Mechanistic-Empirical Structural Design Models for Emulsified

Document Sample
Mechanistic-Empirical Structural Design Models for Emulsified Powered By Docstoc
					Confidential Contract Report CR-2003/44   November 2004



      Mechanistic-Empirical Structural
         Design Models for Emulsified
            Bitumen Treated Materials
DOCUMENT RETRIEVAL PAGE                                                       Report No: CR-2003/44

Title: Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials

Authors: F M Long and H L Theyse
Client:                     Client Reference No:         Date:                Distribution:
SABITA, Gautrans
                                                         October 2004         Restricted
Project No:TIJ90            OE2: 9431: Transport Infrastructure               ISBN:
Abstract:
This report describes the development of mechanistic-empirical structural design transfer functions for
emulsified bitumen treated materials. The models are developed using data from three HVS test sites,
Heilbron (P9/3), Cullinan (D2388) and Vereeniging (P243/1), and from the laboratory testing performed
in conjunction with the Cullinan and Vereeniging HVS tests. Some additional triaxial testing was
performed as part of this project.

The report recommends that emulsified bitumen treated pavement layers are analysed in two phases,
firstly the stiffness reduction phase and secondly permanent deformation. The stiffness reduction phase
life is the number of traffic load applications for the initial high stiffness to reduce to that of a steady state
stiffness. The stiffness reduction transfer function is dependent on the strain ratio, which is the ratio
between the tensile strain at the bottom of the pavement layer and the strain-at-break from the flexural
beam test. The permanent deformation transfer function calculates the accumulation of permanent
deformation in the layer and is a function of the stress ratio, plastic strain and the cement and residual
bitumen contents. The stress ratio is a measure of the shear stress in the pavement and the maximum
allowable shear stress. The stress ratio calculation takes into account the relative density and degree of
saturation of the material. The laboratory tests showed that the addition of active filler (cement)
increases the shear strength, and therefore increases the permanent deformation resistance of the
material. The addition of binder decreases the shear strength, but increases the flexibility of the
material. The increase in flexibility increases the stiffness reduction phase. The binder and cement
contents should be balanced to ensure optimum performance in both stiffness reduction and permanent
deformation.

A material classification system is recommended based on both UCS tests and strain-at-break values
from the flexural beam test. Four material classes are recommended, EB1 to EB4. Using the transfer
functions, design catalogues for new construction and design charts for deep in situ recycling were
developed for the EB2 to EB4 material classes.

Keywords:
Deep in situ recycling, emulsified bitumen, effective fatigue, stiffness reduction transfer function,
permanent deformation model transfer function, equivalent granular state, steady state stiffness,
mechanistic-empirical structural design procedure, Heavy Vehicle Simulator
Proposals for implementation: Report supplied to client for implementation.
Related documents:

Signatures:


                    APT Steering
A McKay             Committee               BMJA Verhaeghe             E van Heerden           PJ Hendricks
Language Editor     Technical Reviewer      Programme Manager          Info Centre             Division Director
EXECUTIVE SUMMARY


            Emulsified bitumen treated materials (EBTM) have been successfully used in South
            Africa for many years and two SABITA manuals1, 2 for their use are available. Existing
            mechanistic-empirical structural design models for these materials are not fully
            appropriate and deficiencies in their use have been identified. This report describes the
            development of such mechanistic-empirical structural design models.

            The models were selected by observations of the behaviour of emulsified bitumen treated
            materials in Heavy Vehicle Simulator (HVS) test sections and from laboratory test results.
            Data from three HVS test sites were analysed: Heilbron (P9/3), Cullinan (D2388) and
            Vereeniging (P243/1). The Heilbron sections were constructed conventionally, the
            Cullinan sections were constructed using labour-intensive construction techniques and
            the Vereeniging sections were constructed using deep in situ recycling. Extensive
            laboratory testing was performed in conjunction with the Cullinan and Vereeniging HVS
            tests. Unfortunately little relevant material data on the Heilbron sections are available.
            The Heilbron HVS data were therefore used to identify and confirm trends in the
            behaviour of the material, but were not used to develop the mechanistic-empirical
            structural design models.

            The modes of distress of emulsified bitumen treated materials identified were stiffness
            reduction and permanent deformation. The stiffness reduction phase is defined as the
            number of load repetitions required for the high initial resilient modulus to reduce to the
            steady state resilient modulus. This steady state is when the resilient modulus is
            equivalent to that of a granular material; the material is not necessarily in a loose,
            particulate or cracked form. The second phase of distress is the accumulation of
            permanent deformation in the emulsified bitumen treated base layer.

            Examination of the laboratory test data identified a gap in the data, in that there was no
            data at varying binder contents using the same parent material, and consequently
            additional laboratory testing was performed on the Vereeniging material. The additional
            testing ensured that data from a wider range of emulsified bitumen and cement contents
            were available. The laboratory testing results showed that an increase in the cement
            content increases the shear strength of the treated material, but an increase in the
            bitumen content decreases the shear strength and the resilient modulus. This implies
            that the addition of cement increases the permanent deformation resistance, but the
            addition of bitumen decreases the permanent deformation resistance. The addition of
            binder does, however, significantly improve the flexibility of the material, which results in
            an increase in the stiffness reduction phase of the pavement life. It is essential that the
            cement and binder contents are balanced to ensure adequate flexibility and permanent
            deformation resistance.

            The stiffness reduction transfer function was developed from the elastic stiffness data
            backcalculated from in-depth MDD deflections. The stiffness reduction phase life is a
            function of the strain ratio, which is the ratio of the tensile strain at the bottom of the
            emulsified bitumen treated layer and the strain-at-break value determined from the
            flexural beam test. The transfer function was developed for four levels of reliability,
            typically associated with Category A to D roads.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 i
Report number CR-2003/44
            The permanent deformation transfer function was developed using permanent
            deformation dynamic triaxial tests and was checked with the HVS permanent deformation
            behaviour. The transfer function is dependent on the stress ratio in the pavement
            calculated at various locations, the cement and binder contents and the plastic strain to
            obtain any level of permanent deformation in the layer. The effect of parent material
            quality, relative density and degree of saturation are accounted for in the calculation of
            the stress ratio. This transfer function was also developed for four levels of reliability,

            Using available laboratory data, a material classification system was developed. Four
            material classes are recommended, EB1 to EB4. The system uses the Unconfined
            Compressive Strength test (UCS) to measure the shear strength and the strain-at-break
            to measure the flexibility of the material. The flexibility component of the material
            classification system is essential to prevent cement being added to increase the UCS, at
            the expense of flexibility of the material. A classification using UCS and Indirect Tensile
            Strength (ITS) values did not provide an acceptable ranking of mixes.

            The stiffness reduction and permanent deformation transfer functions were used to
            develop design catalogues and charts for three of the four material classes, EB2 to EB4.
            Insufficient data were available for the EB1 material class. The development of the
            catalogues and charts are described in the report. The catalogues are used for new
            construction projects and the design charts for deep in situ recycling projects. It is
            recommended that an EB1 material is tested and the stiffness reduction and permanent
            deformation transfer functions modified for this material class.

            The catalogues are in general agreement with those published in SABITA Manual 211,
            which have been validated by experienced consultants. There are no pavement
            structures with an emulsified bitumen treated layer with adequate structural capacity for
            95 per cent reliability (Category A roads) for the ES10 and ES30 traffic classes in the
            catalogues and design charts. However, it is possible that using a good parent material,
            pavements will be able to sustain sufficient traffic loading to fall into these classes.
            Alternatively, a lower level of reliability can be selected.

            Comparisons of the respective design charts show that the structural behaviour of
            emulsified and foamed bitumen treated materials is similar. Where differences occur, the
            emulsified bitumen treated layers are thinner than those required for foamed bitumen
            treated layers. The differences are however, marginal. The structural design ignores any
            effect of differences in the ease of construction of the materials, specifically the
            compaction, and the comparisons are not based on economic factors.

            It is recommended that field tests on existing emulsified bitumen treated pavements are
            evaluated to validate the transfer functions and design catalogues and charts.

            The work described in this report highlighted areas for future work, such as:
            •      Implementation in standard practise of the strain-at-break test to measure
                   flexibility and the Texas triaxial test to measure the shear strength
            •      Measurement and estimation of in situ density and saturation levels.
            •      Several areas in the analyses procedures used in the South African Mechanistic-
                   empirical Design Method require modifying.


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              ii
Report number CR-2003/44
            •       A good quality material, such as crushed stone, and a cohesionless material such
                    as sand should be tested to assess the behaviour of the material when treated
                    with emulsified bitumen. This has already been initiated.
            •       The development of a universal model for all material types treated with
                    bituminous stabilisation.
            •       The implementation of a LTPP program to monitor the behaviour of emulsified
                    bitumen treated pavements under real traffic.

            This report satisfies deliverable 4 of the project, as outlined in the work proposal
            (Appendix A). The structural design method described in this report allows comparison of
            the structural capacities of foamed and emulsified bitumen treated materials in an
            equitable manner.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials           iii
Report number CR-2003/44
ACKNOWLEDGEMENTS


The funding provided for this project by the Gauteng Provincial Government, Department of Public
Transport, Roads and Works, and the South African Bitumen Association, SABITA, is gratefully
acknowledged.

The authors would like to acknowledge the specimen preparation team: Dave Ventura, Nathaniel
Masango, Moosa Sekatane and Piet Botha. Colin Fisher and Joseph Marima ran the triaxial tests.

This report was extensively reviewed by members, or representatives of, the Accelerated Pavement
Testing Committee and the sponsors. Their effort is gratefully acknowledged.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials       iv
Report number CR-2003/44
TABLE OF CONTENTS


EXECUTIVE SUMMARY.......................................................................................................................... i
ACKNOWLEDGEMENTS ...................................................................................................................... iv
LIST OF TABLES ................................................................................................................................. vii
LIST OF FIGURES................................................................................................................................. ix
1.         INTRODUCTION......................................................................................................................... 1
2.         BACKGROUND .......................................................................................................................... 3
        2.1.       Heilbron HVS Tests, 1992 ................................................................................................. 3
        2.2.       SABITA Manual 14, GEMS, 1993...................................................................................... 4
        2.3.       Structural Design Models from Heilbron (P9/3) HVS Data ................................................ 6
        2.4.       SABITA Manual 21, ETB 1999 .......................................................................................... 7
        2.5.       Cullinan HVS Testing, Road D2388, Labour Intensive Construction, 2000 ...................... 8
        2.6.       Vereeniging HVS Testing, Deep In Situ Recycling, 2001.................................................. 8
        2.7.       PIARC Recycling Guideline ............................................................................................. 10
        2.8.       The Way Forward ............................................................................................................ 10
3.         HEAVY VEHICLE SIMULATOR TESTS .................................................................................. 12
        3.1.       Heilbron (P9/3) ................................................................................................................. 12
        3.2.       Cullinan (D2388) .............................................................................................................. 17
        3.3.       Vereeniging (P243/1) ....................................................................................................... 18
        3.4.       Summary of HVS Test Sections ...................................................................................... 22
        3.5.       Selection of Appropriate Transfer Functions ................................................................... 23
4.         EXISTING LABORATORY TESTS .......................................................................................... 24
        4.1.       Calculation of Relative Density and Saturation................................................................ 24
        4.2.       Calculation of Stress Ratio............................................................................................... 24
        4.3.       Heilbron Laboratory Test Data......................................................................................... 25
        4.4.       Cullinan Laboratory Test Data ......................................................................................... 26
        4.5.       Vereeniging (P243/1) ....................................................................................................... 28
        4.6.       Summary.......................................................................................................................... 31
5.         ADDITIONAL LABORATORY TESTING................................................................................. 32
        5.1.       Justification for Additional Laboratory Tests .................................................................... 32
        5.2.       Laboratory Testing Program ............................................................................................ 33
        5.3.       Materials........................................................................................................................... 33
        5.4.       Specimen Preparation ..................................................................................................... 34
        5.5.       Static Triaxial Tests.......................................................................................................... 35
        5.6.       Dynamic Triaxial Tests..................................................................................................... 40
6.         ANALYSES OF LABORATORY TEST DATA......................................................................... 49
        6.1.       Comparison of Flexibility Results..................................................................................... 49
        6.2.       Comparison of Shear Strength Data................................................................................ 49

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                                 v
Report number CR-2003/44
       6.3.       Permanent Deformation Data .......................................................................................... 56
       6.4.       Comparison of Resilient Moduli Data .............................................................................. 60
7.        STIFFNESS REDUCTION TRANSFER FUNCTION................................................................ 62
       7.1.       Stiffness Reduction of Heavy Vehicle Simulator Tests.................................................... 62
       7.2.       Calculation of Strain Ratio ............................................................................................... 67
       7.3.       Stiffness Reduction Transfer Function............................................................................. 68
8.        PERMANENT DEFORMATION TRANSFER FUNCTION ....................................................... 72
       8.1.       Ratio of Cement and Binder Contents ............................................................................. 72
       8.2.       Laboratory Permanent Deformation Model...................................................................... 73
       8.3.       Heavy Vehicle Simulator Permanent Deformation Data.................................................. 76
       8.4.       Permanent Deformation Transfer Function ..................................................................... 83
9.        DESIGN CATALOGUES AND DESIGN CHARTS .................................................................. 85
       9.1.       Material Classification for Emulsified Bitumen Treated Materials.................................... 85
       9.2.       Materials and Input Values .............................................................................................. 87
       9.3.       Mechanistic Analysis........................................................................................................ 89
       9.4.       Pavement Structural Capacity ......................................................................................... 91
       9.5.       Design Catalogues........................................................................................................... 92
       9.6.       Design Charts .................................................................................................................. 98
       9.7.       Validation of Design Models with Field Data ................................................................. 102
10.       CONCLUSIONS AND RECOMMENDATIONS ...................................................................... 103
       10.1.      Laboratory Test Data ..................................................................................................... 103
       10.2.      Stiffness Reduction Transfer Function........................................................................... 104
       10.3.      Permanent Deformation Transfer Function ................................................................... 105
       10.4.      Material Classification, Design Catalogues and Charts................................................. 105
11.       FUTURE WORK ..................................................................................................................... 107
12.       REFERENCES........................................................................................................................ 109
Appendix A: Work Proposal ............................................................................................................ 111
Appendix B: Laboratory Test Data ................................................................................................. 112
       Test Matrix and Specimen Preparation Data Sheet................................................................... 113
       Static Triaxial Test Data ............................................................................................................. 114
       Dynamic Triaxial Test Data ........................................................................................................ 115




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                           vi
Report number CR-2003/44
LIST OF TABLES


Table 1.     ETB Classification from SABITA Manual 211 .................................................................... 7
Table 2.     Variables from HVS test sections and triaxial tests from Cullinan (D2388) ........................
             and Vereeniging (P243/1) ................................................................................................ 11
Table 3.     Heilbron HVS test sections .............................................................................................. 12
Table 4.     Model fits for permanent deformation response, Heilbron............................................... 16
Table 5.     Cullinan HVS tests ........................................................................................................... 17
Table 6.     Model fits for permanent deformation response, Cullinan ............................................... 20
Table 7.     Vereeniging emulsified bitumen treated material HVS tests ........................................... 20
Table 8.     Model fits for permanent deformation response, Vereeniging......................................... 22
Table 9.     Summary of HVS test sections ........................................................................................ 22
Table 10.    Material Properties of Emulsified Bitumen Treated Material from Cullinan ..................... 26
Table 11.    UCS results from Cullinan tests ....................................................................................... 26
Table 12.    Cullinan flexural beam test data....................................................................................... 26
Table 13.    Shear strength parameters for Cullinan (D2388) material with 1 per cent cement ......... 27
Table 14.    Average resilient moduli for Cullinan (D2388) material with 1 per cent cement.............. 27
Table 15.    Optimum moisture content and maximum dry density values for Vereeniging material.. 28
Table 16.    Laboratory testing for Vereeniging (P243/1) emulsified bitumen treated material tests.. 29
Table 17.    Flexural beam test results for Vereeniging material ........................................................ 30
Table 18.    Shear strength parameters for Vereeniging (P243/1) material with 2 per cent cement .....
             and 1.8 per cent residual binder ...................................................................................... 30
Table 19.    Average resilient moduli for Vereeniging (P243/1) material with 2 per cent cement ..........
             and 1.8 per cent residual binder ...................................................................................... 31
Table 20.    Additional triaxial tests in terms of cement and residual binder contents ........................ 33
Table 21.    Test matrix for additional triaxial tests ............................................................................. 33
Table 22.    Cohesion and friction angles of Vereeniging (P243/1) materials with 1 per cent ...............
             cement and 0.9 per cent and 3.0 per cent residual binder .............................................. 37
Table 23.    Model coefficients and statistics for Equation (5), Vereeninging materials with 1% ...........
             cement ............................................................................................................................. 39
Table 24.    Resilient moduli for Vereeniging (P243/1) material with 1 per cent cement.................... 41
Table 25.    Average resilient moduli for Vereeniging (P243/1) material with 1 per cent cement....... 42
Table 26.    Model coefficients for dynamic triaxial permanent deformation data, 3 per cent ...............
             residual binder ................................................................................................................. 45
Table 27.    Model coefficients for dynamic triaxial permanent deformation data, 0.9 per cent ............
             residual binder ................................................................................................................. 46
Table 28.    Model coefficients and statistics for Equation (5), Cullinan ............................................. 50
Table 29.    Model coefficients and statistics for Equation (5), Vereeniging Material ......................... 51
Table 30.    Model coefficients and statistics for Equation (9) ............................................................ 52

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                         vii
Report number CR-2003/44
Table 31.    Model coefficients and statistics for Equation (10) .......................................................... 56
Table 32.    Model coefficients for dynamic triaxial permanent deformation data (Cullinan (D2388), ...
             1 per cent cement, 0.9 per cent and 1.5 per cent residual binder) .................................. 57
Table 33.    Model coefficients for dynamic triaxial permanent deformation Data (Vereeniging, ..........
             2 per cent cement and 1.8 per cent residual binder, low saturation)............................... 58
Table 34.    Model coefficients for dynamic triaxial permanent deformation data, high .........................
             saturation (Vereeniging, 2 per cent cement and 1.8 per cent residual binder, ...................
             high saturation) ................................................................................................................ 59
Table 35.    Average resilient moduli for Cullinan and Vereeniging materials .................................... 60
Table 36.    Estimated steady state stiffnesses for HVS sections from Cullinan and Vereeniging..... 66
Table 37.    Strain-at-break values for HVS test sections ................................................................... 68
Table 38.    Model coefficients and statistics for Equation (15) .......................................................... 75
Table 39.    Material classification of emulsified bitumen treated materials........................................ 87
Table 40.    Asphalt Material Properties.............................................................................................. 87
Table 41.    Emulsified bitumen treated material properties ............................................................... 88
Table 42.    Cemented layer material properties................................................................................. 89
Table 43.    Granular layer material properties ................................................................................... 89




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                      viii
Report number CR-2003/44
LIST OF FIGURES


Figure 1.           Backcalculated stiffnesses from Heilbron HVS Test Sections......................................... 14
Figure 2.           Fits to in-depth MDD permanent deformation (Heilbron Section 374A3A)...................... 16
Figure 3.           Backcalculated stiffnesses from Cullinan (D2388) HVS Test Sections ........................... 19
         Backcalculated stiffnesses from Vereeniging HVS Test Sections .............................................. 21
Figure 4.................................................................................................................................................. 21
Figure 5.           Average ITS results from Heilbron laboratory testing16 ................................................... 25
Figure 6.           UCS and ITS results for Vereeniging material................................................................. 29
Figure 7.           Mohr-Coulomb failure envelopes for Vereeniging materials with 1% cement and .............
                    0.9% and 3.0% residual binder ........................................................................................ 36
Figure 8.           Cohesion and friction angles of Vereeniging (P243/1) materials with 1 per cent ...............
                    cement and 0.9 per cent and 3.0 per cent residual binder .............................................. 38
Figure 9.           Influence of saturation on resilient modulus for Vereeniging (P243/1) material with ..........
                    1 per cent cement ............................................................................................................ 43
Figure 10.          Example dynamic triaxial test data .................................................................................. 44
Figure 11.          Structural capacity of mixes with 3.0 and 0.9 per cent residual binder and 1 per ..............
                    cent cement...................................................................................................................... 47
Figure 12.          Comparison on strain-at-break results from Cullinan and Vereeniging ........................... 49
Figure 13.          Comparison of maximum allowable principal stress from Vereeniging data model fit .... 53
Figure 14.          Comparison of shear strength of untreated Cullinan and Vereeniging materials ............ 54
Figure 15.          Comparison of shear strength of treated Cullinan and Vereeniging materials with ............
                    1% cement ....................................................................................................................... 55
Figure 16.          Resilient moduli as a function of the saturation level....................................................... 61
Figure 17.          Stiffness reduction to steady state stiffness, Cullinan (D2388), 397A4 (Sandstone) ...... 63
Figure 18.          Stiffness reduction to steady state stiffness, Cullinan (D2388), 403A4 (Sandstone) ...... 64
Figure 19.          Stiffness reduction to steady state stiffness, Cullinan (D2388), 407A4, (Sandstone) ..... 64
Figure 20.          Stiffness reduction to steady state stiffness, Cullinan (D2388), 408A4 (Sandstone) ...... 64
Figure 21.          Stiffness reduction to steady state stiffness, Vereeniging (P243/1), 410A4/B4 .............. 65
Figure 22.          Stiffness reduction to steady state stiffness, Vereeniging, 412A4................................... 65
Figure 23.          Stiffness reduction phase life as a function of wheel load ............................................... 66
Figure 24.          Locations to calculate strain ratio in pavement structure under the dual wheels ............ 68
Figure 25.          Stiffness reduction phase lives as a function of the strain ratio....................................... 69
Figure 26.          Comparison of stiffness reduction phase life transfer functions for emulsified ...................
                    bitumen treated materials, foamed bitumen treated materials and lightly ..........................
                    cemented materials.......................................................................................................... 70
Figure 27.          Stiffness reduction phase life for emulsified bitumen treated materials .......................... 71
Figure 28.          Repetitions to 17 per cent plastic strain from dynamic triaxial permanent .........................
                    deformation tests ............................................................................................................. 73

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                                   ix
Report number CR-2003/44
Figure 29.   Laboratory permanent deformation transfer (to arbitrarily selected 17% plastic strain) .. 76
Figure 30.   Heilbron permanent deformation data as a function of wheel load (17 per cent ................
             plastic strain) .................................................................................................................... 77
Figure 31.   Cullinan permanent deformation data as a function of wheel load (17% plastic strain) .. 78
Figure 32.   Vereeniging permanent deformation data as a function of wheel load (17% plastic ..........
             strain) ............................................................................................................................... 78
Figure 33.   HVS permanent deformation data as a function of wheel load (17 per cent plastic ...........
             strain) ............................................................................................................................... 79
Figure 34.   Locations to calculate stress ratio in pavement structure under dual wheels ................. 80
Figure 35.   Structural capacities of HVS test sections as a function of stress ratio (17 per cent .........
             plastic strain) .................................................................................................................... 81
Figure 36.   Laboratory predicted and HVS structural capacities (17 per cent plastic strain)............. 82
Figure 37.   Comparison of predicted vs measured load repetitions to 17% plastic strain ................. 83
Figure 38.   Permanent deformation transfer function for road categories, 17 per cent plastic strain ...
             (3 per cent residual bitumen and 1 per cent cement) ...................................................... 84
Figure 39.   Material classification using UCS and ITS of Cullinan and Vereeniging materials.......... 86
Figure 40.   Material classification using UCS and strain-at-break of Cullinan and ...............................
             Vereeniging materials ...................................................................................................... 86
Figure 41.   Locations to calculate stress ratio in pavement structure................................................ 90
Figure 42.   Design catalogues for emulsified bitumen treated materials, class EB2 ......................... 95
Figure 43.   Design catalogues for emulsified bitumen treated materials, class EB3 ......................... 96
Figure 44.   Design catalogues for emulsified bitumen treated materials, class EB4 ......................... 97
Figure 45.   Design charts for emulsified bitumen treated materials, class EB2................................. 99
Figure 46.   Design charts for emulsified bitumen treated materials, class EB3............................... 100
Figure 47.   Design charts for emulsified bitumen treated materials, class EB4............................... 101




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                              x
Report number CR-2003/44
1.          INTRODUCTION


            Emulsified bitumen treated materials (EBTM) have been successfully used in South
            Africa for many years and two SABITA manuals1, 2 for their use are available. The EBTM
            pavements have, however, not been designed using mechanistic-empirical (ME)
            structural design models. Such ME design models for these materials have been
            developed, but have been found to be not fully appropriate and deficiencies in their use
            have been identified. Previous attempts (1996) to introduce emulsified bitumen treated
            materials into the Technical Recommendations for Highways: Structural Design of
            Flexible Pavements for Interurban and Rural Roads (TRH4)3 failed because the
            mechanistic-empirical design model that was available for EBTM resulted in designs that
            were uneconomical in comparison with other types of pavements. The current mix design
            procedures and the material classification also need to be refined with the latest research
            results. Additionally, the mix design needs to be linked to the structural design.

            Historically, emulsified bitumen treated materials were separated into emulsion
            modification and emulsion stabilization with distinctly different design approaches for the
            two treatment options. It is suggested that a single, consistent approach be developed
            for emulsified bitumen treatment, similar to the approach followed for foamed bitumen
            treatment.

            Recent work on foamed bitumen treated materials showed the potential for such a unified
            approach to the mix and structural design. Mechanistic-empirical structural design
            models and a mix design procedure were developed, and utilized in “Interim Technical
            Guideline: The Design and Use of Foamed Bitumen Treated Materials” (TG2), published
            by the Asphalt Academy4. These guidelines provide the tools from which pavements with
            foamed bitumen can be designed using ME design models with improved confidence. A
            similar approach needs to be applied to emulsified bitumen treated materials. The
            preparation of a similar document for EBTM would benefit the industry in that consultants,
            contractors, suppliers and road owners would have tools that will improve the confidence
            in designs, and thereby reduce the risks. Such a guideline would also allow the
            comparison of foamed bitumen and emulsified bitumen treated materials in a consistent
            manner and the appropriate material could be selected using valid economic appraisals.

            The Gauteng Provincial Government, Department of Public Transport, Roads and Works
            (Gautrans) and the Southern African Bitumen Association (SABITA) funded a project to
            develop mechanistic-empirical structural design models that tie into mix design, and to
            compile a Technical Guideline to complement TG2. The proposal set out a three phase
            work plan:
            •     Phase 1:    Review existing information
            •     Phase 2:    Develop structural design models
            •     Phase 3:    Write technical guideline document

            This report describes Phase 2 of the project. Both laboratory and Heavy Vehicle
            Simulator (HVS) data are used to develop the models, which are the basis of new design
            charts and catalogues. The background information gathered in Phase 1 is included in
            Chapter 2 this report.


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              1
Report number CR-2003/44
            Chapter 3 describes the available HVS data, and establishes trends in the behaviour of
            emulsified bitumen treated materials. Brief descriptions of the materials used and the
            HVS testing variables are described. Justification for the selection of appropriate transfer
            functions, namely stiffness reduction and permanent deformation, are provided.

            In Chapter 4, the laboratory test programs performed in conjunction with the HVS tests
            are described. The laboratory tests attempt to characterize the behaviour of emulsified
            bitumen treated materials. In Phase 1 of this project some limitations in the available
            laboratory data were identified. Consequently, additional laboratory testing, which is
            described in Chapter 5, was performed. Chapter 6 discusses additional analyses that
            were performed on the existing laboratory test data, and summarizes and compares the
            data.

            Chapters 7 and 8 describe the development of the stiffness reduction and permanent
            deformation transfer functions, respectively. The transfer functions are developed for
            Category A to D roads.

            In Chapter 9, a material classification system is suggested, and design catalogues and
            charts are developed. The catalogues are for the design of new construction projects,
            and the design charts for deep in situ recycling (DISR). The catalogues and charts were
            developed for 3 material classes for moderate conditions. It was not possible to develop
            catalogues and charts for the EB1 material class due to the unavailability of sufficient
            data for those material classes.

            Finally, Chapter 10 contains conclusions and recommendations and Chapter 11 suggests
            future work arising from this report. The report satisfies deliverable 4 of the project, as
            detailed in the proposal in Appendix A.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               2
Report number CR-2003/44
2.          BACKGROUND


            Emulsified bitumen treated materials have been successfully used for road construction in
            South Africa for many years. Heavy Vehicle Simulator (HVS) tests have been performed
            on such pavements and laboratory testing was performed in conjunction with the HVS
            tests.

            This chapter identifies the HVS and laboratory tests that have been carried out, and
            identifies data suitable for developing structural design models. The actual data is
            described in detail in Chapters 3 and 4. The existing guideline documents, and previous
            attempts to develop mechanistic-empirical structural design models are also discussed.
            The discussion proceeds in chronological order.

2.1.        Heilbron HVS Tests, 1992

            Six HVS tests were carried out in 1992 on Road P9/3 near Heilbron in the Free State5.
            The test sections all contained emulsified bitumen treated bases, which were classified as
            granular emulsion mixes (GEMS)6.


            The base layer of the HVS test sections was treated with lime and emulsified bitumen.
            The parent base material was a weathered dolerite with a G4 or G5 classification. This
            material is typically found in the Free State Province. All the sections were pre-treated
            with 2.5 per cent lime to reduce the Plasticity Index (PI) of the base material. It is
            possible that the high lime contents resulted in a partially stabilised layer. The emulsified
            bitumen contents of the sections ranged from 1 to 3 per cent, and an anionic emulsion
            was used.

            The majority of the tests were run at an extremely high load, which induced rapid
            deterioration of the pavement. The aim was to induce high horizontal tensile strains at
            the bottom of the base that would lead to fatigue cracking in a relatively short period of
            time. In these situations, it can be difficult to quantify the pavement behaviour between
            the initial and distressed condition and to determine the number of load applications to
            reach the distressed state. Multi-depth deflectometer (MDD) data are available for the
            HVS test sections.

            The data that are available to characterize the materials in the Heilbron HVS test sections
            are limited. The only data that are available are some Indirect Tensile Strength tests, and
            Marshall stability and flow tests. To develop transfer functions for emulsified bitumen
            treated materials in line with current trends with foamed bitumen treated materials4, 7,
            cemented materials and unbound granular materials8, it is, however, necessary to have
            the following data, which are not available:
            •       Shear strength parameters, i.e., cohesion and friction angle
            •       Flexibility measure, i.e., strain-at-break (treated materials only)
            Without these data it is not possible to develop or calibrate complete transfer functions for
            the stiffness reduction phase and permanent deformation from the Heilbron HVS and
            laboratory test data. However, the HVS data can be used to establish and confirm trends
            in the behaviour of emulsified bitumen treated materials, and to compare these trends
            with the behaviour of test sections for which more suitable data are available.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                3
Report number CR-2003/44
2.2.        SABITA Manual 14, GEMS, 1993

            In 1993, SABITA published Manual 14, “GEMS – The design and use of granular
            emulsion mixes”2. Another manual was published in 1999, Manual 21 titled “ETB: The
            design and use of emulsion-treated mixes”1. This manual is discussed in Section 2.4. In
            Manual 14, GEMS are defined as “emulsion modification or stabilization of the following
            granular materials:
            •     Substandard materials (normally those with high PI)
            •     Other granular materials of better quality (up to G1/G2 materials)
            •     Recycled granular bases (may include surfacings)
            •     Recycled cement-and lime-treated bases (in an equivalent granular state)
            •     Combinations of any of the above”

2.2.1       Stabilization and Modification


            The definitions of modification and stabilization in the SABITA manuals1, 2 are different to
            those normally used. Typically modification implies a partial change to a material, which
            may be reversible. Stabilization typically implies the quality of the material has been
            improved and the material has been strengthened. The change should be permanent.
            However, in SABITA Manual 14 (GEMS), the following definitions of modification and
            stabilization are used:

            Stabilization
            This is the process in which GEMS are treated as if they were asphalt mixes. As such,
            the Marshall stability and Indirect Tensile Tests are recommended. Residual binder
            contents generally range between 1.5 and 5 per cent by mass of mix, and the material is
            termed “stabilised GEMS”.

            Modification
            This is the process in which GEMS are treated as if they were gravel mixes. As such, the
            California Bearing Ratio (CBR), Unconfined Compressive Strength (UCS) and Indirect
            Tensile Tests are used. Residual binder contents generally range between 0.6 and 1.5
            per cent by mass of mix, and the material is termed “modified GEMS”.

2.2.2       Structural Behaviour and Design of GEMS


            According to Manual 14, five engineering properties influence the structural behaviour of
            GEMS:
            •    Resistance to fatigue cracking
                 Such cracking is induced by repeated loading, and should be controlled by proper
                 pavement and mix design. The mode of fatigue cracking was considered to be
                 fairly similar to that of asphalt. The manual recommends that the fatigue cracking
                 behaviour of GEMS be assessed with the four-point beam test or the Indirect
                 Tensile Test.
            •    Stiffness
                 The stiffness of the GEMS layer has a significant effect on its load-spreading
                 characteristics. It is recommended that the stiffness is determined by the Indirect
                 Tensile Test or the four-point beam test.


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               4
Report number CR-2003/44
            •     Resistance to permanent deformation
                  Permanent deformation is caused by consolidation and shear, which can be
                  prevented by proper compaction, and a proper mix design, respectively. If an
                  equivalent granular state is reached after excessive cracking, permanent
                  deformation will occur. It is recommended that the evaluation of the potential for
                  premature permanent deformation is done with the Marshall stability, CBR or UCS
                  tests.
            •     Durability
                  Durability of GEMS was identified as important to their behaviour, especially in wet
                  regions. Water sensitivity is assessed with the capillary test.
            •     Resistance to indentation
                  Manual 14 states that GEMS should have adequate resistance to indentation. This
                  is assessed by the grading modulus.

            While these properties are all critical to the behaviour of emulsified bitumen treated
            materials, the focus of the discussion in Manual 14 tends towards treating the material
            like asphalt. Typical EBTM pavements in South Africa do not behave, and are not
            visually similar, to asphalt. For example, EBTMs do not demonstrate the same highly
            temperature dependant, viscoelastic behaviour as asphalt, and they do not have the
            same ability to sustain tension. As such, some of the recommended tests for
            characterizing asphalt are not suitable for EBTMs.

            The recommended method for structural analysis in Manual 14 depends on whether the
            GEM is modified or stabilised. However, for both, it is stated that the “mix design criteria
            recommended … should prevent any permanent deformation from taking place within a
            GEMS base.” It is not a reasonable expectation to prevent permanent deformation. The
            permanent deformation should be controlled, however, rather than prevented completely.

            Stabilised GEMS

            The stabilization structural design is based on work originally proposed by Santucci, but
            simplified9. The method uses quarterly analysis periods, and a base stiffness is used for
            each quarter. Using the stiffness values, the maximum horizontal strains in the stabilised
            GEMS layer and the maximum vertical strain at the top of the subgrade are calculated
            using multi-layer linear elastic (MLLE) theory.

            The failure criteria for pavements with a stabilised GEMS base layer are fatigue cracking
            of the GEMS base and excessive subgrade strain. The transfer function for fatigue
            cracking is from Santucci, and assesses the expected number of equivalent standard
            axles (E80s) that can be applied before failure occurs as a function of the horizontal
            strain, the volumes of voids and the volume of the binder content. This method is
            suggested for mixes containing 5 to 8 per cent of residual binder by mass. The estimated
            structural fatigue capacity can be adjusted for the void content and binder content in the
            mix. The equation used for adjusting these variables is exactly the same as
            recommended for asphalt in the Asphalt Institute design guide10. Emulsified bitumen
            mixes in South Africa are not sufficiently similar enough to asphalt to justify the use of the
            same transfer function, and such high binder contents are unlikely.

            The subgrade strain criteria are the same as those recommended in the South African
            Mechanistic-Empirical Design Method (SAMEDM)8.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 5
Report number CR-2003/44
            Modified GEMS

            The failure criteria for the modification approach are the same as those recommended for
            granular materials in the SAMEDM and are based on the factor of safety method to
            prevent shear deformation. Manual 14 states that for modified GEMS, the cohesion value
            utilized can be obtained by doubling that of the parent material. Suggested values for the
            cohesion and friction angle are given for modified GEMS, based on the parent material
            and the moisture state. The safety factor is determined from the cohesion and friction
            angle, and then used to estimate the number of E80s to failure. The transfer function is
            given in a graphical format only, and the actual failure condition, i.e., how much
            deformation is allowed to occur in the layer, is not clearly defined. Minimum allowable
            safety factors are recommended for each road category and design traffic class.

2.2.3       Summary


            The emphasis in Manual 14 is on the distinction between modified and stabilised GEMS.
            It is strongly recommended that this distinction is dropped, and that the classification of
            EBTMs is dependent on the performance of the material. Similarly, the structural design
            method should not differentiate between modified and stabilised GEMS, and a unified
            approach is recommended. This unified approach should be suitable for all types of
            EBTMs, and it should not be based on a structural design method that is appropriate for
            asphalt.     The structural design approach for modified GEMS is therefore more
            appropriate and a similar approach is used in the analyses described in this report.

2.3.        Structural Design Models from Heilbron (P9/3) HVS Data

            De Beer and Grobler11 determined the structural capacity and developed limiting
            structural design criteria for granular emulsion mixes (GEMS) in terms of fatigue cracking
            and fracture. The analyses were based on data from the Heilbron test sections.


            The structural capacity was derived from rut measurements, and from fatigue and fracture
            concepts for the sections containing 3 per cent emulsified bitumen. The resistance to
            permanent deformation was not considered in the structural design process. The fatigue
            approach was based on the number of repetitions necessary for the stiffness of the
            GEMS layer to reach 50 per cent of the initial stiffness. Fracturing of the GEMS layer
            follows the initial fatigue life, and the layer continues to fracture until its stiffness reduces
            to an equivalent granular state.

            Two transfer functions are available, one for initial fatigue, and another for fatigue and
            total fracture, which equates to approximately 10 mm of rutting. Both are functions of the
            horizontal tensile strain at the bottom of the layer. The transfer functions are not used in
            a phased analysis of the life of the pavement; only one transfer function is used for any
            structural capacity determination. Multipliers for adjusting the structural capacity for
            ageing of the GEMS and for wet areas are suggested. It is stated that GEMS are not
            suitable for pavements with a required structural capacity exceeding 10 million E80s.

            A catalogue of designs was developed from the transfer functions. However, subsequent
            work found that the pavement structures were too conservative and hence too expensive,
            and did not agree with field experience12, 13, 14. These transfer functions are therefore not


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                    6
Report number CR-2003/44
            useful for the design of emulsified bitumen treated pavements and are not considered
            further.

2.4.        SABITA Manual 21, ETB 1999

            SABITA Manual 21, “ETB – The design and use of emulsion-treated materials” was
            published in 19991. In the manual, ETBs are defined as “granular materials (often not
            suitable for use as a base course for the particular application) treated with a small
            percentage of bitumen emulsion (usually less than 2.5 per cent, dependent on the quality
            of the parent material) to render them suitable for base course layers.” This manual
            complements Manual 14, GEMS, discussed in Section 2.2. In Manual 21, no
            differentiation was made between the structural design of modified and stabilised
            materials, and the intention of the manual was to address mixes previously classified as
            modified mixes in Manual 14. Problems had been experienced with the structural design
            procedures in Manual 14. If the same binder content was used in the grey area between
            modified and stabilised mixes, different structural designs were obtained from the two
            methods. The intention of Manual 21 was to solve this problem, although the success of
            this has not been determined.

            The structural design section of Manual 21 is brief, and does not provide mechanistic-
            empirical structural design models. Reference is made to the need for balanced
            pavement structures, and a preference for deep structures is suggested. The fairly stiff
            and brittle nature of ETBs and the increase in fatigue resistance in comparison with
            granular materials is recognised, as is the associated load sensitivity if the ETB layer is
            not well supported.

            A catalogue of pavement designs is recommended, utilising two classes of ETBs, E1 and
            E2. The materials are classified according to either the UCS or CBR values, as shown in
            Table 1.

            Table 1.     ETB Classification from SABITA Manual 211

              Material Code                   Minimum CBR                            Minimum UCS
                   E1              150% at 100% Mod. AASHTO compaction                 1 200 kPa
                   E2              100% at 100% Mod. AASHTO compaction                  700 kPa

            The catalogue provided in Manual 21 was developed using the DCP design method15.
            The intention was to develop mechanistic-empirical transfer functions for emulsified
            bitumen treated materials. However, only data from the Heilbron (P9/3) (see Section 2.1)
            and one of the Cullinan (D2388) (see Section 2.5) HVS test sections were available at the
            time. These data sets were found to be lacking sufficient data from which to develop
            mechanistic-empirical transfer functions15.

            The catalogue developed using the DCP design method was checked against empirical
            catalogues suggested by various consultants experienced with emulsified bitumen treated
            materials16. These catalogues are used as a check of the proposed catalogues
            described in Chapter 9.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              7
Report number CR-2003/44
2.5.        Cullinan HVS Testing, Road D2388, Labour Intensive Construction,
            2000

            HVS and laboratory testing were done as part of a project evaluating labour intensive
            construction on Road D2388 near Cullinan, Gauteng in 2000. Several different materials
            and treatments were used for the base layers, and these were compared with a
            conventionally constructed crushed stone base. One of the treatments was emulsified
            bitumen treatment, and both a sandstone and a Premamix (SASOL ash) were
            treated17, 18. The Premamix showed adequate performance with many of the standard
            laboratory tests, but showed poor performance under the HVS and in the dynamic triaxial
            test, in comparison with the other materials. This is demonstrated by the UCS of the
            material, which was used to classify the Premamix material as an E1 based on the
            SABITA Manual 21 classification1, despite its poor performance in dynamic tests. It
            should only be considered for special applications where a low bearing capacity could be
            tolerated and should not be designed using standard design methods. For the purposes
            of developing general mechanistic-empirical structural design models for emulsified
            bitumen treated materials, the data from the Premamix tests should not be used. The
            Premamix sections have therefore been excluded from the analyses in this report.

            Four HVS tests were performed on emulsified bitumen treated sandstone sections. The
            emulsified bitumen treated bases of all four sandstone test sections were treated with 1.5
            per cent emulsified bitumen (0.9 per cent residual binder) and 1 per cent cement. A 60
            per cent stable grade anionic emulsion was used. Multi-depth deflectometers (MDD)
            were installed in all four test sections and recorded both elastic deflections and
            permanent deformation. The other standard HVS instrumentation was also used, and the
            data are available17, 18. The data from the Cullinan HVS tests have been analysed in
            some detail already13, 17, 18.

            An extensive laboratory testing program was undertaken to complement the Cullinan
            HVS tests13. A full range of tests was performed, including: UCS, ITS, static triaxial,
            dynamic triaxial and flexural beam tests

            The available laboratory and HVS data are sufficient data to develop structural design
            models, in conjunction with the appropriate HVS data. These data sets are discussed in
            detail in Chapters 3 and 4, and used extensively in Chapters 7 and 8 to develop transfer
            functions.

2.6.        Vereeniging HVS Testing, Deep In Situ Recycling, 2001

            HVS testing was performed on Road P243/1 between Vereeniging and Heidelberg in
            Gauteng in 2000 and 200119, 20 on sections constructed using deep in situ recycling
            (DISR) of a previously cement treated ferricrete base layer. Untreated, milled material
            was obtained during the recycling, and was used for laboratory testing12, 21, 22.

            Two sections were constructed, one with foamed bitumen and the other with emulsified
            bitumen. The milled ferricrete recycled in the base was classified as a G7 material prior
            to treatment. The material had a good grading. The EBTMs sections were treated with a
            60 per cent stable grade anionic emulsion. Two HVS test sections were constructed on
            each section, i.e., two foamed bitumen sections (409A4 and 411A4) and two emulsified
            bitumen sections (410A4 and 411A4). MDDs recording in-depth elastic deflections and
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials             8
Report number CR-2003/44
            permanent deformation were installed on all test sections. The data from the foamed
            bitumen HVS sections and the associated laboratory testing were used to develop the
            structural design models in TG22.

            The Vereeniging HVS data have are appropriate for developing transfer functions for
            emulsified bitumen treated materials. However, the HVS testing occurred at only one
            combination of cement and emulsified bitumen. Therefore, if a transfer function were
            developed from these data only, it would strictly only be valid for the same combination of
            cement and emulsified bitumen contents (2% and 1.8% residual binder, respectively).
            Using this data together with data from the Cullinan testing, however, allows the
            development of transfer functions with a wider applicability.

            An extensive laboratory testing program was performed in conjunction with the
            Vereeniging (P243/1) HVS testing21, 22. Untreated material was obtained from behind the
            recycler during construction and was treated in the laboratory. The relevant details and
            results of this laboratory testing are discussed in Chapter 4.

            Strain-at-break and UCS test results showed that the presence of cement greater than
            the initial consumption of cement (ICC) increased the strength of the mix, but reduced the
            flexibility. Conversely, larger quantities of emulsified bitumen improved the flexibility of
            the mix, but reduced the compressive strength. The best mix would optimise the
            compressive strength and flexibility with an appropriate ratio of cement to emulsified
            bitumen.

2.6.1       Structural Design Models for Emulsified Bitumen Treated Materials


            Liebenberg12 provides structural design models for effective fatigue (stiffness reduction)
            and permanent deformation of emulsified bitumen treated materials, based on work
            carried out during secondment to CSIR Transportek. These models have the same form
            as those used for foamed bitumen in TG24. The design models were developed using
            the HVS and associated laboratory data from Vereening (P243/1) only and all these
            sections contained 2 percent cement, which is high for emulsified bitumen treated
            materials. This is a limitation in this work.

            There are some differences in the methods of analyses used for emulsified bitumen
            treated materials by Liebenberg and foamed bitumen treated materials, which were used
            to develop the structural design component of TG24. Limitations in the methods of
            analyses were also identified. For these reasons, it was recommended that with the
            benefit of more HVS and laboratory test data, more reasonable transfer functions could
            be developed.

            Liebenberg also recommended a modified specification for EBTMs, similar to the one
            used for FBTMs in TG2, and suggested four material classes, based on Unconfined
            Compressive Strength (UCS) and Indirect Tensile Strength (ITS) criteria.

2.6.2       Summary


            In summary, an extensive database is available from the Vereeniging (P243/1) laboratory
            and HVS tests from which to develop stiffness reduction and permanent deformation
            transfer functions.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               9
Report number CR-2003/44
2.7.        PIARC Recycling Guideline

            The PIARC, Technical committee 7/8 “Road Pavements”, released a Draft report, Version
            2.3A in May 2002 titled “Cold in-place recycling of pavements with emulsion or foamed
            bitumen”23. The guideline incorporates both emulsified bitumen and foamed bitumen
            treated materials.

