Docstoc

Carbohydrate analysis

Document Sample
Carbohydrate analysis Powered By Docstoc
					NSP
Nordic Standardization Programme
a cooperation between




               Anna Jacobs, Christiane Laine, Kristin Syverud




                        Carbohydrate analysis
           Results and recommendations from the NSP
              Network group Carbohydrate analysis




Contract work for: NSP and NI
Performed: 2001-2003
Reference No:


Contact person NI and NSP: Kerstin Ölander, Programme Manager NSP
Contact person NSP-WG: Anna Jacobs, convenor




                                   CONFIDENTIAL
                                                                  Carbohydrate analysis
NSP                                             Available only to the industry in FI, NO, SE
Nordic Standardization Programme
a cooperation between




               Anna Jacobs, Christiane Laine, Kristin Syverud




                        Carbohydrate analysis
           Results and recommendations from the NSP
              Network group Carbohydrate analysis




 Contract work for: NSP and NI
 Performed: 2001-2003
 Reference No:


 Contact person NI and NSP: Kerstin Ölander, Programme Manager NSP
 Contact person NSP-WG: Anna Jacobs, convenor




                                   CONFIDENTIAL
According to STFI's Confidentiality Policy this report is assigned category 1

Results and recommendations from the NSP Network group Carbohydrate analysis




Acknowledgements
This study was financed by the Nordic Standardization Programme (70 %)
and the Nordic Industrial Fund (30 %).

Marianne Björklund Jansson, STFI, Bengt Eriksson, STFI and Anne Reitan,
PFI, are acknowledged for valuable contributions to this work.

In addition to the authors, the following persons participated in the work of
the NSP Network group Carbohydrate analysis:

Anders Uhlin, STFI
Ann-Cristin Bäckman, Kappa
Annki Karlsson, StoraEnso
Asha Ismail Olsson, STFI
Cherryleen Garcia-Lindgren, M-real
Christina Wedin, MoRe
Elisabeth Olsson, SCA
Mia Tehomaa, KCL
Per Isaksson, Södra
Per Lindgren, MoRe
Tiina Hausalo, KCL
                                    According to STFI's Confidentiality Policy this report is assigned category 1

            Results and recommendations from the NSP Network group Carbohydrate analysis




      Contents
                                                                                                     Page


1     Summary                                                                                                 1

2     Background                                                                                              3

3     Materials and methods                                                                                   4
3.1   Carbohydrate analysis                                                                                   4
3.2   Pulp samples                                                                                            5
3.3   Intercalibrations                                                                                       5

4     Results                                                                                                 7
4.1   Intercalibration I; investigation of internal methods                                                   7
4.2   Intercalibration II; further evaluation of the acid hydrolysis
      procedure                                                                                              7
4.3   Intercalibration III; specified parameters for acid hydrolysis                                        12
4.4   Intercalibration IV, enzymatic hydrolysis                                                             15
4.5   Comparison of intercalibrations                                                                       18

5     Conclusions and recommendations                                                                       23
5.1   Sample pre-treatment                                                                                  23
5.2   Depolymerisation (hydrolysis)                                                                         23
5.3   Carbohydrate determination                                                                            26

6     References                                                                                            28

      STFI Database information                                                                             29
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            1




1            Summary
Carbohydrate analysis of wood pulps is normally accomplished in two steps;
the pulp polysaccharides are first hydrolyzed to mono- and oligosaccharides,
which are subsequently determined by some carbohydrate analysis
technique. In order to study these two steps more in detail, the NSP
Network group Carbohydrate analysis has performed a series of inter-
calibration studies during a period of two years. One of these inter-
calibrations showed that the hydrolysis results were indeed different
between the participating laboratories, most likely due to the different
equipments used. Furthermore, it was obvious that part of the variation was
explained by differences in the carbohydrate determination step, e.g.
different calibrations. However, it was not possible to single out the
influence of each step on the variations in the whole two-step procedure.

The intercalibration study showed that a method applicable for all different
equipments taking part was not revealed and therefore standardization of
carbohydrate analysis appears not to be reasonable at this stage. Instead,
the results from the work of the NSP Network group Carbohydrate analysis
have been utilized to establish a list of recommendations regarding the
different steps of carbohydrate analysis. Some important recommendations
are listed here:

- Step 1 of the acid hydrolysis should be performed according to the TAPPI
  standard method (T249 cm-85).
- The hydrolysis parameters of step 2 of the acid hydrolysis must be care-
  fully tested and optimised at each laboratory and for each equipment (e.g.
  autoclave, microwave oven). The actual hydrolysis time and temperature
  should be in focus upon this optimisation.
- With each set of analyses a standard pulp sample should be included, and
  a control card should be used to keep track of the variations in
  carbohydrate composition obtained upon analysis of this standard pulp.
- Intercalibrations should be repeated with a certain periodicity. It might be
  possible to include carbohydrate analysis in the round-robin analyses
  routinely performed within the pulp and paper industry.
- In order to obtain accurate determination of the absolute carbohydrate
  content, the number of parallel samples to be analysed could be increased.
  Series of samples, the absolute carbohydrate contents of which are going to
  be compared, should be analysed at the same time (i.e., at the same
  calibration and hydrolysis conditions).
- The relative carbohydrate composition values are, as expected, much less
  sensitive to variations in instrument calibration than the corresponding
  absolute compositions. Therefore, if the absolute carbohydrate composition
  is uncertain, the relative carbohydrate compositions could still be studied.
    According to STFI's Confidentiality Policy this report is assigned category 1

2   Results and recommendations from the NSP Network group Carbohydrate analysis



    Suggested future studies
    Due to lack of time, only one intercalibration was performed to study
    enzymatic hydrolysis and thus not many conclusions can be drawn.
    However, the variations in both absolute content of carbohydrates and
    content of glucose were smaller than those obtained for most intercalibra-
    tions using acid hydrolysis and consequently, enzymatic hydrolysis seems
    promising for future studies.

    Some laboratories have shown that the use of a laboratory microwave oven
    may be beneficial for acid hydrolysis. In future investigations, more emphasis
    should be put on studies of acid hydrolysis using a microwave oven.
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            3




2            Background
Carbohydrates are the primary constituents of mechanical and chemical
pulps. During pulping and bleaching, the carbohydrate compositions of
chemical pulps are modified to various extent depending on the cooking and
bleaching chemicals used. The method proposed by the Technical
Association of the Pulp and Paper Industry (TAPPI standard no. T249
cm-85) for determination of the carbohydrate composition comprises acid
hydrolysis of the pulp sample followed by derivatization of the mono-
saccharides and subsequent quantification by gas chromatography.
However, since this standard was adopted in 1985, new demands on
carbohydrate analysis have risen, and new analytical instruments for
carbohydrate determination have been developed. The Nordic Standardiza-
tion Programme board (NSP-PB) therefore initiated a NSP Network group
Carbohydrate analysis, consisting of participants from Kappa, KCL, MoRe,
PFI, SCA, Stora Enso, Södra and STFI. The main tasks of this network
group were:

i) to document and evaluate the methods currently used for carbohydrate
   analysis in the pulp and paper industry in Finland, Norway and Sweden.
ii) to investigate whether it would be beneficial to develop standardized
    carbohydrate analysis methods for pulp samples.

