5;352(1):147–52.
72.   Park JS, Chu JS, Cheng C, Chen F, Chen D, Li S. Differential effects of equiaxial
      and uniaxial strain on mesenchymal stem cells. Biotechnol Bioeng. 2004 Nov 5;
      88(3):359–68.
73.   Sanchez-Ramos J, Song S, Cardozo-Pelaez F et al. Adult bone marrow stromal
      cells differentiate into neural cells in vitro. Exp Neurol . 2000 Aug;164(2):247–56.
74.   Arinzeh TL, Peter SJ, Archambault MP et al. Allogeneic mesenchymal stem cells
      regenerate bone in a critical-sized canine segmental defect. J Bone Joint Surg Am.
      2003 Oct;85-A(10):1927–35.
75.   Schantz JT, Hutmacher DW, Lam CX et al. Repair of calvarial defects with
      customised tissue-engineered bone grafts II. evaluation of cellular efficiency and
      efficacy in vivo. Tissue Eng. 2003;9 Suppl 1:S127–39.
76.   Bruder SP, Kraus KH, Goldberg VM, Kadiyala S. The effect of implants loaded
      with autologous mesenchymal stem cells on the healing of canine segmental bone
      defects. J Bone Joint Surg Am. 1998 Jul;80(7):985–96.
      Development and Application of Naturally Renewable Scaffold Materials for BTE   311

77. Aust L, Devlin B, Foster SJ et al. Yield of human adipose-derived adult stem cells
    from liposuction aspirates. Cytotherapy. 2004;6(1):7–14.
78. Shi C, Zhu Y, Ran X, Wang M, Su Y, Cheng T. Therapeutic potential of chi-
    tosan and its derivatives in regenerative medicine. Journal of Surgical Research.
    2006;133:185–92.
79. Kim IY, Seo SJ, Moon HS et al. Chitosan and its derivatives for tissue engineering
    applications. Biotechnol Adv . 2007 Aug 3.
80. Kong L, Gao Y, Lu G, Gong Y, Zhao N, Zhang X. A study on the bioactivity
    of chitosan/nano-hydroxyapatite composite scaffolds for bone tissue engineering.
    European Polymer Journal . 2006;42:3171–9.
81. Di Martino A, Sittinger M, Risbud MV. Chitosan: A versatile biopolymer for
    orthopaedic tissue-engineering. Biomaterials. 2005 Oct;26(30):5983–90.
82. Inanc B, Elcin AE, Koc A, Balos K, Parlar A, Elcin YM. Encapsulation and osteoin-
    duction of human periodontal ligament fibroblasts in chitosan-hydroxyapatite
    microspheres. Journal of Biomedical Materials Research Part A. 2007;82A(4):
    917–26.
83. Ge ZG, Baguenard S, Lim LY, Wee A, Khor E. Hydroxyapatite-chitin materials as
    potential tissue engineered bone substitutes. Biomaterials. 2004;25(6):1049–58.
84. Malafaya PB, Pedro AJ, Peterbauer A, Gabriel C, Redl H, Reis RL. Chitosan
    particles agglomerated scaffolds for cartilage and osteochondral tissue engineering
    approaches with adipose tissue derived stem cells. Journal of Materials Science-
    Materials in Medicine. 2005;16(12):1077–85.
85. Gravel M, Gross T, Vago R, Tabrizian M. Responses of mesenchymal stem cell
    to chitosan–coralline composites microstructured using coralline as gas forming
    agent. Biomaterials. 2006;27:1899–1906.
86. Larsen NE, Balazs EA. Drug delivery systems using hyaluronan and its derivatives.
    Advanced Drug Delivery Reviews. 1991;7:279–93.
87. Balazs EA. Viscosurgery in the eye. Ocul Inflam Ther. 1983;1(92):93.
88. Chan RW, Titze IR. Hyaluronic acid (with fibropectin) as a bioimplant for the vocal
    fold mucosa. Larygnoscope. 1999;109:1142–9.
89. Pirnazar P, Wolinsky L, Nachnani S, Haake S, Pilloni A, Bernard GW. Bacterio-
    static effects of hyaluronic acid. J Periodont. 1999;70:370–4.
90. Kalish-Siebel H, Gaton DD, Weinberger D, Loya N, Schwartz-Ventik M,
    Solomon A. A comparison of the effect of hyaluronic acid versus gentamicin on
    corneal epithelial healing. Eye. 1998;12:829–33.
91. Doherty, MM, Hughes, PJ, Korszniak, NV, Charman, WN. Prolongation of
    lidocaine-induced epidural anesthesia by medium molecular weight hyaluronic
    acid formulations: pharmacodynamic and pharmacokinetic studies in the rabbit.
    Anesthesia & Analgesia. 1995;80(4):740–6.
92. Moriyama K, Ooya T, Yui N.