            The guideline consists mainly of extracts from various design guides, and is not a
            cohesive guide on either emulsified or foamed bitumen treatment. The document is not
            suitable as a reference for updating the South African emulsified bitumen design
            guidelines.


            One interesting aspect of the guideline is that it gives an indication of the relative
            structural contributions of emulsified and foamed bitumen treated layers, in terms of
            AASHTO layer coefficients. The structural coefficients for foamed bitumen and emulsified
            bitument layers are similar.



2.8.        The Way Forward

            It is recommended that the approach to be followed in the new guideline is to move away
            from differentiating between modification and stabilization of emulsified bitumen treated
            materials. The classification of emulsified bitumen treated materials should focus on the
            performance of the material, and not on a somewhat arbitrary classification based on the
            quantity of emulsified bitumen added. The result of this approach will also render the
            differentiation between GEMs and ETBs null and void. To avoid confusion, the material
            will be referred to as “emulsified bitumen treated material” (EBTM).

            The terms “effective fatigue life” and “equivalent granular state” have caused confusion in
            industry. The perception is that effective fatigue life implies cracking. This is not the
            case. For bituminous treated materials it means a reduction in stiffness. Cracking may
            occur in conjunction with the stiffness reduction, but is not a given. The perception is also
            that at the end of the effective fatigue life the pavement is in a granular state in that it is
            broken up into smaller blocks. The end of the effective fatigue life is, in fact, reached
            when the equivalent granular state has been reached. This equivalent granular state is
            when the stiffness has reached a constant value, and does not decrease further even
            with increases in the traffic load. Because of the confusion these terms cause, they are
            no longer used in this work. The “effective fatigue life” is replaced with the “stiffness
            reduction phase” and the “equivalent granular state” is replaced with the “steady state
            stiffness”.



2.8.1       Suitable HVS and Laboratory Data


            As discussed in Section 2.1, the Heilbron HVS data are not suitable for developing
            structural design models because of the test plan adopted. These data can, however, be
            used to validate trends in the pavement behaviour. An additional limitation of the
            Heilbron data is that insufficient material data are available, and the laboratory test data
            do not include tests now considered to be critical for the structural design of emulsified
            bitumen and foamed bitumen treated materials21.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 10
Report number CR-2003/44
            The majority of data from the other two sets of HVS tests (Cullinan and Vereeniging) are,
            however, suitable for developing structural design models for EBTMs. All the test
            sections used approximately the same test plan, which was set up to be ideally suited to
            developing structural design models. Pavement structures differ among all of the HVS
            test sections. Other variables for the two sets of tests are shown in Table 2.

            With the available data, it will be possible to develop structural design models. However,
            these models may require some extrapolation if used for density, moisture and cement
            and binder combinations that differ from the data used in the modelling.

            Table 2.    Variables from HVS test sections and triaxial tests from Cullinan
                        (D2388) and Vereeniging (P243/1)

                                                             HVS Test Section
                                     Cullinan (D2388)            Vereeniging (P243/1)
            Parent material type     sandstone                   previously cement treated ferricrete
            Cement and               1 % cement                  2 % cement
            emulsified bitumen       0.9 % residual binder       1.8 % residual binder
            contents
            Test loads               40 and 80 kN                 40, 80 and 100 kN
                                     one density level
                                                                  two density levels
                                     one saturation level
            Triaxial tests                                        two saturation levels
                                     two cement and binder
                                                                  one cement and binder combination
                                     combinations




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            11
Report number CR-2003/44
3.          HEAVY VEHICLE SIMULATOR TESTS


            This chapter presents some Heavy Vehicle Simulator data from the Heilbron, Cullinan
            and Vereeniging test sections. The pavement behaviour is discussed, and the data
            provide justification for the development of the transfer functions for the two phases of
            pavement life, namely stiffness reduction and permanent deformation. The data
            presented in this chapter are used in Chapter 7 and 8 to develop the transfer functions.

3.1.        Heilbron (P9/3)

            The Heilbron HVS tests were performed in 1992 on Road P9/3. The parent material was
            a G4/G5 weathered dolerite. The material had a PI of 8.6 and to reduce the PI the
            sections were pre-treated with 2.5 per cent lime5. The emulsified bitumen contents and
            associated residual binder contents are shown in Table 3. A 60 per cent emulsion was
            used. There is little information available on the material used in the emulsified bitumen
            treated layer.

            The pavement structure consisted of a 25 mm asphalt surfacing, 150 mm emulsified
            bitumen treated base, 150 mm cement treated subbase and up to 300 mm of selected
            layers. A 13.2 mm seal was placed on the pavement 9 months after construction, but
            was not placed on the HVS test sections.

            Table 3.       Heilbron HVS test sections

                                     Emulsified bitumen
             HVS Section                                                  Wheel load and tyre    Mode of
                                           content
               number                                                     inflation pressure2   trafficking
                                 (residual bitumen content)
                   372A3                 3 % (1.8%)                         140 kN, 1472 kPa      Wander
                   373A3                 3 % (1.8%)                          60 kN, 775 kPa       Wander
                  374A3A                 2 % (1.2%)                         140 kN, 1472 kPa    Channelized
                  374A3B                 2 % (1.2%)                          40 kN, 614 kPa     Channelized
                  375A31                 1 % (0.6%)                          80 kN, 947 kPa     Channelized
                  376A33                 1 % (0.6%)                         140 kN, 1472 kPa    Channelized
             1.        Data unreliable due to presence of sandy layer in pavement
             2.        Loads applied on a single wheel
             3.        Only two data points available due to very short test


            Six test sections were tested, with the loading, tyre inflation pressures and mode of
            trafficking detailed in Table 3. A range of wheel loads was used on the test sections, and
            the load was applied in both wander and channelized modes. A method of converting
            from wander to channelized traffic is suggested in Reference 5. It is postulated that in the
            channelized mode one load repetition is equivalent to four load repetitions applied in the
            wandering mode. Therefore, in the presentation of the data in this section, the data for
            the 1 and 2 per cent emulsified bitumen sections have been multiplied by four to obtain
            the equivalent number of “wander” repetitions. The transfer functions developed from
            HVS tests are usually in terms of wandering traffic. The one metre wander on the HVS
            test sections is thought to be equivalent to the wander experienced in the wheel paths of
            roads. Channelized traffic is considered to be more severe, because the pavement has
            less time to recover from the load induced deflection and deformation before the next
            load is applied. Although the conversion appears realistic, it has not been fully validated.


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                  12
Report number CR-2003/44
            Because the data are not used to develop the complete transfer functions, the suitability
            of the conversion between wander and channelized traffic is not investigated further.

            Multi-depth deflectometers (MDDs) were installed in all HVS test sections. Both elastic
            deflections and permanent deformation of the MDD modules were recorded. On Sections
            372A3 and 373A3, two MDDs per section have useful data. Sections 374A3A and
            374A3B each had an MDD in the central area of the test section and one in the
            turnaround zone. The data from the turnaround zone cannot be considered
            representative of the pavement behaviour, and was ignored. The elastic deflections are
            used to calculate the stiffnesses of the pavement layers, whereas the in-depth permanent
            deformations of the MDD modules are used to assess the permanent deformation
            resistance of the layers.

3.1.1       Elastic Stiffness


            The in-depth MDD deflection data were used to backcalculate the stiffnesses of all of the
            pavement layers. The stiffnesses are backcalculated using the deflections measured
            under a 40 kN load, regardless of the actual trafficking load. Sections 372A3 and 376A3
            were excluded because of their lack of sufficient 40 kN deflection data. The load
            repetitions from sections trafficked with a channelized mode of loading have been
            converted to the equivalent number of wandering load repetitions. The backcalculated
            stiffnesses for the emulsified bitumen treated base layers from three of the five Heilbron
            HVS test sections are illustrated in Figure 1.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            13
Report number CR-2003/44
                                        600


                                        500

              Resilient Modulus (MPa)
                                        400


                                        300


                                        200
                                                                                   373A3, MDD4, 60 kN, 1.8% binder
                                        100                                        373A3, MDD12, 60 kN, 1.8% binder


                                          0
                                              0   50000 100000 150000 200000 250000 300000 350000 400000 450000 500000
                                                                           Load Repetitions
                                        600


                                        500
              Resilient Modulus (MPa)




                                        400


                                        300


                                        200

                                                                                  374A3A, MDD11, 140 kN, 1.2% binder
                                        100
                                                                                  374A3B, MDD5, 40 kN, 1.2% binder


                                          0
                                              0       500000     1000000       1500000        2000000   2500000        3000000
                                                                           Load Repetitions
                                        600

                                                                                    375A3, MDD4, 80 kN, 0.6% binder
                                        500
                                                                                    375A3, MDD12, 80 kN, 0.6% binder
              Resilient Modulus (MPa)




                                        400


                                        300


                                        200


                                        100


                                          0
                                              0    100000   200000   300000    400000    500000    600000   700000     800000
                                                                           Load Repetitions
            Figure 1.                             Backcalculated stiffnesses from Heilbron HVS Test Sections




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                          14
Report number CR-2003/44
            The section trafficked at 40 kN has a higher stiffness than the other sections; with the
            stiffnesses reducing as the trafficking wheel loads increase. This may indicate that the
            higher wheel loads may be inducing rapid damage to the materials. This trend seems to
            dominate any trend dependent on the residual binder content of the base layer.

            In test sections 373A3 and 375A3, for 3 of the 4 MDDs, the stiffness decreases rapidly
            under loading. This is expected for a stabilized material. However, because of the low
            initial stiffnesses, this reduction in stiffness is relatively small. The stiffnesses stabilize to
            a constant value for most of the test sections. This is the “steady state stiffness”,
            historically termed the “equivalent granular state”. At this stage, the material is not in a
            granular state nor has it cracked to form small, loose blocks of material. The material is
            equivalent to a granular material in stiffness only. The number of load repetitions to reach
            this steady state stiffness has been termed the “stiffness reduction phase” previously
            termed the “effective fatigue life”.

            Because of the lack of material data for the Heilbron test sections, it was not possible to
            use these data to develop an stiffness reduction transfer function. The purpose of
            discussing the data was to illustrate trends such as the reduction in stiffness for the
            Heilbron pavements.



3.1.2       Permanent Deformation Response


            The permanent deformation response was evaluated by fitting a function to the MDD in-
            depth permanent deformation data. The function has the form shown in Equation (1) and
            is used to determine the accumulation of permanent deformation with increasing load
            repetitions for the particular loading, material and pavement conditions. The use of this
            model is discussed in Reference 13. To determine the permanent deformation in the
            base layer, the model is fit individually to the data from the MDD modules at the top and
            bottom of the base. Only the data from the repetitions of the first wheel load applied to
            the test section are used, the permanent deformation behaviour under the second or third
            wheel loads having a load history, which is difficult to model. Data measured after water
            was added to the section during trafficking are not used. Examples of the data and fitted
            models to the top and bottom of the base layer for MDD11 from Section 374A3A are
            shown in Figure 2. The best fit models are determined manually. The model coefficients
            are shown in Table 4. The minimum R2 for the data fits summarized in Table 4 was 0.96.



                                         PD = mN + a(1- e-bN )                                            (1)

                                 where PD      = permanent deformation (mm)
                                       N       = repetitions
                                       a, b, m = model parameters




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                    15
Report number CR-2003/44
                                              6.0
                                                          40mm
                                                          190mm




                 PERMANENT DEFORMATION (mm)
                                                          340mm
                                              5.0         625mm
                                                          900mm
                                                          40 mm model
                                              4.0         190 mm model



                                              3.0


                                              2.0


                                              1.0


                                              0.0
                                                    0     2000   4000     6000   8000    10000   12000   14000     16000    18000
                                                                                 REPETITIONS


            Figure 2.                               Fits to in-depth MDD permanent deformation (Heilbron Section
                                                    374A3A)


            Table 4.                                Model fits for permanent deformation response, Heilbron

                                                Load              Depth      Coefficients for Equation (1)       Mode of
            Section                                       MDD
                                                (kN)              (mm)         a         b           M           Traffic
                                                                    40       4.90    4.22E-05    1.40E-06
                                                           4                                                     Wander
                                                                   190       3.30    4.50E-05    7.50E-07
             372A3                                  140
                                                                    40       1.20    3.00E-04    1.15E-04
                                                           12                                                    Wander
                                                                   190       2.00    1.00E-04    4.40E-05
                                                                    40       1.50    2.00E-05    3.20E-06
                                                           4                                                     Wander
                                                                   190       0.90    1.20E-05    1.20E-06
             373A3                                  60
                                                                    40       2.20    1.00E-05    2.20E-06
                                                           12                                                    Wander
                                                                   190       0.97    1.20E-05    1.00E-06
                                                                    40       4.50    2.80E-04    7.30E-05
            374A3A                                  140    11                                                 Channelized
                                                                   190       2.15    2.40E-04    2.00E-05
                                                                    40       0.41    8.00E-06    4.00E-08
            374A3B                                  40     5                                                  Channelized
                                                                   190       0.30    6.00E-06    5.00E-10
                                                                    40       0.80    9.00E-05    1.10E-05
                                                           4                                                  Channelized
                                                                   190       1.17    3.00E-05    4.40E-06
             375A3                                  80
                                                                    40       2.40    3.00E-05    3.10E-05
                                                           12                                                 Channelized
                                                                   190       1.12    4.00E-05    1.30E-05

            The difference between the two fitted models at a particular number of load repetitions
            gives the permanent deformation in the base layer. Using the permanent deformation in
            the layer, and the distance between the MDD modules, the average plastic strain can be
            calculated. This was done for all of the reasonable data, and the number of load
            repetitions to set levels of plastic strain (1, 4, 7, 10, 13, 17 and 20 per cent) were
            calculated. These data are analysed and discussed further in Chapter 8.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                    16
Report number CR-2003/44
3.2.        Cullinan (D2388)

            The Cullinan HVS tests were performed on road D2388 from 1997 to in 2000. Only the
            HVS test sections with emulsified bitumen treated base layers from the Cullinan series of
            tests are discussed in this section. Five tests were performed, Sections 396A4, 397A4
            and 403A4 were tested as part of Phase 1, and Sections 407A4/A4A and 408A4 as part
            of Phase 2. The pavement structure for the test sections consisted of a 30 mm asphalt
            surfacing, emulsified bitumen treated base of various thicknesses (Table 5), cement
            treated sandstone subbase, and selected layers. Section 397A4 had a sandstone
            selected layer, whereas the other sections had a ferricrete selected layer. The ferricrete
            is of poorer quality than the sandstone material. The test numbers, material properties,
            base layer thickness, estimated moisture content and dry density, and loading sequences
            are shown in Table 5. Four all four test sections shown, the base material was
            sandstone, treated with 1.5 per cent emulsion (0.9 per cent residual binder) and 1 per
            cent cement. The parent base material had a G5 classification. Detailed material data is
            included in Reference 13.

            Table 5.      Cullinan HVS tests

                              Emulsified bitumen
                                 treated base
           HVS Test                                            Selected
                                         Moisture
           Section                                              layer         Loading sequence
                           Thickness content (%),
           number                                              material
                             (mm)       dry density
                                          (kg/m3)
                                                                              200 000 @ 40kN, 520 kPa
                                                 4.9
           397A4               100                            Sandstone       200 000 @ 70 kN, 670 kPa
                                                2101
                                                                              104 500 @ 70 kN, 670 kPa, wet1
                                                                              200 000 @ 40kN, 520 kPa
                                                 4.8
           403A4               150                            Ferricrete      200 000 @ 70 kN, 670 kPa
                                                2096
                                                                              112 522 @ 70 kN, 670 kPa, wet1
                                                                              429 000 @ 80 kN, 800 kPa
                                                 5.1
           407A4/A4A2          150                            Ferricrete      463 859 @ 80 kN, 800 kPa
                                                1968
                                                                              248 292 @ 80 kN, 800 kPa, wet3
                                                 6.2                          765 000 @ 40 kN, 520 kPa
           408A4               150                            Ferricrete
                                                2087                          147 640 @ 40 kN, 520 kPa, wet3
           1.    Water added near points 8 to 12 through core holes
           2.    Section moved 3 metres after 430 000 repetitions due to excessive deformation around one MDD 12
           3.    Water added to whole section with spray bar mounted on the wheels

            MDDs were installed in all the test sections and both elastic and plastic responses were
            measured. The data were used in the same manner as those from Heilbron, described in
            Section 3.1.1 for the elastic stiffness, and Section 3.1.2 for the permanent deformation
            response.

3.2.1       Elastic Stiffness


            The backcalculated elastic stiffness data for the emulsified bitumen treated base layers
            from all MDDs on the Cullinan test sections are shown in Figure 3. Despite the variability,
            reductions in stiffness from the beginning of the tests are shown. The initial stiffnesses
            for some sections are not high, less than 1000 MPa, but even these sections show
            decreases in stiffness under the action of loading. The stiffnesses of the longer tests
            (407A4/A4A and 408A4) stabilise at the end of the test to a value between 200 and
            300 MPa. The stiffnesses at the end of some of the shorter tests are slightly higher,

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                            17
Report number CR-2003/44
            although these may have reduced further under additional loading. The steady state
            stiffness of these sections is greater than the Heilbron sections, which may be due to the
            different parent materials. The Cullinan data also do not show as much load dependency
            as the Heilbron data. These data are discussed further in Chapter 7.



3.2.2       Permanent Deformation


            The permanent deformations measured with the MDDs were used to manually fit the
            model to determine the permanent deformation in the base layer. Equation (1) was used,
            as described in Section 3.1.2. The coefficients for the model fits are shown in Table 6.
            The models were used to calculate the number of load repetitions (structural capacity) to
            specified levels of plastic strain. These data are used in Chapter 8.

3.3.        Vereeniging (P243/1)

            The Vereeniging HVS test sections on Road P243/1 were constructed with 2 per cent
            cement and 1.8 per cent residual binder. Two sections were treated with foamed bitumen
            and the other two with emulsified bitumen (3 per cent emulsion for 1.8 per cent residual
            binder). The 40 kN emulsion treated section (412A4) had stronger support than the other
            sections, whereas the 40 kN foamed bitumen treated section (411A4) had the weakest
            support. The 80 kN emulsion treated section (410A4/B4) had slightly stronger support
            than the 80 kN foamed bitumen treated section (409A4/B4)24. Only the emulsified
            bitumen treated HVS sections are discussed. The test sections were constructed using
            DISR on 8 and 9 June 2000 and HVS testing on Section 410A4 began on 26 September
            2000. This should have been sufficient time for the emulsion to break, which is confirmed
            by the high resilient moduli achieved in the base layer.

            The structure of the rehabilitated pavements comprises 25 mm asphalt concrete
            surfacing, 250 mm of emulsified bitumen treated material, 65 mm cement treated
            ferricrete subbase with a G6 classification, and the natural subgrade. The base layer
            consists of the recycled cement treated ferricrete with portions of the multiple seal
            surfacing and some of the untreated ferricrete subbase. Table 7 gives the actual
            thickness of the base layers for the two emulsified bitumen treated HVS test sections, the
            moisture content and density of the layers (determined from measurements at the start of
            testing or from test pits after testing).




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            18
Report number CR-2003/44
                           1800                                                                                                        1800
                                                                                   397 MDD4 40kN                                                                                                        407 MDD4 80kN
                           1600                                                                                                        1600
                                                                                   397 MDD12 40kN                                                                                                       407 MDD8 80kN
                           1400                                                                                                        1400
                                                                                                                                                                                                        407 MDD12 80kN




                                                                                                             Resilient Modulus (MPa)
 Resilient Modulus (MPa)




                           1200                                                                                                        1200

                           1000                                                                                                        1000

                            800                                                                                                         800

                            600                                                                                                         600

                            400                                                                                                         400

                            200                                                                                                         200

                              0                                                                                                           0
                                  0      100000    200000     300000      400000       500000       600000                                    0      200000         400000      600000         800000      1000000        1200000
                                                            Repetitions                                                                                                       Repetitions
                           1800                                                                                                        1800
                                                                                   403 MDD4 40kN
                           1600                                                    403 MDD8 40kN                                       1600                                                             408 MDD4 40kN
                                                                                   403 MDD12 40kN
                           1400                                                                                                        1400                                                             408 MDD12 40kN
 Resilient Modulus (MPa)




                                                                                                             Resilient Modulus (MPa)
                           1200                                                                                                        1200

                           1000                                                                                                        1000

                            800                                                                                                         800

                            600                                                                                                         600

                            400                                                                                                         400

                            200                                                                                                         200

                              0                                                                                                           0
                                  0      100000    200000     300000      400000       500000       600000                                    0   100000   200000   300000   400000   500000   600000   700000   800000    900000
                                                            Repetitions                                                                                                        Repetitions
                                      Figure 3.   Backcalculated stiffnesses from Cullinan (D2388) HVS Test Sections




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                           Report number CR-2003/44                                                      19
            Table 6.       Model fits for permanent deformation response, Cullinan

                                                   Module                Coefficients for Equation (1)
                           Load
            Section                   MDD          depth
                           (kN)                                         a                 b            M
                                                    (mm)
                                                     25               0.65             5.70E-05     1.32E-06
             397A4          40        MDD8
                                                     125              0.37             2.50E-05     7.00E-07
                                                     25               0.78             2.40E-04     3.50E-06
                                      MDD4
                                                     175              0.10             2.00E-04     2.40E-06
                                                     25               0.83             8.00E-05     1.90E-06
             403A4          40        MDD8
                                                     175              0.25             5.00E-05     1.50E-06
                                                     25               0.77             4.70E-05     1.50E-06
                                     MDD12
                                                     175              0.35             4.00E-05     1.10E-06
                                                     25               8.10             4.00E-05     9.00E-06
                                      MDD4
                                                     175              5.60             4.40E-05     6.50E-06
                                                     25               6.50             4.50E-05     7.30E-06
             407A4          80        MDD8
                                                     175              3.80             4.50E-05     5.70E-06
                                                     25               6.80             4.70E-05     7.50E-06
                                     MDD12
                                                     175              4.60             4.70E-05     5.20E-06
                                                     25               3.79             8.40E-06     1.89E-06
                                      MDD4
                                                     175              1.44             6.10E-06     8.50E-07
             408A4          40
                                                     25               3.39             1.61E-05     4.20E-06
                                      MDD8
                                                     175              1.45             6.00E-06     2.00E-07

            The loading sequence used in the testing is also shown in Table 7. During the last
            approximately 15 000 load repetitions water was applied to the cracked test sections
            through a spray bar attached to the wheel of the HVS.

            Table 7.       Vereeniging emulsified bitumen treated material HVS tests

                    Test         Thickness1      Moisture          Density
                                                                                 Loading sequence
                  number            (mm)        content (%)        (kg/m3)
                   410A4             287                                         295 617 @ 80kN, 800 kPa
                                                    14.47            1947        171 500@ 100 kN, 850 kPa
                  410B42            274
                                                                                 13 907@ 100 kN, 850 kPa, wet3
                                                                                 958 714 @ 40kN, 620 kPa
                  412A4             277             10.43            2028        340 883 @ 80 kN, 800 kPa
                                                                                 17 610 @ 80 kN, 800 kPa, wet3
             1.      Actual thickness measured in test pits
             2.      Section moved 4 meters after completion of Section 410A4
             3.      Water added through spray bar attached to HVS trafficking wheel

            Two MDDs were installed in Sections 410A4 and 410B4 and three in Section 411A4 and
            recorded both elastic deflections and permanent deformation. MDD12 in Section 410A4
            was renamed MDD4 in Section 410B4 as a result of moving the section. The other
            standard HVS instrumentation was also used to measure various pavement
            responses19, 20. Detailed material data are included in References 21 and 22.

3.3.1       Elastic Stiffness


            The in depth deflections measured with the MDDs were used to backcalculate effective
            layer stiffnesses15, 19, 25. The stiffnesses are shown in Figure 1. The test sections showed
            high early stiffnesses in the recycled ETBM base layer, ranging from 1000 to 2500 MPa.
            As the tests progress, the elastic stiffnesses of the base layers reduced to the steady
            states (defined in Section 3.1.1).


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                        20
Report number CR-2003/44
                                       3 500

                                                                                                  410 MDD4 80kN
                                       3 000                                                      410 MDD12/4 80kN/100kN
                                                                                                  410B4 MDD4 100kN
             Resilient Modulus (MPa)
                                       2 500


                                       2 000


                                       1 500


                                       1 000


                                        500

                                                        80 kN     100 kN   wet                    40 kN   80 kN
                                         -
                                               -        200 000     400 000      600 000   800 000    1 000 000   1 200 000   1 400 000
                                                                                    Repetitions

                                       3 500
                                                                                                          412 MDD4 40kN
                                       3 000                                                              412 MDD8 40kN
                                                                                                          412 MDD12 40kN
             Resilient Modulus (MPa)




                                       2 500


                                       2 000


                                       1 500


                                       1 000


                                        500
                                                                                                  40 kN   80 kN

                                         -
                                               -        200 000     400 000      600 000   800 000    1 000 000 1 200 000 1 400 000
                                                                                    Repetitions

            Figure 4.                              Backcalculated stiffnesses from Vereeniging HVS Test Sections

            This reduction was more rapid in the early stages of the test, and then slower during the
            remainder of the test. When the load was increased from 40 kN to 80 kN, the stiffness
            reduced further. For Section 412A4 under the 40 kN load, the stiffnesses continued to
            decrease gradually and did not reach a terminal value. When the load was increased to
            80 kN, the base stiffnesses decreased to between 300 and 500 MPa. The terminal
            values under the 80 kN loads on Sections 409A4 and 410A4 were approximately the
            same as the corresponding 40 kN sections, and increasing the load to 100 kN did not
            result in further decreases in the elastic stiffnesses. This indicates that, regardless of the
            load, the treated materials ultimately reach a steady state stiffness. From the trend in the
            data, it thus seems reasonable to assume that under the 40 kN loads the same steady
            state stiffness would have been reached.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                           21
Report number CR-2003/44
            The load sensitivity of the materials is shown by the trafficking load determining the
            number of repetitions to reach the steady state stiffness. The stiffness data are
            discussed in more detail in Chapter 7.

3.3.2       Permanent Deformation


            The in-depth permanent deformation data measured with the MDDs were used to assess
            the permanent deformation response. The regression models (Equation (1)) determined
            for the data in a previous study were used in the analyses in this report19. It is difficult to
            use the data at high wheel loads from a test section that was initially trafficked at a lower
            wheel load because of the load history. The coefficients for the model fits are given in
            Table 8. These model fits were used to calculate the number of HVS load repetitions to
            reach specified levels of plastic strain. These data are used in Section 8.3.

            Table 8.       Model fits for permanent deformation response, Vereeniging

                                                    Module           Coefficients for Equation (1)
                          Load
            Section                    MDD          depth
                          (kN)                                      a               b               M
                                                     (mm)
                                                      20           1.00         3.10E-05        7.50E-06
                                      MDD4
                                                      275          0.40         5.00E-05        1.20E-06
             410A4          80
                                                      20           0.65         4.00E-05        7.00E-06
                                      MDD12
                                                      275          0.12         4.00E-05        1.60E-06
                                                      20           0.40         9.00E-05        4.80E-05
             410B4         100        MDD12
                                                      275          0.40         9.00E-05        9.00E-06
                                                      20           0.95         5.00E-04        3.50E-07
                                      MDD4
                                                      275          0.03         5.00E-04        1.90E-07
                                                      20           1.30         5.00E-04        4.00E-07
             412A4          40        MDD8
                                                      275          0.11         5.00E-04        3.00E-07
                                                      20           0.11         5.00E-04        4.00E-07
                                      MDD12
                                                      275          0.04         5.00E-04        2.80E-07

3.4.        Summary of HVS Test Sections

            A summary of the HVS test section materials, treatment, pavement structures and the
            number of HVS test sections is shown in Table 9.

            Table 9.       Summary of HVS test sections

                       Property                  Heilbron (P9/3)    Cullinan (D2388)         Vereeniging
                                                                                                (P243/1)
            Material                            Weathered          Sandstone (G5)          Ferricrete (G7)*
                                                dolerite (G4/G5)
            Treatment                           2.5% lime **       1% cement               2% cement
                                                1 – 3 % emulsion   1.5% emulsion           3.0% emulsion
            Pavement          Surfacing         25 mm AC           30 mm AC                25 mm AC
            structure         Base              150 mm EBTM        100-150 mm EBTM         250 mm EBTM
                              Subbase           150 mm CTB         300 mm selected***      65 mm old CTB
                                                300 mm selected    Subgrade                Subgrade
                                                Subgrade
            Number of HVS sections                       6                  4                      3
            * previously cement treated
            ** pretreated
            *** sandstone on some sections, ferricrete on others




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                    22
Report number CR-2003/44
3.5.        Selection of Appropriate Transfer Functions

            The data presented in this chapter provide some justification for the choice of the
            applicable modes of distress, and for the development of the associated transfer
            functions, which are described in Chapters 7 and 8.

            The backcalculated MDD data from the Heilbron, Cullinan and Vereeniging HVS test
            sections show that in almost all cases, the emulsified bitumen treated base layers begin
            with a high initial stiffness, which in most cases, reduces under the action of loading to a
            constant stiffness, termed the steady state stiffness. This occurs relatively early in the life
            of the pavement, and can occur rapidly under heavy traffic loading. A stiffness reduction
            phase life transfer function similar to that of foamed bitumen treated materials is therefore
            appropriate4, 7. This transfer function is discussed in more detail in Chapter 7.

            The data also show the accumulation of permanent deformation in the test sections. A
            permanent deformation transfer function similar to that used for foamed bitumen treated
            materials, lightly cemented materials and granular materials is appropriate4,7. This is
            discussed in more detail in Chapter 8. The accumulation of permanent deformation in the
            emulsified bitumen treated base layer occurs over a much longer time than the stiffness
            reduction phase life. Consequently, most of the permanent deformation occurs while the
            material is in the steady state stiffness, although some small amount occurs during the
            stiffness reduction phase. It is difficult to consider both modes of distress simultaneously
            and, because the permanent deformation occurs predominantly after the stiffness
            reduction phase life, the two phases are considered to occur in sequence in the design
            and analyses procedure. This concurs with the recommended design procedures for
            foamed bitumen and lightly cemented materials4, 8.

            Although the stiffness reduction phase life is not a terminal condition, the reduced
            stiffness increases the tensile strain at the bottom of the asphalt layer, and therefore
            reduces the life of the surfacing. The shorter the stiffness reduction phase life, the
            quicker the surfacing will crack and water will be able to penetrate the base layer. This
            would result in an increase in the maintenance frequency. Routine maintenance for crack
            sealing or the application of an additional surface seal may be necessary prior to reaching
            the terminal structural distress condition, i.e., 20 mm of rutting in the base, subbase or
            subgrade. It is therefore beneficial to use additives that provide a long stiffness reduction
            phase life without sacrificing the permanent deformation resistance. This would be
            achieved by ensuring sufficient emulsified bitumen is used in the mix.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 23
Report number CR-2003/44
4.          EXISTING LABORATORY TESTS


            This chapter describes the laboratory testing that was performed as part of the HVS tests
            discussed in Chapter 3. The discussion focuses on the Cullinan and Vereeniging
            laboratory testing results as little appropriate laboratory test data were available for the
            Heilbron sections. The first part of this chapter discusses the calculation of the relative
            density, saturation and stress ratio values used in the analyses described in this report.

4.1.        Calculation of Relative Density and Saturation

            In the laboratory data described in previous reports7, 13, 21, 22, 26 the relative densities and
            saturation levels were calculated considering the aggregate component only. It is more
            appropriate to calculate these parameters including the aggregate, cement and binder.

            The relative density (RD) refers to the ratio of the dry density to the apparent density, as
            shown in Equation (2). The dry density is calculated using the total mass of the
            aggregate, cement and binder, and the volume of the specimen. The apparent relative
            density (ARDmix) is determined using the individual ARD values for the aggregate, cement
            and binder. The saturation level (S) is calculated using Equation (3). For all the analyses
            in this report, the aggregate, cement and binder will be included in the relative density
            and saturation values, unless otherwise stated.

                                                   DDmix
                                          RD =               × 100                                       (2)
                                                 ARDmix × Dw

                           where RD        = relative density (%)
                                 DDmix     = dry density of mix (kg/m3)
                                 ARD       = apparent relative density of mix including cement,
                                             aggregate and binder
                                    Dw     = density of water (kg/m3)


                                                  MC
                                                     ⋅ DDmix
                                          S=      100
                                                             × 100                                       (3)
                                                 (1 − ) × Dw
                                                        RD
                                                        100



                           where S         = saturation (%)
                                 MC        = moisture content (%)
                                 DDmix     = dry density of mix (kg/m3)
                                 RD        = relative density of mix (Equation (1))



4.2.        Calculation of Stress Ratio

            The stress ratio (SR) is calculated using Equation (4), from the applied major principal
                    a
                     ( )
            stress σ1 , the maximum allowable major principal stress σ1
                                                                      m
                                                                             ( )      and the minor principal

            stress   ( σ3 ) .   The maximum allowable major principal stress and the minor principal
            stress can be replaced with a formulation using the cohesion (C) and friction angle (ϕ), as
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              24
Report number CR-2003/44
            shown in Equation (4). In the triaxial test, the minor principal stress is the confining
            pressure, and the major principal stress is the total vertical pressure applied on the
            specimen. The derivation of the stress ratio is fully discussed by Theyse13.

                                                                  σ1 − σ3
                                                                   a
                                                                                         σ1 − σ3
                                                                                          a
                                                           SR =           =                                                              (4)
                                                                             (     (        ) )           (
                                                                  σ1 − σ3 σ3 tan2 45° + ϕ − 1 + 2C tan 45° +
                                                                   m
                                                                                        2
                                                                                                                 ϕ
                                                                                                                 2   )

4.3.        Heilbron Laboratory Test Data

            The laboratory testing program performed in conjunction with the Heilbron HVS tests was
            geared towards the original criteria for a granular emulsion mix (GEM), i.e., tests typically
            used for asphalt. As such, the laboratory tests included the Indirect Tensile Strength
            (ITS), Marshall stability and flow tests. The Marshall tests are no longer considered
            appropriate for emulsified bitumen treated materials, and are therefore not discussed in
            this report.

            The material was a weathered dolerite with a G4/G5 classification. The high PI (8.6) was
            reduced by pre-treating with 2.5 per cent lime. The emulsified bitumen contents ranged
            between 1 and 3 per cent.

            The ITS results were obtained for various emulsion contents. Three different curing
            methods were used, as shown in Figure 5. The ITS values show an optimum emulsion
            content of 3 per cent. The ITS values show a significant increase with increased curing
            time. The higher ITS values experienced after 1 month of field curing could be due to the
            stabilising effect of the lime. The ITS value at 1.2 per cent residual binder appears
            unreasonably high.


                                               700
                                                                                                                     20 hours at 40C
                                               600                                                                   1 week at 23C
                                                                                                                     1 month field
             Indirect Tensile Strength (kPa)




                                               500


                                               400


                                               300


                                               200


                                               100


                                                0
                                                     0.0           0.5       1.0         1.5        2.0          2.5         3.0       3.5
                                                                                   Residual binder content (%)
            Figure 5.                                       Average ITS results from Heilbron laboratory testing16




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                  25
Report number CR-2003/44
4.4.        Cullinan Laboratory Test Data

            The Cullinan laboratory testing program was performed using material obtained from the
            HVS test sections during construction, and treated with 1 per cent cement and various
            emulsified bitumen contents in the laboratory. The properties of the emulsified bitumen
            treated materials measured in the laboratory are given in Table 10 for the sandstone
            materials. Unconfined compressive strength (UCS), flexural beam tests, and static and
            dynamic triaxial tests were performed.

            Table 10.    Material Properties of Emulsified Bitumen Treated Material from
                         Cullinan

             Material property                          Sandstone
             Maximum dry density (kg/m3)                  2158
             Apparent density (kg/m3)                     2914
             Optimum moisture content (%)                  7.5

4.4.1       Unconfined Compressive Strength (UCS)


            The UCS tests were performed at 0.9 per cent residual binder and 1 per cent cement13.
            The results are shown in Table 11. The material is classified according to the SABITA
            Manual 21 classification (Table 1)1. The material has a UCS lower than the minimum
            requirement for an E2 ETB material.

            Table 11.    UCS results from Cullinan tests

                                          Residual binder (%)       UCS (kPa)        Classification
             Sandstone                           0.9                  511                <E2

4.4.2       Flexural Beam Tests


            Table 12 shows the emulsified bitumen contents and strain-at-break values for the
            flexural beam tests on the Cullinan material. The strain-at-break results show significant
            variability. This is typical of the test, and is one reason why six specimens are normally
            used at each cement and binder content.

            Table 12.    Cullinan flexural beam test data

              Emulsified bitumen content, %             Strain-at-break
               (Residual binder content, %)              (microstrain)
                          1.0 (0.6)                          175
                                                  66, 94, 114, 117, 141, 150
                          1.5 (0.9)
                                                       (average = 114)
                          2.5 (1.5)                          263

            The strain-at-break of the mix with 1.0 per cent emulsion is higher than that of 1.5 per
            cent emulsion. This does not follow the trend observed with similar materials, and the
            trend between the mixes with 1.5 and 2.5 per cent emulsion. Only one flexural beam test
            was tested at 1.0 and 2.5 per cent emulsion and, because of the variability typically
            experienced in strain-at-break test, the result at 1.0 per cent emulsion may not be
            representative.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            26
Report number CR-2003/44
4.4.3       Triaxial Tests


            The Cullinan triaxial tests were performed at two residual binder contents, namely 0.9 and
            1.5 per cent (by mass). The target relative density was 73.8 per cent, based on the
            aggregate content only. The corresponding relative densities including aggregate,
            cement and binder were 76.4 and 77.7 per cent for 0.9 and 1.5 per cent residual binder,
            respectively. The target saturation levels (including all three components) were 39.5 and
            39.7% for 0.9 and 1.5% binder, respectively. This saturation corresponds to the
            approximate saturation in the HVS test sections 397A4 and 403A4. However, the other
            test sections (407A4/A4A and 408A4) were tested immediately after a heavy rainy
            season and the saturation levels were slightly higher for 407A4/A4A and significantly
            higher for 408A4.

            Both static and dynamic triaxial tests were performed. The static triaxial tests were used
            to assess the shear strength of the material. The dynamic triaxial test was used to
            determine the resilient modulus and to measure the permanent deformation response.

            The cohesion and friction angle results from the Cullinan static triaxial tests are shown in
            Table 13. Four specimens were used at each binder content, and the R2 of the fit of the
            failure envelopes to the data is greater than 0.99. The addition of more binder increases
            the cohesion, but decreases the friction angle. The relative density of the mix with 1.5 per
            cent binder is slightly higher, which may contribute slightly to the higher cohesion.

            Table 13.        Shear strength parameters for Cullinan (D2388) material with 1 per cent
                             cement

            Residual binder              Saturation    Relative       Cohesion        Friction angle
              content (%)                   (%)       Density (%)       (kPa)           (degrees)
                  0.9                       39.5         76.4          123.8               45.4
                  1.5                       39.7         77.7          167.8               43.5

            Six dynamic triaxial tests, each on a new specimen, were performed for each mix, at the
            same target relative density and saturation values as the static tests. At each of two
            confining pressures (80 and 140 kPa), three stress ratios (0.2, 0.55, 0.9) were used in the
            testing. These tests are described in more detail by Theyse13. The resilient moduli
            obtained from the tests are shown in Table 14. The values are very similar for both mixes
            and are in the same range as the initial stiffnesses from the HVS test sections (see
            Figure 3). These results are compared with the other mixes in Section 6.4.

            Table 14.        Average resilient moduli for Cullinan (D2388) material with 1 per cent
                             cement

                                         Saturation    Relative
                                                                       Resilient Modulus (MPa)
            Residual binder                 (%)       Density (%)
              content (%)                                                        Standard deviation
                                          Average      Average      Average
                                                                                     (COV,%)*
                      0.9                   39.7         77.7         1017            157 (16)
                      1.5                   39.5         76.4         1059            157 (15)
            * coefficient of variation


            The permanent deformation of the dynamic triaxial specimens is measured by 3 linear
            variable displacement transducers (LVDTs), one placed in the loading section of the
            machine, and the other two on the specimens. It is preferable to place 3 LVDTs on each
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               27
Report number CR-2003/44
            specimen. The best fit models for the permanent deformation data from each LVDT are
            developed in the same manner as described for the HVS permanent deformation data in
            Section 3.1.2. Only the best fits to one LVDT were reported with the laboratory test data
            in reference 13. As part of this project, the model was fit to the remaining data and the
            results are presented in Section 6.3. Discussion of the permanent deformation from the
            dynamic triaxial tests is therefore deferred until that section.


4.5.        Vereeniging (P243/1)

            The material used for the testing was obtained from Road P243/1 and is predominantly
            an old cement treated ferricrete, with some of the multiple surfacing seals and some of
            the untreated ferricrete subbase21.

            In previous work21, 22, 26, the optimum moisture content and maximum dry density values
            used were obtained from the consultant. A sample of the material used for the laboratory
            testing was used to repeat the determination of the optimum moisture content and
            maximum dry density in 2002, at the Transportek laboratory. Comparisons of the
            consultant’s and Transportek’s values are shown in Table 15. It is possible that the
            consultants’s values were not obtained from material behind the recycler, which may
            account for some of the differences in values. Because the Transportek tests were done
            on representative samples of material used for the ongoing specimen preparation, these
            results are more appropriate, and are used in all analyses described in this report. The
            data from all the existing laboratory test results were recalculated using the Transportek
            values.

            Table 15.   Optimum moisture content and maximum dry density values for
                        Vereeniging material

                              Optimum moisture content (%)          Maximum dry density (kg/m3)
             Consultant                  12.5                                 1971
             Transportek                 9.1                                  2025

            Table 16 summarizes the laboratory testing and the residual binder and cement contents
            of the various mixes. Although a wider range of cement and emulsified bitumen contents
            was used in most of the laboratory tests, only one combination was used in the triaxial
            tests. The CBR, UCS, ITS and flexural beam tests were compacted to 100 per cent of
            modified AASHTO maximum dry density at optimum moisture content. The static and
            dynamic triaxial tests were performed at two target densities, 67 and 73 per cent of ARD,
            and two target saturation levels, 45 and 75 per cent. The CBR tests are not discussed in
            this report. Detailed discussion of the results is contained in Reference 22.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            28
Report number CR-2003/44
            Table 16.                 Laboratory testing for Vereeniging (P243/1) emulsified bitumen treated
                                      material tests

                                                              Residual binder content (%)
                                              0.6                 1.8               3.0            5.0
                                                           CBR, UCS, ITS, CBR, UCS, ITS,
                                  0     CBR, UCS, ITS,
             Cement content
                                                            flexural beam     flexural beam
                                                               UCS, ITS,        UCS, ITS,
                                  1        UCS, ITS,                                           flexural beam
                                                            flexural beam     flexural beam
                 (%)


                                                               UCS, ITS,
                                                            flexural beam,      UCS, ITS,
                                  2
                                                             dynamic and      flexural beam
                                                             static triaxial


4.5.1       UCS and ITS Data


            The UCS and ITS values for the combinations of cement and residual binder contents are
            shown in Figure 6. The initial consumption of stabiliser (ICS) for this material was
            approximately 1 per cent. In Reference 22, it was postulated that at cement contents less
            than the ICS, the addition of binder increases the ITS and UCS, whereas when the
            cement content exceeds the ICS, additional binder reduces the ITS and UCS. This
            observation is taken from only two data points, therefore this result may not be a general
            trend. At a fixed binder content, the addition of cement improves both the UCS and ITS.
            It is interesting to note that the addition of binder or cement has the same effect on the
            UCS and ITS, indicating that these tests are measuring the same properties.


                           2500                                                                                350
                                          UCS, no cement
                                          UCS, 1% cement
                                                                                                               300
                           2000           UCS, 2% cement
                                          ITS, no cement
                                          ITS, 1% cement                                                       250
                                          ITS, 2% cement
                           1500
               UCS (kPa)




                                                                                                                        ITS (kPa)
                                                                                                               200


                                                                                                               150
                           1000

                                                                                                               100
                            500
                                                                                                               50


                              0                                                                                0
                                  0         0.5        1         1.5         2           2.5   3         3.5
                                                           Residual binder content (%)
            Figure 6.                 UCS and ITS results for Vereeniging material




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                            29
Report number CR-2003/44
4.5.2       Flexural Beam Data


            The strain-at-break values obtained from the flexural beam tests are summarized in
            Table 17. No results were obtained for the specimens with no cement, because the
            beams broke before testing. The results show that the addition of binder to the mix with 1
            per cent cement significantly increases the strain-at-break, however, the same amount of
            additional binder does not result in such a large increase with the mix with 2 per cent
            cement. This demonstrates that to achieve a flexible mix, the cement content should not
            be too high, in this case it should probably not exceed 1 per cent.


            Table 17.        Flexural beam test results for Vereeniging material

                                                                 Strain-at-break (microstrain)
              Cement content              Residual binder
                                                                                  Standard deviation
                  (%)                       content (%)         Average
                                                                                      (COV,%*)
                                                1.8               243                  39 (16)
                        1
                                                3.0               553                  66 (12)
                                                1.8               178                  31 (18)
                        2
                                                3.0               196                  63 (32)

4.5.3       Triaxial Data


            The Vereeniging triaxial tests were performed on a mix with 2 per cent cement and 1.8
            per cent residual binder at 2 saturation levels and 2 relative densities. The same loading
            conditions as used for the Cullinan triaxial tests were used. These triaxial data showed
            more variability than would typically be expected. This was probably due to the target
            densities and saturation levels not being as closely achieved as for subsequent triaxial
            tests, and some problems were experienced with the loading during the testing21. The
            problems were resolved before later testing.

            The shear strength parameters, cohesion and friction angle, are shown in Table 18. The
            results are difficult to interpret, as at the lower relative densities an increase in saturation
            causes an increase in the cohesion and decrease in the friction angle, whereas the
            opposite is apparent at the higher relative densities. The interaction between the relative
            density, saturation, cohesion and friction angle may be due to model effects in the way
            the cohesion and friction angle values are obtained from the Mohr-Coulomb failure
            envelope. This is discussed further in Section 5.5.1.