Within this study, a series of interlaboratory studies have been performed in
order to compare the methods currently used by the institutes and the
industry. The interlaboratory studies performed were the following:

- Analysis of the carbohydrate compositions of four pulp samples at each
  laboratory using in-house acid hydrolysis procedures and determination
  methods.
- Repeated analyses at each laboratory of one of the pulps in order to deter-
  mine the repeatability and to obtain results less influenced by in-house
  variations.
- Analysis at STFI of acid hydrolyzates from the participating laboratories.
- Analysis at each laboratory of a pulp hydrolyzate prepared by acid hydro-
  lysis at STFI.
- Analysis of the oxygen-delignified softwood kraft pulp and a sample of
  cotton linter using a modified acid hydrolysis protocol.
- Analysis of the oxygen-delignified softwood kraft pulp by enzymatic
  hydrolysis procedure and carbohydrate determination.
    According to STFI's Confidentiality Policy this report is assigned category 1

4   Results and recommendations from the NSP Network group Carbohydrate analysis




    3                        Materials and methods

    3.1                      Carbohydrate analysis
    Carbohydrate analysis of wood pulps is normally accomplished in two steps.
    The pulp polysaccharides are first hydrolyzed to mono- and oligosaccharides,
    which are subsequently determined by some carbohydrate analysis technique,
    for example gas chromatography (after derivatization to alditols) (Sjöström et
    al. 1966; TAPPI 1985; Theander & Westerlund 1986), ion-exchange chromato-
    graphy (Edwards et al. 1987; Hausalo et al. 1995; Wright & Wallis 1996),
    high-performance liquid chromatography (Syverud et al. 1999) or capillary
    zone electrophoresis (after derivatization with a UV-absorbing tag) (El Rassi
    1994; El Rassi 1999; Dahlman et al. 2000).

    The combinations of methods used in this interlaboratory study are listed in
    Table 1. Details concerning the parameters used for hydrolysis of the pulp
    samples are presented in Appendix 1.

        Table 1. Methods used in the interlaboratory studies.

    Method                  Laboratory                 Hydrolysis                   Determination technique
    number
    1                       STFI                       Acid                         Gas Chromatography
    2                       Kappa                      Acid                         Ion Exchange Chromatography
    3                       KCL                        Acid                         Ion Exchange Chromatography
    4                       MoRe                       Acid                         Ion Exchange Chromatography
    5                       PFI                        Acid                         Gas Chromatography
    6                       SCA                        Acid                         Gas Chromatography
    7                       Stora Enso                 Acid                         Ion Exchange Chromatography
    8                       Södra                      Acid                         Gas Chromatography
    9                       PFI                        Enzymatic/Acid               High-Performance Liquid
                                                                                    Chromatography
    10                      STFI                       Enzymatic                    Capillary Zone Electrophoresis
    11                      KCL                        Enzymatic                    Ion Exchange Chromatography
    12                      MoRe                       Enzymatic                    Ion Exchange Chromatography
    13                      Stora Enso                 Enzymatic                    Ion Exchange Chromatography
    14                      PFI                        Enzymatic                    High-Performance Liquid
                                                                                    Chromatography
    15                      Kappa                      Enzymatic                    Ion Exchange Chromatography
    16                      SCA                        Enzymatic                    Gas Chromatography
                                              According to STFI's Confidentiality Policy this report is assigned category 1

                      Results and recommendations from the NSP Network group Carbohydrate analysis                            5




3.2            Pulp samples
Four pulp samples were distributed to the participating laboratories. These
four pulps were:

    Oxygen-delignified hardwood kraft pulp with kappa number 8 from Södra.
    Bleached (TCF) hardwood kraft pulp with kappa number 4 from Södra.
    Oxygen-delignified softwood kraft pulp with kappa number 9 from Östrand
    pulp mill (SCA).
    Softwood thermomechanical pulp (TMP) with kappa number 136 from
    Ortviken (SCA).

The pulp samples were washed until the outlet water was clear, dried at
40°C overnight, and finally ground using an analytical mill.


3.3            Intercalibrations
The NSP Network group has performed a series of interlaboratory studies in
order to compare the methods currently used by the institutes and the
industry. The series of interlaboratory studies performed by the NSP
Network group Carbohydrate analysis are summarized in Table 2.

    Table 2. Summary of the interlaboratory studies performed within the NSP
    Network group.
Intercalibration             Samples                         Hydrolysis                      Determination
I                    All pulp samplesa                   Acid, in-house                        In-house
IIa                  Softwood kraft pulp                 Acid, in-house                        In-house
IIb                  Softwood kraft pulp                 Acid, STFI                            In-house
IIc                  Softwood kraft pulp                 Acid, in-house                        STFI
IIIa                 Softwood kraft pulp                 Acid, modified                        In-house
IIIb                 Cotton linter                       Acid, modified                        In-house
IV                   Softwood kraft pulp                 Enzymatic                             In-house
a) Four pulp samples, which are described in section 3.2.

In the first study, the carbohydrate compositions of all of the four pulp
samples described above were analysed at each laboratory using their in-
house acid hydrolysis procedure and determination method.

In a second interlaboratory study, one of the pulps (an oxygen-delignified
softwood kraft pulp) was subjected to repeated analyses in order to determine
the repeatability in each laboratory and to obtain average analysis results
less influenced by in-house variations. In addition, each laboratory has pre-
ared a pulp hydrolyzate from oxygen-delignified softwood kraft pulp, using
their in-house acid hydrolysis procedure, which was analysed at STFI. Each
laboratory has also analysed the carbohydrate composition of an oxygen-
delignified softwood kraft pulp hydrolyzate prepared by acid hydrolysis at
STFI.
    According to STFI's Confidentiality Policy this report is assigned category 1

6   Results and recommendations from the NSP Network group Carbohydrate analysis




    Finally, a modified protocol for acid hydrolysis was set up by the NSP
    Network group and used for analysis of the oxygen-delignified softwood
    kraft pulp and a sample of cotton linter.

    In addition to the studies on acid hydrolysis, the oxygen-delignified softwood
    kraft pulp was also subjected to repeated analyses using in-house enzymatic
    hydrolysis procedures and determination methods at the different labora-
    tories. In this intercalibration, the laboratories normally not performing
    enzymatic hydrolysis used the method developed at KCL.
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            7




4            Results
In this report we have used comma (not a dot) as decimal sign as it is the
directives within international standardization.

4.1          Intercalibration I; investigation of internal methods
The carbohydrate compositions of the four pulp samples were analysed at
each laboratory using their in-house acid hydrolysis procedure and
determination method. For details regarding this study and the results
thereof, see Appendix 1 (Report from the initial studies of the NSP Network
group Carbohydrate analysis).

The main conclusions from this initial intercalibration were the following:

- There must be differences caused by the hydrolysis procedures at the
  different laboratories.
- It was not possible to find any correlation between the results from the
  carbohydrate analyses and the analysis methods employed. The lack of
  such correlation is most likely due to the fact that the differences in the
  results from the participating laboratories are of the same magnitude as
  the standard deviation of the individual analysis methods.
- It was suggested that in order to further investigate these correlations
  further experiments must be included in the interlaboratory study and
  each laboratory should perform the whole analysis procedure (pre-
  treatment, hydrolysis and carbohydrate determination) for one pulp in a
  series of parallel experiments (i.e. 6-8 replicates).

4.2          Intercalibration II; further evaluation of the acid
             hydrolysis procedure
As concluded from intercalibration I, one of the pulps was subjected to
further intercalibration studies with focus on acid hydrolysis. The purposes
of this further investigation, intercalibration II, were:

1) to find out whether the variations between the participating laboratories
   were caused by differences in the acid hydrolysis step or in the carbo-
   hydrate determinations.
2) to investigate the correlations between the parameters used for acid
   hydrolysis and the result of the carbohydrate analysis.

The oxygen-delignified softwood kraft pulp was chosen for this purpose
because it was the only pulp studied that contained all monosaccharide
components included in the intercalibration (i.e., glucose, mannose, xylose,
arabinose, galactose, 4-O-methylglucuronic acid and hexenuronic acid).
    According to STFI's Confidentiality Policy this report is assigned category 1

8   Results and recommendations from the NSP Network group Carbohydrate analysis



    Intercalibration IIa
    The complete analysis procedure, including pre-treatment, acid hydrolysis
    and carbohydrate determination, was repeated in 5-6 parallel experiments
    in each laboratory. The average analysis result from each laboratory was
    calculated and summarized in Table 3 (absolute carbohydrate compositions)
    and Table 5 (relative carbohydrate compositions). The corresponding
    relative standard deviations obtained at each laboratory for determination
    of each monosaccharide are summarized in Table 4 and 6.

        Table 3. Results from intercalibration IIa; absolute carbohydrate composition
        announced as anhydrosugars of oxygen-delignified softwood kraft pulp (average
        values of 5-6 analyses from each method / laboratory).

                                                     Carbohydrate composition (% of dry weight)
    Methoda               Arabinose             Galactose              Mannose      Xylose   Glucose   Total
    1                           0,9                   0,3                   6,4      7,7      78,6     93,8
    2                           0,8                   0,4                   5,7      7,5      85,6     100,0
    3                           0,7                   0,3                   6,0      7,5      72,4     86,9
    4                           0,7                   0,3                   6,0      7,5      76,5     90,9
    5                           0,8                  n.a.                   6,5      8,0      76,3     91,7
    6                           0,8                   0,7                   6,3      8,4      80,7     96,9
    7                           0,7                   0,3                   6,5      7,9      78,2     93,6
    8                           0,9                   0,3                   6,1      7,1      75,0     89,4
    Average                     0,8                   0,3                   6,2      7,7      77,9     92,9

    RSD (%)                    11,3                  49,9                   4,6      5,1       5,1      4,5

    a) See method descriptions in Table 1.
    n.a. = not analysed


        Table 4. Relative standard deviations obtained at each laboratory upon 5-6
        analyses of the absolute carbohydrate composition of oxygen-delignified softwood
        kraft pulp.
                                                        Relative standard deviation (%)
    Methoda               Arabinose             Galactose   Mannose        Xylose       Glucose        Total
    1                           6,5                  36,4                   0,6       1,9      1,1      1,0
    2                          18,6                  10,2                   2,5       2,5      3,2      3,0
    3                           6,4                  11,7                  10,1      12,2      4,6      4,2
    4                           3,5                   3,9                   2,7       2,6      1,5      1,3
    5                          25.6                  n.a.                   8.6      19.5      1.7      3.7
    6                           9,6                  22,4                   2,5       2,4      1,6      1,5
    7                           2,2                   4,8                   1,5       1,8      1,0      1,1
    8                           5,8                  24,1                   2,5       2,8      3,4      3,2
    a) See method descriptions in Table 1.
    n.a. = not applicable
                                               According to STFI's Confidentiality Policy this report is assigned category 1

                       Results and recommendations from the NSP Network group Carbohydrate analysis                            9




    Table 5. Results from intercalibration IIa; relative carbohydrate composition of
    oxygen-delignified softwood kraft pulp (average values of 5-6 analyses from each
    method / laboratory).
                 Relative carbohydrate composition (% of carbohydrates)
Methoda       Arabinose Galactose       Mannose      Xylose      Glucose
1                0,9           0,4            6,8                 8,4                  83,6
2                0,8           0,4            5,7                 7,5                  85,6
3                0,8           0,3            7,0                 8,6                  83,2
4                0,9           0,5            6,7                 8,4                  83,5
5                0,8           0,1            6,9                 8,2                  84,0
6                0,9           0,7            6,5                 8,7                  83,3
7                0,7           0,3            7,0                 8,6                  83,4
8                1,0           0,4            6,8                 8,0                  83,9
Average          0,8           0,4            6,7                 8,4                  83,7
RSD (%)          9,8          51,3            7,1                 5,5                   1,0
a) See method descriptions in Table 1.



Table 6. Relative standard deviations obtained at each laboratory upon 5-6
analyses of the relative carbohydrate composition of oxygen-delignified
softwood kraft pulp.
                             Relative standard deviation (%)
Methoda       Arabinose    Galactose    Mannose        Xylose                       Glucose
1               6,7       36,6                1,2                 1,1                   0,2
2               0,1       0,04                0,1                0,2                    0,2
3               5,4       12,0                7,8                12,3                   1,8
4               3,5        2,9                3,0                3,0                    0,4
5              22,7        n.a.               5,4                16,2                   2,1
6              10,1       22,4                2,3                 1,6                   0,3
7               1,9        4,5                1,5                0,8                    0,1
8               5,6       23,3                1,0                 2,7                   0,4
a) See method descriptions in Table 1.
n.a.= not applicable
Intercalibration IIb
In order to compare the carbohydrate determination procedures used at the
different laboratories, a hydrolyzate of oxygen-delignified softwood kraft
pulp was prepared at STFI. Aliquots of this hydrolyzate were then subjected
to carbohydrate analysis at the participating laboratories.