            Table 18.        Shear strength parameters for Vereeniging (P243/1) material with 2 per
                             cent cement and 1.8 per cent residual binder

                   Saturation (%)                  Relative Density (%)
                                                                                               Friction
                            Standard                           Standard       Cohesion
                                                                                                angle
               Average      deviation             Average      deviation        (kPa)
                                                                                              (degrees)
                            (COV,%)*                           (COV,%)*
                 64.5        1.94 (3)               72.4        0.65 (1)        541.5            32.5
                 59.0        2.95 (5)               71.5        0.36 (1)        326.8            46.6
                 62.8        5.65 (9)               75.3        0.68 (1)        455.7            48.3
                 58.2        3.49 (6)               74.4        0.67 (1)        579.6            40.1
             * coefficient of variation


            The resilient moduli obtained from the dynamic triaxial tests are summarized in Table 19.
            Because the resilient moduli are high, it can be assumed that the emulsion had broken
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                  30
Report number CR-2003/44
            fully. Such high values would not have been obtained if the bitumen was still in the
            emulsified state. These values are high, and are in the same range as the initial
            stiffnesses determined in the HVS tests, as illustrated in Figure 1. At 79 per cent relative
            density, a decrease in saturation increases the resilient modulus, as expected. However,
            at 76 per cent relative density, decreased saturation results in a lower resilient modulus,
            which is unexpected. These values can be used in structural design.

            Table 19.        Average resilient moduli for Vereeniging (P243/1) material with 2 per
                             cent cement and 1.8 per cent residual binder

                                                                                Resilient Modulus
            Residual               Saturation (%)      Relative Density (%)
                                                                                      (MPa)
             binder
                                           Standard               Standard                Standard
            content
                              Average      deviation   Average    deviation    Average    deviation
               (%)
                                           (COV,%)*               (COV,%)*                (COV,%)*
                                 92.5        5.3 (6)    76.1        0.8 (1)     2683       613 (23)
                                 69.4        3.5 (5)    76.6        0.5 (1)     2367       691 (29)
                1.8
                                104.1        7.3 (7)    79.3        1.2 (2)     1721       621 (36)
                                 88.8        5.0 (6)    79.7        0.1 (0)     2005       266 (13)
            * coefficient of variation


4.6.        Summary

            This chapter presented the laboratory test data available for developing the structural
            design models for emulsified bitumen treated materials prior to initiation of this project.
            Little relevant material data is available for the Heilbron HVS test sections, and therefore
            the data cannot be used to develop the complete transfer functions. Suitable data are
            available from both the Cullinan and Vereeniging test sections, namely strain-at-break
            results, static and dynamic triaxial test results, and resilient moduli. From the data
            described, some limitations in the data were identified, and additional testing was
            performed to supplement the results. The additional testing and results are detailed in
            Chapter 5. Additional analyses on the data presented in this chapter are described in
            Chapter 6, where the laboratory test data are also summarized and compared with the
            new data.

            The UCS and ITS data are used to develop the material classification system in Section
            9.1.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              31
Report number CR-2003/44
5.          ADDITIONAL LABORATORY TESTING


5.1.        Justification for Additional Laboratory Tests

            Although several series of triaxial tests were performed as part of the previous projects,
            described in Chapter 4, some limitations in the data reduce their effectiveness for
            developing permanent deformation structural design models. The Vereeniging (P243/1)
            data were tested at various densities and saturation levels, and can therefore be used to
            fit models as functions of these variables. However, all of the tests were done at one
            cement content (2 per cent) and one residual bitumen content (1.8 per cent), and this
            combination is not necessarily the optimal combination. It is therefore, difficult to justify
            use of the models for other combinations of binder and cement contents with confidence.
            The Cullinan (D2388) triaxial tests were performed at one saturation and density level,
            and are consequently difficult to extrapolate to other density and saturation levels, such
            as the higher saturation level on the last Cullinan section tested.

            An additional set of triaxial data was therefore sought for comparison with existing data to
            quantify the effects of the cement and emulsified bitumen contents. This was possible by
            performing triaxial tests on available Vereeniging material, as no untreated material was
            available from the Cullinan test sections.

            One set of triaxial data was already available for a cement to bitumen ratio of 1.11 for
            both the Cullinan and Vereeniging data, although the actual cement and bitumen
            contents differ. If a set of triaxial tests is performed at a different cement:binder ratio, and
            at two density and two saturation levels, and the previously observed trends in the
            permanent deformation behaviour are apparent, the data from Cullinan and Vereeniging
            can be combined to develop one model. Without additional data, it would not have been
            possible to establish the predominant cause of differences in the permanent deformation
            behaviour at Cullinan and Vereeniging. In particular, the additional set of triaxial tests
            would be useful to determine whether differences were due to the cement and binder
            contents, or to the parent material.

            Two different cement and residual binder contents were thus assessed in the additional
            triaxial tests. One of these combinations was the same as one set of the Cullinan triaxial
            tests, i.e., 1 per cent cement and 0.9 per cent residual binder. The second combination
            was 1 per cent cement and 3 per cent residual binder. These two combinations allow the
            tests on the Vereeniging material to span the Cullinan tests. The second combination (1
            per cent cement, 3 per cent residual binder) allows comparison with the foamed bitumen
            data, which was also tested at these cement and binder contents, although this
            comparison is not described in this report. The matrix of combinations is illustrated in
            Table 20.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                  32
Report number CR-2003/44
            Table 20.       Additional triaxial tests in terms of cement and residual binder
                            contents

                                                           Residual Binder Content (%)
                                      0.9                     1.5              1.8                               3.0
                                   Cullinan
           Cement
           content                                                                                          New tests:
                        1         New tests:                 Cullinan
             (%)                                                                                           Vereeniging
                                  Vereeniging
                        2                                                          Vereeniging

5.2.        Laboratory Testing Program

            There was sufficient untreated material from Vereeniging for approximately 40 triaxial
            tests, and 10 tests are required at each combination of variables. Thus, for two cement
            and bitumen content ratios, only one density, or only one saturation level could be used.
            The Cullinan triaxial tests were only done at a lower saturation level, and not at the higher
            level more representative of the last test during the second phase of the Cullinan HVS
            testing. Because of this, it was considered beneficial to test at two saturation levels, and
            limit the tests to one density level. The two recommended target saturation levels were
            45 and 75 per cent. The recommended density was 73 per cent of apparent density,
            which is similar to the higher level of density used in the previous Vereeniging and
            Cullinan triaxial tests.

            The target test variables used for the additional tests are shown in Table 21. At each
            combination of the variables shown, four static triaxial tests and 6 dynamic triaxial tests
            were carried out. The four static triaxial tests were done at confining pressures of 20, 80,
            140, 200 kPa. The six dynamic triaxial tests were performed at confining pressures of 80
            and 140 kPa; and at three stress ratios for each confining pressure, namely 0.20, 0.55
            and 0.90.

            Table 21.       Test matrix for additional triaxial tests

              Residual binder              Cement content               Relative Density1                Saturation1
                content (%)                    (%)                             (%)                           (%)
                    3.0                          1                              73                            45
                    3.0                          1                              73                            75
                    0.9                          1                              73                            45
                    0.9                          1                              73                            75
            1. Relative density and saturation calculated as a function of the aggregate proportions only. This is consistent
                                            21
            with the foamed bitumen work . In the analyses that follow, the cement and binder are included in the
            calculation of the relative density.


5.3.        Materials

            The aggregate used in the specimen preparation is described in Section 4.5 of this report.
            The emulsion was a NATREF 60 per cent stable grade anionic emulsion and the cement
            was an Alpha Cem II cement with a strength of 32.5 MPa. This is the same emulsion as
            used previously, although a Cem I cement was used before. These cements are similar,
            although the Cem II cement can contain up to 20 per cent limestone filler.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                    33
Report number CR-2003/44
5.4.        Specimen Preparation

            The specimens were prepared in the Transportek soils laboratory, using the same
            procedures as for the previous tests21, 22, 26. This process is briefly outlined below and the
            specimen preparation data sheets are in included in Appendix B.

5.4.1       Mixing


            The one per cent cement was calculated as a percentage of the mass of the dry
            aggregate. The emulsion content was calculated as a percentage of the total mass of the
            dry aggregate and the cement. The percentage of the emulsion is the target residual
            binder content divided by 0.6, where 0.6 represents the binder proportion of the emulsion.
            The moisture added is the difference between the optimum moisture content and the
            emulsion content, calculated as a percentage of the dry aggregate and cement. This
            means that the optimum fluid content is equal to the optimum moisture content, and is not
            equal to the optimum moisture content plus the emulsion content as specified in Manual
            211. For this testing, all specimens were compacted at this optimum fluid content,
            regardless of the target moisture content for testing. This differs from the previous tests,
            which were compacted at this moisture content only if the target testing moisture content
            was equal to or less than the mixing moisture content. Previously, the specimens were
            prepared at the higher moisture content if the target testing moisture content was higher
            than the optimum fluid content.

            The following sequence was used to mix the materials:
            1. Place the dry the aggregate in the pan (concrete) mixer.
            2. Place the cement in the mixer (mass calculated as percentage of mass of dry
               aggregate).
            3. Mix the aggregate and cement with one-third of the additional water required. The
               mass of water is calculated as a percentage of the dry mass of the aggregate and the
               cement.
            4. Pour the emulsion diluted with one-third of the water into the pan and mix. The
               emulsion content is calculated as a percentage of the mass of dry aggregate and the
               cement.
            5. Rinse the emulsion container with the remaining one-third of water and pour into the
               mixer.
            6. Mix all materials for one minute.

5.4.2       Compaction


            The triaxial test specimens were compacted using a vibratory table. Specimens were
            compacted in three equal lifts. The target density could not be achieved on the vibratory
            table, and therefore the target was reduced to a relative density value of 75 per cent
            (includes aggregate, cement and binder, see Section 4.1). After testing, the actual
            density of the specimen was measured by weighing the specimen and recording the
            dimensions, and this value was used in all analyses, unless otherwise stated.

5.4.3       Curing


            After compaction the specimen was placed in a plastic bag. Before testing, the specimen
            was dried back to the lowest target moisture content selected for that residual binder
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials          34
Report number CR-2003/44
            content, and then, where necessary, water was added to reach the target moisture
            content. If a small amount of water needed to be added, the additional water is sprayed
            onto the surface of the specimen. If a fairly large quantity of water was required, the
            specimen was placed within a membrane and sealed at the base with an elastic band.
            The required water was then poured on top of the specimen. All specimens were allowed
            to dry back to facilitate breaking of the emulsion. This was not done previously, although
            it is believed that even without drying back, the emulsion broke. Drying back was an
            additional measure implemented to facilitate breaking. The specimens are weighed until
            the weight associated with the target moisture content is reached and are then placed in
            double plastic bags and sealed at the top and bottom to prevent any changes in the
            moisture content. The specimen was then left for at least 48 hours before testing for the
            moisture to equilibrate throughout the specimen. After testing, the actual moisture
            contents were measured, and these values are used in the analyses.

            Although the target curing time was 28 days, because a maximum of three dynamic
            triaxial tests can be tested in a day, it was difficult to test all specimens on the 28th day.
            All specimens were, however, cured for at least 28 days and no specimens exceeded 31
            days of curing.

5.5.        Static Triaxial Tests

            The static triaxial test was used to determine the shear strength of the material. In this
            sequence of tests, the shear strength was determined as a function of the residual binder
            content, and the saturation level of the material. The complete data set for these tests is
            given in Appendix B. Traditionally, the shear strength is described in terms of cohesion
            and friction angle, which are determined from Mohr-Coulomb failure envelopes. This
            approach is discussed in Section 5.5.1. In Section 5.5.2, the development of a
            regression function that determines the maximum allowable principal stress without
            determining the cohesion and friction angle is discussed.

5.5.1       Shear Strength Parameters, Cohesion and Friction Angle


            The Mohr-Coulomb failure envelopes from the static triaxial tests are shown in 0. The
            relative density and saturation values are the average of the measured values of
            specimens tested at that combination of variables. The variability in the results was
            relatively low.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                35
Report number CR-2003/44
                      1500                                                                                              1500

 Shear Stress (kPa)




                                                                                                   Shear Stress (kPa)
                      1000                                                                                              1000




                       500                                                                                               500




                         0                                                                                                 0
                             0     500       1000           1500             2000   2500   3000                                0      500    1000           1500             2000   2500   3000
                                                    Principal Stress (kPa)                                                                          Principal Stress (kPa)

(a)                      Residual binder content = 3.0%                                           (c)                      Residual binder content = 0.9%
                         Relative density = 79%, Saturation = 65%                                                          Relative density = 75%, Saturation = 69%
                         Cohesion = 111 kPa, Friction angle = 48º, R2 = 0.999                                              Cohesion = 165 kPa, Friction angle = 45º, R2 = 1.000
                      1500                                                                                              1500
 Shear Stress (kPa)




                                                                                                   Shear Stress (kPa)
                      1000                                                                                              1000




                       500                                                                                               500




                         0                                                                                                 0
                             0     500       1000           1500             2000   2500   3000                                0      500    1000           1500             2000   2500   3000
                                                    Principal Stress (kPa)                                                                          Principal Stress (kPa)

(b)                      Residual binder content = 3.0%                                           (d)                      Residual binder content 0.9%
                         Relative density = 77%, Saturation = 34%                                                          Relative density        74%, Saturation    36%
                         Cohesion = 464 kPa, Friction angle = 38º, R2 = 0.992                                              Mohr-Coloumb failure envelope not determined

                                 Figure 7.     Mohr-Coulomb failure envelopes for Vereeniging materials with 1% cement and 0.9% and 3.0% residual binder




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                            Report number CR-2003/44                                       36
            The data in Figure 7(a) and (b) for 3 per cent residual binder mix are good, and show
            strong trends. At the lower saturation level, the major principal stress increases with an
            increase in confining pressure. The cohesion and friction angles are shown in Table 22
            and Figure 8.

            At 0.9 per cent residual binder content, a similar trend of increasing major principal stress
            as a result of the lower saturation level is evident (Figure 7(c) and (d)). The data,
            however, are not as good as for the 3 per cent residual binder content. In Figure 7(c) the
            data at the highest confining stress was not included in the calculation of the cohesion
            and friction angle. This is because the maximum principal stress for this confining
            pressure was less than expected, and a better fit was obtained by excluding these data.

            The data in Figure 7(d) at the lower saturation level are inconsistent in that an increase in
            the confining pressure does not always result in a consistent increase in the maximum
            principal stress. No explanation for these results could be found in the specimen
            preparation or testing of the specimens. To determine which of the data are the most
            representative, a dynamic triaxial test specimen was sacrificed to repeat the test at the
            highest confining stress, and these data are shown as the turquoise line (indicated by the
            arrow) in Figure 7(d). Unfortunately, this specimen also did not provide a conclusive
            answer on which data are the most representative. Therefore, the data were not used to
            calculate a cohesion or friction angle.

            The cohesion and friction angles given in Table 22 and Figure 8 show that an increase in
            cohesion typically results in a decrease in the friction angle. While this may be
            reasonable for some data, this frequently occurs with the Mohr-Coulomb failure envelope
            for any materials and is an effect of how the linear model is fitted. The small friction
            angles are an overcompensation for the large cohesion values. The friction angle value
            of 38° shown in Table 22 for the mix with 3 per cent binder and 34 per cent saturation is
            unrealistically small, and the cohesion value is large (464 kPa).

            Table 22.    Cohesion and friction angles of Vereeniging (P243/1) materials with 1
                         per cent cement and 0.9 per cent and 3.0 per cent residual binder

               Residual          Cement         Relative                                       Friction
                                                              Saturation      Cohesion
            binder content       content        density                                         angle
                                                                 (%)            (kPa)
                 (%)               (%)            (%)                                             (°)
                                                  78.7           65.3            111            48
                   3.0               1
                                                  77.0           34.3            464            38
                                                  75.2           68.6            164            45
                   0.9               1
                                                  74.1           35.7            Data inconclusive




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               37
Report number CR-2003/44
                              500                                                           60

                              450
                                                                                            50
                              400
                              350
             Cohesion (kPa)                                                                 40




                                                                           Friction angle
                              300
                              250                                                           30

                              200
                                                                                            20
                              150
                              100
                                                                                            10
                               50
                               0                                                             0
                                    3% Binder,   3% Binder, 0.9% Binder,                         3% Binder,   3% Binder,   0.9% Binder,
                                       High         Low         High                                High         Low           High
                                    saturation   saturation saturation                           saturation   saturation    saturation

            Figure 8.                  Cohesion and friction angles of Vereeniging (P243/1) materials with 1
                                       per cent cement and 0.9 per cent and 3.0 per cent residual binder


5.5.2       Regression Model to Calculate the Maximum Allowable Principal Stress


            Because of the difficulties in fitting the cohesion and friction angle results, it is not
            recommended that these values be used in the analyses. An alternative approach is
            presented in this section. The purpose of fitting such values is to obtain an estimate of
            the shear strength of the material. This can also be done by developing a model that
            predicts the maximum stress (major allowable principal stress) as a function of the
            confining stress (minor principal stress). Theyse27 has investigated the development of
            such models for untreated granular materials, and has found that the models work well for
            many materials when a model is fitted to each material. The model can also contain
            other variables that influence the shear strength, such as the relative density, saturation
            and also be extended to include the binder and cement contents. A significant advantage
            of these models is that they are valid for a range of relative densities and saturation
            levels, as opposed to the cohesion and friction angles, which are particular to a specific
            relative density and saturation level. In addition, when the variability of relative densities
            and saturation levels within a set of data is large, it is difficult to realistically assign the
            cohesion and friction angle to that set of data, and to report the values as being
            applicable to the average relative density and saturation levels. This introduces errors
            into calculation of the shear strength of the material.

            The model suggested by Theyse is shown in Equation (5) and was applied to the data for
            each residual binder content. The model coefficients and the relevant statistics are
            shown in Table 23. The p-values indicate the statistical significance of the variable; a
            value of less than 0.01 means the variable is statistically significant at a minimum of the
            90 per cent level. The smaller the p-value, the more statistically significant the variable.
            All data for the 3 per cent residual binder mixes were used to fit the model, whereas at
            0.9 per cent residual binder two specimens were excluded from the data set. These data
            points were excluded because they were obvious outliers when comparing the predicted
            and measured values for the maximum principal stress.


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                        38
Report number CR-2003/44
                                  σ1 = c1 + c 2 × RD + c 3 × SAT + c 4 × σ 3
                                                              1
                                                                                                               (5)
                        where σ1         = maximum (major) allowable principal stress (kPa)
                              RD         = relative density (%)
                              SAT        = saturation (%)
                              σ3         = confining stress or minor principal stress (kPa)
                              c1-4       = coefficients



            Table 23.    Model coefficients and statistics for Equation (5), Vereeninging
                         materials with 1% cement

                                            Equation (5)(a)                          Equation (5)(b)
             Residual binder
                                                   3.0%                                    0.9%
                content
             Cement content                        1%                                        1%
                                     Coefficient          p-value              Coefficient          p-value
            c1 (constant)             -413.68             0.1054                -254.27              0.1649
            c2 (relative density)      Statistically insignificant               Statistically insignificant
            c3 (1/saturation)        75 465.30            0.0002               77 332.58             0.0006
            c4 (σ3)                   5.2974              0.0012                4.5299               0.0044
                      R2                          0.95                                      0.98
             Standard error of
                                               157.8                                       118.1
                the estimate
            Models applicable to these values of the variables
            Relative density                76.6 – 79.2                                 73.2 – 75.9
            Saturation                      32.1 – 67.4                                  33 – 71.6
            Confining stress                770 - 2810                                  900 - 2825


            When the relative density term is included, the term is found to be statistically insignificant
            for both sets of the data, and even had an unrealistic sign, i.e., the sign on the coefficient
            was negative, indicating a decrease in the maximum principal stress with an increase in
            density. The ranges in the relative density values are small, 76.6 to 79.2 per cent for the
            3 per cent binder content data and 73.2 to 75.9 per cent for the 0.9 per cent binder
            content data. This could explain the statistical insignificance of this variable. Because of
            this, the density term was omitted from the model. Although the relative density is a
            major factor in the behaviour of the material, using the data available it cannot be
            included in the model without giving confusing and misleading results. The models
            should not be used for relative densities outside of the range indicated.

            The model fits are reasonable, as indicated by the standard error of the estimate and the
            R2. The signs of the terms are also as expected, with an increase in the confining stress
            and decrease in the saturation level resulting in increases in the maximum principal
            stress.

            When Equations (5)(a) and (b) are used to compare the effects of binder content on the
            maximum principal stress, the results are misleading because of the different values of
            the relative density, which are not accounted for in the comparisons.

            A model combining the data at the two binder content ratios was then attempted. This
            model had the same form as shown in Equation (5), but also included terms for the binder
            and cement contents. Because the range of relative density values within each set of
            binder content data was small, and the range differs between the binder contents, it was
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials           39
Report number CR-2003/44
            not possible to fit one model in which both the relative density and binder content terms
            are statistically significant. However, it will be shown in Section 6.2.2 that this could be
            achieved by combining all of the Vereeniging data at the different cement and binder
            contents.

            Equation (6) is traditionally used to determine the maximum allowable principal stress
            before shear failure. This equation is a function of the cohesion and friction angle.
            Where the variability in relative density and saturation level in the data used to calculate
            the cohesion and friction angle is small, the differences between the predicted maximum
            principal stresses calculated using Equation (5) or Equation (6) are negligible. However,
            Equation (5) can be extrapolated to other relative densities and saturation levels. An
            example of where this is particularly useful is when the static triaxial test results are used
            to calculate the stress ratio used in the dynamic triaxial tests. If the saturation and
            relative densities in the dynamic test are not in exact agreement with the static results,
            use of the cohesion and friction angle values through Equation (6) introduces errors in the
            stress ratio calculation. Equation (5) can, however, estimate the stress ratio for a
            particular saturation and relative density of a particular dynamic triaxial test.


                                   (     (
                           σ1 = σ3 tan2 45° +
                            m                   ϕ
                                                2   ) − 1) + 2C tan ( 45° + ) + σ
                                                                          ϕ
                                                                          2         3                 (6)



5.6.        Dynamic Triaxial Tests

            The dynamic triaxial test is used to determine both the resilient modulus and the
            permanent deformation behaviour of a material. This section discusses these aspects.
            In this set of dynamic triaxial tests, the permanent deformation behaviour can be
            determined as a function of the residual binder content and the saturation level of the
            material. As the relative density values are constant within each set of binder content
            data, the permanent deformation and resilient moduli could not be determined as a
            function of relative density for a particular binder content.

            The complete set of dynamic triaxial test results is summarized in Appendix B, and the
            data from each specimen are also given in Appendix B.

5.6.1       Calculation of Stress Ratio


            The dynamic triaxial tests are performed over a range of stress ratios (0.2, 0.55, 0.9).
            Before testing, the test loads are calculated using the target stress ratio and Equation (6),
            using the cohesion and friction angle values determined from the static triaxial tests.
            Because it was not possible to obtain the cohesion and friction angle for the specimens
            with 0.9 per cent residual binder at low saturation levels, the test stress was estimated as
            the stress ratio percentage of the maximum principal stress for the static test with the
            same confining stress as specified for the dynamic test.

            Once the test had been performed, the actual confining stress and test loads measured
            were used to recalculate the stress ratio. This was done using the relevant cohesion and
            friction angles, and using the applicable maximum principal stress model (Equation (5)(a)
            or (b)). The actual saturation levels measured after testing were used in Equation (5).
            The relative densities were similar to those in the static tests. Therefore, Equation (5)
            models were applicable to these data. The actual saturation levels in the static and
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials           40
Report number CR-2003/44
            dynamic specimens differed, which were not accounted for when using the cohesion and
            friction angles to calculate the stress ratios, thus introducing errors. Both the stress ratios
            are shown in the summary data in Appendix B. Because the maximum allowable
            principal stress models (Equation (5)) account for variations in the saturation levels and
            relative densities, the stress ratios calculated using these models are used in all further
            analyses.

5.6.2       Resilient Moduli


            The permanent deformation dynamic triaxial test is used to determine the resilient
            modulus of the specimens. Specific resilient modulus triaxial tests were not performed.
            Three LVDTs record data for each test, and the data are used to calculate three resilient
            modulus values. The resilient moduli recorded for each specimen are selected from
            ranges within the data when the values are constant. The resilient moduli values are
            summarized in Table 24, with the variables that influence the results.

            Table 24.     Resilient moduli for Vereeniging (P243/1) material with 1 per cent
                          cement
             Sample #   Binder    Confining    Bulk    Stress   Relative   Saturation    Data range            Resilient modulus
                        Content    stress     Stress   Ratio    Density                                    LVDT M LVDT 1 LVDT 2
                          (%)       (kPa)     (kPa)     (%)       (%)         (%)         (repetitions)     (MPa)     (MPa)   (MPa)
              FSS17                   79        368     15.4       78          53       30 000 - 50 000      568       1560    1537
              FSS18                   82        752     43.9       78          54       20 000 - 50 000      640       1167    1257
              FSS19                  138       1566     78.5       78          52       10 000 - 50 000      717       1456    1256
                          3.0
              FSS20                  145        573     18.1       78          56       15 000 - 40 000      635       1214    1699
              FSS21                  142       1070     50.3       77          55       15 000 - 50 000      602       1348    1222
              FSS22                  142       1241     59.3       78          53       15 000 - 50 000      751       1700    1353
              FSS23                   82        615     15.6       78          26       10 000 - 20 000      559       1484    1524
              FSS24                   83       1409     52.8       78          31        6 000 - 30 000      598       1286    1230
              FSS25                   80       2129     66.3       78          25          400 - 1000        632       1242    1444
                          3.0
              FSS26                  144        777     15.8       78          26         400 - 15 000       663       2977    2374
              FSS27                  142       1641     43.6       78          26          300 - 5000        560       1497    1449
              FSS28                  143       2407     69.6       78          27           10 - 7 000       676       1300    2155
              FSS29                   82        399     22.8       75          77         100 - 10 000       412       858      745
              FSS30                   81        838     65.3       75          77       15 000 - 50 000      405       640      618
              FSS31                   82       1237    103.3       75          77           10 - 1 000       328        578     545
                          0.9
              FSS32                  140        582     25.2       75          80        2 000 - 10 000      423       915     1292
              FSS33                  141       1124     59.8       76          66        3 000 - 50 000      452       783      900
              FSS34                  140       1591    100.8       75          73                      Data inconclusive
              FSS35                  139        725     19.5       76          38       15 000 - 50 000      556       1756    1874
              FSS36                   81       1387     66.0       75          42         300 - 30 000       450       1634    1231
              FSS37       0.9         85       2098    103.4       76          43            all data        570       1171    1018
              FSS39                  141       1862     81.2       75          45       6 000 to 50 000      492       1541    1490
              FSS40                  144       2783    126.7       75          45             7, 8, 9        520       865     1040




            The data marked “LVDT M” were recorded by the LVDT mounted on the actual testing
            machine, rather than on the specimen itself. The resilient moduli values recorded by this
            LVDT were consistently lower than the two LVDTs mounted on the specimen, because of
            the position of the LVDT. This is consistent with previous testing. The values from
            “LVDT M” were, therefore, not considered in the analyses of the resilient moduli data.

            Some of the stress ratios in Table 24 exceed one. While this is theoretically impossible,
            such high values are sometimes calculated because of inaccuracies in calculating the
            stress ratio or measuring the density and saturation levels. The dynamic effects of the
            test can also prevent immediate shear failure at stress ratios around one.

            Regression models to allow prediction of the resilient modulus as a function of several
            variables were fitted. Data fits were attempted on the 3.0 and 0.9 per cent residual binder
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                               41
Report number CR-2003/44
            content data both individually and in composite models. Several different model forms
            were attempted, with different combinations of the following variables:
            •     Binder content and ratio of the binder to cement contents
            •     Saturation level
            •     Relative density
            •     Stress ratio
            •     Confining stress
            •     Bulk stress

            It was not possible to fit statistically sound models. The goodness-of-fit was poor in
            almost all cases, many of the variables were statistically insignificant and the signs of the
            coefficients often indicated counter-intuitive trends in the data. Because of the poor fits, a
            regression model to predict the resilient modulus as a function of the material and stress
            state is not recommended for use. Recent experience has shown that the resilient
            modulus value obtained from plastic deformation dynamic triaxial tests have not been
            suitable for fitting models as functions of the stress state and material variables.

            One variable that did appear to show some correlation with the resilient modulus was the
            saturation level. This is illustrated in Figure 9. The decrease in the resilient modulus with
            an increase in the saturation level is more obvious for material with 0.9 per cent residual
            binder. Because of the influence of other variables, it was not possible to fit a model
            predicting the resilient modulus as a function of the saturation level only. It is possible
            that the higher the binder content in the material, the less sensitive the resilient modulus
            is to moisture. However, given the variability in the data, a definitive statement on the
            effect of the saturation level should not be made.

            It is suggested that the resilient modulus values used in analyses are selected from the
            averages for each residual binder content shown in Table 25. Also shown in Table 25 are
            the average saturation levels, densities and the standard deviation and coefficient of
            variation of all three variables. Resilient moduli should be rounded to at least the nearest
            10 MPa when used in any analyses.



            Table 25.        Average resilient moduli for Vereeniging (P243/1) material with 1 per
                             cent cement

                                                                                  Resilient Modulus
            Residual               Saturation (%)      Relative Density (%)
                                                                                        (MPa)
             binder
                                           Standard               Standard                  Standard
            content
                              Average      deviation   Average    deviation      Average    deviation
               (%)
                                           (COV,%)*               (COV,%)*                  (COV,%)*
                                 53.8       1.67 (3)    77.6       0.40 (1)       1397       189 (14)
                3.0
                                 26.7       2.33 (9)    77.8       0.23 (0)       1663       547 (33)
                                 75.6       5.33 (7)    75.0       0.57 (1)        787       787 (28)
                0.9
                                 42.6       3.13 (7)    75.4       0.46 (1)       1362       343 (25)
            * coefficient of variation




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                42
Report number CR-2003/44
                                        3500
                                                                                                  3.0% Residual Binder
                                        3000                                                      0.9% Residual Binder
              Resilient Modulus (MPa)
                                        2500


                                        2000


                                        1500


                                        1000


                                        500


                                          0
                                               0               20             40             60             80                  100
                                                                                Saturation (%)


            Figure 9.                              Influence of saturation on resilient modulus for Vereeniging (P243/1)
                                                   material with 1 per cent cement



5.6.3       Permanent Deformation


            The permanent deformation behaviour of a material is determined from dynamic triaxial
            tests run at a range of stress ratios. The test is typically run for 50 000 load repetitions,
            and the permanent deformation of the specimen is measured during the test by the three
            LVDTs. The variables used for permanent deformation testing are given in the test matrix
            in Appendix B.

            The majority of data at 1 per cent cement and 3.0 and 0.9 per cent residual binder show a
            similar trend to specimen FSS33, illustrated in Figure 10. However, a few specimens
            show a trend similar to specimen FSS31, also illustrated in Figure 10. This typically only
            occurs with specimens tested at high stress ratios.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                    43
Report number CR-2003/44
                                          7.0


                                          6.0                                                         Data, FSS31
                                                                                                      Model, FSS31

             Permanent Deformation (mm)
                                                                                                      Data, FSS33
                                          5.0
                                                                                                      Model, FSS33

                                          4.0


                                          3.0


                                          2.0


                                          1.0


                                          0.0
                                                0          10000          20000           30000        40000         50000
                                                                                Repetitions


            Figure 10. Example dynamic triaxial test data

            Non-linear regression models were fitted to the data obtained from each LVDT on each
            specimen. It is also possible to fit the model to the average of the three LVDTs, although
            the method selected should not affect the final result as the model fits from each LVDT
            are used to extrapolate to a higher number of repetitions, and these results are averaged.
            Therefore, the difference between the two methods is when the averaging is done, but
            both average the data.

             Two types of models were used to fit to each LVDT, and are illustrated in Figure 10. The
            model fitted to the FSS31 specimen data is a hyperbolic-linear function (Equation (7)),
            and a double exponential function (Equation (8)) is fitted to the FSS33 data. These
            functions are described in detail in Theyse, 200013. The models were fitted to the data
            manually.

                                                                                     cN
                                                                   PD = mN+                   1
                                                                                                                             (7)
                                                                                ⎡ ⎛ cN ⎞b ⎤ b
                                                                                ⎢1+ ⎜  ⎟ ⎥
                                                                                ⎢ ⎝ a ⎠ ⎥
                                                                                ⎣         ⎦

                                                                   PD = QedN - Ae-bN - Q + A                                 (8)

                                                    where PD                           = permanent deformation
                                                          N                            = load repetitions
                                                          a, b, c, d, m, A, Q          = regression coefficients

            Table 26 and Table 27 show the model used for each particular data set (HL for
            hyperbolic linear and DE for double exponential), and the coefficients of the model fits.
            The R2 and the standard error of the estimate are also shown in the tables.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                      44
Report number CR-2003/44
            Table 26.      Model coefficients for dynamic triaxial permanent deformation data, 3 per cent residual binder
                        Residua
                                   Confining            Relative                                            Model Coefficients                     Statistics
                        l binder               Stress              Saturation                                                                                     End PD
            Sample                  stress              density                 Model   LVDT
                        content                Ratio                  (%)                         a           m              b            c           2            (mm)
                                     (kPa)                (%)                                                                                     R       SEE
                           (%)                                                                    Q           A              d            b
                                                                                                                    -6
                                                                                 HL      M     0.5718    1.598x10          1.050       0.00150   0.792    0.063   0.650
                                                                                                                    -7
             FSS17                   79.3       0.15     78.0         52.8       HL      1     0.5750    9.724x10          1.040       0.00100   0.743    0.072   0.625
                                                                                                                    -7
                                                                                 HL      2     0.2672    7.868x10          1.600       0.00009   0.878    0.031   0.308
                                                                                                                    -6
                                                                                 HL      M     0.8275    2.232x10          1.200       0.00100   0.787    0.102   0.940
                                                                                                                    -6
             FSS18                   82.0       0.44     77.6         53.9       HL      1     0.7982    2.306x10          1.200       0.00150   0.857    0.081   0.914
                                                                                                                    -6
                                                                                 HL      2     0.6465    1.933x10          1.200       0.00100   0.891    0.061   0.744
                                                                                                                    -6
                                                                                 HL      M     1.0100    6.400x10          1.100       0.00200   0.921    0.089   1.299
                                                                                                                    -6
             FSS19                   141.6      0.58     77.9         51.5       HL      1     0.9200    5.769x10          1.100       0.00200   0.914    0.082   1.182
                                                                                                                    -6
                                                                                 HL      2     0.8500    6.700x10          0.950       0.00300   0.968    0.051   1.160
                          3.0                                                                                       -6
                                                                                 HL      M     0.5101    1.415x10          1.050       0.00100   0.906    0.042   0.578
                                                                                                                    -6
             FSS20                   145.2      0.18     77.5         56.0       HL      1     0.4220    1.426x10          1.100       0.00060   0.912    0.037   0.492
                                                                                                                    -6
                                                                                 HL      2     0.4121    1.398x10          1.100       0.00060   0.900    0.038   0.481
                                                                                                                    -6
                                                                                 HL      M     1.2717    4.618x10          1.050       0.00200   0.888    0.122   1.504
                                                                                                                    -6
             FSS21                   142.0      0.50     76.8         55.3       HL      1     1.0124    4.422x10          1.100       0.00100   0.875    0.109   1.231
                                                                                                                    -6
                                                                                 HL      2     1.3701    4.802x10          1.100       0.00200   0.872    0.141   1.609
                                                                                                                    -6
                                                                                 HL      M     1.8500    7.433x10          0.800       0.00350   0.838    0.202   2.180
                                                                                                                    -6
             FSS22                   137.8      0.81     77.6         53.1       HL      1     1.3743    7.261x10          0.950       0.00250   0.696    0.204   1.742
                                                                                                                 -6
                                                                                 HL      2     2.5186   7.890x10           1.150       0.00250   0.972    0.140   2.917
                                                                                                                    -6
                                                                                 HL      M     0.5427    7.727x10          1.000       0.00150   0.930    0.041   0.679
                                                                                                                    -6
             FSS23                   82.0       0.16     77.6         25.5       HL      1     0.4335    1.849x10          1.050       0.00080   0.920    0.036   0.527
                                                                                                                    -6
                                                                                 HL      2     0.4647    2.246x10          1.050       0.00070   0.910    0.042   0.579
                                                                                                                    -5
                                                                                 HL      M     3.6713    1.131x10          1.100       0.00500   0.918    0.307   4.233
                                                                                                                    -5
             FSS24                   83.0       0.53     77.6         31.3       HL      1     3.7284    1.375x10          1.200       0.00400   0.924    0.320   4.408
                                                                                                                    -5
                                                                                 HL      2     3.6009    1.101x10          1.150       0.00500   0.898    0.331   4.145
                                                                                                                                  -4
                                                                                 DE      M     1.0000       2.401        3.950x10      0.00750   0.996    0.095   6.272
                                                                                                                                  -4
             FSS25                   79.7       0.66     77.7         24.8       DE      1     4.5568       1.822        2.032x10      0.00648   0.999    0.056   7.565
                                                                                                                                  -4
                                                                                 DE      2     0.9500       1.650        3.400x10      0.00870   0.998    0.052   4.406
                          3.0                                                                                       -6
                                                                                 HL      M     0.5889    1.962x10          1.200       0.00050   0.446    0.115   0.685
                                                                                                                    -6
             FSS26                   143.6      0.16     77.9         26.1       HL      1     0.5052    1.820x10          1.000       0.00100   0.797    0.059   0.597
                                                                                                                    -6
                                                                                 HL      2     0.3010    1.829x10          1.400       0.00010   0.748    0.053   0.392
                                                                                         M        *                                                               2.373
                                                                                                                    -5
             FSS27                   142.0      0.44     77.9         25.9       HL      1     1.8000    1.400x10          1.000       0.00200   0.886    0.203   2.437
                                                                                                                  -6
                                                                                 HL      2     1.8367    5.312x10          0.950       0.00300   0.887    0.174   2.104
                                                                                                                                  -5
                                                                                 DE      M     1.2040      0.667         5.850x10      0.00059   0.999    0.040   6.431
                                                                                                                                  -5
             FSS28                   143.0      0.70     78.2         26.5       DE      1     1.6000      2.700         5.600x10      0.00030   0.998    0.112   9.823
                                                                                                                                  -5
                                                                                 DE      2     0.4100      0.100         8.000x10      0.00001   0.996    0.053   4.095




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            Report number CR-2003/44                                          45
            Table 27.   Model coefficients for dynamic triaxial permanent deformation data, 0.9 per cent residual binder
                     Residual
                                Confining            Relative                                                Model Coefficients                       Statistics
                      binder                Stress              Saturation                                                                                           End PD
            Sample               stress              density                 Model   LVDT
                     content                Ratio                  (%)                          a             m                 b            c           2            (mm)
                                  (kPa)                (%)                                                                                           R       SEE
                        (%)                                                                     Q             A                 d            b
                                                                                                                    -7
                                                                              HL      M      0.6391       5.670x10           1.100        0.00150   0.767    0.072    0.661
                                                                                                                    -7
             FSS29                81.6       0.23     75.0         77.1       HL      1      0.5148       9.033x10           1.100        0.00100   0.714    0.066    0.551
                                                                                                                    -6
                                                                              HL      2      0.5300       1.011x10           1.100        0.00100   0.615    0.077    0.568
                                                                                                                    -6
                                                                              HL      M      1.7466       2.497x10           1.150        0.00160   0.757    0.231    1.871
                                                                                                                    -6
             FSS30                81.3       0.65     74.7         77.3       HL      1      1.5807       2.927x10           1.250        0.00120   0.779    0.209    1.722
                                                                                                                    -6
                                                                              HL      2      1.3282       2.767x10           1.100        0.00150   0.855    0.143    1.462
                                                                                                                                     -4
                                                                              DE      M      0.2400         1.600          5.000x10       0.00350   0.990    0.152    6.138
                                                                                                                                     -4
             FSS31                81.7       1.03     74.6         76.9       DE      1      0.2757         0.990          4.218x10       0.00580   0.946    0.241    3.957
                                                                                                                                     -4
                                                                              DE      2      0.0406         3.481          8.604x10       0.00073   0.994    0.214   10.521
                        0.9                                                                                         -7
                                                                              HL      M      0.4395       8.688x10           0.950        0.00100   0.801    0.048    0.484
                                                                                                                    -7
             FSS32                139.8      0.25     74.8         80.3       HL      1      0.2767       6.305x10           1.150        0.00040   0.958    0.018    0.310
                                                                                                                    -7
                                                                              HL      2      0.3501       7.691x10           0.950        0.00080   0.532    0.054    0.390
                                                                                                                    -6
                                                                              HL      M      1.1161       4.464x10           1.100        0.00220   0.891    0.103    1.341
                                                                                                                    -6
             FSS33                141.1      0.60     76.0         66.4       HL      1      1.2808       4.910x10           1.100        0.00200   0.888    0.125    1.522
                                                                                                                    -6
                                                                              HL      2      0.8928       4.313x10           1.100        0.00120   0.748    0.126    1.107
                                                                                                                    -5
                                                                              HL      M      6.4243       2.990x10           1.000        0.01500   0.665    0.886    7.904
                                                                                                                    -5
             FSS34                140.1      1.01     74.8         72.9       HL      1      6.8829       3.220x10           1.200        0.00700   0.778    0.923    8.474
                                                                                                                    -5
                                                                              HL      2      5.2268       3.186x10           1.050        0.00700   0.911    0.493    6.800
                                                                                                                    -7
                                                                              HL      M      0.3113       8.551x10           0.950        0.00080   0.768    0.036    0.352
                                                                                                                    -7
             FSS35                138.8      0.20     75.9         37.6       HL      1      0.2265       7.718x10           1.100        0.00050   0.768    0.028    0.266
                                                                                                                    -7
                                                                              HL      2      0.3846       7.567x10           1.050        0.00100   0.809    0.041    0.423
                                                                                                                    -6
                                                                              HL      M      0.6254       5.951x10           1.400        0.00150   0.658    0.103    0.928
                                                                                                                    -6
             FSS36                81.0       0.66     75.4         42.4       HL      1      0.3330       1.574x10           1.000        0.00050   0.805    0.128    0.409
                                                                                                                    -6
                                                                              HL      2      1.1894       2.987x10           1.050        0.00200   0.956    0.075    1.336
                                                                                                                    -4
                        0.9                                                   HL      M      1.3696       5.747x10           1.600        0.01500   0.992    0.049    1.944
                                                                                                                    -5
             FSS37                84.9       1.03     75.9         42.5       HL      1      1.5500       6.479x10           1.400        0.01000   0.984    0.059    1.515
                                                                                                                                     -4
                                                                              DE      2      4.8423         0.774          2.782x10       0.01585   0.999    0.025    2.333
                                                                                                                    -6
                                                                              HL      M      0.8382       3.150x10           1.100        0.00200   0.670    0.117    0.992
                                                                                                                    -6
             FSS39                140.6      0.81     74.9         45.4       HL      1      1.4112       3.243x10           1.200        0.00200   0.856    0.153    1.570
                                                                                                                    -6
                                                                              HL      2      0.6764       3.083x10           0.900        0.00150   0.614    0.105    0.829
             FSS40                144.1      1.27     75.1         45.1                     Specimen failed after 3 Load repetitions




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            Report number CR-2003/44                                             46
            More data are collected in the early stages of the test, than once the behaviour shows a
            more constant trend. Using these model fits, the permanent deformation behaviour is
            extrapolated beyond the range shown, and the dominant factor in the extrapolation is the
            model fit in the latter stages of the test. It is therefore more important that the model fits
            are better at the higher repetitions than at the lower repetitions. Because there is more
            data available at the beginning of the test, the R2 and standard error of the estimate are
            sometimes not as good as desired for typical regression models, but improving these
            values compromises the more important model fit at the higher number of load
            repetitions.

            The model fits are used to determine the load repetitions required to reach set levels of
            permanent deformation. These levels are 1, 4, 7, 10, 13, 17, and 20 per cent of plastic
            strain. The load repetitions (structural capacity) required to reach an arbitrary selected
            level of plastic strain, 17 per cent determined by extrapolating the models fit to each
            specimen, are shown in Figure 11 for both mixes. As the stress ratio increases, the
            structural capacity decreases, with a sharp decrease as the stress ratio approaches one.


                                    100 000 000



                                     10 000 000
              Structural capacity




                                      1 000 000



                                       100 000



                                        10 000
                                                      3.0% Binder, 1% cement
                                                      0.9% Binder, 1% cement

                                          1 000
                                               0.00       0.20       0.40          0.60       0.80   1.00   1.20
                                                                               Stress Ratio


            Figure 11. Structural capacity of mixes with 3.0 and 0.9 per cent residual binder
                       and 1 per cent cement


            The following observations are made from Figure 11:
            •        The mix with 1 per cent cement and 0.9 per cent residual binder has a slightly
                     better structural capacity than the mix with 3 per cent residual binder. This is
                     expected as the mix with 0.9 per cent binder content has a higher shear strength.
            •        The structural capacities of the 3 per cent binder mix between stress ratios of 0.6
                     to 0.7 are very low. These data are for specimens at a high density (78 per cent)
                     and a low saturation level (27 per cent) (FSS23 to FSS28). This saturation level
                     is very low. The double-exponential function (Equation (8)) was also fitted to
                     some of these specimens. This is normally only necessary at higher stress
                     ratios.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              47
Report number CR-2003/44
            The data shown in Table 26 and Table 27, and Figure 11 are used to develop the
            permanent deformation transfer function in Chapter 8.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials   48
Report number CR-2003/44
6.          ANALYSES OF LABORATORY TEST DATA


            In this chapter, the available and suitable laboratory test data discussed in Chapters 4
            and 5 are analysed and compared. The analyses include comparisons of the flexibility,
            shear strength, resilient modulus, and permanent deformation. Where necessary, any
            additional analyses that needed to be performed on the existing laboratory data
            described in Chapter 4 are discussed. The aim of the analyses in this chapter is to
            analyse the data necessary to develop the structural design models. The reports on the
            laboratory tests results give further analyses and comparisons of the data13,21,22.

6.1.        Comparison of Flexibility Results

            As discussed in Chapter 4, flexural beam tests were performed on the Cullinan (D2388)
            and Vereeniging (P243/1) materials to obtain the strain-at-break values. Tests were
            performed at various cement and residual binder contents. The strain-at-break values
            are illustrated in Figure 12. The strain-at-break of the Cullinan and Vereeniging materials
            with 1 per cent cement are in a similar range. An increase in the residual binder content
            increases the flexibility of the material, as long as the cement content is not too high. The
            mixes with 3 per cent residual binder content have a much higher flexibility when only 1
            per cent cement is used. At 2 per cent cement, additional binder shows little increase in
            the flexibility. The same trends were found with foamed bitumen treated materials, which
            highlights the need to carefully select the optimum binder and active filler contents to
            ensure sufficient flexibility for the desired mix behaviour.