As shown in Table 7 the absolute carbohydrate contents determined in most
laboratories were lower than those obtained upon analysis at STFI (lab. no.
1). As we cannot explain this apparent degradation of the samples, these
results should be considered with caution.
     According to STFI's Confidentiality Policy this report is assigned category 1

10   Results and recommendations from the NSP Network group Carbohydrate analysis



         Table 7. Results from intercalibration IIb; absolute carbohydrate composition
         announced as anhydrosugars of an oxygen-delignified softwood kraft pulp
         hydrolyzate prepared at STFI and analysed at the participating laboratories.
                                                   Carbohydrate composition (% of dry weight)
     Lab. Noa              Arabinose             Galactose  Mannose       Xylose       Glucose     Total
     1                          0,8                   0,4                    6,2     7,4    76,5   91,3
     2                          0,6                   0,3                    6,2     6,9    74,7   88,7
     3                          0,6                   0,3                    5,9     6,0    68,9   81,7
     4                          0,6                   0,3                    5,8     6,6    73,9   87,2
     5                          0,5                   n.d.                   5,3     6,5    69,6   81,9
     6                          0,6                   0,5                    5,8     5,6    74,0   86,5
     7                          0,6                   0,3                    6,3     7,2    79,2   93,6
     8                          n.d.                  n.d.                   n.d.    n.d.   n.d.   n.d.
     Average         0,6          0,3         5,9         6,6         73,8       87,3
     RSD (%)        14,8         28,2         6,0         9,7          4,9        5,1
     a) See method descriptions in Table 1. Prehydrolysis of the standards was per-
     formed in method 1, 2, 5, 6, 7, and 8 but not in method 3 and 4.
     n.d.= not determined



         Table 8. Results from intercalibration IIb; relative carbohydrate composition of
         an oxygen-delignified softwood kraft pulp hydrolyzate prepared at STFI and
         analysed at the participating laboratories.
                              Relative carbohydrate composition (% of carbohydrates)
     Lab. Noa              Arabinose Galactose       Mannose      Xylose      Glucose
     1                          0,9                   0,4                    6,8     8,1    83,8
     2                          0,7                   0,3                    7,0     7,8    84,2
     3                          0,8                   0,3                    7,2     7,3    84,4
     4                          0,7                   0,3                    6,7     7,6    84,7
     5                          0,6                   n.d.                   6,5     7,9    85,0
     6                          0,6                   0,6                    6,7     6,5    85,6
     7                          0,7                   0,3                    6,8     7,7    84,6
     8                          n.d.                  n.d.                   n.d.    n.d.   n.d.
     Average         0,7        0,4         6,8                                      7,6    84,6
     RSD (%)        12,2       28,6         3,3                                      7,1     0,7
     a) See method descriptions in Table 1.
     n.d.= not determined

     Intercalibration IIc
     In order to compare the effects of the different parameters for acid hydroysis
     used at the different laboratories, each laboratory prepared a hydrolyzate
     that was analysed at STFI.

     Unfortunately, the transport of one of the samples (from laboratory no. 5) was
     delayed. Upon analysis at STFI, this sample demonstrated a lower carbo-
     hydrate content, primarily of xylose. It is most likely that the sample had
     been damaged by the prolonged transport, and the results are therefore not
     included in the calculation of average carbohydrate content and relative
     standard deviation of the intercalibration.
                                              According to STFI's Confidentiality Policy this report is assigned category 1

                      Results and recommendations from the NSP Network group Carbohydrate analysis                            11



 Table 9. Results from intercalibration IIc; absolute carbohydrate composition
 announced as anhydrosugars of an oxygen-delignified softwood kraft pulp
 hydrolyzate prepared at the participating laboratories and analysed at STFI.
                            Carbohydrate composition (% of dry weight)
Lab. Noa    Arabinose     Galactose  Mannose       Xylose       Glucose                                     Total
1              0,8            0,4            6,5                 7,8                  79,1                  94,6
2              0,9            0,5            6,6                 7,8                  81,5                  97,3
3              0,9            0,4            6,4                 8,3                  77,0                  93,0
4              0,8            0,5            6,1                 7,7                  77,4                  92,5
5              0,8b           0,3b           5,6 b               6,1b                 77,8 b                90,6 b
6              0,9            0,4            6,3                 8,0                  78,8                  94,4
7              0,9            0,4            6,3                 8,5                  78,9                  95,0
8              0,9            0,4            6,3                 7,6                  77,0                  92,2
Average        0,9         0,4         6,4        8,0         78,5         94,1
RSD (%)        5,6        11,4         2,5        4,2          2,0          1,9
a) See method descriptions in Table 1.
b) The sample was degraded during transport and is not included in the calculation
of average and RSD values.


 Table 10. Results from intercalibration IIc; relative carbohydrate composition of
 an oxygen-delignified softwood kraft pulp hydrolyzate prepared at the partici-
 pating laboratories and analysed at STFI.
               Relative carbohydrate composition (% of carbohydrates)
Lab. Noa    Arabinose Galactose       Mannose      Xylose      Glucose
1              0,8            0,4            6,9                 8,2                  83,6
2              0,9            0,5            6,8                 8,0                  83,8
3              1,0            0,4            6,9                 8,9                  82,8
4              0,9            0,5            6,6                 8,3                  83,7
5             0,9 b           0,3 b         6,2 b                6,7 b               85,9 b
6              1,0            0,4            6,7                 8,5                  83,5
7              0,9            0,4            6,6                 8,9                  83,1
8             1,0          0,4         6,8                       8,2                  83,5
Average        0,9         0,5         6,8                       8,5                  83,4
RSD (%)        5,5        11,0         1,8                       4,2                   0,4
a) See method descriptions in Table 1.
b) The sample was degraded during transport and is not included in the
calculation of average and RSD values.
Conclusions from intercalibration II
From intercalibration II the following conclusions could be drawn:

- When the whole carbohydrate analysis procedure was repeated 5-6 times
  the relative standard deviations between the average absolute values for
  glucose, xylose and mannose between laboratories were larger than the
  relative standard deviations obtained within each laboratory, as expected.
  This means that there are differences in the procedures utilized at the
  different laboratories. It was, however, not possible to find any correlations
  between the parameters used for hydrolysis and the results obtained.
     According to STFI's Confidentiality Policy this report is assigned category 1

12   Results and recommendations from the NSP Network group Carbohydrate analysis




     - In the case of arabinose and galactose, the variations within each labora-
       tory were in many cases larger or of the same magnitude as the variations
       between laboratories.
     - In this intercalibration, the studies requiring that acid hydrolyzates were
       transported by mail suffered from some logistical problems. In some cases
       it can be suspected that the acid hydrolyzates were degraded during the
       transport to or from STFI. However, it was obvious that the relative
       standard deviations obtained upon analysis of different acid hydrolyzates
       at STFI were lower than the relative standard deviation between labora-
       tories, indicating that some of the variation is explained by the differences
       in the carbohydrate determination step, e.g. different equipments and
       calibrations.