                                             700
                                                         Cullinan, 1% cement
                                             600         Vereeniging, 1% cement
             Strain-at-break (microstrain)




                                                         Vereeniging, 2% cement
                                             500


                                             400


                                             300


                                             200


                                             100


                                               0
                                                   0.0      0.5       1.0         1.5         2.0       2.5   3.0   3.5
                                                                            Residual binder content (%)

            Figure 12. Comparison on strain-at-break results from Cullinan and Vereeniging


6.2.        Comparison of Shear Strength Data

            In light of the discussion of the disadvantages of fitting the Mohr-Coulomb failure
            envelope to determine the shear strength of the materials (Section 5.5), regression

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                   49
Report number CR-2003/44
                 models were fitted to all available static triaxial data to predict the maximum allowable
                 principal stress. This was done by fitting models of the form shown in Equation (5)
                 (Section 5.5.2) to the existing data. Models were fitted for both the untreated parent
                 materials and for the materials treated with emulsified bitumen and cement.

                 Two of the variables used in Equation (5) are relative density and saturation. The values
                 for these properties reported in other work21 only considered the aggregate component in
                 the density and degree of saturation calculations. These values have thus been
                 recalculated to account for the inclusion of the cement and residual binder on the density,
                 and the results reported in this work include the additives. The formulae for the
                 calculations are given in Section 4.1, Equations (2) and (3).

                 This section describes the development of individual models to predict the maximum
                 principal stress for the Cullinan and Vereeniging materials. While such individual models
                 are not generally applicable, they are an effective means of calculating the maximum
                 allowable principal stress and then the stress ratio for the dynamic triaxial tests of each
                 mix. This section also attempts to develop a single model applicable to both materials,
                 with suggestions as to how the model could be extended to be applicable to all material
                 types.

6.2.1            Cullinan Static Triaxial Test Data


6.2.1.1.         Untreated Material


                 A model to determine the shear strength of the untreated Cullinan material was
                 developed. This allows comparisons of the untreated parent and treated materials from
                 both the Cullinan and Vereeniging HVS test sites. This comparison is discussed in
                 Section 6.2.3. Data from 8 static triaxial tests were used, and the range of saturation
                 levels was sufficient to allow the inclusion of the saturation term. The relative densities of
                 all of the specimens were similar and therefore it was unfortunately necessary to exclude
                 this variable from the model as the variable was found to be statistically insignificant. In
                 these situations including the variable could lead to misleading predictions. The resulting
                 model is only applicable to the range of relative density values used, i.e., 73.6 to 74.1 per
                 cent. The results are shown in Table 28 (Equation (5)(c)).

                 Table 28.        Model coefficients and statistics for Equation (5), Cullinan

                                    Equation (5)(c)           Equation (5)(d)             Equation (5)(e)
            Residual binder                0%                        1.5%                       0.9%
               Cement                      0%                        1.0%                       1.0%
                                 Coefficient p-value Coefficient p-value               Coefficient p-value
               c1 (constant)      -238.462        0.0667    802.313       0.01722        615.85       0.01989
           c2 (relative density)   Stat. insignificant*      Stat. insignificant*       Stat. insignificant*
             c3 (1/saturation)    16 886.83       0.0014     Stat. insignificant*       Stat. insignificant*
                   c4 (σ3)         7.0952         0.0000    5.22538       0.02447        5.8032        0.0136
                     R2                   0.987                     0.927                      0.960
                   SEE**                  62.630                  110.632                      91.54
           Models applicable to these values of the variables
             Relative density        73.6 – 74.1%                   77.7%                      76.4%
                Saturation           19.9 – 38.1%                   39.7%                      39.6%
            Confining stress          20 - 200 kPa             20 - 200 kPa                 20 - 200 kPa
           * Variables excluded due to lack of variation in data
           ** Standard error of the estimate
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                     50
Report number CR-2003/44
6.2.1.2.    Treated Material


            Only four static triaxial tests were available at each of the two binder contents for the
            material from Cullinan (D2388). The actual moisture content and density of the
            specimens were not determined after the test. Therefore, the target moisture content and
            density were used to recalculate the relative density and degree of saturation. The
            recalculated values, which include the cement and residual binder, are shown in Table
            28. Equation (5) was fit to the four data points at each binder content, and the results are
            shown in Table 28. The model fits are good, although the models are only applicable to
            the relative densities and saturation levels shown in Table 28. Equations (5)(d) and (e)
            were used to calculate the stress ratio for the Cullinan dynamic triaxial test data at 1.5
            and 0.9 per cent residual binder, respectively.

            A model that combines all of the data for both binder contents was attempted. However,
            because the relative density and saturation values differ in the two sets of data, as do the
            residual binder contents, it was not possible to differentiate the effects of these three
            variables. It is not possible, therefore, to fit one model including all the variables in
            Equation (5) with additional variables to account for the residual binder and cement
            contents with the available data sets.

6.2.2       Vereeniging Static Triaxial Test Data


6.2.2.1.    Untreated Material


            To fit Equation (5) to the untreated Vereeniging material, one specimen was excluded as
            the data was an obvious outlier. Sufficient ranges of relative density and saturation were
            available to allow the inclusion of both terms, as shown in Table 29 (Equation (5)(f)).

            Table 29.    Model coefficients and statistics for Equation (5), Vereeniging Material

                                           Equation (5)(f)                        Equation (5)(g)
             Residual binder                    0%                                     1.8%
                Cement                          0%                                     2.0%
                                    Coefficient        p-value             Coefficient        p-value
                c1 (constant)        -6 749.94          0.0010             -12 384.00          0.0000
            c2 (relative density)     89.0071           0.0000               181.68            0.0000
              c3 (1/saturation)     37 712.59           0.0011             70 649.79           0.0650
                    c4 (σ3)           5.4323            0.0010               4.7068            0.0000
                      R2                        0.92                                   0.93
             Standard error of
                                               107.62                                  107.6
                the estimate
            Models applicable to these values of the variables
              Relative density              66.5 – 71.9%                           71.2 – 75.6%
                 Saturation                  43 – 75.3%                            54.1 – 67.1%
            Confining pressure              20 – 200 kPa                           20 – 200 kPa


6.2.2.2.    Treated Material


            The earlier Vereeniging triaxial data included two saturation levels and two relative
            densities, for the mix with 2 per cent cement and 1.8 per cent residual binder (Chapter 4

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              51
Report number CR-2003/44
            and Reference 21).           These data were used to fit a model of the form shown in
            Equation (5).

            The data from four of the sixteen specimens were excluded from the data used to fit the
            models as they were considered unreasonable. Typically, it was not possible to
            determine the failure stress from these data. The results of the model fit are shown in
            Table 29 (Equation (5)(g)).

6.2.2.3.    Combined Model for Vereeniging Data


            The Vereeniging data (27 specimens) were combined to fit a single model. This included
            the data containing 2 per cent cement and 1.8 per cent residual binder, discussed in
            Section 6.2.2 above, and the data at 1 per cent cement and both 0.9 and 3.0 per cent
            residual binder, discussed earlier in Section 5.5.

            Various model forms were investigated. The primary difference between the models is
            the variables for the cement and residual binder contents. The cement and residual
            binder content were combined as a ratio, and as individual variables. The best fit was
            obtained with the individual variables, as shown in Equation (9). The statistics of the
            model are shown in Table 30. The model fit is significantly improved using the individual
            cement and residual binder content variables rather than a ratio term. The model fit is
            good, and each variable is statistically significant at more than the 99 per cent confidence
            level. The signs of the coefficients are all as expected.

                          σ1 = c1 + c 2 × RD + c 3 × SAT + c 4 × σ 3 + c 5 × CEM+ c 6 × BIN
                                                      1
                                                                                                              (9)


                          where σ1             = maximum (major) allowable principal stress (kPa)
                                RD             = relative density (%)
                                SAT            = saturation (%)
                                σ3             = confining stress or minor principal stress (kPa)
                                CEM            = cement content (%)
                                BIN            = residual binder content (%)
                                c1-6           = coefficients

            Table 30.       Model coefficients and statistics for Equation (9)

                                                                              Equation (9)
                                                                                                  Range of
                                                           Coefficient         p-value
                                                                                              applicable values
            c1 (constant)                                  -12 920.69           0.0000               n/a
            c2 (relative density)                            148.19             0.0000            71 – 79%
            c3 (1/saturation)                               91 225.65           0.0000           32 – 71%*
            c4 (σ3)                                          4.4326             0.0000          20 – 205 kPa
            c5 (cement content)                              1599.48            0.0000             1 – 2%
            c6 (residual binder content)                     -295.08            0.0000            0.9 – 3%
            R2                                                                     0.951
            Standard error of the estimate                                        145.54
            * Saturation values, 1/saturation used in the model


            The data fit to Equation (9) can be used to identify the influences of the different
            variables, particularly the cement and residual binder content. The model coefficients
            shown in Table 30 were used to calculate the maximum allowable principal stresses
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                           52
Report number CR-2003/44
            shown in Figure 13 for the treated material, and using Equation (5)(f) for the untreated
            material. A confining stress of 120 kPa was used, and the high and low relative density
            values are 70 and 74 per cent, and the high and low saturation values are 45 and 65 per
            cent.


                                                         3000.0
                                                                                                      Untreated
              Maximum allowable principal stress (kPa)


                                                                                                      1.8% binder, 1% cement
                                                                                                      1.8% binder, 2% cement
                                                         2500.0                                       3.0% binder, 1% cement
                                                                                                      3.0% binder, 2% cement

                                                         2000.0


                                                         1500.0


                                                         1000.0


                                                          500.0


                                                            0.0
                                                                  Low saturation,   Low saturation,   High saturation, High saturation,
                                                                    low relative     high relative      low relative     high relative
                                                                      density          density            density          density


            Figure 13. Comparison of maximum allowable principal stress from Vereeniging
                       data model fit

            Figure 13 illustrates the expected behaviour in that an increase in the saturation and/or
            decrease in the relative density reduces the maximum allowable principal stress that the
            material can sustain. An increase in the cement content increases the maximum
            allowable principal stress, whereas an increase in the residual binder content reduces the
            maximum allowable principal stress. This means that an increase in the residual binder
            content decreases the shear strength of the material, and thereby decreases its
            resistance to permanent deformation. Unfortunately no data are available for mixes
            treated with binder only, therefore it is not possible to determine the effect of adding
            binder only on the shear strength.

            The untreated material shows a maximum allowable shear strength that is consistently
            lower than the mixes with 2 per cent cement. However, the mixes with only 1 per cent
            cement show similar shear strength to the untreated material. The mix with 1 per cent
            cement and 3 per cent binder has a lower shear strength than the untreated material.
            Similar trends were found with foamed bitumen treated materials7 using the same parent
            material. Increasing the binder content for the same cement content decreases the shear
            strength. This is evident for the mixes with both 1 and 2 per cent cement.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                           53
Report number CR-2003/44
6.2.3       Comparison of Shear Strength of Cullinan and Vereeniging Materials


            Using the models in Table 28 and Table 29, and a range of values for the confining
            stress, relative density and saturation, the following observations are made:
            •        The maximum allowable shear stress is consistently higher for the Vereeniging
                     material than the Cullinan material. This is demonstrated in Figure 14 for the
                     untreated material and Figure 15 for the treated materials. The same values for
                     saturation and relative density were used to calculate the maximum allowable
                     principal stress for each set of data shown in the two figures (where such terms
                     are included in the models). The higher shear strength of the Vereeniging
                     material may be due to the better grading.
            •        The data in Figure 15 show that for the Cullinan material an increase in binder
                     results in an increase in shear strength. This is because the relative densities of
                     the mix with 1.5 per cent binder are higher. The models used to calculate the
                     values shown in the figure do not include a term for the relative density. This was
                     discussed in Section 6.2.1.2.


                                                        2000
             Maximum allowable principal stress (kPa)




                                                        1800                    Cullinan untreated material

                                                        1600                    Vereeniging untreated material

                                                        1400

                                                        1200

                                                        1000

                                                         800

                                                         600

                                                         400

                                                         200

                                                           0
                                                               Low saturation         High saturation

            Figure 14. Comparison of shear strength of untreated Cullinan and Vereeniging
                       materials




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                          54
Report number CR-2003/44
                                                         2000
                                                                            RD = 76.4%                                   RD = 77.7%
                                                                            Sat = 39.6%                                  Sat = 39.7%



              Maximum allowable principal stress (kPa)
                                                         1800

                                                         1600

                                                         1400
                                                                   Cullinan
                                                         1200
                                                                   Vereeniging
                                                         1000

                                                         800

                                                         600

                                                         400

                                                         200

                                                            0
                                                                                0.9% Binder                               1.5% Binder

            Figure 15. Comparison of shear strength of treated Cullinan and Vereeniging
                       materials with 1% cement


6.2.4       Combined Cullinan and Vereeniging Shear Strength Model for Treated Materials


            A combined model to predict the maximum allowable principal stress that is valid for both
            the Cullinan and Vereeniging models was developed. Equation (9) could not be
            successfully fitted. This is probably because the parent materials differ, and Equation (9)
            does not account for these differences. A modified version of Equation (9) including an
            indicator variable, equal to 1 or 0 for the Cullinan and Vereeniging materials, respectively,
            was more successful. This revised model is shown in Equation (10) and the coefficients
            and statistics are shown in Table 31.

                                                         σ1 = c1 + c 2 × RD + c 3 × SAT + c 4 × σ 3 + c 5 × CEM+ c 6 × BIN + c 7 × MAT
                                                                                     1
                                                                                                                                         (10)


                                                                where σ1            = maximum (major) principal stress (kPa)
                                                                      RD            = relative density (%)
                                                                      SAT           = saturation (%)
                                                                      σ3            = confining stress or minor principal stress (kPa)
                                                                      CEM           = cement content (%)
                                                                      BIN           = residual binder content (%)
                                                                      MAT           = material indicator
                                                                                    = 1 for Cullinan material
                                                                                    = 0 for Vereeniging material
                                                                         c1-7       = coefficients




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                    55
Report number CR-2003/44
            Table 31.       Model coefficients and statistics for Equation (10)

                                                                         Equation (10)
                                                                                    Range of applicable
                                                           Coefficient   p-value
                                                                                          values
            c1 (constant)                                  -1 3217.67    0.0000             n/a
            c2 (relative density)                            151.35      0.0000        71.2 – 79.2%
            c3 (1/saturation)                               90 815.54    0.0000        32.1 – 71.6%*
            c4 (σ3)                                          4.6903      0.0000        20 – 205 kPa
            c5 (cement content)                              1615.67     0.0000           1 – 2%
            c6 (residual binder content)                     -286.96     0.0000         0.9 – 3.0%
            c7 (material indicator)                         -1206.20     0.0000            0 or 1
                             R2                                              0.961
             Standard error of the estimate                                  139.71
            * Saturation values, 1/saturation used in the model


            Because of the indicator variable, this model is not applicable to a wider range of
            materials. It does however indicate that a variable that accounts for the type of material
            needs to be included when predicting the shear strength of a material. With such a
            variable, a good fit is possible. More work needs to be done, using data from a range of
            material types, to determine the most appropriate properties to capture the material type.
            The appropriate property will probably relate to the particle shape, Atterberg limits or the
            grading. It was not possible to determine an appropriate material property with the
            available data because there are only two materials and therefore any property would be
            statistically significant. This work demonstrates that this method has potential for
            predicting the maximum allowable principal stress of all materials using one model, and in
            this manner eliminating the need to calculate the cohesion and friction angle.

6.3.        Permanent Deformation Data

            To develop a laboratory permanent deformation model (Section 8.2), it is necessary to fit
            the permanent deformation model (Equation (7)) as discussed in Section 5.6.3 to all the
            available dynamic triaxial test data. The model fits for the Vereeniging material tested
            with 1 per cent binder content are given in Table 26 and Table 27, Section 5.6.3. As
            described earlier, models were fitted to the dynamic triaxial permanent deformation test
            data for the Cullinan material (1 per cent cement, 0.9 per cent and 1.5 per cent residual
            binder) and Vereeniging material (2 per cent cement, 1.8 per cent residual binder). The
            parameters for the model fits are shown in Table 32 for the Cullinan data and in Table 33
            and Table 34 for the Vereeniging data. In Table 33 and Table 34, a large amount of data
            is noted as unreasonable. This is because at the time of this testing, some of the
            machine control settings were set incorrectly. Data were excluded when the permanent
            deformation decreases or remains constant with increasing load repetitions.

            All data are shown in the tables, although some were later excluded from the data set
            used for development of the permanent deformation transfer function. The models for the
            Vereeniging material were refitted in this project, and therefore differ from those
            described in References 15 and 21. The stress ratios shown in Table 32, Table 33 and
            Table 34 were calculated using the models predicting the maximum allowable principal
            stress. The individual model specific to each material, cement and binder content was
            used, i.e., Equation (5)(a) and (b) (Table 23), Equation (5)(d) and (e) (Table 28), and
            Equation (5)(g) (Table 29).



Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               56
Report number CR-2003/44
            Table 32.     Model coefficients for dynamic triaxial permanent deformation data (Cullinan (D2388), 1 per cent cement, 0.9 per cent and
                          1.5 per cent residual binder)

                      Residual
                                  Confining              Relative                                         Model Coefficients                           Statistics
                       binder                  Stress               Saturation
            Sample                 stress                density                  LVDT                                                                                 End PD (mm)
                      content                  Ratio                   (%)                                                                         2
                                    (kPa)                  (%)                                  A             M            b          c          R             SEE
                         (%)
                                                                                                                   -6
                                                                                    M        0.3600       1.120x10       0.35      0.0200      0.992           0.008       0.367
             ETB09                   80.3        0.20                               1           *             *            *          *          *               *         0.095
                                                                                    2           *             *            *          *          *               *        0.120**
                                                                                                                   -6
                                                                                    M        0.7000       1.600x10       0.52      0.0200      0.989           0.017       0.758
                                                                                                                   -6
             ETB10                   80.4        0.52                               1        0.5500       9.800x10       0.90      0.0040      0.959           0.026       0.593
                                                                                                                   -6
                                                                                    2        0.5817       1.522x10       1.00      0.0020      0.824           0.059       0.654
                                                                                                                   -6
                                                                                    M        1.1700       2.000x10       0.64      0.0200      0.994           0.022       1.232
                                                                                                                   -6
             ETB11                   80.0        0.73                               1        0.8737       1.370x10       0.95      0.0040      0.946           0.050       0.942
                                                                                                                   -6
                                                                                    2        0.9689       1.446x10       0.95      0.0050      0.952           0.052       1.041
                         0.9                               76.4        39.5                                        -6
                                                                                    M        0.2900       1.300x10       0.50      0.0200      0.952           0.017       0.336
                                                                                                                   -7
             ETB12                   140.4       0.18                               1        0.1671       9.161x10       1.04      0.0010      0.966           0.008       0.211
                                                                                                                   -7
                                                                                    2        0.1734       5.867x10       0.70      0.0050      0.985           0.005       0.201
                                                                                                                   -6
                                                                                    M        0.4800       1.700x10       0.50      0.0200      0.992           0.010       0.547
                                                                                                                   -6
             ETB13                   139.7       0.47                               1        0.3497       1.377x10       0.95      0.0025      0.965           0.017       0.421
                                                                                                                   -6
                                                                                    2        0.2604       1.100x10       0.80      0.0100      0.939           0.016       0.318
                                                                                                                   -6
                                                                                    M        0.7000       1.900x10       0.60      0.0200      0.992           0.015       0.773
                                                                                                                   -6
             ETB14                   141.0       0.67                               1        0.4565       1.330x10       0.80      0.0050      0.980           0.016       0.521
                                                                                                                   -6
                                                                                    2        0.4643       1.501x10       1.00      0.0020      0.909           0.035       0.539
                                                                                                                   -6
                                                                                    M        0.3700       1.900x10       0.45      0.0200      0.984           0.012       0.448
                                                                                                                   -7
             ETB15                   80.2        0.22                               1        0.1540       4.630x10       1.05      0.0010      0.967           0.007       0.177
                                                                                                                   -7
                                                                                    2        0.1070       5.900x10       0.64      0.0100      0.911           0.008       0.137
                                                                                                                   -6
                                                                                    M        0.6300       2.600x10       0.80      0.0050      0.976           0.026       0.763
                                                                                                                   -7
             ETB16                   79.9        0.57                               1        0.3000       8.140x10       0.80      0.0050      0.975           0.011       0.340
                                                                                                                   -6
                                                                                    2        0.4400       1.300x10       0.70      0.0050      0.988           0.012       0.500
                                                                                                                   -6
                                                                                    M        0.8300       2.900x10       0.70      0.0070      0.961           0.041       0.934
                                                                                                                   -6
             ETB17       1.5         79.7        0.80      77.7        39.7         1        0.5400       1.300x10       0.75      0.0050      0.985           0.017       0.601
                                                                                                                   -6
                                                                                    2        0.5700       1.200x10       0.80      0.0030      0.940           0.035       0.628
                                                                                                                   -7
                                                                                    M        0.1800       7.000x10       0.45      0.0090      0.942           0.013       0.217
             ETB18                   139.3       0.21                               1           *             *            *          *          *               *         0.051
                                                                                                                   -7
                                                                                    2        0.1050       7.000x10       0.70      0.0009      0.963           0.007       0.140
                                                                                                                   -6
                                                                                    M        0.6500       2.500x10       0.60      0.0200      0.980           0.024       0.784
                                                                                                                   -6
             ETB19                   140.3       0.54                               1        0.5400       1.700x10       0.70      0.0090      0.989           0.014       0.624
                                                                                                                   -7
                                                                                    2        0.3100       8.600x10       0.90      0.0030      0.963           0.014       0.355
            * Data unreasonable
            ** LVDT M measures more permanent deformation than LVDTs 1 and 2 because it measures across the whole specimen and not across a portion of the specimen as do the on
            specimen mounted Lvdts.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                    Report number CR-2003/44                                                 57
            Table 33.     Model coefficients for dynamic triaxial permanent deformation Data (Vereeniging, 2 per cent cement and 1.8 per cent
                          residual binder, low saturation)

                      Residual
                                  Confining            Relative                                         Model Coefficients                Statistics
                       binder                 Stress              Saturation                                                                             End PD
            Sample                 stress              density
                      content                 Ratio                  (%)                                                                  2               (mm)
                                    (kPa)                (%)                                     A          M           b        c       R       SEE
                         (%)                                                   LVDT   Model*
                                                                                M      HL         *         *           *         *       *        *        *
             ETF17                  79.5       0.18     75.4         87.7        1     HL         *         *           *         *       *        *        *
                                                                                 2     HL         *         *           *         *       *        *        *
                                                                                M      HL         *         *           *         *       *        *      6.239
             ETF18                  80.0       0.47     76.1         89.4        1     HL         *         *           *         *       *        *      0.293
                                                                                                                 -6
                                                                                 2     HL      0.4430   1.840x10      1.000   0.025     0.881   0.040     0.544
                                                                                M      HL         *         *                                            10.783
                                                                                                                 -6
             ETF19                  80.8       0.78     76.4         93.5        1     HL      0.9700   2.500x10      1.100    0.0020   0.936    0.072    1.095
                                                                                                                 -6
                                                                                 2     HL      1.1500   2.600x10      0.700    0.0200   0.961    0.053    1.275
                         1.8
                                                                                M      HL         *         *           *         *       *        *        *
             ETF20                  140.1      0.16     77.1         99.1        1     HL         *         *           *         *       *        *        *
                                                                                 2     HL         *         *           *         *       *        *        *
                                                                                M      HL         *         *           *         *       *        *      5.939
                                                                                                                 -6
             ETF21                  140.5      0.46     76.6         98.1        1     HL      0.2700   2.600x10      0.700    0.0200   0.932    0.018    0.354
                                                                                                                 -6
                                                                                 2     HL      0.2800   2.720x10      0.600    0.0200   0.955    0.016    0.352
                                                                                M      HL         *         *           *         *       *        *      9.526
             ETF22                  141.9      0.81     74.9         87.0        1     HL         *         *           *         *       *        *      4.835
                                                                                                                 -6
                                                                                 2     HL      1.1600   3.700x10      1.100    0.0100   0.977    0.047    1.345
                                                                                M      HL         *         *           *         *       *        *      3.179
             ETF23                  82.2       0.15     76.0         66.7        1     HL         *         *           *         *       *        *      0.394
                                                                                                                 -7
                                                                                 2     HL      0.4300   5.470x10      0.750   0.0100    0.631    0.054    0.450
                                                                                M      HL         *         *           *         *       *        *      6.646
                                                                                                                 -7
             ETF24                  81.7       0.42     75.9         73.2        1     HL      0.9000   9.900x10      0.800    0.0070   0.741    0.097    0.950
                                                                                                                 -6
                                                                                 2     HL      1.2000   2.400x10      0.850    0.0100   0.730    0.133    1.294
                                                                                M      HL         *         *           *         *       *        *      9.540
                                                                                                                 -6
             ETF25                  80.9       0.64     76.5         70.3        1     HL      2.4500   7.200x10      0.800    0.0070   0.885    0.218    2.716
                                                                                 2     HL         *         *           *         *       *        *      1.501
                         1.8
                                                                                M      HL         *         *           *         *       *        *        *
             ETF26                  140.2      0.15     76.9         65.6        1     HL         *         *           *         *       *        *        *
                                                                                 2     HL         *         *           *         *       *        *        *
                                                                                M      HL         *         *           *         *       *        *      7.162
             ETF27                  143.2      0.42     76.7         73.7        1     HL         *         *           *         *       *        *      1.043
                                                                                                                 -7
                                                                                 2     HL      0.8800   9.700x10      0.850    0.0030   0.974    0.042    0.894
                                                                                M      HL         *         *           *         *       *        *     11.637
                                                                                                                 -6
             ETF28                  142.7      0.66     77.3         67.1        1     HL      1.9500   5.700x10      1.000    0.0100   0.939    0.127    2.226
                                                                                                                 -6
                                                                                 2     HL      1.2110   1.700x10                                          1.296
            * Data unreasonable




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials          Report number CR-2003/44                                    58
            Table 34.   Model coefficients for dynamic triaxial permanent deformation data, high saturation (Vereeniging, 2 per cent cement and
                        1.8 per cent residual binder, high saturation)

                     Residual
                                Confining            Relative                                         Model Coefficients               Statistics
                      binder                Stress              Saturation                                                                            End PD
            Sample               stress              density                 LVDT   Model*
                     content                Ratio                  (%)                                                                 2               (mm)
                                  (kPa)                (%)                                     A          M           b       c       R       SEE
                        (%)
                                                                              M      HL         *         *           *        *       *        *      3.130
                                                                                                               -7
             ETF29                80.4       0.18     80.2        102.7       1      HL      0.3720   4.100x10      1.000   0.0035   0.435    0.055    0.392
                                                                                                               -7
                                                                              2      HL      0.5298   9.400x10      0.850   0.0100   0.487    0.074    0.572
                                                                              M      HL         *         *           *        *       *        *      9.047
             ETF30                81.3       0.46     80.8         92.0       1      HL         *         *           *        *       *        *      0.532
                                                                                                               -6
                                                                              2      HL      1.1200   1.500x10      0.900   0.0060   0.984    0040     1.189
                                                                              M      HL         *         *           *        *       *        *        *
             ETF31      1.8       135.6      0.17     79.4        109.2       1      HL         *         *           *        *       *        *        *
                                                                              2      HL         *         *           *        *       *        *        *
                                                                              M      HL         *         *           *        *       *        *     4.200
                                                                                                               -6
             ETF32                140.5      0.21     78.3        109.1       1      HL      0.7600   2.200x10      0.900   0.0080   0.964    0.037   0.808
                                                                                                               -6
                                                                              2      HL      0.4000   1.900x10        *        *       *        *     0.429
                                                                              M      HL         *         *           *        *       *        *     11.744
                                                                                                               -5
             ETF33                138.5      0.59     77.9        107.6       1      HL      1.9300   1.230x10      1.000   0.0040   0.875    0.211    2.524
                                                                                                               -5
                                                                              2      HL      1.9500   1.560x10      0.900   0.0090   0.992    0.059    2.730
                                                                              M      HL         *         *           *        *       *        *        *
             ETF35                81.8       0.19     79.6         86.4       1      HL         *         *           *        *       *        *        *
                                                                              2      HL         *         *           *        *       *        *        *
                                                                              M      HL         *         *           *        *       *        *     9.812
                                                                                                               -6
             ETF36                79.7       0.53     79.7         95.4       1      HL      1.1000   1.600x10      0.655   0.0100   0.909    0.076   1.140
                                                                              2      HL         *         *           *        *       *        *     0.860
                        1.8
                                                                              M      HL         *         *           *        *       *        *        *
             ETF38                139.9      0.21     79.6         89.5       1      HL         *         *           *        *       *        *        *
                                                                              2      HL         *         *           *        *       *        *        *
                                                                              M      HL         *         *           *        *       *        *     11.197
                                                                                                               -6
             ETF39                140.3      0.56     79.7         83.7       1      HL      1.8900   5.400x10      0.950   0.0200   0.920    0.124    2.159
                                                                                                               -6
                                                                              2      HL      1.6000   6.100x10      1.400   0.0300   0.959    0.072    1.891
            * Data unreasonable




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials        Report number CR-2003/44                                   59
            The models were used to estimate the load repetitions to specified levels of plastic strain,
            i.e., 1, 4, 7, 10, 13, 17, 20 per cent. It is difficult to use the data to make observations of
            the permanent deformation performance before models are fitted to the data. Discussion
            is therefore reserved until the laboratory permanent deformation model is described in
            Section 8.2.

6.4.        Comparison of Resilient Moduli Data

            A summary of the resilient moduli data for the Cullinan and Vereeniging materials are
            given in Table 35, and are illustrated in Figure 16 as a function of the saturation level.
            The resilient moduli are obtained from the permanent deformation (plastic strain) triaxial
            test, and not from a resilient modulus dynamic triaxial test. The results show that an
            increase in the saturation generally results in a decrease in the resilient modulus. A
            comparison of the resilient moduli at 1 per cent cement and 0.9 per cent residual binder
            at the lower saturation levels shows that the Vereeniging material has a higher stiffness
            than the Cullinan material, despite the Vereeniging relative density and saturation being
            lower and higher, respectively. This may be due to the good grading of the Vereeniging
            material. The actual difference in values, though, is not significantly large. The mix with
            2 per cent cement has the highest stiffness values, whereas the mixes with the least
            bitumen (0.9 per cent) have the lowest stiffnesses. As discussed earlier, it was not
            possible to develop models to predict the resilient modulus.

            Table 35.     Average resilient moduli for Cullinan and Vereeniging materials

                            Cement        Residual                                            Resilient
                                                            Saturation       Relative
              Material      content    binder content                                         Modulus
                                                               (%)          Density (%)
                              (%)           (%)                                                (MPa)
                                              1.5               39.5            76.4            1059
              Cullinan         1
                                              0.9               39.7            77.7            1017
                                                                53.8            77.6            1397
                                              3.0
                                                                26.7            77.8            1663
                               1
                                                                75.6            75.0            787
                                              0.9
                                                                42.6            75.4            1362
            Vereeniging
                                                                92.5            76.1            2683
                                                                69.4            76.6            2367
                               2              1.8
                                                               104.1            79.3            1721
                                                                88.8            79.7            2005




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 60
Report number CR-2003/44
                                         3000
                                                    Cullinan, 1% cement
                                                    Vereeniging, 1% cement
                                         2500
                                                    Vereeniging, 2% cement

               Resilient modulus (MPa)
                                         2000


                                         1500


                                         1000


                                          500


                                            0
                                                0      20         40          60          80   100   120
                                                                         Saturation (%)

            Figure 16. Resilient moduli as a function of the saturation level




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                    61
Report number CR-2003/44
7.          STIFFNESS REDUCTION TRANSFER FUNCTION


            In this chapter, a transfer function for the first phase in the life of an emulsified bitumen
            treated pavement layer is developed, namely the stiffness reduction phase life, previously
            termed the effective fatigue life. The model was developed using the elastic stiffness
            data from the HVS tests. The flexibility data from the flexural beam test were used to
            expand the transfer function for wider applicability to general pavement design.

7.1.        Stiffness Reduction of Heavy Vehicle Simulator Tests

            The stiffness data backcalculated from in-depth deflection measurements during HVS
            tests were used to develop the stiffness reduction transfer function. This transfer function
            predicts the number of repetitions to reach the steady state stiffness (defined in Section
            3.1.1). To determine the stiffness reduction phase life, models were fitted to the stiffness
            data.

            Two models were fitted to the stiffness data to predict the number of load repetitions to
            reach the steady state stiffness. The first is a simple log-log model (Equation (11)), the
            second is a non-linear model (Equation (12)). In the equations, N is the number of load
            repetitions, MR the stiffness, and a, b and c are regression coefficients.

                                              MR = a × Nb                                              (11)

                                                             1
                                                        ⎛ N ⎞b
                                           log MR = c − ⎜ ⎟                                            (12)
                                                        ⎝a⎠

            Only the Cullinan and Vereeniging data are used, because of the lack of suitable data for
            the Heilbron sections. The stiffnesses used were backcalculated from the deflections
            measured under a 40 kN wheel load. The differences in these stiffnesses and those
            backcalculated from higher wheel loads on the same section were not significant. The
            models were only fitted to data from the first wheel load on each test, unfortunately it is
            not possible to fit models for the reduction in the initial modulus to the steady state
            stiffness for the later load levels on each test, because the initial modulus is dependant
            on the previous loading. It was difficult to fit the models to the data, due to the variability.
            Equation (12) gave better fits, and therefore was used to determine the stiffness
            reduction phase life. The model fits using this equation are shown in Figure 17 to Figure
            22 for the Cullinan and Vereeniging HVS tests.

            It was difficult to fit a good model to the data MDD12 of Section 397A4 (Figure 17) that
            would reach the steady state stiffness in a reasonable number of load repetitions. The
            stiffness data for all 3 MDDs on Section 403A4 (Figure 18) are variable, and it was not
            possible to fit a model to the MDD4 data. The models fitted to MDD8 and MDD12 data
            are not good fits, and therefore the data from this section were not used to develop the
            stiffness reduction transfer function. Should they be included, they will not significantly
            affect the transfer function, and will not significantly affect the catalogues and design
            charts developed in this report. The surfacing around MDD12 on Section 407A4
            (Figure 19) failed early in the test, which necessitated moving the HVS to avoid this area.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                  62
Report number CR-2003/44
            The stiffnesses calculated from this MDD are low from the start of the test, which are
            expected to be due to the localised weak area. The data from this MDD were therefore
            excluded from the analyses. The data and model fits from Sections 408A4, 410A4/B4
            and 412A4 (Figure 20, Figure 21 and Figure 22) were all suitable for use in developing
            the transfer function. The Vereeniging sections have higher stiffnesses than the Cullinan
            sections, which agrees with the resilient moduli measured in the dynamic triaxial test.


                                       1600
                                                                                                   397 MDD8 40kN
                                                      40 kN, dry        70 kN, dry                 397 MDD12 40kN
                                       1400
                                                                                                   MDD8 function
                                                                                                   MDD12 function
                                       1200
             Resilient Modulus (MPa)




                                       1000
                                                                                  70 kN, dry   70 kN, wet
                                        800

                                        600

                                        400

                                        200

                                          0
                                              0   100000      200000         300000       400000       500000       600000
                                                                           Repetitions
            Figure 17. Stiffness reduction to steady state stiffness, Cullinan (D2388), 397A4
                       (Sandstone)



                                       1600
                                                                                                     403 MDD4 40kN
                                                           40 kN   70 kN
                                       1400                                                          403 MDD8 40kN
                                                                                                     403 MDD12 40kN
                                       1200                                                          MDD8 model
             Resilient Modulus (MPa)




                                                                                                     MDD12 model
                                       1000

                                        800

                                        600

                                        400

                                        200
                                                                                          70 kN 70 kN, wet
                                          0
                                              0   100000       200000         300000        400000      500000        600000
                                                                            Repetitions



Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                               63
Report number CR-2003/44
            Figure 18. Stiffness reduction to steady state stiffness, Cullinan (D2388), 403A4
                       (Sandstone)

                                       1600
                                                                                                    407 MDD4 80kN
                                       1400                                                         407 MDD8 80kN
                                                                                                    407 MDD12 80kN
                                                              80 kN   80 kN, HVS moved              MDD4 model
                                       1200                                                         MDD8 model
             Resilient Modulus (MPa)



                                                                                                    MDD12 model
                                       1000
                                                                                                    80 kN, wet
                                        800

                                        600

                                        400

                                        200

                                          0
                                              0      200000     400000      600000         800000     1000000    1200000
                                                                          Repetitions

            Figure 19. Stiffness reduction to steady state stiffness, Cullinan (D2388), 407A4,
                       (Sandstone)



                                       1600
                                                                                        408 MDD4 40kN
                                       1400                                             408 MDD12 40kN
                                                                                        MDD4 model
                                                                                        MDD12 model
                                       1200
             Resilient Modulus (MPa)




                                       1000

                                        800

                                        600
                                                                                                    40 kN   40 kN, wet
                                        400

                                        200

                                          0
                                              0   100000 200000 300000 400000 500000 600000 700000 800000 900000
                                                                          Repetitions

            Figure 20. Stiffness reduction to steady state stiffness, Cullinan (D2388), 408A4
                       (Sandstone)




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                            64
Report number CR-2003/44
                                       3500
                                                                                                 410A4 MDD4 40kN
                                                                                                 410A4 MDD12 40kN
                                       3000
                                                                                                 410B4 MDD4 40kN
             Resilient Modulus (MPa)                                                             410A4 MDD4 model
                                       2500                                                      410A4 MDD12 model
                                                                                                 410B4 MDD4 model
                                       2000


                                       1500


                                       1000


                                       500
                                                                                 80 kN    100 kN                       wet
                                         0
                                              0   50000 100000 150000 200000 250000 300000 350000 400000 450000 500000
                                                                           Repetitions

            Figure 21. Stiffness reduction to steady state stiffness, Vereeniging (P243/1),
                       410A4/B4



                                       2500

                                                                                                           412 MDD4 40kN
                                                                                                           412 MDD8 40kN
                                       2000                                                                412 MDD12 40kN
                                                                                                           MDD4 model
             Resilient Modulus (MPa)




                                                                                                           MDD8 model
                                       1500                                                                MDD12 model




                                       1000



                                        500

                                                                                         40 kN     80 kN
                                          0
                                              0     200000    400000    600000     800000        1000000     1200000   1400000
                                                                           Repetitions

            Figure 22. Stiffness reduction to steady state stiffness, Vereeniging, 412A4


            From the data shown in Figure 17 to Figure 22, steady state stiffnesses were estimated
            and are shown in Table 36 with relevant section data.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                 65
Report number CR-2003/44
            Table 36.                               Estimated steady state stiffnesses for HVS sections from Cullinan and
                                                    Vereeniging

                                                                                      Residual       Cement       Base          Steady state
        Section                                                    Material            binder        content   thickness          stiffness
                                                                                     content (%)       (%)       (mm)              (MPa)
                                                  397A4                                                           100
                                                  403A4
        Cullinan                                                   Sandstone             0.9            1                             275
                                                  407A4                                                             150
                                                  408A4
                                                  410A4/B4                                                                            600
        Vereeniging                                                Ferricrete            1.8            2           250
                                                  412A4                                                                               400

            The models shown in the figures were used to determine the exact number of load
            repetitions required to reach the steady state stiffnesses shown in Table 36 for all of the
            suitable data from the MDDs on each section. For many sections, particularly those
            trafficked at 40 kN, this required extrapolation beyond the actual repetitions tested at the
            first wheel load. The load repetitions to reach the steady state stiffness are termed the
            stiffness reduction phase life, and are shown in Figure 23 as a function of the HVS
            trafficking wheel load. The stiffness reduction phase lives appear realistic, in that the life
            decreases with an increase in the trafficking load. It is however, fairly difficult to compare
            the data because of the different base layer thicknesses, support condition, and material
            types. These differences are not accounted for in Figure 23.


                                              1.0E+07
             Stiffness reduction phase life




                                              1.0E+06
                    (load repetitions)




                                              1.0E+05


                                                             Cullinan, 397A4 (100mm), 40 kN
                                                             Cullinan, 403A4 (100mm), 40kN
                                              1.0E+04        Cullinan, 407A4 (150mm), 80 kN
                                                             Cullinan, 408A4 (150mm), 40 kN
                                                             Vereeniging, 410A4, 80 kN
                                                             Vereeniging, 410B4, 100 kN
                                                             Vereeniging, 412A4, 40 kN
                                              1.0E+03
                                                        20    30      40        50       60     70      80     90         100   110
                                                                                      Wheel Load (kN)

            Figure 23. Stiffness reduction phase life as a function of wheel load


            To develop a stiffness reduction transfer function for general pavement analysis, it is
            necessary to replace the wheel load in Figure 23 with a generic stress or strain
            parameter. The parameter that has been used for foamed bitumen treated materials and
            lightly cemented materials is the strain ratio (Equation (13)), where the tensile strain ( ε t )
            is calculated from an analysis of the pavement and the strain-at-break ( εb ) is determined
            from the flexural beam test. The process to convert the wheel load to the strain ratio is
            described in the next section.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                 66
Report number CR-2003/44
                                                            εt
                                           Strain ratio =                                            (13)
                                                            εb



7.2.        Calculation of Strain Ratio

            To calculate the strain ratio it is necessary to determine the tensile strain at the bottom of
            the emulsified bitumen treated base layer for the HVS pavement structures and
            representative strain-at-break values. The strain is calculated using multi-layer linear
            elastic theory. Only the Cullinan and Vereeniging data are used, because of the lack of
            adequate input data for the Heilbron sections.

7.2.1       Multi-layer Linear Elastic Analyses of HVS Pavement Structures


            To calculate the stresses using multi-layer linear elastic (MLLE) analyses, the pavement
            layer thicknesses, stiffnesses, Poisson ratios, wheel load and tyre pressure must be
            known. The software program CIRCLY was used for the MLLE calculations. The
            stiffnesses were determined from the back-calculated MDD deflections. The stiffnesses
            of the emulsified bitumen treated base layer have been presented earlier in this report,
            but the stiffnesses used for all layers were calculated in the same back-calculation
            procedure. The stiffnesses at the beginning of the tests were used, i.e., the initial
            condition, known as the N10 readings in HVS terminology. The Poisson’s ratios were the
            same as used for the backcalculation of the stiffnesses (0.25 for all layers). The
            pavement structures in the HVS tests were determined from test pits. The thicknesses of
            layers beneath the bottom of the test pit were the same as those used for the
            backcalculation. This included the depth of the rigid layer, which was included in this
            forward-calculation. The wheel loads and tyre pressures used to calculate the stresses
            were the same as those used to traffic the HVS test sections during the first phase of
            loading (see Table 5 for Cullinan tests, and Table 7 for Vereeniging tests).

            The strains were determined at 2 locations in the base layer, under and between the dual
            wheels at a depth of 1 mm above the bottom of the base layer. These positions are
            illustrated in Figure 24. The critical value is selected from the maximum of the two
            positions. Some MLLE software packages can sometimes give unrealistic results at the
            layer interfaces, and calculating the strain 1 mm above the interface ensures a realistic
            value. The difference between a correct calculation at the interface and 1 mm above the
            interface is insignificant. The principal tensile strain, which is always horizontal or very
            close to horizontal, is used.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                67
Report number CR-2003/44
                                                                                       Surfacing

                                                                                       Base


                                                                                       Subbase



            Figure 24. Locations to calculate strain ratio in pavement structure under the dual
                       wheels


7.2.1.1.    Tensile Strain Results for Actual HVS Sections


            The critical principal tensile strain values ranged from 90 to 470 microstrain, with one
            value reaching 660 microstrain. For Cullinan Section 407A4/B4, using the stiffnesses
            back-calculated from the 80 kN load, unrealistically high tensile strains were calculated.
            This is because of low back-calculated values for the support layers. Using the 40 kN
            back-calculated stiffnesses gave more reasonable results. The stiffnesses of the base
            layer for both the 40 and 80 kN loads were very similar and therefore it is acceptable to
            use the 40 kN back-calculated stiffnesses to calculate the tensile strain. As no
            reasonable values were obtained for the steady state stiffness of Cullinan Section 403A4,
            no strain values were calculated.

7.2.2       Representative Strain-At-Break Values


            The representative strain-at-break values used to calculate the strain ratio are shown in
            Table 37. These average values were selected from the laboratory test results
            (Chapter 4). These values, and the critical tensile strain values were used to calculate
            the strain ratio.

            Table 37.    Strain-at-break values for HVS test sections

                                         Residual binder         Cement content         Strain-at-break
                    Material
                                           content (%)               (%)                 (microstrain)
            Cullinan (Sandstone)               0.9                    1                       114
            Vereeniging                        1.8                     2                      178

7.3.        Stiffness Reduction Transfer Function

            In Figure 25 the wheel loads used in Figure 23 have been replaced with the strain ratio
            calculated for each HVS pavement structure. The stiffness reduction phase life is then
            represented by the number of load repetitions of the wheel load used to calculate the
            tensile strain. The strain ratios range from 0.5 to 3.4. Strain ratios exceeding 1 are often
            found. This is because the boundary conditions in a MLLE pavement analysis (a
            simulation of the pavement behaviour) and the flexural beam test are different, and
            therefore the strain ratio is not necessarily a one-to-one comparison of the tensile strain in


Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                68
Report number CR-2003/44
            the pavement and the maximum allowable tensile strain.                                            Despite these differences, the
            strain ratio is a valid parameter.


                                                1.0E+07

               Stiffness reduction phase life


                                                1.0E+06
                      (load repetitions)




                                                1.0E+05




                                                1.0E+04
                                                                Cullinan, 397A4 (100mm), 40 kN            Cullinan, 407A4 (150mm), 80 kN
                                                                Cullinan, 408A4 (150mm), 40 kN            Vereeniging, 410A4, 80 kN
                                                                Vereeniging, 410B4, 100 kN                Vereeniging, 412A4, 40 kN
                                                                all data                                  excluding SR>3
                                                1.0E+03
                                                          0.0        1.0           2.0                  3.0          4.0           5.0
                                                                                         Strain ratio

            Figure 25. Stiffness reduction phase lives as a function of the strain ratio


            The data in Figure 25 show that as the strain ratio increases, due to either an increase in
            the tensile strain or a decrease in the strain-at-break, the stiffness reduction phase life is
            reduced. A log-linear model was fitted to the all of the data, and is shown as the dotted
            line in the figure. Without data at strain ratio of 4 and 5, it is not possible to determine if
            this log-linear model is correct for this range of strain ratios. The strain ratio value
            exceeding 3 seems unreasonably high, especially since it is from a section tested with a
            40 kN load, and is considerably different to the other MDD on the section (408A4). It was
            thus excluded from the data set and the resulting model fit is shown as the solid line in
            Figure 25. It was felt that this transfer function was more reasonable, and was therefore
            used. Given the variability in the data, it is not possible to fit the transfer functions
            following only statistical guidelines; engineering judgement is essential.

            The stiffness reduction transfer function for emulsified bitumen treated materials is
            compared with those of foamed bitumen treated materials and lightly cemented materials
            in Figure 26. It must be noted that the end of the stiffness reduction phase life for lightly
            cemented materials is defined as 2 mm of deformation and 0.5 to 0.75 mm of elastic
            deflection. This definition is different to bitumen treated materials




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                    69
Report number CR-2003/44
                                                         1.0E+07

                                                                                                          Emulsified bitumen


             Effective fatigue life (load repetitions)
                                                                                                          Foamed bitumen
                                                         1.0E+06                                          Lightly cemented




                                                         1.0E+05




                                                         1.0E+04




                                                         1.0E+03
                                                                   0.0   1.0   2.0                  3.0          4.0                5.0
                                                                                     Strain Ratio


            Figure 26. Comparison of stiffness reduction phase life transfer functions for
                       emulsified bitumen treated materials, foamed bitumen treated materials
                       and lightly cemented materials


            At higher strain ratios the emulsified bitumen treated materials have a longer stiffness
            reduction phase life than the other materials shown. The emulsified bitumen treated
            material consistently has a longer stiffness reduction phase life than the foamed bitumen
            treated materials. Feedback from consultants using the foamed bitumen stiffness
            reduction function indicates that the predicted stiffness reduction phase lives are
            conservative. Analyses of the Vereeniging data used to develop the foamed bitumen and
            emulsified bitumen treated materials identified that the support under the emulsified
            bitumen treated materials was better than that under the foamed bitumen treated
            materials24. This support probably resulted in the conservative stiffness reduction transfer
            function. It is therefore reasonable that the emulsified bitumen transfer function predicts
            longer stiffness reduction phase lives. A transfer function developed for EBTM using only
            the Vereeniging data would have predicted even longer stiffness reduction phase lives,
            due to the very good support conditions. Including the Cullinan data has resulted in a
            more general stiffness reduction transfer function.

            The emulsified bitumen stiffness reduction transfer function was extended to include
            reliability, as defined in the SAMEDM8. This was done by analysing the residuals
            (differences between the predicted and measured values) used to fit the function. For a
            given level of reliability, e.g. 95 per cent, that percentile of the residuals must be
            negative, that is the prediction is greater than the measured value and the measured data
            will lie below the fitted line. The resulting transfer functions for 95 per cent reliability
            (typically associated with Category A roads), 90 per cent (Category B), 80 per cent
            (Category C) and 50 per cent (Category D) are given in Equation (14) and are illustrated
            in Figure 27. The 50 per cent reliability transfer function gives considerably longer lives
            than the other levels. This is because it is at a low reliability level, and is essentially the
            deterministic case fitted by linear regression.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                        70
Report number CR-2003/44
                                                                                         ⎡              ⎛ε     ⎞⎤
                                                                                         ⎢ 6.254 − 0.612⎜ t
                                                                                                        ⎜      ⎟⎥
                                                                                                               ⎟
                                                                                                        ⎝ εb
                                                                 Category A : Nsr = 10 ⎢
                                                                                       ⎣                       ⎠⎥⎦

                                                                                         ⎡              ⎛ε     ⎞⎤
                                                                                         ⎢ 6.279 − 0.612⎜ t
                                                                                                        ⎜      ⎟⎥
                                                                                                               ⎟
                                                                                                        ⎝ εb
                                                                 Category B : Nsr = 10 ⎢
                                                                                       ⎣                       ⎠⎥⎦
                                                                                                                                                 (14)
                                                                                         ⎡              ⎛ε     ⎞⎤
                                                                                         ⎢ 6.349 − 0.612⎜ t
                                                                                                        ⎜ε     ⎟⎥
                                                                                                               ⎟
                                                                                         ⎢                     ⎠⎥
                                                                 Category C : Nsr = 10   ⎣              ⎝ b      ⎦

                                                                                         ⎡              ⎛ε     ⎞⎤
                                                                                         ⎢ 6.659 − 0.612⎜ t
                                                                                                        ⎜ε     ⎟⎥
                                                                                                               ⎟
                                                                                         ⎢                     ⎠⎥
                                                                 Category D : Nsr = 10   ⎣              ⎝ b      ⎦




                                                    where Nsr          = stiffness reduction phase life
                                                          εt           = maximum tensile strain at bottom or layer ( µε )
                                                                εb     = strain-at-break from laboratory testing ( µε )
                                                                εt
                                                                       = strain ratio
                                                                εb


                                                1.0E+07

                                                                                                                     95% Reliability (Cat A)
                                                                                                                     90% Reliability (Cat B)
               Stiffness reduction phase life




                                                1.0E+06                                                              80% Reliability (Cat C)
                                                                                                                     50% Reliability (Cat D)



                                                1.0E+05




                                                1.0E+04




                                                1.0E+03
                                                          0.0          1.0              2.0               3.0                 4.0              5.0
                                                                                              Strain ratio

            Figure 27. Stiffness reduction phase life for emulsified bitumen treated materials


            The stiffness reduction transfer function is sensitive to the strain ratio. This is illustrated
            with the following example. For example, with a tensile strain of 200 microstrain, by
            doubling the strain-at-break from 100 to 200, the stiffness reduction phase life increases
            from approximately 100 000 to 400 000 load repetitions, a four-fold increase. This
            example illustrates the importance of measuring the strain-at-break for the particular
            material and not relying on historical data or from data obtained from a table of
            recommended values, especially when the material is significantly different.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                              71
Report number CR-2003/44
8.          PERMANENT DEFORMATION TRANSFER FUNCTION


            This chapter describes the development of the permanent deformation transfer function
            for the second phase in the pavement life of emulsified bitumen treated materials. The
            second phase occurs when the material exhibits a steady state stiffness, at the end of the
            stiffness reduction phase life. The transfer function was developed from the dynamic
            triaxial tests, and then validated with HVS in-depth MDD data. It is desirable that the
            transfer function contains a term to account for the binder and cement contents.

8.1.        Ratio of Cement and Binder Contents

            In TG2, the permanent deformation transfer function included a variable for the ratio of
            the cement to binder content. This accounts for the different mix behaviour at different
            cement and residual binder contents. The use of a ratio was more appropriate than a
            product, for example, a cement content of 1 per cent and a binder content of 2 per cent
            gives a ratio of 0.5 and a product of 2, whereas a cement content of 2 per cent and binder
            content of 1 per cent gives a ratio of 2, but also a product of 2. It is not reasonable that
            the same value for the variable is used for these two situations, therefore the product of
            the cement and binder contents is not the best parameter to use.

            With the cement to binder content ratio, the cement content is the numerator, to ensure
            that if no cement is used the term is not undefined. However, if no cement is used, the
            ratio equals zero regardless of the residual binder content. Another problem with the
            cement to binder content ratio is that if, for example, both the cement and binder contents
            are doubled, the ratio remains the same, which is also not desirable.

            A number of additional options for a ratio of the cement and binder contents were
            investigated, including:
            •      (     binder
                       cement + x   ) , where x is an integer
            •      (   binder + x
                       cement + y   ) , where x and y are integers
            •             binder
                        cement + x
                                     , where x is an integer
            Of the ratios listed above, the first one gives the most reasonable values for a variable to
            describe the effects of the cement and binder interaction. This ratio is never undefined
            and will only equal zero when no binder is included. The ratio is linear for increases in
            the binder or cement contents. The ratio decreases with an increase in the cement
            content, and increases with an increase in the residual binder content.

            A value of 2 for the integer x gives a reasonable range of values for the ratio. The range
            of values is 0 to 1.5 for cement contents of 0 to 2 per cent and 0 to 3 per cent residual
            binder. Where a ratio of the binder and cement contents is used in the transfer function,
            the ratio    (     binder
                             cement + 2   )   is thus used. The transfer functions should not be used for materials
            that contain no binder and cement, or cement only. The amount of binder and cement
            that should be included in the mix must be determined from a mix design procedure. The
            nature of different material is accounted for in the calculation of the stress ratio, and the
            binder and cement ratio determines the effects of adjustments of the binder or cement
            contents.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                         72
Report number CR-2003/44
            When data for mixes using inert or no fillers is obtained, this ratio will not longer be valid,
            and an appropriate means of including the different filler types will be investigated.

8.2.        Laboratory Permanent Deformation Model

            The dynamic triaxial permanent deformation laboratory tests were used to develop a
            permanent deformation transfer function. This model is known as the laboratory
            permanent deformation model, and its development is described in this section.

            The model fits to the permanent deformation data from the dynamic triaxial tests were
            described in Sections 5.6.3 and 6.3, and the details are given in Table 26, Table 27,
            Table 32, and Table 33. These model fits were used to determine the number of load
            repetitions to reach set levels of plastic strain, i.e., 1, 4, 7, 10, 13, 17 and 20 per cent.
            These data are shown in Figure 28 for an arbitrarily selected 17 per cent plastic strain.
            The varying relative densities and saturation levels are not identified on the figure.


                                    1.E+09

                                    1.E+08

                                    1.E+07
              Structural capacity
               (load repetitions)




                                    1.E+06

                                    1.E+05

                                    1.E+04

                                                   Vereeniging 2% Cement, 1.8% Binder
                                    1.E+03         Vereeniging 1% Cement, 3.0% Binder
                                                   Vereeniging 1% Cement, 3.0% Binder (omitted)
                                    1.E+02         Vereeniging 1% Cement, 0.9% Binder
                                                   Cullinan 1% Cement, 0.9% Binder
                                                   Cullinan 1% Cement, 1.5% Binder
                                    1.E+01
                                             0.0   0.1   0.2   0.3   0.4   0.5   0.6   0.7   0.8   0.9   1.0   1.1    1.2
                                                                            Stress ratio
            Figure 28. Repetitions to 17 per cent plastic strain from dynamic triaxial
                       permanent deformation tests


            Theoretically the stress ratio should not exceed 1. However, with the variability
            introduced into measurement of the density and moisture contents and calculation of the
            stress ratio, values slightly exceeding 1 are sometimes found, as seen in Figure 28. The
            Vereeniging data at 1 per cent cement and 3 per cent binder between stress ratios of 0.6
            and 0.7 are unreasonably low. These low values could not be explained and were
            excluded from the data set from which the permanent deformation transfer function was
            developed.

            It is clear from Figure 28 that at the stress ratios close to 1, the structural capacity
            decreases dramatically. This can be expected because at these high stress ratios the
            material approaches its shear strength, and therefore cannot sustain many repetitions
            before failure.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                              73
Report number CR-2003/44
            Fitting a conventional log-linear model through the data in Figure 28 results in the
            structural capacity being overestimated at high stress ratios. A more appropriate model is
            a cubic polynomial, which can accommodate both the low structural capacities at the
            higher stress ratios and the gradual decrease in structural capacity between stress ratios
            of 0.2 to 0.8. To this end, two different cubic polynomials were attempted. The first
            included terms for the stress ratio raised to the power of 3, 2 and 1 (SR3, SR2, SR),
            whereas the second included only the stress ratio to the power 3 (SR3). The first model
            resulted in an increase in the structural capacity as the stress ratio increased between 0.4
            and 0.7 that is physically not realistic. The second cubic polynomial always decreases
            with an increase in the stress ratio, and this rate of decrease accelerates as the stress
            ratio exceeds 0.8. This model does not give as high structural capacities at the low stress
            ratios as the first model. This is, however, conservative.

            Once the type of cubic polynomial was selected, other variables were also included in the
            model, such as the relative density, saturation, plastic strain and the ratio of binder and
            cement contents     (     binder
                                    cement + 2   ).   It was found that the relative density and saturation
            components were not statistically significant and should therefore be excluded from the
            laboratory permanent deformation model. The relative density values ranged from 73 to
            81 per cent and the saturation levels ranged from 25 to 109 per cent, which is a large
            enough range for any trends to be detected statistically. Therefore, the statistical
            insignificance does not imply that the density and saturation of a material have no
            influence on the behaviour. Rather, these influences are accounted for in the calculation
            of the maximum allowable principal stress, and hence the stress ratio, which depends on
            the saturation and relative density levels (Sections 5.5.2 and 6.2). In any estimation of
            the structural capacity the stress ratio must be calculated at the expected density and
            saturation. This means that either a model such as that used for the Vereeniging and
            Cullinan materials must be used to calculate the maximum allowable principal stress as a
            function of the saturation and relative density, or the cohesion and friction angle values
            should be appropriate for the saturation and relative density. It is preferable to actually
            measure the shear strength with static triaxial tests performed at the expected relative
            densities and saturation level.

            A material indicator variable for the type of material was also included in a trial model to
            assess the significance of the material type. This was done in the same manner as for
            the models predicting the maximum allowable principal stress (Equation (10), Section
            6.2.4). While it was found that the type of material was statistically significant, the effect
            on the structural capacity was in the same order of magnitude as the variability of the
            data. It was thus not reasonable to include such a term in a generally applicable
            permanent deformation transfer function. Dependence on the material type is, however,
            an area for further investigation, and every effort should be made to develop a universally
            applicable model.

            The final form of the laboratory permanent deformation function is shown in Equation
            (15), and the fitted model coefficients and p-values are given in Table 38. This model is
            applicable to both the Vereeniging and Cullinan materials. The differences in the quality
            of the parent materials are captured in the calculation of the stress ratio from the static
            triaxial tests.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 74
Report number CR-2003/44
                                                       ⎛ binder ⎞
                            log N = c1 + c 2 SR3 + c 3 ⎜            ⎟ + c 4 PS                     (15)
                                                       ⎝ cement + 2 ⎠

                         where N         = log of load repetitions
                               SR        = stress ratio
                               binder    = binder content (%)
                               cement    = cement content (%)
                               PS        = plastic strain (%)
                               c1-4      = regression coefficients

            Table 38.     Model coefficients and statistics for Equation (15)

                                                           Equation (15)
                                                    Coefficient       p-value
            c1 (constant)                             6.9956          0.0000
            c2 (stress ratio)                        -2.3477          0.0000
            c3 ( cement + 2 )
                   binder
                                                      -0.6558             0.0000
            c4 (plastic strain)                       0.0679              0.0000
                              R2                                 0.716
             Standard error of the estimate                      0.498

            Figure 29 shows the data and the model fit for 17 per cent plastic strain. The model is
            demonstrated for the same combinations of cement and emulsified bitumen as the
            original data set. The model shows a fairly rapid reduction in the structural capacity (load
            repetitions to reach the selected level of plastic strain) at stress ratios larger than 0.8.
            The worst fit is experienced between 0.7 and 0.8. Under field conditions, stress ratios in
            this range, and higher, only occur in unusual circumstances.

            The lowest structural capacity is estimated for a mix with the highest binder content. As
            the binder content decreases at the same cement content, the structural capacity
            increases. The mixes with binder contents less than 3 per cent all give very similar
            performance. For a given binder content, an increase in the cement content will result in
            an increase in the structural capacity. These data are consistent with trends in the shear
            strength of the materials, and trends observed with foamed bitumen treated materials7.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              75
Report number CR-2003/44
                                     1.E+09


                                     1.E+08


                                     1.E+07
               Structural Capacity
                (load repetitions)
                                     1.E+06

                                                          Vereeniging 2% Cement, 1.8% Binder
                                     1.E+05               Vereeniging 1% Cement, 3% Binder
                                                          Vereeniging 1% Cement, 0.9% Binder
                                                          Cullinan 1% Cement, 0.9% Binder
                                     1.E+04               Cullinan 1% Cement, 1.5% Binder
                                                          Model, 2% Cement, 1.8% Binder
                                                          Model, 1% Cement, 3.0% Binder
                                     1.E+03
                                                          Model, 1% Cement, 0.9% Binder
                                                          Model, 1% Cement, 1.5% Binder
                                     1.E+02
                                              0.0   0.1   0.2   0.3   0.4   0.5   0.6   0.7    0.8   0.9   1.0   1.1   1.2
                                                                             Stress Ratio
            Figure 29. Laboratory permanent deformation transfer (to arbitrarily selected 17%
                       plastic strain)


8.3.        Heavy Vehicle Simulator Permanent Deformation Data

            In this section, the MDD in-depth permanent deformation data from the Heilbron, Cullinan
            and Vereeniging HVS test sections are compared and analysed, and then compared with
            the laboratory permanent deformation transfer function. The HVS data were discussed in
            Chapter 3. The transfer functions are developed from the permanent deformation that
            occurred from the beginning of the test. However, because the majority of the permanent
            deformation occurs after the end of the stiffness reduction phase life, the transfer function
            is used for Phase 2 of the pavement life. In pavement design using the transfer functions
            a small amount of deformation is assumed to occur during Phase 1 (stiffness reduction
            phase life).

8.3.1       Permanent Deformation as a Function of Wheel Load


            To analyse the HVS permanent deformation data, the load repetitions to reach set levels
            of plastic strain as a function of the wheel load were investigated. These analyses are
            discussed for the Heilbron, Cullinan and Vereeniging sites individually, and then
            combined.

            The results for the Heilbron HVS test sections are shown in Figure 30, where the number
            of load repetitions to reach 17 per cent plastic strain are shown as a function of the wheel
            load. Each data point represents the data from one MDD. Note that the sections with 1
            and 2 per cent emulsion were tested in the channelized trafficking mode, and the results
            shown here were converted to the equivalent wandering traffic using the method
            described in Section 3.1.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                  76
Report number CR-2003/44
                                    1.E+10



                                    1.E+09
              Structural Capacity
              (Load Repetitions)
                                    1.E+08



                                    1.E+07



                                    1.E+06       Heilbron, 1% emulsion
                                                 Heilbron, 2% emulsion
                                                 Heilbron, 3% emulsion
                                    1.E+05
                                             0       20      40          60         80      100   120   140   160
                                                                              Wheel load (kN)

            Figure 30. Heilbron permanent deformation data as a function of wheel load (17
                       per cent plastic strain)


            The lowest number of load repetitions to 17 per cent plastic strain was achieved with the
            mix containing 3 per cent emulsion (1.8 per cent residual binder). This is consistent with
            similar observations from laboratory tests and from tests on foamed bitumen treated
            materials, where the addition of binder at a given active filler content reduces the shear
            strength and permanent deformation resistance.

            The Cullinan results are shown in Figure 31. The result for MDD8 at 40 kN from Section
            408A4 (150 mm) section was considerably lower than the other two MDDs from the same
            section.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                       77
Report number CR-2003/44
                                    1.E+10



                                    1.E+09
              Structural Capacity
              (Load Repetitions)
                                    1.E+08



                                    1.E+07



                                    1.E+06       Cullinan, 100mm

                                                 Cullinan, 150mm
                                    1.E+05
                                             0      10        20   30      40      50      60   70     80    90
                                                                        Wheel load (kN)

            Figure 31. Cullinan permanent deformation data as a function of wheel load (17%
                       plastic strain)


            The Vereeniging data are shown in Figure 32. These data follow the expected trend, and
            no outliers are apparent.



                                    1.E+10



                                    1.E+09
              Structural Capacity
              (Load Repetitions)




                                    1.E+08



                                    1.E+07



                                    1.E+06



                                    1.E+05
                                             0           20        40         60          80     100        120
                                                                        Wheel load (kN)
            Figure 32. Vereeniging permanent deformation data as a function of wheel load
                       (17% plastic strain)


            The results from all the HVS test sections are combined in Figure 33. The numbers of
            load repetitions to 17 per cent plastic strain are shown as a function of wheel load. Fitting
            a function through these data would result in a transfer function. However, this function

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                     78
Report number CR-2003/44
            would not be useful for general design and analysis as it would be specific to the
            particular materials, thicknesses and loads on the pavements, and the density and
            saturation of the materials involved. Accounting for these differences should reduce the
            scatter apparent on the figure.


                                    1.E+10
                                                                              Vereeniging
                                                                              Cullinan, 100mm
                                                                              Cullinan, 150mm
                                    1.E+09
                                                                              Heilbron, 3% emulsion
                                                                              Heilbron, 2% emulsion
              Structural Capacity
              (Load Repetitions)




                                                                              Heilbron, 1% emulsion
                                    1.E+08



                                    1.E+07



                                    1.E+06



                                    1.E+05
                                             0   20   40   60     80      100        120      140     160
                                                            Wheel load (kN)

            Figure 33. HVS permanent deformation data as a function of wheel load (17 per
                       cent plastic strain)

            In order to make use of the HVS permanent deformation data described above to
            calibrate the laboratory transfer function, it is necessary to convert the HVS data to the
            number of load repetitions required to produce specified levels of plastic strain as a
            function of the stress ratio, rather than the HVS wheel load. The procedure to convert the
            data is described in the next section.

8.3.2       Calculation of Stress Ratios in HVS Pavement Structures


            The stress ratio in the actual HVS pavement structures is calculated using multi-layer
            linear elastic theory. Only the Cullinan and Vereeniging data were used, as insufficient
            material data were available to calculate the stress ratios for the Heilbron sections.

8.3.2.1.    Multi-layer Linear Elastic Analyses Inputs


            To calculate the stresses using MLLE analysis, it was necessary to know the pavement
            layer thicknesses, stiffnesses and Poisson’s ratios, wheel load and tyre pressure. The
            stiffnesses were determined from the back-calculated MDD in-depth deflections. The
            stiffnesses at the beginning of the tests were used, i.e., the same stiffnesses used in the
            stiffness reduction transfer function calculations (Section 7.2.1), except for the emulsified
            bitumen treated base, which used the steady state stiffness (Section 7.1). This concurs
            with the use of the permanent deformation transfer function in the design and analysis of
            pavement structures. The steady state stiffnesses are 275 MPa for the Cullinan sections
            and 600 MPa and 400 MPa for the 410A4/B4 and 412A4 Vereeniging sections,
            respectively. The Poisson’s ratios, layer thicknesses, wheel loads and tyre pressures are
            the same as those used for the stiffness reduction transfer function calculations.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                     79
Report number CR-2003/44
            The stresses are determined at 12 locations in the base layer, four horizontal positions
            and three depths. The horizontal positions are between the dual tyres, under one tyre,
            and at the inner and outer edges of one tyre. The distance between the wheels is
            constant, but the position of the inner and outer edges depends on the wheel load and
            tyre pressure. The three depths are at one-quarter, half, and three-quarters of the base
            layer thicknesses. These positions are illustrated in Figure 34. The stress ratio is
            calculated at each position and the critical (maximum) value selected.




                                                                                           Surfacing

                                                                                           Base


                                                                                           Subbase


            Figure 34. Locations to calculate stress ratio in pavement structure under dual
                       wheels


8.3.2.2.    Stress Ratio Calculation


            The stress ratio was calculated using Equation (4), using the major and minor principal
            stresses determined from MLLE analyses, and the maximum allowable major principal
            stress calculated using Equation (5) (Section 5.5.2). The models developed for each
            individual material and specific binder and cement contents were used, i.e.,
            Equation (5)(e) for the Cullinan sandstone materials, and Equation (5)(g) for the
            Vereeniging materials.

            The models to determine the maximum allowable principal stress typically contain terms
            for the density and saturation of the materials. The dry density and the moisture contents
            measured at the actual test sections were used to calculate the relative density and
            saturation through Equations (2) and (3). These densities and moisture contents were
            obtained from construction data, nuclear density and moisture measurements taken at
            the beginning of the tests, or in the worst case, from test pit data. The test pit data could
            be misleading after water was added to the test section during trafficking. None of the
            values seemed to indicate that the materials were unreasonably wet, however. The
            maximum allowable principal stress models for Cullinan do not contain saturation and
            relative density terms, and therefore could introduce some error in the stress ratio
            calculation if the actual values differ from the values used to develop the models.

            MLLE analyses sometimes calculate tensile stresses, which are not physically possible
            for granular materials. This is a common problem and in the SAMEDM it is
            recommended that the tensile stresses are reduced to zero, and the compressive stress
            (maximum principal stress) increased by the corresponding amount. In this way, the
            deviator stress remains the same, i.e., for the principal stresses, the difference between
            the major and minor principal stress remains constant. This is not a foolproof method of
            counteracting the tensile stresses, but was selected to be consistent with the SAMEDM.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials             80
Report number CR-2003/44
            For comparison, the stress ratios were also calculated using the cohesion and friction
            angles for the relevant materials and binder and filler combinations, using Equation (5).
            The agreement in the two stress ratio values was good. An exact match is not expected
            when the actual density and saturation values differ from those used to calculate the
            shear strength parameters. The stress ratio was also calculated using the model for the
            maximum allowable principal stress for both materials (Equation (10)). These values
            showed fair agreement with the Vereeniging materials, but poor agreement for the
            Cullinan materials.

8.3.3       Structural Capacity as a Function of Stress Ratio in Pavement


            The structural capacities (load repetitions to selected level of plastic strain) from the HVS
            data, as a function of the stress ratios calculated from the MLLE pavement analyses are
            shown in Figure 35. Both the Vereeniging data and Cullinan data are shown together as
            the determination of the stress ratio should account for the different material properties,
            allowing the development of one transfer function, applicable for both materials. It was
            noted during the analyses that the critical stresses occurred at three locations in the
            various pavement structures, at one-quarter depth under the tyre and at the outside edge
            of the tyre and under the tyre in the middle of the base layer (see Figure 34). Only for the
            thinner, 100 mm base layer, pavements was the typical critical location in the middle of
            the layer, under the tyres.


                                     1.E+10
                                                                                               Vereeniging
                                     1.E+09                                                    Cullinan, 100mm
                                                                                               Cullinan, 150mm
               Structural Capacity
               (Load Repetitions)




                                     1.E+08


                                     1.E+07


                                     1.E+06


                                     1.E+05


                                     1.E+04
                                              0.0   0.1   0.2   0.3   0.4   0.5   0.6   0.7   0.8   0.9   1.0    1.1   1.2
                                                                             Stress Ratio

            Figure 35. Structural capacities of HVS test sections as a function of stress ratio
                       (17 per cent plastic strain)


            The data in Figure 35 shows that the slope of the Vereeniging data is steeper than that of
            the Cullinan data. This may be due to the high cement content of this material, although
            this trend was not apparent in the laboratory data. The stresses calculated through linear
            elasticity are predominantly a function of the contact stress between the tyre and
            pavement, and the depth in the pavement structure. The actual magnitude of the load
            has a lesser effect.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                 81
Report number CR-2003/44
            Although the different materials and material properties contribute to the total variability,
            some of the variability evident in Figure 35 may be due to the locations at which the
            stress ratios were calculated. The calculation points depend on the thickness of the
            layer, and so different layer thicknesses may result in significantly different stress ratios.
            The SAMEDP calculates the permanent deformation as an average across the whole
            layer, therefore the approach used is correct. However, in pavements, the permanent
            deformation does not occur uniformly through the layer, but occurs predominately in the
            upper portion of the layer. Attempts to adjust the calculation locations did not produce a
            significant improvement, and this should be investigated further.

            Predictions from the laboratory permanent deformation transfer function are
            superimposed on the HVS data in Figure 36. The structural capacity predicted from the
            laboratory transfer function for the specific stress ratio experienced in the HVS tests at
            the actual binder and cement contents are shown in Figure 37 for 17 per cent plastic
            strain. The correlation between the Cullinan data and the laboratory permanent
            deformation transfer function is good. The fit to the Vereeniging data is not as good, but
            the range of results is still reasonable because the stress ratios are in the range where
            the laboratory transfer function has a relatively flat slope. It is not possible to identify
            either the Cullinan or Vereeniging as more correct. It was very difficult to fit a model to
            the permanent deformation data for Section 412A4 (40kN, therefore low stress ratios)
            from Vereeniging, and the resulting model fit may introduce errors into estimates of the
            structural capacity. It must be remembered that large variability was also evident in the
            test results used to fit the laboratory permanent deformation model. In addition,
            Vereeniging base layers are considerably thicker than the Cullinan data, and averaging
            the permanent deformation over the whole layer may be contributing to the disagreement
            in the results.

                                    1.E+10
                                                                                             Vereeniging
                                                                                             Cullinan, 100mm
                                    1.E+09                                                   Cullinan, 150mm
                                                                                             1% Cement, 3.0% Binder
                                                                                             2% Cement, 1.8% Binder
              Structural capacity




                                                                                             1% Cement, 0.9% Binder
               (load repetitions)




                                    1.E+08
                                                                                             1% Cement, 1.5% Binder

                                    1.E+07


                                    1.E+06


                                    1.E+05


                                    1.E+04
                                             0.0   0.1   0.2   0.3   0.4   0.5   0.6   0.7   0.8   0.9   1.0   1.1    1.2
                                                                            Stress Ratio

            Figure 36. Laboratory predicted and HVS structural capacities (17 per cent plastic
                       strain)




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                               82
Report number CR-2003/44
                                               1E+09
                                                               Vereeniging
                                                               Cullinan, 100mm



               Predicted structural capacity
                                                               Cullinan, 150mm
                                                               Line of equality

                     (load repetitions)
                                               1E+08




                                               1E+07




                                               1E+06
                                                   1E+06           1E+07                       1E+08                      1E+09
                                         Actual HVS data
                                        (load repetitions)
            Figure 37. Comparison of predicted vs measured load repetitions to 17% plastic
                       strain


8.4.        Permanent Deformation Transfer Function

            Because of the acceptable correlation with the Cullinan and Vereeniging data, and the
            amount of data available for the laboratory permanent deformation model, it was decided
            to use the laboratory model as the permanent deformation transfer function. The transfer
            function was used to develop design catalogues and charts in the next chapter.

            The South African Mechanistic-Empirical Design Method specifies the reliability levels for
            four different categories of road (Category A to D). The transfer function shown in
            Equation (15) and in Figure 29 was modified by ranking the residuals and determining the
            residual values for the 95, 90, 80 and 50th percentiles, in the same manner as for the
            stiffness reduction transfer function. The resulting transfer functions for all four categories
            are shown in Equation (16). An example of the effect of the various levels of reliability is
            demonstrated in Figure 38 for 3 per cent bitumen and 1 per cent cement. For a given
            stress ratio, the predicted structural capacity is lowest for Category A, which has the
            highest reliability.

                                                                      ⎡                  3          ⎛ binder ⎞                  ⎤
                                                                      ⎢6.2395 − 2.3477 SR − 0.65583 ⎜            ⎟ + 0.06791 PS ⎥
                                                                                                    ⎝ cement + 2 ⎠
                                               Category A: NPD = 10 ⎣                                                           ⎦

                                                                      ⎡                  3          ⎛ binder ⎞                  ⎤
                                                                      ⎢6.3794 − 2.3477 SR − 0.65583 ⎜            ⎟ + 0.06791 PS ⎥
                                                                                                    ⎝ cement + 2 ⎠
                                               Category B: NPD = 10 ⎣                                                           ⎦
                                                                                                                                    (16)
                                                                      ⎡                  3          ⎛ binder ⎞                  ⎤
                                                                      ⎢6.6242 − 2.3477 SR − 0.65583 ⎜            ⎟ + 0.06791 PS ⎥
                                                                                                    ⎝ cement + 2 ⎠
                                               Category C: NPD = 10   ⎣                                                         ⎦

                                                                      ⎡                  3          ⎛ binder ⎞                  ⎤
                                                                      ⎢6.9526 − 2.3477 SR − 0.65583 ⎜            ⎟ + 0.06791 PS ⎥
                                                                                                    ⎝ cement + 2 ⎠
                                               Category D: NPD = 10 ⎣                                                           ⎦




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                               83
Report number CR-2003/44
                                      where NPD          = structural capacity
                                            SR           = stress ratio
                                            binder       = binder content (%)
                                            cement       = cement content (%)
                                            PS           = plastic strain



                                    1.E+10

                                    1.E+09

                                    1.E+08
              Structural capacity
               (load repetitions)




                                    1.E+07

                                    1.E+06

                                                     Category A
                                    1.E+05
                                                     Category B
                                                     Category C
                                    1.E+04
                                                     Category D
                                    1.E+03
                                             0.0   0.1   0.2   0.3    0.4    0.5   0.6     0.7   0.8   0.9   1.0   1.1
                                                                            Stress Ratio


            Figure 38. Permanent deformation transfer function for road categories, 17 per
                       cent plastic strain (3 per cent residual bitumen and 1 per cent cement)




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                            84
Report number CR-2003/44
9.          DESIGN CATALOGUES AND DESIGN CHARTS


            In this chapter, a classification system for four classes of emulsified bitumen treated
            materials is suggested. The system attempts to consider both the flexibility and shear
            strength of the material. Also in this chapter, using the transfer functions for stiffness
            reduction and permanent deformation and the South African Mechanistic-Empirical
            Design Method (SAMEDM), design catalogues for new construction and design charts for
            deep in situ recycling were developed. With deep in situ recycling projects, the support
            condition is not changed, and the pavement needs to be designed for the existing
            support. It was therefore not possible to use catalogues of designs that assume a
            standard support for these projects.

9.1.        Material Classification for Emulsified Bitumen Treated Materials

            Emulsified bitumen treated materials have, in the past, typically been classified according
            to their UCS or CBR strengths. However, such a classification ignores the flexibility of the
            materials, which contributes to the pavement life through the stiffness reduction phase.
            Classifying the material in terms of the shear strength only, results in flexibility being
            sacrificed for compressive strength, and this is not desirable for an optimal mix. Similar
            issues were identified for foamed bitumen treated materials; hence a classification system
            that accounts for both the flexibility and compressive strength was introduced in TG2.
            This system uses the UCS for the shear strength, and the ITS as a surrogate test for
            flexibility.

            In this work, two classification systems were developed, one using UCS and ITS tests,
            and the other using UCS and strain-at-break tests. Figure 39 shows the classification
            using UCS and ITS. Four material classes were identified. In comparison with the
            foamed bitumen treated materials, the UCS limits of the material classes have been
            modified. The UCS limit of the EB1 and EB2 materials has been reduced from the TG2
            UCS value of 1400 kPa to 1200 kPa to be in line with that suggested in Manual 211. The
            TG2 1400 kPa limit for foamed bitumen treated materials was selected on limited data.
            The lower UCS limit of the EB3 and EB4 material classes has been raised to 500 kPa,
            based on input from consultants experienced with emulsified bitumen treated materials.
            The other adaptation from the TG2 classification is the removal of the upper limits for the
            UCS of EB1 and EB2 classes, and ITS of EB1 and EB3 classes. Some materials exceed
            the upper limits, and while they may not be optimal mixes, they will perform adequately
            and should therefore not be excluded. It is recommended that the same revisions be
            made to the foamed bitumen treated material classification system.

            It is well known that the ITS test is not a good test of the flexibility of a material. The
            addition of cement usually results in an increased ITS, but its flexibility is reduced.
            However, currently there is no better routine test available for general use. A better
            measure of the flexibility of a material can be gained from the strain-at-break value
            determined by the monotonic flexural beam test. Data from this test have been used in
            this report, and were used in the stiffness reduction transfer function. It is recommended
            that this test be incorporated into standard practice. It is a relatively simple test, and
            should not require expensive, complex equipment. Implementation of the test into
            standard practice should be investigated further. A similar classification to that in

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              85
Report number CR-2003/44
            Figure 39 has been developed using strain-at-break instead of ITS, and is shown in
            Figure 40. Also shown in the figures are the values of the materials used in this report.
            No ITS results were available for the Cullinan materials, and strain-at-break results were
            not available for all materials.


                         2400
                                      EBITS2                                  EBITS1
                         2200
                         2000
                                                                                               1% Cement, 0.9%
                         1800                                                                  Binder, Vereeniging
                                                                                               1% Cement, 1.8%
                         1600                                                                  Binder, Vereeniging
             UCS (kPa)




                         1400                                                                  1% Cement, 3.0%
                                                                                               Binder, Vereeniging
                         1200                                                                  2% Cement, 1.8%
                         1000                                                                  Binder, Vereeniging
                                                                                               2% Cement, 3.0%
                          800                                                                  Binder, Vereeniging
                          600         EBITS4                                  EBITS3

                          400
                          200
                            0
                                0   100        200      300       400       500         600
                                                     ITS (kPa)

            Figure 39. Material classification using UCS and ITS of Cullinan and Vereeniging
                       materials



                         2400
                                      EB2                                     EB1              1% Cement, 0.9%
                         2200                                                                  Binder, Vereeniging
                         2000                                                                  1% Cement, 1.8%
                         1800                                                                  Binder, Vereeniging

                         1600                                                                  1% Cement, 3.0%
                                                                                               Binder, Vereeniging
             UCS (kPa)




                         1400
                                                                                               2% Cement, 1.8%
                         1200                                                                  Binder, Vereeniging
                                      EB4
                         1000                                                                  2% Cement, 3.0%
                                                                                               Binder, Vereeniging
                          800
                                                                                               1% Cement, 0.9%
                          600                                                     EB3
                                                                                               Binder, Cullinan
                          400
                          200
                            0
                                0   100        200      300        400      500          600
                                            Strain-at-break (microstrain)

            Figure 40. Material classification using UCS and strain-at-break of Cullinan and
                       Vereeniging materials


            When the catalogues were developed, there was a discernible difference between the
            pavement structures required for the mix with 3 per cent binder and 1 per cent cement
            and the other mixes in class EBITS4, and therefore the UCS and strain-at-break
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials       86
Report number CR-2003/44
            classification is more reasonable because this mix is in a different class. The mix with 1.8
            per cent binder and 2 per cent cement is in the highest class in Figure 39, whereas it is
            classified as EB2 in Figure 40. This material was not of a high quality, and the binder and
            cement contents are not optimal, and therefore it is not reasonable that it has the highest
            ranking. For these reasons, it is recommended that the material classification for
            emulsified bitumen treated materials be based on the UCS and strain-at-break tests, for
            the limits shown in Table 39 and Figure 40. It is recommended that a UCS and strain-at-
            break classification system be investigated for foamed bitumen treated materials. Use of
            this classification system in a guideline document will require the strain-at-break to be
            used in standard laboratories, the implications of which need to be investigated further.

            Table 39.    Material classification of emulsified bitumen treated materials

                                                                Strain-at-break (microstrain)
                                                               100 to 300             > 300
                                         > 1200                   EB2                  EB1
                 UCS (kPa)
                                       400 to 1200                EB4                  EB3

            The UCS test is also not a perfect test for measuring the shear strength of a material,
            although it is a better measure of shear strength than the ITS is of flexibility. It may be
            possible in future to replace the UCS value in the classification by some measure of
            shear strength, possibly obtained from the Texas triaxial test.

            As the classification system is based purely on the behaviour of the treated material in
            terms of measured engineering parameters; there is no consideration for the quantity of
            binder added. Differentiation between modification and stabilization is no longer
            recommended for emulsified bitumen treated materials, and should alleviate some of the
            current confusion surrounding the use of these terms.

9.2.        Materials and Input Values

            The pavement structures recommended in the catalogues contain asphalt, emulsified
            bitumen treated, lightly cement treated and granular materials. This section discusses
            the material properties of these materials, including the recommended ranges of material
            property values and the specific values used in the analyses. The values used for the
            asphalt, cemented and granular materials were the same as those used in the analyses
            performed to revise TRH428.

9.2.1       Surfacing
            Two types of surfacing are used in the pavement designs, a thin asphalt layer (30 mm) or
            a thin bituminous seal. In the analyses, the material properties used for the asphalt layer
            are shown in Table 40. A thin bituminous seal is a non-structural layer. When a seal is
            utilized, a layer thickness of 10 mm is assumed with the same material properties used
            for the base.

            Table 40.    Asphalt Material Properties

             Material                              Asphalt
             Elastic modulus (MPa)                  3000
             Poisson’s ratio                        0.44



Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              87
Report number CR-2003/44
9.2.2       Emulsified Bitumen Treated Materials


            The catalogues were initially developed using the data from 5 mixes. Three of these
            mixes used the Vereeniging material, and two used the Cullinan material. The
            catalogues and design charts were developed for a moderate climatic condition. Should
            other climatic conditions be prevalent, the full ME design procedure should be followed.

            Observations of the pavement structures developed for 5 mixes allowed some mixes to
            be combined, and allowed the finalisation of the material classification system described
            in Section 9.1 above. The input values used for all five mixes are shown in Table 41,
            together with their material classifications. The binder and cement contents listed for
            each mix are not prescribed quantities for the particular material class.

            Table 41.     Emulsified bitumen treated material properties

            Material class                     EB2               EB3                              EB4
            Parent material                Vereeniging       Vereeniging        Vereeniging       Cullinan      Cullinan
            Binder content (%)                  1.8                3                0.9             0.9           1.5
            Cement content (%)                   2                 1                 1               1             1
                                              2680              1360               1660            1020          1060
            Stiffness* (MPa)
                                              (500)             (500)              (500)           (275)         (275)
            Strain-at-break (µε)               178               553                160             114           263
            Relative density (%)               73.5             na***              na***           na***         na***
            Equation**(5)                       (g)               (a)               (b)             (e)           (d)
            *       Initial stiffness, steady state stiffness in parenthesis
            **      Model used to calculate the maximum allowable principal stress for the stress ratio
            ***     Maximum principal stress model is not a function of density for these particular materials because of
                    an insufficient range of relative density data from which to fit the model


            A density of 98 per cent of Mod AASHTO maximum dry density was assumed, and at
            least this value should therefore be specified in the construction specifications. Failure to
            achieve this density could result in inadequate structural capacity of the pavement. A
            saturation level of 55 per cent was assumed for the moderate conditions. This saturation
            value, and the relative density shown in the Table are calculated incorporating the
            aggregate, cement and binder (see Section 4.1). A Poisson’s ratio of 0.35 was assumed
            for the material in both the initial and steady states. This value is assumed for most
            pavement analyses of granular or treated materials, and was assumed in the calculations
            of the catalogues in TRH43.

9.2.3       Cemented Materials
            In the catalogue, only one lightly cemented material class was used, namely C4. The
            assumed equivalent granular state of a C4 material is an EG5, for the purpose of
            developing the catalogue. The material properties shown in Table 42 were used for this
            material. These properties were used in the TRH4 revision analyses and are within the
            ranges recommended in the SAMEDM.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                88
Report number CR-2003/44
            Table 42.     Cemented layer material properties

            Material                                 Cemented, C4     Equivalent Granular, EG5
            Elastic modulus (MPa)                       1500                    200
            Poisson’s ratio                             0.35                    0.35
            Strain-at-break (µε)                         145
            Cohesion (kPa)                                                         29
            C-term                                                                100*
            Friction angle (kPa)                                                   40
            φ-term                                                                2.8*
            * Moderate moisture condition assumed


9.2.4       Granular Materials
            Several categories of granular materials were used to develop the catalogue. The
            material properties for these materials are given in Table 43. Material classes below G7
            were analysed as subgrade layers using the subgrade transfer function, not the granular
            layer transfer function. The subgrade transfer function is a function of the vertical
            compressive strain, and therefore does not directly account for a material’s shear strength
            properties8.

            Table 43.     Granular layer material properties

             Material                                G6     G7       G8     G9       G10
             Elastic modulus (MPa)                  200    120       90     70        45
             Poisson’s ratio                        0.35   0.35     0.35   0.35      0.35
             Cohesion (kPa)                          27     24
             C-term                                 84*    66*
             Friction angle (degrees)                37     35
             φ-term                                 2.3*   1.9*
             * Moderate moisture condition assumed




9.3.        Mechanistic Analysis

            This section describes the mechanistic analyses to determine the parameters used in the
            pavement life calculations. The analyses were based on MLLE analyses. The inputs to
            the software are the material properties (discussed above in Section 9.2), layer thickness,
            loading, and the locations in the pavement at which the stresses and strains are
            evaluated.

9.3.1       Loading
            The same loading as in the TRH4 revision analyses was used, which is a single axle load
            of 80 kN, i.e., dual wheels with 20 kN on each wheel and a tyre pressure of 520 kPa.
            MLLE programs assume a circular load area. The centres of the tyres are 350 mm apart.

9.3.2       Layer thickness
            A minimum layer thickness of 100 mm was used in the catalogue analyses. Increments
            in layer thickness of 25 mm were used; it was felt that smaller increments are impractical.
            The maximum layer thickness was 300 mm, and for this thickness the layer may need to
            be compacted in two lifts. No rigid layer is assumed in the analyses. The thicknesses
            shown in the catalogues and design charts are minimum thicknesses. Construction
            tolerances should be added to ensure that the minimum thickness is obtained under all
            circumstances.
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials             89
Report number CR-2003/44
9.3.3       Critical parameters and evaluation locations
            The stress or strain parameter used to determine the pavement life depends on the type
            of material in the layer, and the phase. For all layers, except the emulsified bitumen
            treated layers, the stress or strain parameters prescribed in the SAMEDM were used8.
            These parameters and locations are briefly described below:

            •     Emulsified bitumen treated
                  In the stiffness reduction phase, the principal tensile strains at the bottom of the
                  layer both directly beneath and between the dual tyres were calculated. The
                  values were actually calculated 1 mm above the bottom of the layer to eliminate
                  erroneous calculations at the interface. The principal tensile strain was typically
                  horizontal, or close to horizontal. In the steady state stiffness, the major and minor
                  principal stresses were determined at 9 locations in the pavement: one quarter
                  below the top of the layer, the middle of the layer and one quarter above the
                  bottom of the layer, under, between and at the outside edge of the wheels, as
                  shown in Figure 41. Note that in comparison with Figure 34, the locations beneath
                  the inside edge of the tyre have been eliminated, as they were not found to be
                  critical locations.
            •     Cemented treated
                  In the cemented phase of the pavement life, the tensile strain 1 mm above the
                  bottom of the layer, both underneath and between the tyres was evaluated. Again
                  the actual value was calculated 1 mm above the layer interfaces. The maximum
                  principal tensile strain was used in further analyses. In the equivalent granular
                  phase of the pavement life the major and minor principal stresses were evaluated
                  in the middle of the layer, underneath and between the wheels.
            •     Granular
                  Similar to the equivalent granular state of cemented materials, the major and minor
                  principal stresses were calculated underneath and between the wheels, in the
                  middle of the layer.
            •     Subgrade
                  The vertical compressive strain was evaluated at 1 mm below the top of the
                  subgrade layers under and between the tyres, with the maximum value used to
                  calculate the pavement life.




                                                                                         Surfacing

                                                                                         Base


                                                                                         Subbase


            Figure 41. Locations to calculate stress ratio in pavement structure




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              90
Report number CR-2003/44
9.4.        Pavement Structural Capacity

            The next step in designing or analysing pavement structures is to calculate the structural
            capacity of the entire pavement. The procedure was described in detail by Theyse8. The
            structural capacity was determined by the number of load repetitions to reach the terminal
            distress of the critical layer.

            The terminal distress conditions assumed are 20 mm of rutting or shear failure in the
            critical layer or fatigue cracking on the surface of the pavement8. The extent of the
            distress depends on the road category. Five percent of the total design section length is
            allowed to have failed at the end of the structural design life of the pavement for road
            category A, whereas road categories B, C and D allow 10, 20 and 50 percent,
            respectively8.