     4.3                      Intercalibration III; specified parameters for acid
                              hydrolysis
     In order to try out a hydrolysis procedure to be used at all participating
     laboratories, the NSP Network group set up a common protocol for acid
     hydrolysis. In each laboratory, the new hydrolysis procedure was utilized for
     hydrolysis of oxygen-delignified softwood kraft pulp and pure cotton linter.
     The carbohydrate compositions of these samples were subsequently deter-
     mined using the in-house technique. The parameters for acid hydrolysis
     were the following:

     Sample weight: 100 mg
     First hydrolysis step: 3 ml 72% sulphuric acid, 30°C, 60 minutes
     Second hydrolysis step: Add 84 ml distilled water, 125°C, 20 minutes

     Intercalibration IIIa; analysis of oxygen-delignified softwood kraft pulp
     The complete analysis procedure, including pre-treatment, acid hydrolysis
     and carbohydrate determination, was repeated in 3 parallel experiments in
     each laboratory. The average analysis result from each laboratory was
     calculated and the results from all laboratories summarized in Table 11
     (absolute carbohydrate compositions) and Table 13 (relative carbohydrate
     compositions). The corresponding relative standard deviations obtained at
     each laboratory for determination of each monosaccharide are summarized
     in Table 12 and 14.
                                               According to STFI's Confidentiality Policy this report is assigned category 1

                       Results and recommendations from the NSP Network group Carbohydrate analysis                            13




    Table 11. Results from intercalibration III; absolute carbohydrate composition
    announced as anhydrosugars of oxygen-delignified softwood kraft pulp (average
    values of 3 analyses from each method/laboratory).
                              Carbohydrate composition (% of dry weight)
Methoda       Arabinose     Galactose   Mannose      Xylose       Glucose                                    Total
1             0,9          0,3          5,3         7,3         77,7       91,8
2             0,8          0,4          5,6         6,7         74,7       88,2
3             0,8          0,3          5,7         7,4         73,0       87,2
4             0,7          0,3          5,9         7,1         75,6       89,5
5             0,8          0,1          6,5         8,5         83,8       99,4
6             0,8          0,7          7,2         8,6         81,0       99,2
7             0,7          0,3          6,5         7,7         79,2       94,4
8             0,9          0,3          6,6         8,4         86,3       102,7
Average       0,8          0,3          6,2         7,7         78,9       94,1
RSD (%)      10,5         51,1         10,4         9,3          5,8        6,2
a) See method descriptions in Table 1. Prehydrolysis of the standards was per-
formed in method 1, 5, 6, and 7 but not in method 2, 3, 4 and 8.

    Table 12. Relative standard deviations obtained at each laboratory upon 3 ana-
    lyses of the absolute carbohydrate composition of oxygen-delignified softwood
    kraft pulp.
                                   Relative standard deviation (%)
Methoda       Arabinose    Galactose   Mannose        Xylose       Glucose                                   Total
1                2,7           3,7            2,7                 1,9                   1,5                   2,7
2                5,8           7,4            4,4                 5,8                   2,9                   5,8
3                7,9          15,7            3,0                 2,1                   2,2                   7,9
4                3,9           3,4            1,6                 5,3                   1,7                   3,9
5                2,8           n.a.           1,1                 1,9                   0,8                   2,8
6               11,3           8,8            6,0                 2,9                   1,9                  11,3
7                2,5           8,8            0,3                 0,5                   0,9                   2,5
8                2,2           6,0            2,0                 0,6                   0,1                   2,2
a) See method descriptions in Table 1.
n.a.= not applicable
     According to STFI's Confidentiality Policy this report is assigned category 1

14   Results and recommendations from the NSP Network group Carbohydrate analysis




         Table 13. Results from intercalibration III; relative carbohydrate composition of
         oxygen-delignified softwood kraft pulp (average values of 3 analyses from each
         method/laboratory).
                              Relative carbohydrate composition (% of carbohydrates)
     Methoda               Arabinose Galactose       Mannose      Xylose      Glucose
     1                           0,7                   0,4                   5,8     8,1     85,0
     2                           0,7                   0,7                   6,3     7,6     84,7
     3                           0,9                   0,4                   6,5     8,7     83,6
     4                           0,8                   0,3                   6,6     8,1     84,3
     5                           0,8                  n.a.                   6,5     8,7     84,0
     6                           0,9                   1,5                   7,2     8,9     81,5
     7                           0,7                   0,3                   6,9     8,1     83,9
     8                           0,9                   0,4                   6,4     8,2     84,0
     Average                     0,8                   0,6                   6,5     8,3     83,9
     RSD (%)                    13,1                 71,1                    6,3     5,2      1,3
     a) See method descriptions in Table 1.
     n.a.= not applicable

         Table 14. Relative standard deviations obtained at each laboratory upon 3
         analyses of the relative carbohydrate composition of oxygen-delignified softwood
         kraft pulp.
                                                   Relative standard deviation (%)
       Methoda             Arabinose             Galactose    Mannose        Xylose         Glucose
             1                   1,2                   3,5                   1,8     0,3      0,1
             2                  n.a.                  n.a.                  n.a.     n.a.    n.a.
             3                   4,9                 15,1                    4,0     0,7      0,2
             4                   5,4                   2,1                   0,3     3,4      0,3
             5                   2,7                  n.a.                   0,4     1,9      0,2
             6                  11,8                   8,3                   7,0     2,6      0,8
             7                   2,2                   8,2                   0,8     0,5      0,1
             8                   2,5                   6,2                   1,8     0,7      0,1
     a) See method descriptions in Table 1.
     n.a.= not applicable

     Cotton linter
     The modified hydrolysis procedure was also used on cotton linter in 2
     parallel experiments in each laboratory. The average analysis result from
     each laboratory and the corresponding relative standard deviations are
     summarized in Table 15.
                                            According to STFI's Confidentiality Policy this report is assigned category 1

                    Results and recommendations from the NSP Network group Carbohydrate analysis                            15




 Table 15. Results from intercalibration III; absolute carbohydrate composition,
 announced as anhydrosugars, of cotton linter (average values of 2 analyses from
 each method/laboratory).
Methoda       Carbohydrate composition           RSD Glucose
                  (% of dry weight)
                Glucose         Total                     (%)
1                  97,7           97,9                    1,1
2                  87,5           87,5                    1,3
3                  86,3           86,3                    5,2
4                  93,9           93,9                   n.a.
5                 101,4          101,4                   n.a.
6                  95,9           97,0                   n.a.
7                  95,4           95,4                    0,6
8                 103,8          104,4                    4,0
Average            95,2           95,5                   n.a.
RSD (%)              6,4        6,5                      n.a.
a) See method descriptions in Table 1.
n.a.= not applicable

In addition to glucose, some laboratories detected minor amounts (typically
less than 0,5 % of the dry weight) of xylose (method 1 and 8), arabinose
(method 8), galactose (method 6) and mannose (method 6).

Conclusions from intercalibration III
In some cases, low yield of glucose can partially be explained by the fact that
the analyzing laboratories not subjected their monosaccharide standards to
pre-hydrolysis. Another explanation to the variations in glucose yield is that
the hydrolysis equipments used at different laboratories differ, and that
some of them may not be working optimally at the specified conditions. The
conclusions from intercalibration III are:

- The hydrolysis parameters must be carefully tested and optimised at each
  laboratory.
- The effect of pre-hydrolysis of the monosaccharide standards must be
  evaluated at each laboratory.

4.4          Intercalibration IV, enzymatic hydrolysis
Acid hydrolysis for the depolymerisation of pulp polysaccharides has certain
drawbacks. Acid hydrolysis releases the uronic acid residues present to only a
limited extent, whereas the hexenuronic acid residues are completely de-
stroyed (Lindberg et al. 1973, Teleman et al 1996). In addition, losses of
liberated monosaccharides occur in the course of the hydrolysis due to decom-
position in the harsh conditions (Saeman et al. 1954, Sjöström et al. 1966,
Biermann 1988). An alternative method for depolymerisation of pulp samples
is enzymatic hydrolysis. This method enables also the determination of uronic
acids, including hexenuronic acid. Most of the participating laboratories have
in-house methods for enzymatic hydrolysis. The details regarding these
     According to STFI's Confidentiality Policy this report is assigned category 1

16   Results and recommendations from the NSP Network group Carbohydrate analysis




     enzymatic hydrolysis methods are summarized in Appendix 2. In inter-
     calibration IV, laboratories not having an in-house method for enzymatic
     hydrolysis used the procedure developed at KCL.

     Enzymatic hydrolysis has shown to be applicable for chemical pulp samples
     with kappa numbers up to at least 70 (Dahlman et al 2000). In the present
     intercalibration, repeated analyses of the oxygen-delignified softwood kraft
     pulp with kappa number 9 was performed.

     The complete analysis procedure, including enzymatic hydrolysis and
     carbohydrate determination, was repeated in 5-6 parallel experiments in
     each laboratory. The average analysis result from each laboratory was
     calculated and the results from all laboratories summarized in Table 16
     (absolute carbohydrate composition). The relative carbohydrate com-
     positions are presented in Table 18. The corresponding relative standard
     deviations obtained at each laboratory for determination of each mono-
     saccharide are summarized in Tables 17 and 19.

         Table 16. Results from intercalibration IV; absolute carbohydrate composition,
         announced as anhydrosugars, of oxygen-delignified softwood kraft pulp (average
         values of 5-6 analyses from each method / laboratory).
                                    Carbohydrate composition (% of dry weight)
       Methoda             Arabinose Galactose Mannose Xylose         Glucose                                    Total
        10                       0,7                 0,3                 7,6         9,0           78,7          97,0
        11                      n.a.                 n.a.               n.a.         n.a.          n.a.          n.a.
        12                       0,7                 0,3                 7,1         7,7           79,5          95,3
        13                       0,7                 0,3                 6,8         7,7           83,0          98,6
        14                       1,2                 0,8                 6,1         8,2           77,0          93,2
        15                       0,7                 0,3                 4,0         5,8           79,0          89,6
        16                       0,7                 0,3                 5,0         6,9           75,8          88,8
      Average                    0,8                 0,4                 6,1         7,6           78,8          93,7
      RSD (%)                   24,0                52,3                22,5         14,8          3,1           4,2
     a) See method descriptions in Table 1.
     n.a. = not analyzed
         Table 17. Relative standard deviations obtained at each laboratory upon 5-6
         analyses of the absolute carbohydrate composition of oxygen-delignified softwood
         kraft pulp.
                                                         Relative standard deviation (%)
     Methoda               Arabinose             Galactose   Mannose        Xylose       Glucose                         Total
     10                         9,5                   13,4                   6,7            2,2           2,2            2,2
     11                         n.a.                  n.a.                   n.a.           n.a.          n.a.           n.a.
     12                          3,3                   9,8                   1,4            0,4           0,6             0,4
     13                          3,4                   8,6                   1,9            1,4           0,7             0,8
     14                         17,3                   9,6                   3,3            2,0           1,4             1,2
     15                         n.a.                  n.a.                  n.a.            n.a.          n.a.           n.a.
     16              6,2       16,0         7,1                                             2,4           5,7             5,1
     a) See method descriptions in Table 1.
     n.a.= not applicable
                                                According to STFI's Confidentiality Policy this report is assigned category 1

                        Results and recommendations from the NSP Network group Carbohydrate analysis                            17



  Table 18. Results from intercalibration IV; relative carbohydrate composition of
  oxygen-delignified softwood kraft pulp (average values of 5-6 analyses from each
  method/laboratory).
                Relative carbohydrate composition (% of carbohydrates)
Methoda      Arabinose Galactose       Mannose      Xylose      Glucose
10              0,8            0,3            7,8                 9,3                   81,2
11              n.a.           n.a.           n.a.                n.a.                  n.a.
12              0,7            0,3            7,5                 8,1                   83,3
13              0,7            0,3            6,9                 8,0                   84,1
14              0,8            1,3            6,6                 8,8                   82,5
15              1,2            0,8            6,4                 8,8                   82,9
16              0,8            0,3            5,7                 7,8                   85,4
Average          0,8            0,5            6,8                8,5                   83,2
RSD (%)         22,8           71,4           11,4                6,8                    1,7
a) See method descriptions in Table 1.



  Table 19. Relative standard deviations obtained at each laboratory upon 5-6
  analyses of the relative carbohydrate composition of oxygen-delignified softwood
  kraft pulp.
                              Relative standard deviation (%)
Methoda      Arabinose     Galactose    Mannose      Xylose                       Glucose
10              8,7            12,6           6,2                  1,0                  0,5
11              n.a.           n.a.           n.a.                 n.a.                 n.a.
12               2,8            8,3            1,0                 0,3                   0,1
13               3,6            6,8            1,0                 4,6                   0,4
14              9,7            17,7           2,9                  1,8                  0,5
15              n.a.           n.a.           n.a.                 n.a.                 n.a.
16              12,0           18,3            3,2                 7,0                   0,8
a) See method descriptions in Table 1.
n.a.= not applicable

In addition to neutral carbohydrates, the amount of hexenuronic acid in the
oxygen delignified hardwood and softwood pulp samples was also analysed.
The results from the hexenuronic acid determinations are summarized in
Table 20.

Table 20. Results from intercalibration IV; determination of hexenuronic acid in
oxygen delignified hardwood and softwood kraft pulp

              Hexenuronic acid content,        Relative standard deviation,
                    (mmol/kg pulp)                          (%)
 Methoda       Softwood      Hardwood           Softwood        Hardwood
               kraft pulp    kraft pulp         kraft pulp      kraft pulp
     10          38,0             n.a.               6,3                      n.a.
     11          31,0             49,2               2,0                      0,7
     13          32,0             49,3               0,8                      0,9
a) See method descriptions in Table 1. N.a.= not analysed.
     According to STFI's Confidentiality Policy this report is assigned category 1

18   Results and recommendations from the NSP Network group Carbohydrate analysis




     4.5                                                 Comparison of intercalibrations
     For comparison, the results from the intercalibrations were summarized in
     four diagrams, displaying the total carbohydrate content (Figure 1) as well
     as the total and relative content of glucose (Figure 2), mannose (Figure 3)
     and xylose (Figure 4) determined in each of the participating laboratories. In
     each diagram, the symbols correspond to the number of the method used for
     the carbohydrate analysis, according to Table 2. As pointed out earlier, in
     intercalibration IIb the hydrolyzate aliquots sent out from STFI had been
     partially degraded during the transport to the participating laboratories,
     and are consequently left out in the comparison.