            For thin asphalt surfacings or seals, the structural capacity of the layer is not considered
            in the determination of the critical layer. If it was, it would often be the critical layer
            resulting in a reduced structural capacity. However, regular surface maintenance retards
            distress in this layer, preventing the layer from reaching a terminal distress condition. It
            was therefore reasonable to neglect this layer in the structural capacity estimate, but
            surface maintenance should be done at regular intervals. This was also assumed in the
            development of the 1996 revision of TRH4.

9.4.1       Pavement Behaviour Phases


            When treated materials, such as emulsified bitumen or cemented materials, are included
            in the pavement structure, the analysis is performed in phases. During the first phase,
            the layer is in an intact, undamaged condition with a relatively high initial stiffness. This
            phase ends when the layer reaches a steady state stiffness, with the time to reach this
            state defined as the stiffness reduction phase life. The term “steady state stiffness” is
            used to describe the material behaviour phase after the loss in resilient modulus
            (stiffness) and is comparable with granular materials only in the stiffness and not in the
            physical state of the materials. The material is in a loose condition consisting of individual
            particles.

            When two stabilised layers are included in the pavement structure, the pavement may
            need to be analysed in three phases. In the first phase, all layers are in the stiffness
            reduction phase. The critical layer is the one with the shortest stiffness reduction phase
            life, and this life defines the end of Phase 1. In Phase 2, the modulus of that critical layer
            is reduced to the steady state stiffness and the pavement is reanalysed. During this
            phase, the stiffness reduction phase life of the second stabilised layer is typically
            reached. In Phase 3, the steady state stiffness both treated layers are used. The
            pavement life is not a simple sum of the lives of each phase. The method of determining
            the structural capacity at the end of each phase and the total pavement structural
            capacity is described by Theyse8.

            Although the stiffness reduction phase life is not a terminal condition, the reduced
            stiffness increases the tensile strain at the bottom of the asphalt layer, and therefore
            reduces the life of the surfacing. The shorter the stiffness reduction phase life, the
            quicker the surfacing will crack and water will be able to penetrate the base layer. This
            means that the maintenance frequencies are increased. Routine maintenance for crack
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                91
Report number CR-2003/44
            sealing, or the application of an additional surface seal, may be necessary prior to
            reaching the terminal structural distress condition, i.e., 20 mm of rutting in the base,
            subbase or subgrade. It is therefore beneficial to use additives that provide a long
            stiffness reduction phase life without sacrificing permanent deformation resistance.

9.4.2       Transfer Functions


            The transfer function for the stiffness reduction phase of emulsified bitumen treated
            materials was discussed in Section 7.3 and is given in Equation (14), and the permanent
            deformation transfer function is given in Equation (16), Section 8.4. In the permanent
            deformation analyses, the plastic strain is the terminal deformation expressed as a
            percentage of the layer thickness.        The terminal deformation in the permanent
            deformation phase was assumed to be 18 mm, 2 mm of deformation is assumed to occur
            during the stiffness reduction phase. This agrees with findings on with cemented
            materials where the end of the stiffness reduction (effective fatigue) phase is defined as 2
            mm of deformation and 0.5 to 0.75 mm of elastic deflection under an 80kN axle load at
            the top of the cemented layer8.

            The transfer function for the stiffness reduction phase of cemented layers relates the ratio
            of the tensile strain at the bottom of the layer and the strain-at-break from a flexural beam
            test to the number of load repetitions to reach the steady state stiffness. In the equivalent
            granular phase, the transfer function for granular materials is used. This transfer function
            is discussed by Theyse8. Crushing of the cemented layers was ignored because the
            cemented layers are not at the top of the pavement and therefore not at risk of crushing.

            The transfer function for granular layers relates the factor of safety calculated from the
            stresses in the pavement to the number of load repetitions to shear failure8. The c- and
            φ-terms in Table 42 and Table 43 are used to calculate the factor of safety. The minimum
            factor of safety under and between the tyres was used in the transfer function.

            The life of the subgrade support layers is calculated with a transfer function that relates
            the vertical compressive strain at the top of the layer to the number of load repetitions to
            reach 20 mm of permanent deformation in the layer. This transfer function is also
            discussed by Theyse8.

9.5.        Design Catalogues

            The design catalogues for new construction are given in Figure 42, Figure 43 and Figure
            44 for the EB2, EB3 and EB4 material classes, respectively. The thicknesses shown in
            the catalogues are the minimum thicknesses required for the required structural capacity,
            construction tolerances should be added. Similarly, the asphalt thickness indicated in the
            catalogue should also be increased to accommodate both a levelling and a wearing
            course when required.


            Design catalogues are typically shown for the four road categories, Category A to D,
            which are associated with decreasing levels of reliability3,4. When TG2 was published,
            the design catalogues were criticised because there were no pavement structures in the 3
            to 10 million E80s (ES10) or higher traffic classes for the FB2 and FB3 materials. It is
            however possible to design such pavements structures at lower levels of reliability for

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               92
Report number CR-2003/44
            these material classes. In light of this, it seems more appropriate to highlight the chosen
            level of reliability, rather than the road category in the catalogues.

            Separate catalogues were developed for all three materials classified as EB4. However,
            the catalogues were very similar, and were therefore combined into one catalogue,
            shown in Figure 44.

            The EB2 material is not necessarily a better material than the EB3 material, as the
            materials may have similar structural capacities.         The EB2 material has more
            compressive strength, and therefore more permanent deformation resistance, whereas
            the EB3 material has more flexibility, which lengthens the stiffness reduction phase. The
            selection of the appropriate material class depends on the required pavement
            performance and available materials.

            It was not possible to identify a pavement structure that will satisfy the ES10 traffic class
            for Category A roads. Simply increasing the thickness does not always increase the
            pavement life. Because the permanent deformation transfer function is dependent on the
            plastic strain, as the layer thickness increases the plastic strain to reach the terminal rut
            depth decreases, and hence the structural capacity is reached quicker. This only occurs
            with thicker layers. Similar observations were made with the catalogues for foamed
            bitumen treated materials. These observations should be validated with data from the
            long-term performance of pavements.

            It is interesting to note that the catalogues for lightly cemented materials in TRH4 also do
            not have pavement structures adequate for Category A roads and ES10 and higher traffic
            classes28. The only material with pavement structures adequate for these high class
            roads is crushed stone28. These are possible because the granular material transfer
            function seems to overestimate the bearing capacity at high factors of safety (low stress
            ratios). The granular material transfer function was also not calibrated for 20 mm of
            permanent deformation, as has been done for foamed and emulsified bitumen treated
            materials. It is recommended that a new transfer function be developed for granular
            materials.

            It may be possible to achieve longer pavement lives for pavement structures utilizing
            good quality parent materials, i.e., an EB1 material. This will be assessed after laboratory
            testing of the N7 crushed stone material treated with emulsified bitumen has been
            completed.

            A catalogue for emulsified bitumen treated materials was developed using the DCP
            design method15. As part of that project, several consultants, including Adrian Bergh,
            were requested to develop a catalogue based on their judgement and experience. These
            catalogues were used to calibrate the one developed using the DCP design method.
            Comparison of that catalogue and the ones developed from the mechanistic-empirical
            transfer functions show that the pavement structures are very similar for Categories B to
            D for the EB2 and EB3 material class. The EB2 and EB3 pavement structures
            recommended in Figure 42 and Figure 43 are the same, very similar to those developed
            using the DCP method. The designs for Category A roads do not agree, in that the DCP
            catalogues have structures in the ES10 and ES30 traffic classes. The reasons that these
            structures cannot be obtained for the mechanistic-empirical catalogues were discussed
            above.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               93
Report number CR-2003/44
            Because the DCP catalogue was validated using the experience of leading consultants,
            and it agrees with the mechanistic-empirical catalogues, it is felt that the catalogues
            recommended in this report are reasonable.

            Comparison of the catalogues for foamed bitumen and emulsified bitumen show that
            pavements with these materials essentially have the same structural capacity. It is
            therefore envisaged that it may, with additional data, be possible to develop one universal
            model for bitumen stabilized materials and to have one set of catalogues and design
            charts applicable to both foamed and emulsified bitumen treated materials.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials             94
Report number CR-2003/44
Design reliability                            PAVEMENT CLASS AND DESIGN BEARING CAPACITY (80 kN AXLES/LANE)
                        ES0,003       ES0,01      ES0,03       ES0,1         ES0,3          ES1           ES3        ES10        ES30         ES100       Foundation
(Road category)       0,1–0,3x10
                                 4
                                     0,3–1x10
                                              4
                                                  1–3x10
                                                         4
                                                              3–10x10
                                                                      4
                                                                          0,1–0,3x10
                                                                                     6
                                                                                         0,3–1x10
                                                                                                  6
                                                                                                        1–3x10
                                                                                                               6
                                                                                                                   3–10x10
                                                                                                                           6
                                                                                                                               10–30x10
                                                                                                                                        6
                                                                                                                                            30–100x10
                                                                                                                                                      6


                                                                                         Seal         30 AC
95% approximate
                                                                                         150 EB2      150 EB2
design reliability
                                                                                         150 C4       200 C4
(A: Major
interurban
                                                                                         Seal         30 AC
freeways and
                                                                                         200 EB2      200 EB2
roads)
                                                                                         200 G6       200 G6
                                                                          Seal           Seal         30 AC
90% approximate                                                           100 EB2        125 EB2      150 EB2
design reliability                                                        100 C4         150 C4       175 C4
                                                                                                                                                             150 G7
(B: Interurban
                                                                                                                                                             150 G9
collectors and                                                            Seal           Seal         30 AC
                                                                                                                                                               G10
major rural roads)                                                        125 EB2        175 EB2      200 EB2
                                                                          125 G6         175 G6       200 G6
                                                                          Seal           Seal         30 AC
80% approximate                                                           100 EB2        125 EB2      125 EB2
design reliability                                                        100 C4         125 C4       175 C4
(C: Lightly
trafficked rural                                             Seal         Seal           Seal         30 AC
roads and strategic                                          100 EB2      100 EB2        175 EB2      175 EB2
roads)                                                       100 G6       125 G6         175 G6       175 G6

50% approximate
design reliability
(D: Light                                                                                                                                                    150 G9
pavement                                                                                                                                                       G10
                                                             Seal
structures, rural
                                                             100 EB2
access roads)
                                                             100 G8

                Figure 42. Design catalogues for emulsified bitumen treated materials, class EB2




  Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials             Report number CR-2003/44                                   95
                                              PAVEMENT CLASS AND DESIGN BEARING CAPACITY (80 kN AXLES/LANE)
 Road Category          ES0,003       ES0,01       ES0,03       ES0,1         ES0,3          ES1           ES3        ES10        ES30         ES100       Foundation
                                 4            4           4            4              6            6            6           6            6             6
                      0,1–0,3x10     0,3–1x10      1–3x10      3–10x10     0,1–0,3x10     0,3–1x10       1–3x10     3–10x10     10–30x10     30–100x10
                                                                                          Seal         30 AC
95% approximate
                                                                                          125 EB3      150 EB3
design reliability
                                                                                          150 C4       200 C4*
(A: Major
interurban
                                                                                          Seal
freeways and
                                                                                          175 EB3
roads)
                                                                                          175 G6
                                                                           Seal           Seal         30 AC
90% approximate                                                            100 EB3        125 EB3      150 EB3
design reliability                                                         100 C4         125 C4       175 C4                                                 150 G7
(B: Interurban                                                                                                                                                150 G9
collectors and                                                             Seal           Seal         30 AC                                                    G10
major rural roads)                                                         125 EB3        175 EB3      175 EB3
                                                                           125 G6         175 G6       200 G6
                                                                           Seal           Seal         30 AC
80% approximate
                                                                           100 EB3        100 EB3      125 EB3
design reliability
                                                                           100 C4         125 C4       175 C4
(C: Lightly
trafficked rural
                                                              Seal         Seal           Seal         30 AC
roads and strategic
                                                              100 EB3      100 EB3        150 EB3      150 EB3
roads)
                                                              100 G6       150 G6         150 G6       200 G6
50% approximate
design reliability
(D: Light                                                                                                                                                     150 G9
pavement                                                                                                                                                        G10
                                                  Seal                     Seal
structures, rural
                                                  100 EB3                  125 EB3
access roads)
                                                  100 G8                   125 G8
                * compact in two lifts


                Figure 43. Design catalogues for emulsified bitumen treated materials, class EB3




  Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              Report number CR-2003/44                                   96
                                              PAVEMENT CLASS AND DESIGN BEARING CAPACITY (80 kN AXLES/LANE)
 Road Category          ES0,003       ES0,01       ES0,03        ES0,1         ES0,3          ES1           ES3        ES10        ES30         ES100       Foundation
                                 4            4           4             4              6            6            6           6            6             6
                      0,1–0,3x10     0,3–1x10      1–3x10       3–10x10     0,1–0,3x10     0,3–1x10       1–3x10     3–10x10     10–30x10     30–100x10
95% approximate                                                                            Seal         30 AC
design reliability                                                                         150 EB4      175 EB4
(A: Major                                                                                  200 C4*      200 C4*
interurban
freeways and
roads)
                                                                            Seal           Seal         30 AC
90% approximate                                                             100 EB4        150 EB4      175 EB4
design reliability                                                          125 C4         150 C4       175 C4*
                                                                                                                                                               150 G7
(B: Interurban
                                                                                                                                                               150 G9
collectors and                                                              Seal           Seal
                                                                                                                                                                 G10
major rural roads)                                                          150 EB4        200 EB4
                                                                            175 G6         200 G6
                                                                            Seal           Seal
80% approximate
                                                                            100 EB4        100 EB4
design reliability
                                                                            125 C4         200 C4*
(C: Lightly
trafficked rural
                                                              Seal          Seal           Seal
roads and strategic
                                                              100 EB4       150 EB4        175 EB4
roads)
                                                              125 G6        150 G6         200 G6
50% approximate
design reliability
(D: Light                                                                                                                                                      150 G9
pavement                                                                                                                                                         G10
                     Seal           Seal          Seal          Seal
structures, rural
                     125 EB4        125 EB4       150 EB4       150 EB4
access roads)
                     125 G8         125 G7        150 G8        150 G7
                * Construct in two lifts. Increase thickness, if necessary, for ease of construction.


                Figure 44. Design catalogues for emulsified bitumen treated materials, class EB4




  Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               Report number CR-2003/44                                   97
9.6.        Design Charts

            The design charts for deep in situ recycling projects are given in Figure 45, Figure 46 and
            Figure 47 for material classes EB2, EB3 and EB4. The charts assume a thin bituminous
            surfacing seal. To use the charts, it is necessary to have an estimate of the stiffness of
            the combined support below the recycled layer. This support layer is of semi-infinite
            thickness. An iterative process is required to determine this stiffness, as it is influenced by
            the layers that are not disturbed during recycling. For example, if only some of the
            subbase is recycled, the stiffness of the support will be higher than if all the subbase was
            recycled with the base. The thickness of the recycled layer is determined using the
            stiffness of the support, and the contours of the traffic classes.

            The layer thicknesses range from 75 to 300 mm. If a 300 mm layer is used, it may need
            to be compacted in two lifts to ensure that the required density levels are reached
            throughout the layer. The thicknesses shown are minimum thicknesses, and therefore
            construction tolerances should be added.

            The design charts allow a more direct comparison of emulsified and foamed bitumen
            treated materials. Such a comparison showed that the materials are very similar, but
            when differences in the structural capacity do occur, the emulsified bitumen has a slightly
            higher structural capacity. This is expected due to the trends that were observed in the
            laboratory and HVS data used to develop both transfer functions. The stiffness reduction
            transfer function for foamed bitumen treated materials is probably conservative. Hence, if
            this function is revised it is possible that the design charts for both materials may be
            essentially the same. This supports the development of a universal model for bituminous
            stabilized materials.

            The design charts should be used with great care, and every effort should be made to
            determine the relevant input data for the specific project. It is recommended that the
            design charts are used to check a design done using the full mechanistic-empirical design
            procedure.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                 98
Report number CR-2003/44
                      300                                                                                                                             300
                                                                                                                                                                         3 000 000          Design traffic, E80s
                                       3 000 000            Design traffic, E80s                  95% Design Reliability                              275                                                                    90% Design Reliability
                      275                                                                                                                                                                                                            EB2
                                                                                                          EB2
                                       1 000 000                                                                                                      250
                      250                                                                                                                                                1 000 000
                                                                                                                                                      225




                                                                                                                                 Thickness (mm)
                      225
  Thickness (mm)




                      200                                                                                                                             200
                                           300 000                                                                                                                        300 000
                      175                                                                                                                             175

                      150                                                                                                                             150

                                           100 000                                                                                                    125
                      125                                                                                                                                                 100 000
                                            30 000                                                                                                                         30 000
                      100                                                                                                                             100                  10 000
                                            10 000
                                             3 000                                                                                                                          3 000
                              75                                                                                                                               75
                                   0                  50            100            150        200            250           300                                      0                  50            100              150    200          250          300
                                                                            Support (MPa)                                                                                                                    Support (MPa)

                              300                                                                                                                                  300
                                                                                                                                                                          10 000 000
                                        3 000 000           Design traffic, E80s                                                                                                               Design traffic, E80s
                                                                                                  80% Design Reliability                                                                                                     50% Design Reliability
                              275                                                                                                                                  275
                                                                                                          EB2                                                                                                                        EB2
                                                                                                                                                                          3 000 000
                              250                                                                                                                                  250
                                        1 000 000
                                                                                                                                                                   225




                                                                                                                                                  Thickness (mm)
                              225
             Thickness (mm)




                                                                                                                                                                          1 000 000
                              200                                                                                                                                  200

                              175                                                    10 000 000                                                                    175
                                            300 000
                              150                                                                                                                                  150
                                                                                                                                                                             300 000
                              125                                                                                                                                  125
                                            100 000
                                                                                                                                                                             100 000
                              100            30 000                                                                                                                100        30 000
                                             10 000                                                                                                                           10 000
                                              3 000                                                                                                                            3 000
                               75                                                                                                                                   75
                                       0               50            100        150           200            250           300                                           0              50            100             150    200         250          300
                                                                            Support (MPa)                                                                                                                    Support (MPa)

                                             Figure 45. Design charts for emulsified bitumen treated materials, class EB2




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                                                        Report number CR-2003/44                                                     99
                   300                                                                                                               300

                                                                                     95% Design Reliability                          275                                                                 90% Design Reliability
                   275
                                                                                             EB3                                                                                                                 EB3
                                                  Design traffic, E80s                                                                              Design traffic, E80s
                   250                                                                                                               250

                             1 000 000                                                                                               225




                                                                                                                    Thickness (mm)
                   225
  Thickness (mm)




                                                                                                                                               1 000 000
                                                                                                                                                                              3 000 000
                   200                                                                                                               200

                   175
                                                                                                                                     175
                                                                                                                                                300 000
                                 300 000                                                                                                        100 000
                                                                                                                                     150
                   150           100 000
                                                                                                                                                 30 000
                                  30 000                                                                                                         10 000
                                  10 000                                                               3 000 000                     125
                   125
                                   3 000                                                                                                           3 000
                                                                                                                                     100
                   100

                                                                                                                                      75
                    75
                         0                  50             100           150        200          250          300
                                                                                                                                           0               50          100            150            200            250           300
                                                                    Support (MPa)                                                                                                 Support (MPa)

                   300                                                                                                               300

                                                                                     80% Design Reliability                                                                                                50% Design Reliability
                   275                                                                                                               275
                                                                                             EB3                                                                                                                   EB3
                                             Design traffic, E80s
                   250                                                                                                               250
                                 3 000 000                                                                                                                 Design traffic, E80s
                   225                                                                                                               225




                                                                                                                    Thickness (mm)
  Thickness (mm)




                                 1 000 000                                                                                                     3 000 000
                   200                                                                                                               200
                                                                                                                                                                                            10 000 000
                   175                                                                                                               175
                                  300 000                                                                                                      1 000 000
                   150            100 000                                                                                            150         300 000
                                                                                                                                                 100 000
                                   30 000
                   125                                                                                                               125          30 000
                                   10 000                                                                                                         10 000
                                    3 000                                                                                                          3 000
                   100                                                                                                               100

                    75                                                                                                                75
                             0               50            100         150          200          250          300                          0               50           100            150           200             250            300
                                                                   Support (MPa)                                                                                                  Support (MPa)

                                   Figure 46. Design charts for emulsified bitumen treated materials, class EB3




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                             Report number CR-2003/44                                                        100
                                                                                                                                                                300
                                    300
                                              1 000 000                                                                                                                              1 000 000
                                                                   Design traffic, E80s                                                                                                300 000              3 000 000                                  90% Design Reliability
                                                300 000                                                   95% Design Reliability                                275
                                    275                                                                                                                                                                                                                        EB4
                                                100 000                                                           EB4                                                                    100 000
                                                                                                                                                                250                                                       Design traffic, E80s
                                    250

                                                30 000                                                                                                          225




                                                                                                                                          Thickness (mm)
                                    225                                                                                                                                                   30 000
                   Thickness (mm)




                                    200
                                                                                                                                                                200
                                                10 000
                                                                                                                                                                                          10 000

                                    175
                                                                                                                                                                175

                                    150
                                                                                                                                                                150
                                                 3 000                                                                                                                                     3 000
                                    125
                                                                                                                                                                125

                                    100
                                                                                                                                                                100

                                     75
                                                                                                                                                                            75
                                          0               50           100           150            200            250             300                                           0                   50          100             150             200            250             300
                                                                                Support (MPa)                                                                                                                              Support (MPa)
                         300
                                              300 000                                                                                                                       300
                                                               1 000 000     Design traffic, E80s          80% Design Reliability                                                        1 000 000         3 000 000
                         275                                                                                       EB4                                                                                                                                 80% Design Reliability
                                                                                                                                                                            275
                                                                       3 000 000                                                                                                          300 000                      Design traffic, E80s                    EB4
                                              100 000
                         250                                                                                                                                                250
                                                                                                                                                                                          100 000
                         225                                                                                                                                                225
  Thickness (mm)




                                                                                                                                                           Thickness (mm)
                                               30 000
                         200                                                                                                                                                200
                                                                                                                                                                                           30 000
                                               10 000
                         175                                                                                                                                                175
                                                                                                                                                                                           10 000

                         150                                                                                                                                                150
                                                                                                                                                                                            3 000
                                                3 000
                         125                                                                                                                                                125

                                                                                                                                                                            100
                         100
                                                                                                                                                                             75
                                    75
                                                                                                                                                                                     0                50          100            150          200              250          300
                                         0                50           100         150              200              250            300
                                                                                                                                                                                                                            Support (MPa)
                                                                               Support (MPa)


                                                 Figure 47. Design charts for emulsified bitumen treated materials, class EB4




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                                                                                                     Report number CR-2003/44                                                              101
9.7.        Validation of Design Models with Field Data

            Emulsifed bitumen treated materials have been used in South African pavements for
            many years, with few reported failures. Ideally, this experience should be utilised to
            calibrate the structural design models developed in this project. It is difficult to use field
            sections to actually develop structural design models, as the level of material and
            pavement data required are typically more than are available. The lack of sufficient data
            also makes it difficult to use the data to calibrate the models. Despite the lack of data,
            such valuable field experience should not be ignored.

            In the late 1980s a project to evaluate the long-term pavement performance of EBTM
            pavements was begun on several test sections identified throughout the country29.
            Unfortunately, the project was stopped after two years and monitoring of the test sections
            was terminated. The material data from the first round of tests is very limited30.

            One of the test sections identified for a potential LTPP site is the N12, previously denoted
            as S12. In 1968 several test sections were constructed on this section of road. The first
            two test sections had a cement-treated base, which were rehabilitated in 1974. The
            rehabilitation involved replacement of the cement-treated base with an emulsified bitumen
            treated base. No cement was added, and 0.9 per cent residual binder was used in the
            treatment. This pavement has performed well since 1974, and may have had some
            maintenance treatments, but no further rehabilitation has been performed31. Because
            this section was initially constructed as a test section, more data are available than for
            most field pavements.

            This section could possibly be used to validate the structural design models developed for
            EBTMs. However, this will require very detailed data, which are not currently available. It
            is recommended that an in depth study of the test section be performed to obtain all data
            that could be relevant. This study should include an investigation into the construction of
            the sections, any data obtained during the life of the pavement, and an assessment of the
            current condition of the pavement. If this investigation provides data that are usable for a
            quantitative calibration of the structural design models, it is recommended that the
            procedure be repeated for other emulsified bitumen treated field pavements, such as
            those identified in the late 1980s.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               102
Report number CR-2003/44
10.         CONCLUSIONS AND RECOMMENDATIONS


            Until this project, no reasonable mechanistic-empirical structural design models existed
            for emulsified bitumen treated materials. This report describes the development of such
            models and thereby satisfies Deliverable 4 of the project (Appendix A). The models were
            based on the behaviour of emulsified bitumen treated materials in HVS test sections and
            sophisticated laboratory test results. Data from three HVS test sites were analysed:
            Heilbron, Cullinan and Vereeniging.          The Heilbron sections were constructed
            conventionally, Cullinan was constructed using labour-intensive methods and the
            Vereeniging sections were constructed using deep in situ recycling. Extensive laboratory
            testing was performed in conjunction with the Cullinan and Vereeniging HVS tests.
            Unfortunately little relevant material data on the Heilbron sections were available. The
            Heilbron HVS data were therefore, where possible, used to identify and confirm trends in
            the behaviour of the material, specifically the stiffness reduction, and permanent
            deformation accumulation in the HVS test sections. These data were not used to develop
            the mechanistic-empirical structural design models.

            The modes of distress of emulsified bitumen treated materials identified were stiffness
            reduction and permanent deformation. The stiffness reduction phase life is the number of
            load repetitions required before high initial stiffness reduces to a steady state stiffness.
            This steady state stiffness is reached when the stiffness is equivalent to a granular
            material; the material is not in a loose, particulate or cracked form. The second phase of
            distress is the accumulation of permanent deformation.

            Previous guidelines1,2 for emulsified bitumen treated materials differentiated between
            modification and stabilisation, depending on the quantity of emulsion added. In the work
            described in this report, this distinction is no longer used. The materials are differentiated
            only on their behaviour, and not on their binder quantity present. It is recommended that
            future guidelines for these materials do not differentiate between modification and
            stabilisation.

10.1.       Laboratory Test Data

            Examination of the laboratory test data identified a deficiency gap in the available data,
            and consequently additional laboratory testing was performed. The additional testing
            ensured that data from a range of emulsified bitumen and cement contents were
            available. The laboratory testing showed that an increase in the cement content
            increased the shear strength of the treated material, but an increase in the binder content
            decreased the shear strength and the resilient modulus. This implies that the addition of
            cement increases the permanent deformation resistance, but the addition of binder has
            little effect on, or decreases, the permanent deformation resistance. The addition of
            binder does, however, significantly improve the flexibility of the material, which results in
            an increase in the stiffness reduction phase of the pavement life. It is essential that the
            cement and binder contents are balanced to ensure adequate flexibility and permanent
            deformation resistance.

            Analyses of the static triaxial test data showed that the cohesion and friction angles could
            be misleading measures of the shear strength because they are unduly affected by the

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               103
Report number CR-2003/44
            method of calculation used to determine the values. These shear strength parameters
            are also only valid for a particular relative density and saturation condition. However,
            one representative value is not applicable to all situations. For these reasons, it was
            decided to use regression models that predict the maximum allowable principal stress as
            a function of the confining stress, relative density and saturation for each material at each
            binder and cement content. These models negate the need to calculate the shear
            strength parameters, and can account for variations in relative density and saturation. A
            model for all materials that included terms for the binder and cement content and the
            parent material was also attempted, and showed promise although it is not yet suitable for
            wide application. It is recommended that this be investigated further. In the analyses in
            this report, the models developed for each material at each combination of binder and
            cement contents were used. Unfortunately, for some of the materials there were
            insufficient variations in relative density or saturation levels for these variable to be
            statistically significant, which resulted in them being excluded from the model. Including
            the terms in the model when they are statistically insignificant can lead to misleading
            results. The resulting model is strictly only applicable to the density and saturation level
            ranges in the data set used to develop the model. Although the relative density and
            saturation are statistically insignificant for some the data sets, this does not imply that
            these terms do not influence material behaviour.

10.2.       Stiffness Reduction Transfer Function

            The stiffness reduction transfer function was developed from the elastic stiffness data
            backcalculated from in-depth MDD deflections. The stiffness reduction phase life is a
            function of the strain ratio, which is the ratio of the tensile strain at the bottom of the
            emulsified bitumen treated layer and the strain-at-break value determined from the
            flexural beam test. This transfer function is given in Equation (14), Section 7.3.

10.2.1      Limitations


            There are limitations in the stiffness reduction transfer function. One of the major
            limitations is the variability in the data used to develop the function, particularly the
            Cullinan data, which is most likely due to the method of construction (labour–intensive).
            Because of the variability, engineering judgement was essential in developing the
            functions. These issues are, however, always present and were not unique to the
            development of transfer functions in this report. The data used for the development of
            the transfer functions in the South African Mechanistic-Empirical Design Method8 also
            had large variability, which highlights the importance of carefully evaluating any answer
            obtained from any transfer function before implementing it.

            Another limitation in the stiffness reduction transfer function is the use of the strain-at-
            break parameter from the flexural beam test. This is not a standard test, but
            unfortunately no other readily available test adequately measures the flexibility of a
            material. It should be simple for the flexural beam test to be performed in standard
            laboratories and widely adopted in practise.

            The stiffness reduction transfer function does not contain a variable for neither material
            type, nor binder and cement contents. Unfortunately, because the development of this
            transfer function relies on HVS data, insufficient data are available from which to develop
            a model which includes these variables. The strain-at-break and consequently the strain
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials              104
Report number CR-2003/44
            ratio is very dependant on these material properties, therefore the predicted phase life is
            influenced by the material properties, although they do not appear in the transfer function.

            The data sets used to develop the transfer functions all contained cement. This work
            does not intend to advocate the use of cement. It is recommended that the influence of
            other fillers is investigated.

10.3.       Permanent Deformation Transfer Function


            The permanent deformation transfer function was developed using permanent
            deformation dynamic triaxial tests. The transfer function was checked against the HVS
            permanent deformation behaviour. The transfer function is dependent on the stress ratio
            in the pavement, calculated at various locations, the cement and binder contents and the
            plastic strain to obtain any level of permanent deformation in the layer. The transfer
            function is given in Equation (16), Section 8.4.

10.3.1      Limitations


            A limitation in the analyses performed to develop the permanent deformation transfer
            function is the use of linear elastic theory. In the calculation of the stress ratio, the major
            and minor principal stresses were calculated through linear elasticity. The theory
            sometimes predicts tensile horizontal stresses, which are not physically possible in
            granular materials. This is a common problem, and is not unique to the analyses
            presented in this report. Until improved material models and calculation techniques are
            available, this problem cannot be corrected. However, techniques have been developed
            to counteract the tensile stresses, and the method used in this work, described in Section
            8.3.2.2, is the method recommended in the South African Mechanistic-Empirical Design
            Method.

            Another limitation is the calculation of the relative density and saturation. It is difficult to
            obtain accurate measurements of the in situ density and moisture content, and poor
            estimations could lead to large errors. It is recommended that this is investigated further.

            A third limitation is that no general model is available for predicting the maximum
            allowable principal stress for all emulsified bitumen treated materials. The models fitted
            to the individual materials are difficult to expand to other materials, and some of the
            models presented in this report do not account for changes in the relative density and
            saturation because of a lack of sufficient data. The development of a single, universal
            model is an area for further research.

            A final limitation in the use of only one filler type, cement. The behaviour of emulsified
            bitumen treated materials with other fillers should be investigated.

10.4.       Material Classification, Design Catalogues and Charts

            Using available laboratory data, a material classification system was developed. Four
            material classes are recommended, EB1 to EB4. The system uses both the UCS to
            measure the shear strength and the strain-at-break to measure the flexibility of the
            material. The flexibility component of the material classification system is essential to
Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                105
Report number CR-2003/44
            prevent cement being added to increase the UCS, at the expense of the flexibility of the
            material. A classification using UCS and ITS values did not provide a reasonable ranking
            of mixes and it is therefore necessary that the strain-at-break test is adopted as a routine
            test for emulsified bitumen treated materials. It is recommended that the TG2
            classification system be revised to UCS and strain-at-break criteria.

            The stiffness reduction and permanent deformation transfer functions were used to
            develop design catalogues and charts for three of the four material classes, EB2 to EB4.
            Insufficient data are available for the EB1 material class. The catalogues are for new
            construction and the design charts for deep in situ recycling. It is strongly recommended
            that an EB1 material is tested and the stiffness reduction and permanent deformation
            transfer functions modified for this material class.

            The design charts show that the structural behaviour of emulsified bitumen treated
            materials is very similar to foamed bitumen treated materials. When differences occur,
            the emulsified bitumen treated layers are thinner than those required for foamed bitumen.
            The differences are, however, minimal. Reasons for these differences are discussed in
            Section 9.6. The catalogues ignore any effect of differences in the ease construction of
            the materials, specifically the ease of compaction. Also, the comparisons are not based
            on economic factors. It is recommended that field tests on existing emulsified bitumen
            treated pavements are evaluated to validate the transfer functions and design catalogues
            and design charts and confirm their transferability over a wider range of materials and
            road classes.

            The catalogues are in general agreement with those published in SABITA Manual 211,
            which were validated by experienced consultants. No pavements with adequate
            structural capacity for 95 per cent reliability (Category A roads) for the ES3 and ES10
            traffic classes could be developed, unless a higher risk (lower reliability) is accepted. It is,
            however, possible that by using a good parent material, pavements will be able to sustain
            sufficient traffic loading to fall into these classes.

            The TRH4 catalogues for lightly cemented materials also do not have pavement
            structures adequate for Category A roads and ES10 and higher traffic classes28. The
            only material with pavement structures adequate for these high class roads is crushed
            stone28. These are possible because the granular material transfer function seems to
            overestimate the bearing capacity at high factors of safety (low stress ratios). It is
            recommended that a new transfer function be developed for granular materials.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials                106
Report number CR-2003/44
11.         FUTURE WORK


            The work described in this report, and the conclusions drawn and recommendations
            made highlight several areas for future work.

            The strain-at-break test is recommended for measuring the flexibility of the materials, and
            is used in the stiffness reduction transfer function and in the material classification
            system. The implementation of this test in standard practice must therefore be
            investigated.

            The UCS test is also used in the material classification, as a measure of the shear
            strength of a mix. In the permanent deformation transfer function, it is necessary to
            measure the shear strength of the mix, which is currently performed with the dynamic
            triaxial test. The test equipment required is costly and intricate and therefore difficult to
            implement in standard practise. It is recommended that the Texas triaxial test be
            investigated for routine applications, and to possibly replace the UCS test in the material
            classification system.

            Another area for future work is the estimation and measurement of the density and
            saturation levels. Both these properties have a large influence on the material behaviour,
            but they are currently difficult to accurately measure in the field and to estimate for design
            purposes.

            Several deficiencies in the analyses procedures used in the South African Mechanistic-
            empirical Design Method, particularly in the use of multi-layer linear elastic (MLLE)
            theory, have been highlighted:
            •       MLLE calculates tensile stresses for granular type materials. Emulsified bitumen
                    treated materials typically have low tensile strength. The method of modifying the
                    tensile stress to a more realistic value must be further investigated.
            •       The permanent deformation in a layer is averaged across the layer, and the
                    stress ratio evaluation locations used to calculate the stress ratio are relative to
                    the whole layer. However, permanent deformation typically occurs in the upper
                    portions of the layer. Therefore analyses techniques that do not average the
                    permanent deformation should be investigated.
            •       The material properties of a layer are obtained from the initial state of the
                    material, except for the steady state stiffness. This does not account for
                    changing material properties during a phase. In addition, two or more phases are
                    analysed discreetly, also not accounting for changing in the layers in a
                    continuous manner. It is recommended that a recursive method of analyses,
                    where the material properties can be modified with each load repetition, is
                    investigated.

            A good quality material, such as crushed stone, and a cohesionless material such as
            sand should be tested to assess the behaviour of the material when treated with
            emulsified bitumen. This has already been initiated with an extensive laboratory testing
            project using a G2 crushed stone hornfels from the Western Cape and a sand/calcrete
            blend from Kwazulu-Natal. In additional, fly ash is being investigated as an alternative
            filler. Valuable information will be gained from this project.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials               107
Report number CR-2003/44
            In the analyses in this report, the development of a universal model for all material types
            was investigated. The results indicated that such a model is feasible, however insufficient
            data are currently available. The new laboratory testing project will provide additional
            data with which the development of such a model can attempted. The catalogues and
            design charts for foamed bitumen and emulsified bitumen are similar, indicating that it
            may be possible to develop one universal model applicable to bitumen stabilised
            materials. It is recommended that this be investigated on completion of the new
            laboratory testing. It is also recommended that the publication of a guideline document
            with the structural design models and associated catalogues and design charts is delayed
            until the models are updated with the additional laboratory testing data.

            The reliability of structural design models developed from HVS and laboratory data can
            be significantly enhanced by validating the models with long-term pavement performance
            under real traffic. It is therefore recommended that an LTPP program to monitor
            pavements with emulsified bitumen treatment is implemented, specifically related to HVS
            test sections or the sections where extensive construction data are available, such as on
            the N12 in Gauteng.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            108
Report number CR-2003/44
12.         REFERENCES


1.    SABITA, ETB: The Design and Use of Emulsion-treated Bases, Manual 21, 1999.
2.    SABITA, GEMS - The Design and Use of Granular Emulsion Mixes, Manual 14, 1993.
3.    Technical Recommendations for Highways, Structural Design of Flexible Pavements for
      Interurban and Rural Roads, TRH4, Draft, 1996.
4.    Asphalt Academy, Interim Technical Guideline: The Design and Use of Foamed Bitumen
      Treated Materials, TG2, September 2002.
5.    De Beer, M. and Grobler, J., ETBs: Heavy Vehicle Simulator (HVS) Evaluation of the Heilbron
      Sections, PR 92/2/1, August 1993.
6.    du Plessis, J.L., ETBs: The Design and Construction of the Heilbron Test Sections, First Draft,
      ETB/8 or I/AT/11/92, February 1992.
7.    Long, F.M., The Development of Structural Design Models for Foamed Bitumen Treated
      Pavement Layers, Transportek, CSIR, Contract Report CR-2001/76, 2002.
8.    Theyse, H.L., Overview of the South African Mechanistic Pavement Design Method, South
      African Transport Conference, July 2000.
9.    Santucci, L.E., Thickness design procedure for asphalt and emulsified asphalt mixes, Fourth
      International Conference on the Structural Design of Asphalt Pavements, Ann Arbor, 1977.
10.   The Asphalt Institute, Research and Development of The Asphalt Institute’s Thickness Design
      Manual (MS-1) Ninth Edition, Research Report No. 82-2, 1982.
11.   De Beer, M. and Grobler, J., “Towards Improved Structural Design Criteria for Granular
      Emulsion Mixes (GEMS)”, 6th Conference on Asphalt Pavements for Southern Africa, 1994.
12.   Liebenberg, J.J.E., A Structural Design Procedure for Emulsion Treated Pavement Layers,
      Master’s Thesis, University of Pretoria, 2002.
13.   Theyse, H.L., Laboratory design models for materials suited to labour-intensive construction,
      CR-99/038, June 2000.
14.   Steyn, W.J.vdM., Summary of field and laboratory test results on selected Emulsion Treated
      Base (ETB) pavements, Transportek, CSIR, Contract Report CR-98/073, 1997.
15.   Theyse, H.L., Towards Structural Guidelines for the Structural Design of Pavements with
      Emulsion-treated Layers, CR-97/045, CSIR Transportek, 1998.
16.   Grobler, J.E., du Plessis, J.L., ETBs: Interim Recommendations on Correlations between Field
      and Laboratory Properties, Draft, ETB/10 or I/AT/5/93, January 1993.
17.   Mancotywa, W.S., First Level Analysis Report: Phase 1 HVS Testing of Experimental Sections
      on Road R2388 near Cullinan, CR-99/011, November 2001.
18.   Mancotywa, W.S., First Level Analysis Report: 2nd Phase HVS Testing of the Emulsion Treated
      Natural Gravel Base Sections on Road P2388 near Cullinan, CR-2000/47, November 2000.
      DRAFT.
19.   Theyse, H.L., and W.S. Mancotywa, First Level Analysis Report: 2nd phase HVS Testing of the
      Emulsion Treated Gravel and Foam Treated Gravel Base Sections on Road P243/1 near
      Vereeniging, Transportek, CSIR, Contract Report CR-2001/53, 2001.
20.   Steyn, W.J.vdM., Level one data analysis of HVS tests on Foam Treated Gravel and Emulsion
      Treated Gravel on Road P243-1: 80 kN and 100 kN test sections, Transportek, CSIR, Contract
      Report CR-2001/5, 2001.
21.   Long, F.M. and H.L.Theyse, Laboratory Testing for the HVS Sections on Road P243/1,
      Transportek, CSIR, Contract Report CR-2001/22, 2001.
22.   Liebenberg, J.J.E., The Influence of various emulsion and cement contents on an emulsion
      treated ferricrete from the HVS Test Sections on Road P243/1, Transportek, CSIR, Contract
      Report, CR-2001/77, 2002.

Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials          109
Report number CR-2003/44
23.   PIARC, Technical committee 7/8 “Road Pavements”, Cold in-place recycling of pavements with
      emulsion or foamed bitumen, Draft report, Version 2.3A, May 2002.
24.   Long, F.M. and H.L. Theyse, Second level analysis of the HVS Data from Road P243/1, CSIR,
      Contract Report, CR-2002/23.
25.   Long, F.M., H.L. Theyse, S. Robroch, J.J.E. Liebenberg, Performance models for deep in situ
      recycled, bitumen stabilised pavements under accelerated traffic, Ninth International
      Conference on Asphalt Pavements, Denmark, August 2002.
26.   Robroch, S., Laboratory Testing on Foamed Bitumen and Cement Treated Materials from the
      HVS Test Sections on Road P243/1, Transportek, CSIR, Contract Report, CR 2001/69, 2001.
27.   Theyse, H.L., Ph.D research. Unpublished.
28.   Theyse, H.L., TRH4 Revision (1995) Phase II: Mechanistic Design Analysis of the Pavement
      Structures Contained in the TRH4 (1995): Pavement Design Catalogue, Transportek, CSIR,
      Contract Report I/PA/14/95, 1995.
29.   Wright, B.G., S.C. Lacante, A.J. Laatz and G.D. du Toit, Long term performance of emulsion
      treated base pavements, CSIR, RDAC Report 88/014, March 1991.
30.   Steyn, W.J.vdM., Summary of field and laboratory test results on selected Emulsion Treated
      Base (ETB) pavements, CSIR, Internal Technical Report TR-97/049, 1997.
31.   Hughes, M.R., H.C. Thompson, J.A. du Plessis, Managing and designing pavements into the
      next decade, Using information from previous decades: N12 a case in point, SATC, 1990.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials      110
Report number CR-2003/44
APPENDIX A: WORK PROPOSAL




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials   111
Report number CR-2003/44
                SABITA/GAUTRANS/CSIR RESEARCH PROPOSAL:
                         PROJECT DETAIL SUMMARY

                                        in terms of:
    SABITA:          Memorandum of Agreement of 5 September 1991
    Gautrans:        Agreement for Specialist Consulting Engineering and Contracting Services for
                     HVS Testing (D01/07/TC)

1   Project title:
    Mechanistic-Empirical Structural Design and the Development of Guidelines for the
    Design and Use of Bituminous Emulsion Treated Materials



2   Background and Motivation

    Bituminous emulsion treated materials have been used in South Africa for many years and
    SABITA guidelines for their use are available. However, no mechanistic-empirical structural
    design models for these materials have been developed and the mix design procedure needs
    to be refined with the latest research results. In addition, the mix design needs to be linked to
    the structural design. Previous attempts (1996) to introduce emulsion treated materials into
    TRH4 failed because the mechanistic-empirical design method used was far too conservative.

    The structural design method used in the SABITA guidelines is an empirical method, which
    does not allow for the development of a universal design model applicable to both foamed
    bitumen and emulsion treatment. Structural design models, and mix and structural design
    procedures should be based on the fundamental engineering behaviour of the material. The
    mix design should also be aimed at optimising the desired material behaviour and be related
    to the structural design. By developing such structural design models and refining the mix
    design procedures, the confidence in pavement designs using bituminous emulsions should
    be significantly improved. An increase in confidence should reduce the risks associated with
    using emulsion treated materials, that is, the risk of premature failure or overdesign.

    Recent work on foamed bitumen treated materials showed the potential for such a unified
    approach to mix and structural design. Mechanistic-empirical structural design models and a
    mix design procedure were compiled in an interim guideline document for the design and use
    of foamed bitumen treated materials. These guidelines provide the tools from which
    pavements with foamed bitumen can be designed with improved confidence. A similar
    approach needs to be applied to emulsion treated materials. The preparation of a similar
    document for emulsion treated materials is of benefit to the industry in that consultants,
    contractors, suppliers and road owners will have access to design tools that will improve the
    confidence in designs, and thereby reduce the risks. Such a guideline will also allow the
    comparison of foamed bitumen and bituminous emulsion treated materials in a consistent
    manner and the appropriate material to be selected based on valid economic appraisals.


    This proposal is for the development of mechanistic-empirical structural design models and
    refinement of the mix design procedure for bituminous emulsion treated materials based on
    available research results and with additional testing in cases where the existing data are


                                              -1-
    insufficient. The structural design will be linked to the mix design procedure. The structural
    design models will be based on fundamental material and pavement behaviour. This should
    facilitate the later combination of the design methods for foamed bitumen and emulsion
    treatment into a single universal model.

    The mix and structural design procedures will be incorporated into a guideline document on
    the design and use of emulsion treated materials. The document will be a companion
    document to that prepared for foamed bitumen treated materials. By preparing the guidelines
    for emulsion treated materials, the existing experience can be disseminated, and used in the
    industry, facilitating validation of the mix design procedure and structural design models. This
    should promote information and experience sharing. These experiences will contribute to the
    preparation of final guidelines with validated structural design models. Once sufficient
    databases have been developed for both foamed bitumen and bituminous emulsion treated
    materials, the design methods will be incorporated into the South African Mechanistic-
    Empirical Design Method and TRH4.

    The work covered in this proposal supports the objectives of a framework to improve cold-
    treatment technology. The informal partners in this framework are Gautrans, SANRAL and
    SABITA. CSIR, Transportek is in ongoing discussion with these parties and will ensure that all
    efforts are supplementary.

3   Funding

    To date the majority of research providing the data to be used in this project has been
    sponsored by the Gauteng Provincial Government, Department of Transport and Public
    Works (Gautrans). These data are from HVS test sections near Cullinan and Vereeniging and
    associated laboratory testing. Additional data from HVS tests near Heilbron, funded by the
    National Department of Transport and SABITA will also be investigated.