                                                 105.0



                                                 100.0                                                1
                                                                                                      2
            Carbohydr. cont. (% of dry weight)




                                                                                                      3
                                                                                                      4
                                                  95.0
                                                                                                      5
                                                                                                      6
                                                                                                      7
                                                  90.0                                                8
                                                                                                      14
                                                                                                      10
                                                                                                      11
                                                  85.0
                                                                                                      12
                                                                                                      13
                                                                                                      16
                                                  80.0                                                av.




                                                  75.0
                                                            I        IIa             IIc   III   IV


         Figure 1. Total carbohydrate content of oxygen-delignified softwood kraft pulp
         determined in the intercalibrations. Intercalibration I = in –house acid
         hydrolysis, in-house determination. Intercalibration IIa = in –house acid
         hydrolysis, in-house determination, average of 3 replicates. Intercalibration
         Intercalibration IIc = in –house acid hydrolysis, determination at STFI.
         Intercalibration III = modified acid hydrolysis, in-house determination.
         Intercalibration IV = in –house enzymatic hydrolysis, in-house determination.
         The symbols correspond to the method numbers which are explained in Table 2.
         The average value for each intercalibration is represented by “•”


     As shown in Figure 1, the value for the total carbohydrate content varied
     quite extensively (i.e. between 82% and 102% of the dry weight) between the
     different determinations. However, the average value for the total carbo-
     hydrate content (•) was quite similar in all the intercalibrations. Since
     approximately 80% of the pulp sample consisted of glucose, the result for
     glucose in the different intercalibrations (Figure 2) resembles the results for
     total carbohydrate content quite closely. The values obtained for glucose in
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            19



the different intercalibrations varied between 73 and 87 % of the dry weight.
In contrast, the relative content of glucose (% of the carbohydrates in the
sample) showed much smaller variations.

As expected, the variations were reduced when the analyses were performed
5-6 times (Intercalibration IIb) compared to the initial intercalibration (I) in
which two parallel analyses were performed.

The smallest variations in both absolute carbohydrate content (Figure 1)
and absolute content of glucose (Figure 2) were obtained when samples
hydrolysed at the participating laboratories were analysed at STFI
(Intercalibration IIc), which indicates that the end determination step,
including the calibration of the analysis instrument, is vital for the result of
the analysis.

In contrary to the intentions of the NSP Network group, analysis using the
same conditions for acid hydrolysis (intercalibration III) gave only slightly
improved variations compared to the initial intercalibration.

Five of the intercalibrations involved acid hydrolysis whereas, due to lack of
time, only one intercalibration was performed to study enzymatic hydrolysis
(intercalibration IV). Since only one intercalibration has been performed not
many conclusion can be drawn. However, the variations in both absolute
content of carbohydrates and content of glucose were smaller than those
obtained for most intercalibrations using acid hydrolysis and consequently,
enzymatic hydrolysis seems promising for future studies.
     According to STFI's Confidentiality Policy this report is assigned category 1

20   Results and recommendations from the NSP Network group Carbohydrate analysis




                                               90.0                                                                  90.0



                                               85.0                                                                  85.0                              1
                                                                                                                                                       2
          Carbohydr. cont. (% of dry weight)




                                                                                 Relative carbohydrate content (%)
                                                                                                                                                       3
                                               80.0                                                                                                    4
                                                                                                                     80.0
                                                                                                                                                       5
                                                                                                                                                       6
                                                                                                                                                       7
                                               75.0                                                                  75.0                              8
                                                                                                                                                       9
                                                                                                                                                       10
                                               70.0                                                                  70.0                              11
                                                                                                                                                       12
                                                                                                                                                       13
                                                                                                                                                       14
                                               65.0                                                                  65.0                              av.



                                               60.0                                                                  60.0
                                                      I   IIa   IIc   III   IV                                              I   IIa   IIc   III   IV


         Figure 2. Total and relative content of glucose in oxygen-delignified softwood
         kraft pulp determined in the intercalibrations (see explanation of intercali-
         brations in connection with Figure 1). The symbols correspond to the method
         numbers which are explained in Table 2. The average value for each inter-
         calibration is represented by “•”.


     Xylose (Figure 3) and mannose (Figure 4) comprise approximately 8% and
     6%, respectively, of the dry weight of the pulp sample. Consequently, the
     difference in variation between the absolute and relative content of xylose
     and mannose are less pronounced than in the case of the glucose deter-
     minations. However, as expected, the variations are in all cases smaller for
     the relative content of xylose and mannose than are the corresponding
     variations for the absolute content.
                                                                                              According to STFI's Confidentiality Policy this report is assigned category 1

                                                      Results and recommendations from the NSP Network group Carbohydrate analysis                                            21




                                     10.0                                                                        10.0

                                      9.5                                                                         9.5
                                                                                                                                                                1
                                      9.0                                                                         9.0                                           2
Carbohydr. cont. (% of dry weight)




                                                                             Relative carbohydrate content (%)
                                                                                                                                                                3
                                      8.5                                                                         8.5                                           4
                                                                                                                                                                5
                                      8.0                                                                         8.0                                           6
                                                                                                                                                                7
                                      7.5                                                                         7.5                                           8
                                                                                                                                                                9
                                      7.0                                                                         7.0                                           10
                                                                                                                                                                11
                                                                                                                                                                12
                                      6.5                                                                         6.5
                                                                                                                                                                13
                                                                                                                                                                14
                                      6.0                                                                         6.0
                                                                                                                                                                av.

                                      5.5                                                                         5.5

                                      5.0                                                                         5.0
                                            I   IIa    IIc   III    IV                                                  I   IIa   IIc      III       IV


           Figure 3. Total and relative content of xylose in oxygen-delignified softwood
           kraft pulp determined in the intercalibrations (see explanation of intercali-
           brations in connection with Figure 1). The symbols correspond to the method
           numbers which are explained in Table 2. The average value for each inter-
           calibration is represented by “•”.
     According to STFI's Confidentiality Policy this report is assigned category 1

22   Results and recommendations from the NSP Network group Carbohydrate analysis




                                              8.0                                                                  8.0


                                              7.5                                                                  7.5

                                                                                                                                                    1
         Carbohydr. cont. (% of dry weight)




                                                                               Relative carbohydrate content (%)
                                              7.0                                                                  7.0                              2
                                                                                                                                                    3
                                                                                                                                                    4
                                              6.5                                                                  6.5
                                                                                                                                                    5
                                                                                                                                                    6
                                              6.0                                                                  6.0                              7
                                                                                                                                                    8
                                                                                                                                                    9
                                              5.5                                                                  5.5                              10
                                                                                                                                                    11
                                                                                                                                                    12
                                              5.0                                                                  5.0
                                                                                                                                                    13
                                                                                                                                                    14
                                              4.5                                                                  4.5                              av.



                                              4.0                                                                  4.0
                                                    I   IIa   IIc   III   IV                                             I   IIa   IIc   III   IV


         Figure 4. Total and relative content of mannose in oxygen-delignified softwood
         kraft pulp determined in the intercalibrations (see explanation of intercali-
         brations in connection with Figure 1). The symbols correspond to the method
         numbers which are explained in Table 2. The average value for each inter-
         calibration is represented by “•”.