    The funding requested in this proposal covers the development of a structural design method,
    refinement of the mix design method and the development of a guideline document.

4   Project Goals

    The use of the cold treatment processes with bituminous emulsion and foamed bitumen,
    specifically with the deep in situ recycling technology for the rehabilitation of roads is being
    increasingly used in South Africa. The interim guideline document for the design and use of
    foamed bitumen treated materials includes mechanistic-empirical structural design models for
    foamed bitumen treated materials and a comprehensive mix design procedure and material
    classification system based on both ITS and UCS tests. While guidelines for emulsion treated
    materials are available, “ETB: The design and use of emulsion-treated bases” (Manual 21)
    and “GEMS: The design and use of granular emulsion mixes” (Manual 14), these guidelines
    were not developed using mechanistic-empirical structural design models and do not
    incorporate the latest research results. The manuals therefore need to be updated to reflect
    the latest research results.




                                             -2-
    The purpose of this work is to develop mechanistic-empirical structural design models for
    bituminous emulsion treated materials and to incorporate them into a guideline document.
    Also included in the guideline document will be an updated mix design procedure and a new
    material classification system based on the parameters that need to be optimised for the best
    structural performance. This guideline document will increase the confidence in the mix and
    structural design of pavements with emulsion treated materials, and facilitate the comparison
    of emulsion and foamed bitumen treated materials in a consistent manner, enabling valid
    economic appraisals of the materials.


5   Methodology

    The project will be completed in three phases. The first phase will investigate the available
    data and determine which data can be used to perform the necessary analyses. Data are
    available from HVS testing on sections constructed with emulsion treated materials at
    Vereeniging (P243/1), Cullinan (R2388) and Heilbron (P9/3), tested in 2000/2001, 1997/1998
    and 1992/1993, respectively. Some laboratory data are available for the materials used in
    these test sections. The data include aspects such as:
        • The engineering properties (such as the bearing strength, permeability and erodibility)
            of the material;
        • The mechanical properties (such as the stiffness, shear strength and strain at break)
            of the material;
        • The material and pavement behaviour and performance of the material under
            repeated loading, and
        • All aspects that impact on the above such as design, construction and maintenance.
    Additional data that can be used in the analyses will also be sought. During this phase, a
    detailed proposal for the analyses in Phase 2 will be determined.

    The second phase involves the preparation of mechanistic-empirical structural design models
    for bituminous emulsion treated materials using the applicable HVS and laboratory data
    (determined in Phase 1). The structural design models will include a design methodology for
    the different types of structural distress experienced by bituminous emulsion treated materials.
    Data from HVS test sections and some laboratory data are available for the materials used in
    these test sections, and at wider ranges of density and saturation and at different
    combinations of active filler and emulsion contents. The triaxial test data will be used, in
    conjunction with the HVS test data, to develop the permanent deformation models.

    The permanent deformation transfer function for the foamed bitumen treated material is
    sensitive to the foamed bitumen and active filler contents. This was possible because triaxial
    data were available for one material at two different combinations of additive contents. To
    achieve a model with the same degree of sensitivity for the emulsion treated materials,
    additional triaxial testing needs to be performed. The additional triaxial testing will provide
    data at another combination of active filler and emulsion, and tests will be run at two density
    and two saturation levels, resulting in four combinations. Four static and six dynamic triaxial
    tests are run at each combination of density and saturation level. This is the same test plan
    as used in previous triaxial testing on the emulsion and foamed bitumen treated materials. By
    performing the same laboratory tests, at the same bitumen and active filler contents, as was

                                             -3-
done for the foamed bitumen treated materials, a comparison of the performance of the two
materials can be made. This will provide insight into the differences and similarities in the two
materials and to establish whether a single design procedure can be used for both materials.

The third phase involves preparation of an interim guideline document for the design and use
of bituminous emulsion treated materials, such as that recently produced for foamed bitumen
treated materials. This guideline shall include sections on selection criteria for suitable
projects, mix design and structural design. Construction aspects shall be included, however
this section will be taken from the existing SABITA manuals with minor modification. Updates
may occur as part of the review committee feedback. The compilation of the guideline will
draw on the existing SABITA ETB and GEMS manuals. The guideline will supercede the
existing manuals. The guideline will move away from the distinction between ETBs and
GEMS, and the associated distinction between modification and stabilization.

The work for all three phases will proceed in several tasks, listed below. Also shown are the
subtasks, or a description of what the task involves.

Phase 1: Review existing information

Collect and review the available information from the literature, laboratory and HVS testing.
This involves collection and review of the available literature and the available HVS and
laboratory testing results. This work will be summarized in a brief, but comprehensive report
listing all the works consulted even those that will not be used in the final development of the
guideline documents (in which case reasons will be provided for not using the information). In
order to prevent delays and reduce cost, this report will not be reviewed externally from the
project team but the important findings will be included in the final project report at which time
it will be thoroughly reviewed.

The scope and approach of the project will be assessed during Phase 1 and may result in
adjustments to these at the end of Phase 1. Subsequent phases will only continue with
approval from the client. A detailed proposal for the laboratory testing and analyses to be
performed in Phase 2 will be developed as an output from Phase 1.

Phase 2: Develop structural design models

Phase 2, Task 1: Perform additional laboratory testing.
Additional static and dynamic triaxial tests will be performed at various emulsion and active
filler contents. The static strength parameters and the permanent deformation behaviour will
be obtained from these tests. The results will be presented in the final report.

Phase 2, Task 2: Combine HVS data and laboratory data to develop an interim rational
mechanistic-empirical design method for bituminous emulsion treated materials.
This involves investigating and evaluating the basic behaviour of the material, the failure
modes and developing transfer functions. This design method will be regarded as an interim
design method as it will be based on a limited database. The specific subtasks are:
• Develop effective fatigue model from HVS test sections
• Develop effective fatigue model from laboratory test data

                                          -4-
•   Combine effective fatigue models from HVS and laboratory tests
•   Develop permanent deformation model from HVS test sections
•   Develop permanent deformation model from laboratory test data
•   Combine permanent deformation models from HVS and laboratory tests
•   Identify limitations

The analyses performed in Task 3 and the final design models will be documented and
discussed in the final report, which will also briefly describe the data from which the models
are developed.

Phase 2,Task 3: Develop interim design catalogues and charts
A collection of pavements will be designed for the various subgrade support conditions,
material classifications, and for different types of surfacings. This task also involves
developing or modifying software algorithms to perform the calculations. The catalogue and
design charts will be prepared in graphical format, as in TRH4 and similar documents.

Phase 2,Task 4: Prepare final report
A final report describing the results and analyses will be compiled. The report will include a
summary of the data used in the analyses, the transfer functions and the development thereof,
the recommended design procedure and catalogues and design charts will be documented.



Phase 3: Write guideline document

Phase 3, Task 1: Review existing guidelines
Review the available information and outline guideline document.

Phase 3, Task 2: Write Introductory Sections
Write the introductory sections for the guideline document, including guidelines for the
selection criteria for suitable projects.

Phase 3, Task 3: Mix Design Section
Prepare the mix design procedure, including new information.           Develop the material
classification system.

Phase 3, Task 4: Structural Design Section
Prepare the structural design chapter, which includes the catalogues and design charts. An
appendix detailing the information necessary to perform a mechanistic-empirical structural
design will also be prepared.

Phase 3, Task 5: Complete Guideline Document and Internal Review
Complete the guideline document, including all relevant Appendices. As part of this task the
document will be reviewed internally by Transportek, Gautrans, Sabita and the Asphalt
Academy.




                                        -5-
    Phase 3, Task 6: External Reviews and Finalization of Guideline Document
    This task involves external reviews by parties nominated by SABITA, Gautrans, the Asphalt
    Academy and Transportek. Provision is made for updating the document and incorporating
    the comments suggested by the review panel.


6   Project Deliverables

    The project will have 6 deliverables, discussed below:
    Deliverable 1: Brief, comprehensive report discussing the available data investigated in
                      Phase 1. The proposal for the Phase 2 testing and analyses will also be
                      refined.
    Deliverable 2: Presentation on laboratory testing results.
    Deliverable 3: Presentation on the development of the structural design models.
    Deliverable 4: Report presenting and discussing the results from Phase 2.
    Deliverable 5: Draft guideline document
    Deliverable 6: Final guideline document, incorporating comments from external review
                      panel. The document will include:
                      1. Guidelines for the selection criteria for the use of bituminous emulsion
                           treated materials.
                      2. A material classification for bituminous emulsion treated material and
                           guidelines on mix design.
                      3. Structural design guidelines:
                           • Engineering principles;
                           • Recommendations on load equivalency for traffic conversions, and
                           • Pavement design catalogues.
                      4. Construction guidelines, taken from the existing SABITA manuals, with
                            possible updates from the review process.




                                            -6-
7           Programme

                                                                                                                                             Start date:
PROJECT TITLE: Mechanistic-Empirical Structural Design and the Development
                                                                                                                                               August
of Guidelines for Bituminous Emulsion Treated Materials
                                                                                                                                                2002




                                                                                                                                  Month 10



                                                                                                                                                        Month 12

                                                                                                                                                                   Month 13
                                                                                                                                             Month 11
                                        Month 1

                                                  Month 2

                                                            Month 3

                                                                      Month 4

                                                                                Month 5

                                                                                          Month 6

                                                                                                    Month 7

                                                                                                              Month 8

                                                                                                                        Month 9
Phase




          Task description


          Review available          P
Phase 1




          information               A
                                    P             X
          Deliverable 1
                                    A
                                    P
          1. Laboratory testing
                                    A
                                    P                                 X
          Deliverable 2
                                    A
          2. Develop structural     P
          design models             A
Phase 2




                                    P                                           X
          Deliverable 3
                                    A
          3. Develop design         P
          catalogue                 A
          4. Document design        P
          procedure and catalogue   A
                                    P                                                               X
          Deliverable 4
                                    A
          1. Review existing        P
          guidelines                A
          2. Write introductory     P
          sections                  A
                                    P
          3. Mix design section
                                    A
          4. Structural design      P
Phase 3




          section                   A
          5. Complete draft and     P
          internal review           A
                                    P                                                                                                                   X
          Deliverable 5
                                    A
          6. External review and    P
          finalize document         A
                                    P                                                                                                                              X
          Deliverable 6
                                    A




                                                            -7-
8    Implementation

     The dissemination of the information provided in the Interim Guideline document to industry
     will be done by the Asphalt Academy through seminars and workshops. The cost of
     implementation is not included in this proposal.

9    Project Team

     The project will be completed by Dr Fenella Long, Mr Hechter Theyse and Mr Colin Fisher,
     with support from Transportek, CSIR personnel. Mr Adrian Bergh and other practicing
     engineers will be involved in the peer review process and will be consulted throughout the
     project. The project will be managed by the Asphalt Academy, on behalf of SABITA and
     Gautrans.

10   Quality Control

     The Transportek laboratory is currently preparing for ISO 9001 certification for the standard
     laboratory tests in early 2003. Protocols for the advanced laboratory testing are being
     prepared for Gautrans and will be available in August 2002. All laboratory work will be closely
     supervised by the project engineers.


11   Estimated Cost

     The cost of Phase 1 is shown in Table 1, and the estimated cost of Phases 2 and 3 are shown
     in Table 2 to 4. The estimated total project cost is shown in Table 5. The final costs of, and
     approval for, each Phase will be determined at the end of the preceeding phase. These costs
     do not include the management of the project by the Asphalt Academy.

     Table 1.   Estimated costs for Phase 1
      Task                                                    Person                     Cost
                                                            Fenella Long
      Task 1 Review available information                 Hechter Theyse                  R 51 360
                                                            Colin Fisher
                                   Total Cost of Phase (excluding 14% VAT)                R 51 360


     Table 2.   Estimated costs for Phase 2
                             Task                                  Person                Cost
                                                                Fenella Long
                                                               Hechter Theyse             R 23 280
      Task 1 Laboratory testing                                 Colin Fisher
                                                             Laboratory testing*          R 62 200
                                                            Fenella Long
      Task 2 Develop structural design models                                             R 60 960
                                                          Hechter Theyse
                                                            Fenella Long
      Task 3 Develop design catalogue and charts                                          R 23 520
                                                          Hechter Theyse
      Task 4 Final report                                   Fenella Long                  R 49 800
                                   Total Cost of Phase (excluding 14% VAT)               R 219 760
     * see Table 3




                                              -8-
Table 3.     Estimated laboratory testing costs for Phase 2, Task 1
 Item                                                 Number Cost per                Cost
                                                      of tests     Test
 Specimen preparation                                                                  R 3 000
 Static triaxial tests                                   16        1000               R 16 000
 Dynamic triaxial tests                                  24        1800               R 43 200
                              Total Cost of Testing (excluding 14% VAT)               R 62 200


Table 4. Estimated Costs for Phase 3
 Task                                                        Person                  Cost
 Task 1 Review existing guidelines                         Fenella Long               R 21 120
                                                           Fenella Long
 Task 2    Write introductory sections                                                R 14 400
                                                          Hechter Theyse
                                                           Fenella Long
 Task 3    Mix design section                                                         R 25 440
                                                          Hechter Theyse
                                                           Fenella Long
 Task 4    Structural design section                                                  R 14 400
                                                          Hechter Theyse
                                                           Fenella Long
 Task 5    Complete draft and internal review                                         R 43 300
                                                          Hechter Theyse
                                                           Fenella Long
 Task 6    External review and finalize document                                      R 45 700
                                                          Hechter Theyse
                                  Total Cost of Phase (excluding 14% VAT)            R 164 360

Table 5.      Total Project Cost
 Phase                                                                               Cost
 Phase 1:     Review available information                                            R 51 360
 Phase 2:     Develop structural design models                                       R 219 760
 Phase 3:     Write guideline document                                               R 164 360
                                            Total Cost (excluding 14% VAT)           R 435 480
                                                                   14% VAT            R 61 967
                                                   TOTAL PROJECT COST                R 496 447

Pro-forma payments will be made, as per Table 6:

Table 6. Pro-forma Payments
Task/Deliverable                                  Payment Amount            Organisation
Deliverable 1, end of Phase 1                              R 51 360             SABITA
Deliverable 2, laboratory testing                         R 84 200*             SABITA
Deliverable 4, end of Phase 2                            R 135 560*            Gautrans
Deliverable 5, Draft guideline document
                                                   To be decided on approval of Phase 3
Deliverable 6, Final guideline document
* Estimated, to be finalised and approved at the end of Phase 1, on receipt of the detailed plan
for Phase 2

Submitted by Fenella Long on behalf of CSIR.




Signature :                                       Date:




                                         -9-
12   Approval of project:

     Phase 1 of the project entitled "Mechanistic-Empirical Structural Design and the Development
     of Guidelines for Bituminous Emulsion Treated Materials" is approved in terms of the
     Sabita/CSIR Research Agreement ~ Memorandum of Agreement of 5 September 1991 and
     the Gautrans/CSIR agreement for specialist consulting engineering and contracting services
     for HVS testing (D01/07/TC).

     Signed at                      on this             day of                                 2002.

     As witness:                                        For and on behalf of the CSIR



     1.




     2.                                                 ........................................................................



     Signed at                      on this             day of                                 2002.

     As witness:                                        For and on behalf of Sabita



     1.




     2.                                                 ........................................................................



     As witness:                                        For and on behalf of Gautrans



     1.




     2.                                                 ........................................................................




13   Commencement Date :                      ...............1 August 2002...........




                                              - 10 -
APPENDIX B: LABORATORY TEST DATA




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials   112
Report number CR-2003/44
                       Test Matrix and Specimen Preparation Data Sheet




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials   113
Report number CR-2003/44
Summary Triaxial test data for the       Emulsion treated ferricrete                                                          This is adjusted for what
                                         1% cement, two different emulsion contents                                           was actually obtained in the lab.

                    MDD:       2025           Cement %             1         0.010 G agg                   3.15                                                                                     x = agg *
                    OMC:         9.1          Binder               3         0.030 G bit                   1.01            ARD incl 3.0% bit         2.6452                                          1.04030
                ARD cem:      2.777           Binder             0.9         0.009                                         ARD incl 0.9% bit         2.7375                                          1.01909
 Sample #   Dry density Compaction Moisture Saturation Residual         Target         Actual                                              Static test results                                                    Cohesion                             Actual        Actual    Compaction Saturation
            of aggregate             Content               Binder     Confining      Confining    Failure load Failure stress   Stiffness       Sigma 1                        p                    q             Friction                              MC         Dry density
             (kg/cub m)    (%)         (%)    (%)        Content (%) Stress (kPa) Stress (kPa)         kN           (kPa)         (MPa)           (kPa)                      (kPa)                (kPa)          angle & R2                             (%)        (kg/cub m)     (%)        (%)
FSS01               2055     75.1%         10      82.4%        3.00            20             22          13.6            749           72              771                          397                  375                                              7.01        2063.7      78.0%      65.8%
FSS02                                                                           80             84          17.5            965           93             1049                          567                  483     110.9                                    6.20        2094.9      79.2%      62.5%
FSS03                                                                          140            143          25.8           1423          128             1566                          855                  712      47.7                                    6.89        2082.8      78.7%      67.4%
FSS04                                                                          200            199          30.8           1697          175             1896                         1048                  849     0.999                                    6.71        2080.9      78.7%      65.5%
                                                                                                                                                                                                                                                                  Average           78.7%      65.3%
                                                                                                                                                                                                                                                                  COV                 0.6%      3.1%
FSS05                2055            1           6     49.50%         3.00               20              20            35.9              1978            231      1998               1009                  989                                              3.93        2032.6      76.8%      34.4%
FSS06                                                                                    80              41            38.2              2103            254      2144               1093                 1052     463.6                                    3.76        2025.9      76.6%      32.5%
FSS07                                                                                   140             139            41.4              2282            217      2421               1280                 1141      38.4                                    4.25        2042.6      77.2%      38.1%
FSS08                                                                                   200             199            47.3              2607            228      2806               1503                 1304     0.992                                    3.54        2047.3      77.4%      32.1%
                                                                                                                                                                                                                                                                  Average           77.0%      34.3%
                                                                                                                                                                                                                                                                  COV                 0.5%      8.0%
FSS09                2052        75.0%          10      82.0%         0.90               20              20            16.1               885            139       905                463                  443                                              7.90        2064.5      75.4%      66.3%
FSS10                                                                                    80              85            21.6              1189            165      1274                680                  595     164.8                                    9.00        2037.5      74.4%      71.6%
FSS11                                                                                   140             138            26.2              1444            192      1582                860                  722      44.7                                    8.36        2051.3      74.9%      68.3%
FSS12                                                                                   200             203            28.4              1565            195      1768                986                  783     1.000                                    7.91        2078.2      75.9%      68.2%
                                                                                                                                                                                                                                                                  Average           75.2%      68.6%
                                                                                                                                                                                                                                                                  COV                 0.9%      3.2%
FSS13                2052        75.0%           6      49.2%         0.90               20              21            32.5              1793            247      1814                918                  897                                              4.52        2019.5      73.8%      34.8%
FSS14                                                                                    80              85            47.1              2595            389      2680               1383                 1298     328.7                                    4.21        2040.1      74.5%      33.7%
FSS15                                                                                   140             144            48.6              2680            278      2824               1484                 1340      47.0                                    4.33        2013.9      73.6%      33.0%
FSS16                                                                                   200             205            42.9              2364            278      2569               1387                 1182      0.92                                    5.05        2003.2      73.2%      37.7%
FSS38                                                                                   200             201            45.0              2479            240      2680               1441                 1240     0.000                                    4.70        2064.0      75.4%      39.4%
                                                                                                                                                                                                                                                                  Average           74.1%      35.7%
                                                                                                                                                                                                                                                                  COV                 1.2%      7.6%
 Sample #    Dry density Compaction Moisture Saturation Residual        Target dynamic test conditions                                      Actual dynamic test conditions                           Dynamic test results                              Actual        Actual    Compaction Saturation
                                    Content              Binder     Confining    Test stress       Stress                       Confining     Test stress Stress Ratio     Stress ratio Plastic deform    PD rate       Plastic strain    PS rate       MC         Dry density
             (kg/cub m)     (%)       (%)       (%)    Content (%) stress (kPa)     (kPa)          Ratio                       stress (kPa)      (kPa)        C, phi     Sigma 1 model       (mm)       (mm/million)         (%)        (%/million)     (%)        (kg/cub m)     (%)        (%)
FSS17               2055      77.7%       10     92.1%        3.00            80         205.4           0.20                            79.3        209.2          0.20       0.15                0.53           1.21             0.18         0.41          5.6       2063.4      78.0%      52.8%
FSS18                                                                         80         564.8           0.55                            82.0        587.9          0.57       0.44                0.87           2.12             0.29         0.71          5.9       2051.8      77.6%      53.9%
FSS19                                                                         80         924.3           0.90                           137.8       1290.7          0.95       0.79                1.21           3.86             0.41         1.29          5.5       2060.1      77.9%      51.5%
FSS20                                                                        140         273.5           0.20                           145.2        282.7          0.20       0.18                0.52           0.96             0.17         0.32          6.1       2050.1      77.5%      56.0%
FSS21                                                                        140         752.0           0.55                           142.0        785.6          0.57       0.50                1.45           4.10             0.48         1.37          6.3       2032.8      76.8%      55.3%
FSS22                                                                        140       1230.6            0.90                           141.6        957.4          0.70       0.59                2.28           8.89             0.76         2.98          5.8       2053.3      77.6%      53.1%
                                                                                                                                                                                                                                                                  Average           77.6%      53.8%
                                                                                                                                                                                                                                                                  COV                 0.5%      3.1%
FSS23                2055        77.7%           6      55.3%         3.00               80           435.6            0.20               82.0         450.8       0.21      0.16                         0.59           2.68         0.20     0.90           2.8       2053.0      77.6%      25.5%
FSS24                                                                                    80          1197.8            0.55               83.0        1242.5       0.57      0.53                         4.26           9.97         1.43     3.36           3.4       2051.8      77.6%      31.3%
FSS25                                                                                    80          1960.1            0.90               79.7        1969.6       0.90      0.66                         6.08        1614.49         2.04   541.00           2.7       2056.6      77.7%      24.8%
FSS26                                                                                   140           474.8            0.20              143.6         489.6       0.21      0.16                         0.56           1.86         0.19     0.62           2.8       2061.9      77.9%      26.1%
FSS27                                                                                   140          1305.8            0.55              142.0        1356.6       0.57      0.44                         2.30           3.87         0.77     1.30           2.8       2061.9      77.9%      25.9%
FSS28                                                                                   140          2136.7            0.90              143.0        2120.6       0.89      0.70                         6.78         317.03         2.28   106.52           2.8       2068.5      78.2%      26.5%
                                                                                                                                                                                                                                                                  Average           77.8%      26.7%
                                                                                                                                                                                                                                                                  COV                 0.3%      8.7%
FSS29                2052        75.0%          10      82.0%         0.90               80           233.6            0.20               81.6         236.2       0.20      0.23                         0.59          -0.56         0.20    -0.19           9.4       2052.2      75.0%      77.1%
FSS30                                                                                    80           642.4            0.55               81.3         675.0       0.57      0.65                         1.68           1.73         0.56     0.58           9.6       2045.1      74.7%      77.3%
FSS31                                                                                    80          1051.1            0.90               81.7        1073.9       0.91      1.03                         6.87        2082.05         2.31   700.06           9.6       2043.3      74.6%      76.9%
FSS32                                                                                   140           290.4            0.20              139.8         302.4       0.21      0.25                         0.39           0.86         0.13     0.29           9.9       2046.5      74.8%      80.3%
FSS33                                                                                   140           798.6            0.55              141.1         841.6       0.58      0.60                         1.32           3.65         0.44     1.23           7.7       2079.5      76.0%      66.4%
FSS34                                                                                   140          1306.8            0.90              140.1        1311.2       0.90      1.01                         7.73          25.83         2.59     8.67           9.0       2046.8      74.8%      72.9%
                                                                                                                                                                                                                                                                  Average           75.0%      75.2%
                                                                                                                                                                                                                                                                  COV                 0.7%      6.5%
FSS35                2052        75.0%           6      82.0%         0.90                80             536.0         0.20              138.8        447.23                 0.20                         0.35           1.16         0.12     0.39           4.4       2078.3      75.9%      37.6%
FSS36                                                                                     80            1474.0         0.55               81.0       1224.95                 0.66                         0.89           3.11         0.30     1.03           5.1       2064.1      75.4%      42.4%
FSS37                                                                                     80            2412.0         0.90               84.9       1928.21                 1.03                         1.93         805.97         0.65   270.90           4.9       2076.4      75.9%      42.5%
FSS38                                                                        repeat static test as if specimen FSS16                                                                                                                                          4.7       2064.0      75.4%      39.4%
FSS39                                                                                    140            1553.2         0.55              140.6       1580.39                 0.81                       1.13              2.66        0.38      0.89          5.6       2050.0      74.9%      45.4%
FSS40                                                                                    140            2541.6         0.90              144.1       2494.33                 1.27           Specimen failed after three repetitions                           5.5       2056.7      75.1%      45.1%
                                                                                                                                                                                                                                                                  Average           75.4%      42.1%
                                                                                                                                                                                                                                                                  COV                 0.5%      7.3%




                                                                                                                                                                                                                                                           13/03/2003   Report table   Emulsion,triax,1%cem #5.xls
Sheet 1 Sample preparation data for the               Emulsion treated ferricrete           Cement Content:            1.00 %
         Sample dimensions:                           OMC:                     9.1 %
             Height:            0.305 m                                                          RD of bitumen         1.01
           Diameter:            0.152 m
                                                                                                                         784
Sample # Dry density      Moisture       Residual       Dry mass       Cement        Mass of       Volume of     Volume of     Volume of   Mass of    Wet sample    Dyring      Target mass Sample dimensions        Actual        Actual           Days of
         (agg only)    content (MC)       Binder                                     residual     compaction     additional    emulsion    emulsion     mass       out weight    before test Average    Average       MC         Dry density         curing
         (kg/cub m)    at testing (%)   Content (%)        (kg)           (g)      bitumen (g)     water (ml)    water (ml)       (ml)       (g)         (kg)         (kg)          (kg)      height   diameter       (%)        (kg/cub m)
FSS01          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.303       0.152          7.0          2064                29
FSS02          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.302       0.152          6.2          2095                29
FSS03          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.301       0.152          6.9          2083                28
FSS04          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.301       0.152          6.7          2081                28
FSS05          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.307       0.152          3.9          2033                28
FSS06          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.308       0.152          3.8          2026                28
FSS07          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.305       0.152          4.2          2043                28
FSS08          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.306       0.152          3.5          2047                28
FSS09          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.305       0.152          7.9          2064                28
FSS10          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.307       0.152          9.0          2038                28
FSS11          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.306       0.152          8.4          2051                28
FSS12          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.303       0.152          7.9          2078                28
FSS13          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.311       0.152          4.5          2020                28
FSS14          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.309       0.152          4.2          2040                28
FSS15          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.312       0.152          4.3          2014                28
FSS16          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.311       0.152          5.0          2003                28
FSS17          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.306       0.152          5.6          2063                28
FSS18          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.307       0.152          5.9          2052                28
FSS19          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.305       0.152          5.5          2060                29
FSS20          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.307       0.152          6.1          2050                29
FSS21          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.306       0.152          6.3          2033                30
FSS22          1976               10          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       12.147     0.305       0.152          5.8          2053                30
FSS23          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.304       0.152          2.8          2053                31
FSS24          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.302       0.152          3.4          2052                31
FSS25          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.305       0.152          2.7          2057                28
FSS26          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.304       0.152          2.8          2062                28
FSS27          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.304       0.152          2.8          2062                29
FSS28          1976                6          3.00          10.936       109.36        328.08           1005           786           541        547       12.379       11.705       11.705     0.303       0.152          2.8          2068                29
FSS29          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.304       0.152          9.4          2052                30
FSS30          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.304       0.152          9.6          2045                30
FSS31          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.305       0.152          9.6          2043                31
FSS32          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.304       0.152          9.9          2046                31
FSS33          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.304       0.152          7.7          2079                28
FSS34          2014               10          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       12.382     0.305       0.152          9.0          2047                28
FSS35          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.304       0.152          4.4          2078                29
FSS36          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.304       0.152          5.1          2064                30
FSS37          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.303       0.152          4.9          2076                30
FSS38          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.305       0.152          4.7          2064                31
FSS39          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.306       0.152          5.6          2050                31
FSS40          2014                6          0.90          11.146       111.46        100.31           1024           958           166        167       12.382       11.932       11.932     0.305       0.152          5.5          2057                31




                                                                                                                                                                                                            12/03/2003   Sample prep    Emulsion,triax,1%cem #5.xls
                               Static Triaxial Test Data




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials   114
Report number CR-2003/44
                                                     STATIC TRIAXIAL TEST
Material: Ferricrete                                                 Treatment      1 % cement & 3% bitumen
Sample #:      FSS01
Dry Density (kg/cub m): 1976                                         Confining pressure (kPa): 22
Moisture (%): 0.0                                                    Linear stiffness (MPa): 72
                                                                     Maximum deviator stress (kPa): 749


                    800.00

                    700.00

                    600.00
  Stress (kPa)




                    500.00

                    400.00

                    300.00

                    200.00

                    100.00

                      0.00
                              0     5000     10000         15000         20000     25000     30000    35000    40000
                                                            Strain (microstrain)


                    140

                    120
  Stiffness (MPa)




                    100

                     80

                     60

                     40

                     20

                      0
                          0       100      200       300           400       500       600      700      800     900
                                                             Bulk stress (kPa)




                                                                                                 FSS01.xls 12/03/2003
                                                  STATIC TRIAXIAL TEST
Material: Ferricrete                                         Treatment       1 % cement & 3% bitumen
Sample #:      FSS02
Dry Density (kg/cub m): 1976                                 Confining pressure (kPa): 84
Moisture (%): 0.0                                            Linear stiffness (MPa): 93
                                                             Maximum deviator stress (kPa): 965


                    1200.00


                    1000.00
  Stress (kPa)




                     800.00


                     600.00


                     400.00


                     200.00


                       0.00
                              0   10000   20000     30000     40000         50000     60000    70000    80000
                                                     Strain (microstrain)


                    160

                    140

                    120
  Stiffness (MPa)




                    100

                     80

                     60

                     40

                     20

                      0
                          0       200      400         600            800           1000       1200      1400
                                                     Bulk stress (kPa)




                                                                                           FSS02.xls 12/03/2003
                                                  STATIC TRIAXIAL TEST
Material: Ferricrete                                          Treatment     1 % cement & 3% bitumen
Sample #:      FSS03
Dry Density (kg/cub m): 1976                                  Confining pressure (kPa): 143
Moisture (%): 0.0                                             Linear stiffness (MPa): 128
                                                              Maximum deviator stress (kPa): 1423


                    1600.00

                    1400.00

                    1200.00
  Stress (kPa)




                    1000.00

                     800.00

                     600.00

                     400.00

                     200.00

                       0.00
                              0         10000   20000     30000      40000      50000      60000      70000
                                                        Strain (microstrain)


                    250


                    200
  Stiffness (MPa)




                    150


                    100


                     50


                      0
                          0       200     400   600     800   1000    1200     1400     1600   1800    2000
                                                        Bulk stress (kPa)




                                                                                        FSS03.xls 12/03/2003
                                            STATIC TRIAXIAL TEST
Material: Ferricrete                                     Treatment    1 % cement & 3% bitumen
Sample #:      FSS04
Dry Density (kg/cub m): 1976                             Confining pressure (kPa): 199
Moisture (%): 0.0                                        Linear stiffness (MPa): 175
                                                         Maximum deviator stress (kPa): 1697


                    1800.00

                    1600.00

                    1400.00
  Stress (kPa)




                    1200.00

                    1000.00

                     800.00

                     600.00

                     400.00

                     200.00

                       0.00
                              0   10000   20000      30000     40000      50000      60000     70000
                                                   Strain (microstrain)


                    300

                    250
  Stiffness (MPa)




                    200


                    150


                    100


                     50

                      0
                          0         500           1000           1500             2000          2500
                                                  Bulk stress (kPa)




                                                                                  FSS04.xls 12/03/2003
                                           STATIC TRIAXIAL TEST
Material: Ferricrete                                   Treatment   1 % cement & 3% bitumen
Sample #:      FSS05
Dry Density (kg/cub m): 1976                           Confining pressure (kPa): 20
Moisture (%): 0.0                                      Linear stiffness (MPa): 231
                                                       Maximum deviator stress (kPa): 1978


                    2500.00


                    2000.00
  Stress (kPa)




                    1500.00


                    1000.00


                     500.00


                       0.00
                              0     5000       10000           15000        20000         25000
                                               Strain (microstrain)


                    350

                    300
  Stiffness (MPa)




                    250

                    200

                    150

                    100

                     50

                      0
                          0       500         1000             1500          2000            2500
                                              Bulk stress (kPa)




                                                                             FSS05.xls 12/03/2003
                                         STATIC TRIAXIAL TEST
Material: Ferricrete                               Treatment     1 % cement & 3% bitumen
Sample #:      FSS06
Dry Density (kg/cub m): 1976                       Confining pressure (kPa): 41
Moisture (%): 0.0                                  Linear stiffness (MPa): 254
                                                   Maximum deviator stress (kPa): 2103


                    2500.00


                    2000.00
  Stress (kPa)




                    1500.00


                    1000.00


                     500.00


                       0.00
                              0   5000   10000      15000          20000     25000      30000
                                             Strain (microstrain)


                    350

                    300
  Stiffness (MPa)




                    250

                    200

                    150

                    100

                     50

                      0
                          0       500       1000            1500           2000            2500
                                             Bulk stress (kPa)




                                                                           FSS06.xls 12/03/2003
                                                 STATIC TRIAXIAL TEST
Material: Ferricrete                                        Treatment    1 % cement & 3% bitumen
Sample #:      FSS07
Dry Density (kg/cub m): 1976                                Confining pressure (kPa): 139
Moisture (%): 0.0                                           Linear stiffness (MPa): 217
                                                            Maximum deviator stress (kPa): 2282


                    2500.00


                    2000.00
  Stress (kPa)




                    1500.00


                    1000.00


                     500.00


                       0.00
                              0   5000   10000      15000    20000      25000    30000     35000   40000
                                                     Strain (microstrain)


                    400

                    350

                    300
  Stiffness (MPa)




                    250

                    200

                    150

                    100

                     50

                      0
                          0       500        1000           1500          2000           2500       3000
                                                     Bulk stress (kPa)




                                                                                    FSS07.xls 12/03/2003
                                                    STATIC TRIAXIAL TEST
Material: Ferricrete                                             Treatment    1 % cement & 3% bitumen
Sample #:      FSS08
Dry Density (kg/cub m): 1976                                     Confining pressure (kPa): 199
Moisture (%): 0.0                                                Linear stiffness (MPa): 228
                                                                 Maximum deviator stress (kPa): 2607


                    3000.00


                    2500.00
  Stress (kPa)




                    2000.00


                    1500.00


                    1000.00


                     500.00


                       0.00
                              0   5000   10000   15000   20000    25000   30000   35000   40000     45000   50000
                                                         Strain (microstrain)


                    450
                    400
                    350
  Stiffness (MPa)




                    300

                    250

                    200

                    150

                    100

                     50
                      0
                          0       500        1000         1500         2000        2500           3000       3500
                                                         Bulk stress (kPa)




                                                                                          FSS08.xls 12/03/2003
                                                 STATIC TRIAXIAL TEST
Material: Ferricrete                                         Treatment     1 % cement & 3% bitumen
Sample #:      FSS09
Dry Density (kg/cub m): 1976                                 Confining pressure (kPa): 20
Moisture (%): 7.9                                            Linear stiffness (MPa): 139
                                                             Maximum deviator stress (kPa): 885


                    1000.00
                     900.00
                     800.00
                     700.00
  Stress (kPa)




                     600.00
                     500.00
                     400.00
                     300.00
                     200.00
                     100.00
                       0.00
                              0         5000   10000     15000      20000      25000         30000     35000
                                                       Strain (microstrain)


                    300

                    250
  Stiffness (MPa)




                    200


                    150


                    100


                     50

                      0
                          0       100    200   300     400    500        600   700     800       900    1000
                                                       Bulk stress (kPa)




                                                                                       FSS09.xls 12/03/2003
                                                    STATIC TRIAXIAL TEST
Material: Ferricrete                                          Treatment     1 % cement & 3% bitumen
Sample #:      FSS10
Dry Density (kg/cub m): 1976                                  Confining pressure (kPa): 85
Moisture (%): 9.0                                             Linear stiffness (MPa): 165
                                                              Maximum deviator stress (kPa): 1189


                    1400.00

                    1200.00

                    1000.00
  Stress (kPa)




                     800.00

                     600.00

                     400.00

                     200.00

                       0.00
                              0     5000    10000     15000    20000      25000   30000   35000    40000
                                                        Strain (microstrain)


                    300

                    250
  Stiffness (MPa)




                    200


                    150


                    100


                     50

                      0
                          0       200      400        600      800        1000     1200    1400       1600
                                                        Bulk stress (kPa)




                                                                                      FSS10.xls 12/03/2003
                                                             STATIC TRIAXIAL TEST
Material: Ferricrete                                                      Treatment       1 % cement & 3% bitumen
Sample #:      FSS11
Dry Density (kg/cub m): 1976                                              Confining pressure (kPa): 138
Moisture (%): 8.4                                                         Linear stiffness (MPa): 192
                                                                          Maximum deviator stress (kPa): 1444


                    1600.00

                    1400.00

                    1200.00
  Stress (kPa)




                    1000.00

                     800.00

                     600.00

                     400.00

                     200.00

                       0.00
                              0         5000         10000     15000    20000     25000    30000    35000    40000   45000
                                                                   Strain (microstrain)


                    300

                    250
  Stiffness (MPa)




                    200


                    150


                    100


                     50

                      0
                          0       200          400       600      800      1000     1200     1400     1600    1800    2000
                                                                   Bulk stress (kPa)




                                                                                                     FSS11.xls 12/03/2003
                                           STATIC TRIAXIAL TEST
Material: Ferricrete                                    Treatment    1 % cement & 3% bitumen
Sample #:      FSS12
Dry Density (kg/cub m): 1976                            Confining pressure (kPa): 203
Moisture (%): 7.9                                       Linear stiffness (MPa): 195
                                                        Maximum deviator stress (kPa): 1565


                    1800.00

                    1600.00

                    1400.00
  Stress (kPa)




                    1200.00

                    1000.00

                     800.00

                     600.00

                     400.00

                     200.00

                       0.00
                              0   5000   10000      15000     20000      25000      30000     35000
                                                  Strain (microstrain)


                    300

                    250
  Stiffness (MPa)




                    200


                    150


                    100


                     50

                      0
                          0        500           1000           1500             2000          2500
                                                 Bulk stress (kPa)




                                                                                 FSS12.xls 12/03/2003
                                                  STATIC TRIAXIAL TEST
Material: Ferricrete                                           Treatment      1 % cement & 3% bitumen
Sample #:      FSS13
Dry Density (kg/cub m): 1976                                    Confining pressure (kPa): 21
Moisture (%): 4.5                                               Linear stiffness (MPa): 247
                                                                Maximum deviator stress (kPa): 1793


                    2000.00
                    1800.00
                    1600.00
                    1400.00
  Stress (kPa)




                    1200.00
                    1000.00
                     800.00
                     600.00
                     400.00
                     200.00
                       0.00
                              0         10000         20000      30000          40000      50000          60000
                                                          Strain (microstrain)


                    500
                    450
                    400
  Stiffness (MPa)




                    350
                    300
                    250
                    200
                    150
                    100
                     50
                      0
                          0       200   400     600      800    1000     1200     1400   1600      1800    2000
                                                          Bulk stress (kPa)




                                                                                         FSS13.xls 12/03/2003
                                                STATIC TRIAXIAL TEST
Material: Ferricrete                                          Treatment       1 % cement & 3% bitumen
Sample #:      FSS14
Dry Density (kg/cub m): 1976                                  Confining pressure (kPa): 85
Moisture (%): 4.2                                             Linear stiffness (MPa): 389
                                                              Maximum deviator stress (kPa): 2595


                    3000.00


                    2500.00
  Stress (kPa)




                    2000.00


                    1500.00


                    1000.00


                     500.00


                       0.00
                              0   2000   4000     6000      8000      10000    12000   14000   16000   18000
                                                         Strain (microstrain)


                    500
                    450
                    400
  Stiffness (MPa)




                    350
                    300
                    250
                    200
                    150
                    100
                     50
                      0
                          0        500          1000           1500            2000        2500         3000
                                                       Bulk stress (kPa)




                                                                                        FSS14.xls 12/03/2003
                                                      STATIC TRIAXIAL TEST
Material: Ferricrete                                                 Treatment    1 % cement & 3% bitumen
Sample #:      FSS15
Dry Density (kg/cub m): 1976                                         Confining pressure (kPa): 144
Moisture (%): 4.3                                                    Linear stiffness (MPa): 278
                                                                     Maximum deviator stress (kPa): 2680


                    3000.00


                    2500.00
  Stress (kPa)




                    2000.00


                    1500.00


                    1000.00


                     500.00


                       0.00
                              0         5000          10000           15000        20000       25000      30000
                                                          Strain (microstrain)


                    400

                    350

                    300
  Stiffness (MPa)




                    250

                    200

                    150

                    100

                     50

                      0
                          0       500          1000           1500         2000       2500       3000       3500
                                                          Bulk stress (kPa)




                                                                                             FSS15.xls 12/03/2003
                                                  STATIC TRIAXIAL TEST
Material: Ferricrete                                         Treatment    1 % cement & 3% bitumen
Sample #:      FSS16
Dry Density (kg/cub m): 1976                                 Confining pressure (kPa): 205
Moisture (%): 5.0                                            Linear stiffness (MPa): 278
                                                             Maximum deviator stress (kPa): 2364


                    2500.00


                    2000.00
  Stress (kPa)




                    1500.00


                    1000.00


                     500.00


                       0.00
                              0         5000         10000           15000           20000          25000
                                                     Strain (microstrain)


                    400

                    350

                    300
  Stiffness (MPa)




                    250

                    200

                    150

                    100

                     50

                      0
                          0       500      1000       1500         2000       2500           3000    3500
                                                     Bulk stress (kPa)




                                                                                     FSS16.xls 12/03/2003
                               Dynamic Triaxial Test Data




             Note: The stress ratios indicated on the data printouts are the target values, the actual
                        values are recalculated after testing and are used in the analyses.




Mechanistic-Empirical Structural Design Models for Emulsified Bitumen Treated Materials            115
Report number CR-2003/44
Material:                                                Ferricrete                                                  Sample #: FSS17                                                                                        Sample Good
Density:                                                        1976 kg/ cub m                Confining pressure:                                                                                      79.3 kPa
Moisture:                                                        5.6 %                                     Stress Ratio                                                                                    20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          1000                                                                              2000
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               800                                                                               1500




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          600
                  300                                                                                                                                                            1000
                  200                                                                          400
                                                                                               200                                                                                      500
                  100
                    0                                                                          0                                                                                               0
                                               0              20000           40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                   0.8
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   0.6                                                                                                                       cycles)           1000
                                                   0.4                                                                                                                                             100
                                                   0.2                                                                                                                                             10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.25
                                                0.2                                                                                                                                           1000
                                               0.15
                                                0.1                                                                                                                                           100
                                               0.05
                                                                                                                                                                                               10
                                                  0
                                                          0           20000           40000        60000                                                                                           1
                                                                         Load cycles                                                                                                                   0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M          487          447                                                            362                            562           0.65          0.9                                                                       0.21          0.3
  LVDT #1         225         1017                                                            134                           1514           0.63          1.5                                                                       0.21          0.5
  LVDT #2         208         1040                                                            127                           1598           0.31          1.2                                                                       0.10          0.4
Avg.Lvdt 1&2      217         1028                                                            131                           1556           0.53          1.2                                                                       0.18          0.4
Material:                                                Ferricrete                                                  Sample #: FSS18                                                                                        Sample Good
Density:                                                       1976 kg/ cub m                 Confining pressure:                                                                                      82.0 kPa
Moisture:                                                        5.9 %                                     Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          1500                                                                              1500
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          1000                                                                              1000
                  300
                  200                                                                          500                                                                                      500
                  100
                    0                                                                          0                                                                                               0
                                               0             20000            40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                     1
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                                                   0.8
                 Permanent




                                                                                                                                                                                               1000
                                                   0.6
                                                                                                                                                                             cycles)

                                                                                                                                                                                                   100
                                                   0.4
                                                   0.2                                                                                                                                             10
                                                     0                                                                                                                                                 1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.4
                                                                                                                                                                                              1000
                                               0.3
                                               0.2                                                                                                                                            100
                                               0.1
                                                                                                                                                                                               10
                                                    0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
           Average for test                                                               Final values                                  Final values
                 Elastic    Resilient                                                        Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain     Modulus                                                          Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         1062          558                                                            927                            619           0.94          1.6                                                                       0.31          0.5
  LVDT #1         625          957                                                            491                           1168           0.91          2.6                                                                       0.31          0.9
  LVDT #2         579         1029                                                            468                           1225           0.74          2.2                                                                       0.25          0.7
                  602          993                                                            480                           1196           0.87          2.1                                                                       0.29          0.7
Material:                                                Ferricrete                                                  Sample #: FSS19                                                                                        Sample Good
Density:                                                       1976 kg/ cub m                 Confining pressure:                                                                                   141.6 kPa               INCORRECT CELL PRESS.
Moisture:                                                        5.5 %                                     Stress Ratio                                                                                    90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          2000                                                                              2000
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               1500                                                                              1500




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400
                  300                                                                          1000                                                                              1000
                  200
                                                                                               500                                                                                      500
                  100
                    0                                                                          0                                                                                               0
                                               0             20000            40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                   1.5
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                    1
                                                                                                                                                                             cycles)           1000
                                                                                                                                                                                                   100
                                                   0.5
                                                                                                                                                                                                   10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.5
                                               0.4                                                                                                                                            1000
                                               0.3
                                               0.2                                                                                                                                            100
                                               0.1
                                                                                                                                                                                               10
                                                 0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         1463          658                                                           1248                            771           1.30          3.7                                                                       0.43          1.2
  LVDT #1         738         1313                                                            642                           1498           1.18          3.5                                                                       0.40          1.2
  LVDT #2         838         1147                                                            726                           1326           1.16          4.4                                                                       0.39          1.5
Avg.Lvdt 1&2      788         1230                                                            684                           1412           1.21          3.9                                                                       0.41          1.3
Material:                                                Ferricrete                                                 Sample #: fss20                                                                                         Sample Good
Density:                                                       1976 kg/ cub m                  Confining pressure:                                                                                 145.2 kPa
Moisture:                                                        6.1 %                                    Stress Ratio                                                                                     20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                            800                                                                            2500
   Applied stress (kPa)




                  500




                                                                                                                                                                Esec (MPa)
                                                                                                 600                                                                            2000




                                                                                                         Elastic Strain
                                                                                                         (microstrain)
                  400                                                                                                                                                           1500
                  300                                                                            400
                                                                                                                                                                                1000
                  200
                                                                                                 200                                                                             500
                  100
                    0                                                                             0                                                                                0
                                               0              20000           40000           60000                                                                                                0               20000         40000         60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M          LVDT (1)          LVDT (2)                                                                                       LVDT M                 LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                Permanent Deformation Rate
              deformation (mm)




                                                   0.8
                                                                                                                                                                     Rate (mm/million load
                                                                                                                                                                                             10000
                 Permanent




                                                   0.6                                                                                                                      cycles)           1000
                                                   0.4                                                                                                                                         100
                                                   0.2                                                                                                                                             10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000           60000                                                                                                0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                Rate (% /million load cycles)




                                                                                                                                                                                      1000
                          Plastic Strain (%)




                                                    0.2
                                               0.15                                                                                                                                          100
                                                    0.1
                                               0.05                                                                                                                                          10
                                                     0
                                                          0           20000           40000           60000                                                                                   1
                                                                         Load cycles                                                                                                               0                20000         40000         60000
                                                                                                                                                                                                                      Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                   LVDT M              LVDT (1)           LVDT (2)


       Elastic response results                                                                                                 Plastic response results
            Average for test                                                             Final values                                  Final values
                 Elastic     Resilient                                                      Elastic                       Resilient    Permanent      Rate of                                                                     Plastic       Rate of
                 Strain      Modulus                                                        Strain                        Modulus      Deformation       PD                                                                       Strain          PS
              (microstrain)   (Mpa)                                                      (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                    (%)        (%/million)
  LVDT M          497          577                                                           397                            706           0.58          1.1                                                                        0.19           0.3
  LVDT #1         268         1074                                                           185                           1514           0.49          0.9                                                                        0.17           0.3
  LVDT #2         196         1475                                                           142                           1971           0.48          1.0                                                                        0.16           0.3
Avg.Lvdt 1&2      232         1275                                                           164                           1742           0.52          1.0                                                                        0.17           0.3
Material:                                                Ferricrete                                                  Sample #: fss21                                                                                        Sample Good
Density:                                                       1976 kg/ cub m                 Confining pressure:                                                                                   142.0 kPa
Moisture:                                                        6.3 %                                     Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          2500                                                                              1500
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               2000




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          1500
                                                                                                                                                                                 1000
                  300
                  200                                                                          1000                                                                                     500
                  100                                                                          500
                    0                                                                          0                                                                                               0
                                               0             20000            40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                    2
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   1.5                                                                                                                       cycles)           1000
                                                    1                                                                                                                                              100
                                                   0.5                                                                                                                                             10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.6
                                                                                                                                                                                              1000
                                               0.4
                                                                                                                                                                                              100
                                               0.2
                                                                                                                                                                                               10
                                                    0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         1406          562                                                           1268                            620           1.50          4.6                                                                       0.49          1.5
  LVDT #1         649         1220                                                            581                           1352           1.23          3.5                                                                       0.42          1.2
  LVDT #2         688         1144                                                            637                           1235           1.61          4.2                                                                       0.55          1.4
Avg.Lvdt 1&2      668         1182                                                            609                           1294           1.45          4.1                                                                       0.48          1.4
Material:                                                  Ferricrete                                                 Sample #: FSS22                                                                                        Sample Good
Density:                                                         1976 kg/ cub m                Confining pressure:                                                                                   137.8 kPa
Moisture:                                                          5.8 %                                    Stress Ratio                                                                                    90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                                Resilient modulus

                  600                                                                           3000                                                                              2000
   Applied stress (kPa)




                  500                                                                           2500




                                                                                                                                                                  Esec (MPa)
                                                                                                                                                                                  1500




                                                                                                           Elastic Strain
                                                                                                           (microstrain)
                  400                                                                           2000
                  300                                                                           1500                                                                              1000
                  200                                                                           1000
                                                                                                                                                                                         500
                  100                                                                           500
                    0                                                                           0                                                                                               0
                                               0               20000           40000        60000                                                                                                   0               20000         40000        60000
                                                                  Load cycles                                                                                                                                          Load cycles


                                          Stress                LVDT M          LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                            Permanent Deformation                                                                                                                   Permanent Deformation Rate
              deformation (mm)




                                                   4
                                                                                                                                                                       Rate (mm/million load
                                                                                                                                                                                               10000
                 Permanent




                                                   3                                                                                                                          cycles)           1000
                                                   2                                                                                                                                                100
                                                   1                                                                                                                                                10
                                                   0                                                                                                                                                    1
                                                       0            20000              40000         60000                                                                                                  0     10000 20000 30000 40000 50000
                                                                         Load cycles                                                                                                                                     Load cycles

                                               LVDT M               LVDT (1)             LVDT (2)                                                                                                   LVDT M             LVDT (1)         LVDT (2)



                                                                  Plastic Strain                                                                                                                                Plastic Strain Rate
                                                                                                                                  Rate (% /million load cycles)




                                                                                                                                                                                        10000
                          Plastic Strain (%)




                                               1.5
                                                                                                                                                                                               1000
                                                       1
                                                                                                                                                                                               100
                                               0.5
                                                                                                                                                                                                10
                                                       0
                                                           0           20000           40000         60000                                                                                          1
                                                                          Load cycles                                                                                                                   0            20000         40000        60000
                                                                                                                                                                                                                        Load cycles
                                                   LVDT M               LVDT (1)          LVDT (2)                                                                                                   LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                   Plastic response results
            Average for test                                                               Final values                                  Final values
                 Elastic     Resilient                                                        Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                          Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                        (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         2313          571                                                            1721                            744           2.18          9.0                                                                       0.71          3.0
  LVDT #1         980         1369                                                             761                           1681           1.74          8.6                                                                       0.59          2.9
  LVDT #2        1099         1191                                                             983                           1302           2.92          9.1                                                                       0.99          3.1
Avg.Lvdt 1&2     1040         1280                                                             872                           1491           2.28          8.9                                                                       0.76          3.0
Material:                                                Ferricrete                                                  Sample #: FSS23                                                                                      Sample Good
Density:                                                        1976 kg/ cub m                Confining pressure:                                                                                    82.0 kPa
Moisture:                                                        2.8 %                                     Stress Ratio                                                                                  20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                             Resilient modulus

                  600                                                                          1500                                                                              2500
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                                                                                                                 2000




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          1000                                                                              1500
                  300
                                                                                                                                                                                 1000
                  200                                                                          500
                  100                                                                                                                                                             500
                    0                                                                          0                                                                                    0
                                               0              20000           40000        60000                                                                                                 0               20000         40000        60000
                                                                Load cycles                                                                                                                                         Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                         LVDT M              LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                 Permanent Deformation Rate
              deformation (mm)




                                                   0.8
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   0.6                                                                                                                       cycles)           1000
                                                   0.4                                                                                                                                          100
                                                   0.2                                                                                                                                           10
                                                    0                                                                                                                                                1
                                                         0            20000           40000        60000                                                                                                 0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                   Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                  LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                               Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.25
                                                0.2                                                                                                                                           1000
                                               0.15
                                                0.1                                                                                                                                           100
                                               0.05
                                                                                                                                                                                               10
                                                  0
                                                          0           20000           40000        60000                                                                                        1
                                                                         Load cycles                                                                                                                 0            20000         40000        60000
                                                                                                                                                                                                                     Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                 LVDT M              LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                  Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                    Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                 (%)       (%/million)
  LVDT M          845          540                                                            717                            632           0.68          3.0                                                                     0.22          1.0
  LVDT #1         316         1441                                                            238                           1907           0.53          2.3                                                                     0.18          0.8
  LVDT #2         312         1462                                                            222                           2043           0.58          2.8                                                                     0.20          0.9
Avg.Lvdt 1&2      314         1451                                                            230                           1975           0.59          2.7                                                                     0.20          0.9
Material:                                                  Ferricrete                                                 Sample #: FSS24                                                                                       Sample Good
Density:                                                         1976 kg/ cub m                Confining pressure:                                                                                      83.0 kPa
Moisture:                                                          3.4 %                                    Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                             Resilient modulus

                  600                                                                           4000                                                                              1500
   Applied stress (kPa)




                  500




                                                                                                                                                                  Esec (MPa)
                                                                                                3000




                                                                                                           Elastic Strain
                                                                                                           (microstrain)
                  400                                                                                                                                                             1000
                  300                                                                           2000
                  200                                                                                                                                                                    500
                                                                                                1000
                  100
                    0                                                                           0                                                                                               0
                                               0               20000           40000        60000                                                                                                   0             20000         40000        60000
                                                                  Load cycles                                                                                                                                        Load cycles


                                          Stress                LVDT M          LVDT (1)        LVDT (2)                                                                                            LVDT M           LVDT (1)         LVDT (2)


                                                            Permanent Deformation                                                                                                                   Permanent Deformation Rate
              deformation (mm)




                                                   5
                                                                                                                                                                       Rate (mm/million load
                                                                                                                                                                                               100000
                                                   4
                 Permanent




                                                                                                                                                                                                10000
                                                   3
                                                                                                                                                                              cycles)
                                                                                                                                                                                                 1000
                                                   2                                                                                                                                              100
                                                   1                                                                                                                                               10
                                                   0                                                                                                                                                1
                                                       0            20000              40000         60000                                                                                                  0    10000 20000 30000 40000 50000
                                                                         Load cycles                                                                                                                                   Load cycles

                                               LVDT M               LVDT (1)             LVDT (2)                                                                                                   LVDT M           LVDT (1)         LVDT (2)



                                                                  Plastic Strain                                                                                                                             Plastic Strain Rate
                                                                                                                                  Rate (% /million load cycles)




                                                                                                                                                                                        10000
                          Plastic Strain (%)




                                                       2
                                                                                                                                                                                               1000
                                               1.5
                                                       1                                                                                                                                       100
                                               0.5
                                                                                                                                                                                                10
                                                       0
                                                           0           20000           40000         60000                                                                                          1
                                                                          Load cycles                                                                                                                   0          20000         40000        60000
                                                                                                                                                                                                                      Load cycles
                                                   LVDT M               LVDT (1)          LVDT (2)                                                                                                   LVDT M           LVDT (1)          LVDT (2)


       Elastic response results                                                                                                   Plastic response results
            Average for test                                                               Final values                                  Final values
                 Elastic     Resilient                                                        Elastic                       Resilient    Permanent      Rate of                                                                  Plastic      Rate of
                 Strain      Modulus                                                          Strain                        Modulus      Deformation       PD                                                                    Strain         PS
              (microstrain)   (Mpa)                                                        (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                 (%)       (%/million)
  LVDT M         2313          543                                                            2019                            621           4.23          9.4                                                                     1.40          3.1
  LVDT #1        1016         1229                                                            1002                           1253           4.41         11.6                                                                     1.49          3.9
  LVDT #2        1089         1149                                                            1047                           1198           4.14          8.9                                                                     1.40          3.0
Avg.Lvdt 1&2     1052         1189                                                            1025                           1225           4.26         10.0                                                                     1.43          3.4
Material:                                                  Ferricrete                                                  Sample #: FSS25                                                                                           Sample Failed
Density:                                                           1976 kg/ cub m              Confining pressure:                                                                                       79.7 kPa
Moisture:                                                           2.7 %                                    Stress Ratio                                                                                    90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                              Resilient modulus

                  600                                                                           5000                                                                               2000
   Applied stress (kPa)




                  500




                                                                                                                                                                   Esec (MPa)
                                                                                                4000                                                                               1500




                                                                                                            Elastic Strain
                                                                                                            (microstrain)
                  400                                                                           3000
                  300                                                                                                                                                              1000
                  200                                                                           2000
                                                                                                1000                                                                                      500
                  100
                    0                                                                           0                                                                                                0
                                               0           1000    2000    3000     4000     5000                                                                                                    0           1000    2000    3000      4000         5000
                                                                   Load cycles                                                                                                                                           Load cycles


                                          Stress                  LVDT M         LVDT (1)        LVDT (2)                                                                                            LVDT M              LVDT (1)          LVDT (2)


                                                            Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                   8
                                                                                                                                                                        Rate (mm/million load
                                                                                                                                                                                                100000
                 Permanent




                                                   6                                                                                                                           cycles)           10000
                                                                                                                                                                                                  1000
                                                   4
                                                                                                                                                                                                   100
                                                   2                                                                                                                                                10
                                                   0                                                                                                                                                 1
                                                       0       1000       2000     3000      4000     5000                                                                                                   0     10000 20000 30000 40000 50000
                                                                          Load cycles                                                                                                                                      Load cycles

                                               LVDT M                 LVDT (1)          LVDT (2)                                                                                                     LVDT M              LVDT (1)          LVDT (2)



                                                                   Plastic Strain                                                                                                                             Plastic Strain Rate
                                                                                                                                   Rate (% /million load cycles)




                                                                                                                                                                                         10000
                          Plastic Strain (%)




                                               3
                                                                                                                                                                                                1000
                                               2
                                                                                                                                                                                                100
                                               1
                                                                                                                                                                                                 10
                                               0
                                                       0       1000       2000     3000      4000     5000                                                                                           1
                                                                          Load cycles                                                                                                                    0        1000    2000      3000       4000      5000
                                                                                                                                                                                                                          Load cycles
                                               LVDT M                 LVDT (1)          LVDT (2)                                                                                                      LVDT M              LVDT (1)          LVDT (2)


       Elastic response results                                                                                                    Plastic response results
            Average for test                                                                Final values                                  Final values
                 Elastic     Resilient                                                         Elastic                       Resilient    Permanent      Rate of                                                                     Plastic           Rate of
                 Strain      Modulus                                                           Strain                        Modulus      Deformation       PD                                                                       Strain              PS
              (microstrain)   (Mpa)                                                         (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                    (%)            (%/million)
  LVDT M         3375          593                                                             3429                            572           6.27        1696.5                                                                       2.06              556.2
  LVDT #1        1600         1235                                                             1845                           1063           7.56        1987.4                                                                       2.56              673.7
  LVDT #2        1428         1387                                                             1488                           1318           4.41        1159.7                                                                       1.49              393.1
Avg.Lvdt 1&2     1514         1311                                                             1667                           1190           6.08        1614.5                                                                       2.04              541.0
Material:                                                Ferricrete                                                  Sample #: FSS26                                                                                        Sample Good
Density:                                                        1976 kg/ cub m                Confining pressure:                                                                                   143.6 kPa
Moisture:                                                        2.8 %                                     Stress Ratio                                                                                    20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          1200                                                                              4000
   Applied stress (kPa)




                  500                                                                          1000




                                                                                                                                                                 Esec (MPa)
                                                                                                                                                                                 3000




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          800
                  300                                                                          600                                                                               2000
                  200                                                                          400
                                                                                                                                                                                 1000
                  100                                                                          200
                    0                                                                          0                                                                                               0
                                               0              20000           40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                   0.8
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   0.6                                                                                                                       cycles)           1000
                                                   0.4                                                                                                                                             100
                                                   0.2                                                                                                                                             10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.25
                                                0.2                                                                                                                                           1000
                                               0.15
                                                0.1                                                                                                                                           100
                                               0.05
                                                                                                                                                                                               10
                                                  0
                                                          0           20000           40000        60000                                                                                           1
                                                                         Load cycles                                                                                                                   0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M          755          657                                                            583                            840           0.68          1.9                                                                       0.23          0.6
  LVDT #1         228         2311                                                            154                           3180           0.60          1.9                                                                       0.20          0.7
  LVDT #2         250         1983                                                            189                           2595           0.39          1.8                                                                       0.13          0.6
Avg.Lvdt 1&2      239         2147                                                            172                           2887           0.56          1.9                                                                       0.19          0.6
Material:                                                Ferricrete                                                  Sample #: FSS27                                                                                        Sample Good
Density:                                                       1976 kg/ cub m                 Confining pressure:                                                                                   142.0 kPa
Moisture:                                                        2.8 %                                     Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          3000                                                                              2000
   Applied stress (kPa)




                  500                                                                          2500




                                                                                                                                                                 Esec (MPa)
                                                                                                                                                                                 1500




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          2000
                  300                                                                          1500                                                                              1000
                  200                                                                          1000
                                                                                                                                                                                        500
                  100                                                                          500
                    0                                                                          0                                                                                               0
                                               0             20000            40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                     3
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                                                   2.5
                 Permanent




                                                     2
                                                                                                                                                                             cycles)           1000
                                                   1.5                                                                                                                                             100
                                                     1
                                                   0.5                                                                                                                                             10
                                                     0                                                                                                                                                 1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                                 1
                                               0.8                                                                                                                                            1000
                                               0.6
                                               0.4                                                                                                                                            100
                                               0.2
                                                                                                                                                                                               10
                                                 0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         2433          560                                                           1993                            682           2.37          3.9                                                                       0.78          1.3
  LVDT #1         959         1424                                                           1002                           1357           2.44          1.7                                                                       0.83          0.6
  LVDT #2         929         1465                                                            805                           1689           2.10          6.0                                                                       0.71          2.0
Avg.Lvdt 1&2      944         1444                                                            903                           1523           2.30          3.9                                                                       0.77          1.3
Material:                                                  Ferricrete                                                  Sample #: FSS28                                                                                            Sample Failed
Density:                                                         1976 kg/ cub m                Confining pressure:                                                                                    143.0 kPa
Moisture:                                                            2.8 %                                   Stress Ratio                                                                                     90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                                  Resilient modulus

                  600                                                                             4000                                                                             2500
   Applied stress (kPa)




                  500




                                                                                                                                                                   Esec (MPa)
                                                                                                  3000                                                                             2000




                                                                                                            Elastic Strain
                                                                                                            (microstrain)
                  400                                                                                                                                                              1500
                  300                                                                             2000
                                                                                                                                                                                   1000
                  200
                                                                                                  1000                                                                              500
                  100
                    0                                                                            0                                                                                    0
                                               0            10000       20000     30000      40000                                                                                                    0            10000      20000       30000    40000
                                                                    Load cycles                                                                                                                                             Load cycles


                                          Stress                LVDT M           LVDT (1)         LVDT (2)                                                                                        LVDT M                   LVDT (1)          LVDT (2)


                                                            Permanent Deformation                                                                                                                 Permanent Deformation Rate
              deformation (mm)




                                                   12
                                                                                                                                                                        Rate (mm/million load
                                                                                                                                                                                                10000
                                                   10
                 Permanent




                                                    8
                                                                                                                                                                               cycles)           1000
                                                    6                                                                                                                                             100
                                                    4
                                                    2                                                                                                                                                 10
                                                    0                                                                                                                                                     1
                                                        0        10000          20000     30000       40000                                                                                                   0     10000 20000 30000 40000 50000
                                                                          Load cycles                                                                                                                                        Load cycles

                                               LVDT M                 LVDT (1)            LVDT (2)                                                                                                    LVDT M                LVDT (1)         LVDT (2)



                                                                  Plastic Strain                                                                                                                                  Plastic Strain Rate
                                                                                                                                   Rate (% /million load cycles)




                                                                                                                                                                                         1000
                          Plastic Strain (%)




                                                   4
                                                   3                                                                                                                                            100
                                                   2
                                                   1                                                                                                                                            10
                                                   0
                                                       0        10000        20000        30000       40000                                                                                      1
                                                                         Load cycles                                                                                                                  0             10000      20000       30000    40000
                                                                                                                                                                                                                            Load cycles
                                                   LVDT M               LVDT (1)           LVDT (2)                                                                                                   LVDT M                 LVDT (1)         LVDT (2)


       Elastic response results                                                                                                    Plastic response results
            Average for test                                                                Final values                                  Final values
                 Elastic     Resilient                                                         Elastic                       Resilient    Permanent      Rate of                                                                        Plastic     Rate of
                 Strain      Modulus                                                           Strain                        Modulus      Deformation       PD                                                                          Strain         PS
              (microstrain)   (Mpa)                                                         (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                       (%)      (%/million)
  LVDT M         3188          666                                                             3537                            592           6.43         316.7                                                                          2.12        104.5
  LVDT #1        1702         1255                                                             2186                            958           9.82         400.5                                                                          3.33        135.7
  LVDT #2         997         2128                                                             1053                           1988           4.09         233.9                                                                          1.39         79.3
Avg.Lvdt 1&2     1350         1691                                                             1619                           1473           6.78         317.0                                                                          2.28        106.5
Material:                                                Ferricrete                                                 Sample #: FSS29                                                                                        Sample Good
Density:                                                       2014 kg/ cub m                  Confining pressure:                                                                                    81.6 kPa
Moisture:                                                        9.4 %                                    Stress Ratio                                                                                    20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                              Resilient modulus

                  600                                                                            800                                                                            1500
   Applied stress (kPa)




                  500




                                                                                                                                                                Esec (MPa)
                                                                                                 600




                                                                                                         Elastic Strain
                                                                                                         (microstrain)
                  400                                                                                                                                                           1000
                  300                                                                            400
                  200                                                                                                                                                                  500
                                                                                                 200
                  100
                    0                                                                             0                                                                                           0
                                               0              20000           40000           60000                                                                                               0               20000         40000        60000
                                                                Load cycles                                                                                                                                          Load cycles


                                          Stress              LVDT M          LVDT (1)          LVDT (2)                                                                                          LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                   Permanent Deformation Rate
              deformation (mm)




                                                   0.8
                                                                                                                                                                     Rate (mm/million load
                                                                                                                                                                                             10000
                 Permanent




                                                   0.6                                                                                                                      cycles)           1000
                                                   0.4                                                                                                                                            100
                                                   0.2                                                                                                                                            10
                                                    0                                                                                                                                                 1
                                                         0            20000           40000           60000                                                                                               0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                    Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                   LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                Plastic Strain Rate
                                                                                                                                Rate (% /million load cycles)




                                                                                                                                                                                      10000
                          Plastic Strain (%)




                                               0.25
                                                0.2                                                                                                                                          1000
                                               0.15
                                                0.1                                                                                                                                          100
                                               0.05
                                                                                                                                                                                              10
                                                  0
                                                          0           20000           40000           60000                                                                                       1
                                                                         Load cycles                                                                                                                  0            20000         40000        60000
                                                                                                                                                                                                                      Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                   LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                 Plastic response results
           Average for test                                                              Final values                                  Final values
                 Elastic    Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain     Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                      (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M          575          412                                                           596                            399           0.66         (0.5)                                                                      0.22         (0.2)
  LVDT #1         276          863                                                           263                            902           0.55         (0.8)                                                                      0.19         (0.3)
  LVDT #2         317          750                                                           296                            803           0.57         (0.4)                                                                      0.19         (0.1)
                  296          807                                                           279                            852           0.59         (0.6)                                                                      0.20         (0.2)
Material:                                                Ferricrete                                                  Sample #: FSS30                                                                                          Sample Good
Density:                                                       2014 kg/ cub m                 Confining pressure:                                                                                      81.3 kPa
Moisture:                                                        9.6 %                                     Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          2500                                                                              800
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               2000                                                                              600




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          1500
                  300                                                                                                                                                            400
                  200                                                                          1000
                                                                                               500                                                                               200
                  100
                    0                                                                          0                                                                                              0
                                               0             20000            40000        60000                                                                                                  0               20000          40000         60000
                                                                Load cycles                                                                                                                                          Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                           LVDT M              LVDT (1)          LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                    2
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   1.5                                                                                                                       cycles)              1000
                                                    1                                                                                                                                             100
                                                   0.5                                                                                                                                                10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                       Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)           LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.8
                                                                                                                                                                                              1000
                                               0.6
                                               0.4                                                                                                                                            100
                                               0.2
                                                                                                                                                                                                  10
                                                    0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000         60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                      LVDT M           LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic       Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain          PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)        (%/million)
  LVDT M         1875          362                                                           1592                            420           1.87          2.1                                                                       0.62           0.7
  LVDT #1        1187          574                                                            999                            670           1.72          1.5                                                                       0.58           0.5
  LVDT #2        1194          570                                                           1027                            652           1.46          1.6                                                                       0.50           0.5
Avg.Lvdt 1&2     1191          572                                                           1013                            661           1.68          1.7                                                                       0.56           0.6
Material:                                               Ferricrete                                                  Sample #: FSS31                                                                                         Sample Failed
Density:                                                           2014 kg/ cub m           Confining pressure:                                                                                       81.7 kPa
Moisture:                                                           9.6 %                                 Stress Ratio                                                                                    90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                           Resilient modulus

                  600                                                                         6000                                                                              800
   Applied stress (kPa)




                  500                                                                         5000




                                                                                                                                                                Esec (MPa)
                                                                                                                                                                                600




                                                                                                         Elastic Strain
                                                                                                         (microstrain)
                  400                                                                         4000
                  300                                                                         3000                                                                              400
                  200                                                                         2000
                                                                                                                                                                                200
                  100                                                                         1000
                    0                                                                         0                                                                                              0
                                               0            2000      4000       6000      8000                                                                                                  0            2000       4000        6000         8000
                                                                   Load cycles                                                                                                                                       Load cycles


                                          Stress               LVDT M         LVDT (1)          LVDT (2)                                                                                         LVDT M              LVDT (1)          LVDT (2)


                                                            Permanent Deformation                                                                                                                Permanent Deformation Rate
              deformation (mm)




                                                   12
                                                                                                                                                                     Rate (mm/million load
                                                                                                                                                                                             100000
                                                   10
                 Permanent




                                                                                                                                                                                              10000
                                                    8
                                                                                                                                                                            cycles)
                                                                                                                                                                                               1000
                                                    6
                                                    4                                                                                                                                           100
                                                    2                                                                                                                                            10
                                                    0                                                                                                                                             1
                                                        0          2000       4000       6000       8000                                                                                                  0    10000 20000 30000 40000 50000
                                                                          Load cycles                                                                                                                                   Load cycles

                                               LVDT M                LVDT (1)           LVDT (2)                                                                                                  LVDT M             LVDT (1)          LVDT (2)



                                                                   Plastic Strain                                                                                                                          Plastic Strain Rate
                                                                                                                                Rate (% /million load cycles)




                                                                                                                                                                                      10000
                          Plastic Strain (%)




                                               4
                                                                                                                                                                                             1000
                                               3
                                               2                                                                                                                                             100
                                               1
                                                                                                                                                                                                 10
                                               0
                                                    0           2000         4000        6000       8000                                                                                         1
                                                                          Load cycles                                                                                                                 0         2000      4000         6000       8000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)           LVDT (2)                                                                                                    LVDT M            LVDT (1)         LVDT (2)


       Elastic response results                                                                                                 Plastic response results
            Average for test                                                             Final values                                  Final values
                 Elastic     Resilient                                                      Elastic                       Resilient    Permanent      Rate of                                                                     Plastic      Rate of
                 Strain      Modulus                                                        Strain                        Modulus      Deformation       PD                                                                       Strain          PS
              (microstrain)   (Mpa)                                                      (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                    (%)       (%/million)
  LVDT M         3559          308                                                          5052                            206            6.14       1543.5                                                                       2.01         506.1
  LVDT #1        1946          556                                                          2484                            418            3.96        278.5                                                                       1.34          94.4
  LVDT #2        2214          499                                                          3186                            326           10.52       4424.1                                                                       3.57        1499.7
Avg.Lvdt 1&2     2080          527                                                          2835                            372            6.87       2082.1                                                                       2.31         700.1
Material:                                                Ferricrete                                                  Sample #: FSS32                                                                                        Sample Good
Density:                                                        2014 kg/ cub m                Confining pressure:                                                                                   139.8 kPa
Moisture:                                                        9.9 %                                     Stress Ratio                                                                                    20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          1200                                                                              2000
   Applied stress (kPa)




                  500                                                                          1000




                                                                                                                                                                 Esec (MPa)
                                                                                                                                                                                 1500




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          800
                  300                                                                          600                                                                               1000
                  200                                                                          400
                                                                                                                                                                                        500
                  100                                                                          200
                    0                                                                          0                                                                                               0
                                               0              20000           40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                   0.6
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                                                   0.5
                 Permanent




                                                   0.4
                                                                                                                                                                             cycles)           1000
                                                   0.3                                                                                                                                             100
                                                   0.2
                                                   0.1                                                                                                                                             10
                                                     0                                                                                                                                                 1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                                    0.2
                                                                                                                                                                                              1000
                                               0.15
                                                    0.1                                                                                                                                       100
                                               0.05
                                                                                                                                                                                               10
                                                     0
                                                          0           20000           40000        60000                                                                                           1
                                                                         Load cycles                                                                                                                   0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M          738          413                                                            644                            459           0.48          0.4                                                                       0.16          0.1
  LVDT #1         328          946                                                            286                           1036           0.31          1.4                                                                       0.11          0.5
  LVDT #2         268         1207                                                            196                           1512           0.39          0.8                                                                       0.13          0.3
Avg.Lvdt 1&2      298         1076                                                            241                           1274           0.39          0.9                                                                       0.13          0.3
Material:                                                Ferricrete                                                  Sample #: FSS33                                                                                      Sample Good
Density:                                                       2014 kg/ cub m                 Confining pressure:                                                                                141.1 kPa
Moisture:                                                        7.7 %                                     Stress Ratio                                                                                  55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                             Resilient modulus

                  600                                                                          2500                                                                              1000
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               2000                                                                               800




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          1500                                                                               600
                  300
                                                                                               1000                                                                               400
                  200
                  100                                                                          500                                                                                200
                    0                                                                          0                                                                                    0
                                               0             20000            40000        60000                                                                                                 0               20000         40000        60000
                                                                Load cycles                                                                                                                                         Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                         LVDT M              LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                 Permanent Deformation Rate
              deformation (mm)




                                                    2
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   1.5                                                                                                                       cycles)           1000
                                                    1                                                                                                                                           100
                                                   0.5                                                                                                                                           10
                                                    0                                                                                                                                                1
                                                         0            20000           40000        60000                                                                                                 0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                   Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                  LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                               Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.6
                                                                                                                                                                                              1000
                                               0.4
                                                                                                                                                                                              100
                                               0.2
                                                                                                                                                                                               10
                                                    0
                                                         0           20000            40000        60000                                                                                        1
                                                                        Load cycles                                                                                                                  0            20000         40000        60000
                                                                                                                                                                                                                     Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                 LVDT M              LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                  Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                    Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                 (%)       (%/million)
  LVDT M         2024          419                                                           1811                            460           1.34          3.7                                                                     0.44          1.2
  LVDT #1        1049          803                                                           1092                            762           1.52          3.7                                                                     0.52          1.3
  LVDT #2         961          877                                                            920                            905           1.11          3.5                                                                     0.38          1.2
Avg.Lvdt 1&2     1005          840                                                           1006                            834           1.32          3.7                                                                     0.44          1.2
Material:                                                  Ferricrete                                                 Sample #: FSS34                                                                                     Sample Good
Density:                                                         2014 kg/ cub m               Confining pressure:                                                                                 140.1 kPa
Moisture:                                                               9 %                                 Stress Ratio                                                                                  90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                           Resilient modulus

                  600                                                                        12000                                                                                1000
   Applied stress (kPa)




                  500                                                                        10000




                                                                                                                                                                  Esec (MPa)
                                                                                                                                                                                   800




                                                                                                           Elastic Strain
                                                                                                           (microstrain)
                  400                                                                        8000                                                                                  600
                  300                                                                        6000
                                                                                                                                                                                   400
                  200                                                                        4000
                  100                                                                        2000                                                                                  200
                    0                                                                        0                                                                                       0
                                               0               20000          40000      60000                                                                                                    0             20000         40000        60000
                                                                 Load cycles                                                                                                                                       Load cycles


                                          Stress                LVDT M          LVDT (1)        LVDT (2)                                                                                         LVDT M            LVDT (1)         LVDT (2)


                                                            Permanent Deformation                                                                                                                Permanent Deformation Rate
              deformation (mm)




                                                   10
                                                                                                                                                                       Rate (mm/million load
                                                                                                                                                                                               100000
                                                    8
                 Permanent




                                                                                                                                                                                                10000
                                                    6
                                                                                                                                                                              cycles)
                                                                                                                                                                                                 1000
                                                    4                                                                                                                                             100
                                                    2                                                                                                                                              10
                                                    0                                                                                                                                               1
                                                        0              20000          40000         60000                                                                                                 0    10000 20000 30000 40000 50000
                                                                          Load cycles                                                                                                                                Load cycles

                                               LVDT M               LVDT (1)            LVDT (2)                                                                                                  LVDT M           LVDT (1)         LVDT (2)



                                                                  Plastic Strain                                                                                                                           Plastic Strain Rate
                                                                                                                                  Rate (% /million load cycles)




                                                                                                                                                                                        10000
                          Plastic Strain (%)




                                                   4
                                                                                                                                                                                               1000
                                                   3
                                                   2                                                                                                                                           100
                                                   1
                                                                                                                                                                                                10
                                                   0
                                                       0           20000              40000         60000                                                                                        1
                                                                         Load cycles                                                                                                                  0          20000         40000        60000
                                                                                                                                                                                                                    Load cycles
                                                   LVDT M               LVDT (1)         LVDT (2)                                                                                                 LVDT M            LVDT (1)          LVDT (2)


       Elastic response results                                                                                                   Plastic response results
            Average for test                                                               Final values                                  Final values
                 Elastic     Resilient                                                        Elastic                       Resilient    Permanent      Rate of                                                                Plastic      Rate of
                 Strain      Modulus                                                          Strain                        Modulus      Deformation       PD                                                                  Strain         PS
              (microstrain)   (Mpa)                                                        (microstrain)                     (Mpa)          (mm)      (mm/million)                                                               (%)       (%/million)
  LVDT M         4063          330                                                            3865                            342           7.90         24.6                                                                   2.59          8.1
  LVDT #1        2269          614                                                            1985                            666           8.47         26.8                                                                   2.87          9.1
  LVDT #2        2033          674                                                            1724                            767           6.80         26.1                                                                   2.31          8.8
Avg.Lvdt 1&2     2151          644                                                            1854                            717           7.73         25.8                                                                   2.59          8.7
Material:                                                Ferricrete                                                  Sample #: FSS35                                                                                         Sample Good
Density:                                                        2014 kg/ cub m                Confining pressure:                                                                                   138.8 kPa                INCORRECT CELL PRESS
Moisture:                                                        4.4 %                                     Stress Ratio                                                                                     20 % of failure stress


                                               Applied stress and elastic strain                                                                                                                                Resilient modulus

                  600                                                                          1500                                                                              2500
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                                                                                                                 2000




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400                                                                          1000                                                                              1500
                  300
                                                                                                                                                                                 1000
                  200                                                                          500
                  100                                                                                                                                                             500
                    0                                                                          0                                                                                    0
                                               0              20000           40000        60000                                                                                                    0               20000         40000         60000
                                                                Load cycles                                                                                                                                            Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                         LVDT M                 LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                 Permanent Deformation Rate
              deformation (mm)




                                                   0.5
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                                                   0.4
                 Permanent




                                                                                                                                                                                               1000
                                                   0.3
                                                                                                                                                                             cycles)

                                                                                                                                                                                                100
                                                   0.2
                                                   0.1                                                                                                                                              10
                                                     0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                    0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                      Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                     LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                  Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       1000
                          Plastic Strain (%)




                                                    0.2
                                               0.15                                                                                                                                           100
                                                    0.1
                                               0.05                                                                                                                                           10
                                                     0
                                                          0           20000           40000        60000                                                                                       1
                                                                         Load cycles                                                                                                                0                20000         40000         60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M              LVDT (1)           LVDT (2)


       Elastic response results                                                                                                  Plastic response results
           Average for test                                                               Final values                                  Final values
                 Elastic    Resilient                                                        Elastic                       Resilient    Permanent      Rate of                                                                     Plastic       Rate of
                 Strain     Modulus                                                          Strain                        Modulus      Deformation       PD                                                                       Strain          PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                    (%)        (%/million)
  LVDT M          878          515                                                            839                            533           0.35          1.5                                                                        0.12           0.5
  LVDT #1         284         1589                                                            250                           1785           0.27          1.0                                                                        0.09           0.4
  LVDT #2         263         1706                                                            240                           1866           0.42          0.9                                                                        0.14           0.3
                  273         1648                                                            245                           1826           0.35          1.2                                                                        0.12           0.4
Material:                                                Ferricrete                                                  Sample #: FSS36                                                                                        Sample Good
Density:                                                       2014 kg/ cub m                 Confining pressure:                                                                                      81.0 kPa
Moisture:                                                        5.1 %                                     Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          4000                                                                              2000
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               3000                                                                              1500




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400
                  300                                                                          2000                                                                              1000
                  200
                                                                                               1000                                                                                     500
                  100
                    0                                                                          0                                                                                               0
                                               0             20000            40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                   1.5
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                    1
                                                                                                                                                                             cycles)           1000
                                                                                                                                                                                                   100
                                                   0.5
                                                                                                                                                                                                   10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.5
                                               0.4                                                                                                                                            1000
                                               0.3
                                               0.2                                                                                                                                            100
                                               0.1
                                                                                                                                                                                               10
                                                 0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         2685          457                                                           2176                            567           0.93          6.1                                                                       0.31          2.0
  LVDT #1         792         1572                                                            695                           1776           0.41          0.8                                                                       0.14          0.3
  LVDT #2         968         1272                                                           1006                           1226           1.34          2.4                                                                       0.45          0.8
Avg.Lvdt 1&2      880         1422                                                            850                           1501           0.89          3.1                                                                       0.30          1.0
Material:                                                Ferricrete                                                Sample #: FSS37                                                                                        Sample Failed
Density:                                                       2014 kg/ cub m              Confining pressure:                                                                                       84.9 kPa
Moisture:                                                        4.9 %                                   Stress Ratio                                                                                    90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                          Resilient modulus

                  600                                                                        4000                                                                              1500
   Applied stress (kPa)




                  500




                                                                                                                                                               Esec (MPa)
                                                                                             3000




                                                                                                        Elastic Strain
                                                                                                        (microstrain)
                  400                                                                                                                                                          1000
                  300                                                                        2000
                  200                                                                                                                                                                 500
                                                                                             1000
                  100
                    0                                                                        0                                                                                               0
                                               0              500           1000          1500                                                                                                   0              500           1000             1500
                                                                Load cycles                                                                                                                                       Load cycles


                                          Stress              LVDT M         LVDT (1)        LVDT (2)                                                                                            LVDT M           LVDT (1)          LVDT (2)


                                                          Permanent Deformation                                                                                                                  Permanent Deformation Rate
              deformation (mm)




                                                   2.5
                                                                                                                                                                    Rate (mm/million load
                                                                                                                                                                                            100000
                                                     2
                 Permanent




                                                                                                                                                                                             10000
                                                   1.5
                                                                                                                                                                           cycles)
                                                                                                                                                                                              1000
                                                     1                                                                                                                                         100
                                                   0.5                                                                                                                                          10
                                                     0                                                                                                                                           1
                                                         0            500          1000           1500                                                                                                   0    10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                   Load cycles

                                               LVDT M               LVDT (1)        LVDT (2)                                                                                                     LVDT M           LVDT (1)          LVDT (2)



                                                                Plastic Strain                                                                                                                            Plastic Strain Rate
                                                                                                                               Rate (% /million load cycles)




                                                                                                                                                                                     10000
                          Plastic Strain (%)




                                                 1
                                               0.8                                                                                                                                          1000
                                               0.6
                                               0.4                                                                                                                                          100
                                               0.2
                                                                                                                                                                                             10
                                                 0
                                                         0            500          1000           1500                                                                                           1
                                                                        Load cycles                                                                                                                  0            500          1000            1500
                                                                                                                                                                                                                   Load cycles
                                                   LVDT M             LVDT (1)        LVDT (2)                                                                                                    LVDT M           LVDT (1)          LVDT (2)


       Elastic response results                                                                                                Plastic response results
            Average for test                                                            Final values                                  Final values
                 Elastic     Resilient                                                     Elastic                       Resilient    Permanent      Rate of                                                                  Plastic      Rate of
                 Strain      Modulus                                                       Strain                        Modulus      Deformation       PD                                                                    Strain          PS
              (microstrain)   (Mpa)                                                     (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                 (%)       (%/million)
  LVDT M         3383          570                                                         3453                            558           1.94         623.7                                                                    0.64         204.5
  LVDT #1        1649         1171                                                         1549                           1244           1.51          62.0                                                                    0.51          21.0
  LVDT #2        1901         1018                                                         2095                            920           2.33        1732.2                                                                    0.79         587.2
Avg.Lvdt 1&2     1775         1095                                                         1822                           1082           1.93         806.0                                                                    0.65         270.9
Material:                                                Ferricrete                                                  Sample #: FSS39                                                                                        Sample Good
Density:                                                       2014 kg/ cub m                 Confining pressure:                                                                                   140.6 kPa
Moisture:                                                        5.6 %                                     Stress Ratio                                                                                    55 % of failure stress


                                               Applied stress and elastic strain                                                                                                                               Resilient modulus

                  600                                                                          4000                                                                              2000
   Applied stress (kPa)




                  500




                                                                                                                                                                 Esec (MPa)
                                                                                               3000                                                                              1500




                                                                                                          Elastic Strain
                                                                                                          (microstrain)
                  400
                  300                                                                          2000                                                                              1000
                  200
                                                                                               1000                                                                                     500
                  100
                    0                                                                          0                                                                                               0
                                               0             20000            40000        60000                                                                                                   0               20000         40000        60000
                                                                Load cycles                                                                                                                                           Load cycles


                                          Stress              LVDT M           LVDT (1)        LVDT (2)                                                                                            LVDT M             LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                    Permanent Deformation Rate
              deformation (mm)




                                                    2
                                                                                                                                                                      Rate (mm/million load
                                                                                                                                                                                              10000
                 Permanent




                                                   1.5                                                                                                                       cycles)           1000
                                                    1                                                                                                                                              100
                                                   0.5                                                                                                                                             10
                                                    0                                                                                                                                                  1
                                                         0            20000           40000        60000                                                                                                   0     10000 20000 30000 40000 50000
                                                                        Load cycles                                                                                                                                     Load cycles

                                               LVDT M             LVDT (1)             LVDT (2)                                                                                                    LVDT M             LVDT (1)         LVDT (2)



                                                                Plastic Strain                                                                                                                                 Plastic Strain Rate
                                                                                                                                 Rate (% /million load cycles)




                                                                                                                                                                                       10000
                          Plastic Strain (%)




                                               0.6
                                                                                                                                                                                              1000
                                               0.4
                                                                                                                                                                                              100
                                               0.2
                                                                                                                                                                                               10
                                                    0
                                                         0           20000            40000        60000                                                                                           1
                                                                        Load cycles                                                                                                                    0            20000         40000        60000
                                                                                                                                                                                                                       Load cycles
                                                   LVDT M             LVDT (1)          LVDT (2)                                                                                                    LVDT M             LVDT (1)          LVDT (2)


       Elastic response results                                                                                                  Plastic response results
            Average for test                                                              Final values                                  Final values
                 Elastic     Resilient                                                       Elastic                       Resilient    Permanent      Rate of                                                                    Plastic      Rate of
                 Strain      Modulus                                                         Strain                        Modulus      Deformation       PD                                                                      Strain         PS
              (microstrain)   (Mpa)                                                       (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                   (%)       (%/million)
  LVDT M         3234          489                                                           3224                            492           0.99          2.9                                                                       0.32          1.0
  LVDT #1        1106         1434                                                            972                           1634           1.57          2.5                                                                       0.53          0.9
  LVDT #2        1068         1489                                                           1088                           1459           0.83          2.5                                                                       0.28          0.9
Avg.Lvdt 1&2     1087         1462                                                           1030                           1546           1.13          2.7                                                                       0.38          0.9
Material:                                                Ferricrete                                                   Sample #: FSS40                                                                                        Sample Failed
Density:                                                          2014 kg/ cub m              Confining pressure:                                                                                       144.1 kPa
Moisture:                                                          5.5 %                                    Stress Ratio                                                                                   90 % of failure stress


                                               Applied stress and elastic strain                                                                                                                            Resilient modulus

                  600                                                                             6000                                                                            1500
   Applied stress (kPa)




                  500                                                                             5000




                                                                                                                                                                  Esec (MPa)
                                                                                                           Elastic Strain
                                                                                                           (microstrain)
                  400                                                                             4000                                                                            1000
                  300                                                                             3000
                  200                                                                             2000                                                                                   500
                  100                                                                             1000
                    0                                                                             0                                                                                                0
                                               0             2     4       6        8        10                                                                                                        0        2      4         6         8       10
                                                                  Load cycles                                                                                                                                         Load cycles


                                          Stress                 LVDT M         LVDT (1)        LVDT (2)                                                                                               LVDT M        LVDT (1)         LVDT (2)


                                                          Permanent Deformation                                                                                                                        Permanent Deformation Rate
              deformation (mm)




                                                   0.4
                                                                                                                                                                       Rate (mm/million load
                                                                                                                                                                                               10
                 Permanent




                                                   0.3                                                                                                                        cycles)

                                                   0.2
                                                   0.1
                                                    0                                                                                                                                              1
                                                         0        2        4         6          8           10                                                                                         0     10000   20000   30000    40000     50000
                                                                          Load cycles                                                                                                                                Load cycles

                                               LVDT M                 LVDT (1)           LVDT (2)                                                                                                      LVDT M        LVDT (1)         LVDT (2)



                                                                  Plastic Strain                                                                                                                            Plastic Strain Rate
                                                                                                                                  Rate (% /million load cycles)




                                                                                                                                                                                        10
                          Plastic Strain (%)




                                               0.15

                                                    0.1

                                               0.05

                                                     0
                                                          0        2        4        6          8          10                                                                                  1
                                                                          Load cycles                                                                                                              0            2     4          6         8        10
                                                                                                                                                                                                                     Load cycles
                                                   LVDT M              LVDT (1)           LVDT (2)                                                                                                      LVDT M        LVDT (1)         LVDT (2)


       Elastic response results                                                                                                   Plastic response results
            Average for test                                                               Final values                                  Final values
                 Elastic     Resilient                                                        Elastic                       Resilient    Permanent      Rate of                                                                  Plastic        Rate of
                 Strain      Modulus                                                          Strain                        Modulus      Deformation       PD                                                                    Strain           PS
              (microstrain)   (Mpa)                                                        (microstrain)                     (Mpa)          (mm)      (mm/million)                                                                 (%)         (%/million)
  LVDT M         4794          520                                                           #DIV/0!                           0            0.00          0.0                                                                     0.00            0.0
  LVDT #1        2883          865                                                           #DIV/0!                           0            0.00          0.0                                                                     0.00            0.0
  LVDT #2        2399         1040                                                           #DIV/0!                           0            0.00          0.0                                                                     0.00            0.0
Avg.Lvdt 1&2     2641          953                                                           #DIV/0!                           0            0.00          0.0                                                                     0.00            0.0