     Conclusions from comparison of the intercalibrations
     - The value for the total carbohydrate content varied quite extensively (i.e.
       between 82% and 102% of the dry weight) but the average value for the
       total carbohydrate content was quite similar in all the intercalibrations.
     - The absolute values obtained for glucose in the different intercalibrations
       varied between 73% and 87% of the dry weight. In contrast, the relative
       content of glucose (% of the carbohydrates in the sample) showed much
       smaller variations. For xylose and mannose, the difference in variation
       between the absolute and relative content were less pronounced but the
       variations were still in all cases smaller for the relative content than for
       the absolute content.
     - The smallest variations were obtained when samples hydrolysed at the
       participating laboratories were analysed at STFI (Intercalibration IIc),
       which indicates that the end determination step, including the calibration
       of the analysis instrument, is vital for the result of the analysis.
     - Intercalibration using the same conditions for acid hydrolysis gave only
       slightly improved variations compared to the initial intercalibration.
     - Even though only one intercalibration has been performed and not many
       conclusions can be drawn, enzymatic hydrolysis seems promising for
       future studies.
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            23




5           Conclusions and recommendations
Carbohydrate analysis of wood pulps is normally accomplished in two steps;
the pulp polysaccharides are first hydrolyzed to mono- and oligosaccharides,
which are subsequently determined by some carbohydrate analysis technique.
Intercalibration II showed that the hydrolysis results were indeed different
between the participating laboratories, most likely due to the different
equipments used. Furthermore, it was obvious that part of the variation was
due to the hydrolysis step and another part explained by differences in the
carbohydrate determination step, e.g. different calibrations. However, it was
not possible to single out the influence of each step on the variations in the
whole two-step procedure. The intercalibration study showed that a method
applicable for all different equipments taking part was not revealed and
therefore standardization of carbohydrate analysis appears not to be
reasonable at this stage. Instead, the methods applied at the different
participating laboratories have been optimised in accordance with each
individual set of equipment. Nevertheless, the work of the NSP Network
group Carbohydrate analysis has resulted in some conclusions and
recommendations concerning the different steps in analysis of the carbo-
hydrate compositions of pulp samples:


5.1         Sample pre-treatment
In this study, all pulp samples were washed until the outlet water was clear.
The effect of washing was not studied separately here, but it is recom-
mended that the pulp samples be thoroughly washed prior to the carbo-
hydrate analysis.

The samples were also dried and ground using an analytical mill. In
connection with acid hydrolysis, grinding is necessary in order to achieve
complete depolymerisation of the pulps. On the contrary, grinding is not
necessary for complete depolymerisation of the pulps by enzymatic
hydrolysis. However, in the analytical procedures used for enzymatic
hydrolysis in the current study, the samples taken out for analysis contain
between 10 mg and 100 mg of pulp and, since pulp samples may be
inhomogeneous by nature, milling of a larger amount of pulp (>1 g) from
which the sample is taken enhances the accuracy of the analysis.


5.2         Depolymerisation (hydrolysis)
Depolymerisation of the pulp can be performed by acid or enzymatic hydro-
lysis. Acid hydrolysis can be utilized for depolymerisation of all pulp
samples. Drawbacks of this method are, however, that acid hydrolysis
decomposes the liberated monosaccharides to various extent (Saeman et al.
1954, Sjöström et al. 1966, Biermann 1988), while releasing the uronic acid
residues present only to a limited extent. At the same time, the hexenuronic
     According to STFI's Confidentiality Policy this report is assigned category 1

24   Results and recommendations from the NSP Network group Carbohydrate analysis




     acids present are completely destroyed by acid hydrolysis (Lindberg et al.
     1973, Teleman et al. 1996).

     Enzymatic hydrolysis, on the other hand, has proven useful for analysis of
     all reducing sugars, including uronic acid and hexenuronic acid residues, in
     chemical pulps with kappa numbers up to at least 70 (Dahlman et al. 2000).
     It is, however, not possible to obtain complete depolymerisation of lignin-
     rich pulp samples, such as thermomechanical pulp, by enzymatic hydrolysis.
     Acid hydrolysis
     Acid hydrolysis is normally accomplished in two steps. In the first step the
     pulp is swelled in 72% sulphuric acid at 30°C for 60 minutes. Thereafter the
     sulphuric acid is diluted to 4% and kept at elevated temperature (120-125°C)
     for a certain time (20–60 minutes). Descriptions of the parameters used in
     the different participating laboratories are presented in Appendix 1.

     As mentioned earlier, it is well known that under the conditions of acid
     hydrolysis some monosaccharides decompose to various extents whereas
     others are released only to a limited extent. A large part of the work of the
     present study was therefore focused on acid hydrolysis. For several reasons,
     it was not possible to find any correlations between the parameters used for
     hydrolysis and the results obtained. First of all, the differences in the
     results from the participating laboratories are often of the same magnitude
     as the standard deviation of the individual analysis methods. This makes it
     difficult to study the variation connected to variations in acid hydrolysis
     alone. Secondly, many of the parameters important for the hydrolysis result
     are difficult to control. For example, it is possible to control the time at
     which the sample is kept at the desired hydrolysis temperature in step 2 ,
     but it is much more difficult to control the total time that the sample is
     heated, (i.e. including the time required for heating and cooling). The fact
     that the equipment used for hydrolysis differs widely between the partici-
     pating laboratories makes this even more complicated. As a consequence,
     the effort to use a common protocol for acid hydrolysis (intercalibration III)
     did not turn out well; most likely because of the different hydrolysis equip-
     ments used at the different laboratories, and the fact that some of them may
     not be working optimally at the specified conditions.

     Recommendations regarding acid hydrolysis:
     - All laboratories performed step 1 of the acid hydrolysis according to the
       TAPPI standard method (T249 cm-85).
     - The hydrolysis parameters of step 2 of the acid hydrolysis must be carefully
       tested and optimised at each laboratory and for each equipment (e.g. auto-
       clave, microwave oven). The actual hydrolysis time and temperature are the
       parameters that seem to vary the most between the participating labora-
       tories/equipments used (temperature 120-125°C, time 20–60 minutes).
       These parameters should therefore be in focus upon optimisation of the acid
       hydrolysis.
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            25



- The pulps and cotton linter samples used for intercalibrations in the
  present study are suitable for this optimisation of the acid hydrolysis. The
  pulp samples could also be used for repeated intercalibrations (for example
  every year).
- The effect of pre-hydrolysis of the monosaccharide standards must be
  evaluated at each laboratory, see chapter 4.3.
- Series of samples, the absolute carbohydrate contents of which are going to
  be compared, should be analysed at the same time (i.e., at the same calibra-
  tion and hydrolysis conditions).

Enzymatic hydrolysis
Due to lack of time, only one intercalibration was performed to study
enzymatic hydrolysis (intercalibration IV). Since only one intercalibration
has been performed not many conclusion can be drawn. However, the
variations in both absolute content of carbohydrates and content of glucose
were smaller than those obtained for most intercalibrations using acid
hydrolysis and consequently, enzymatic hydrolysis seems promising for
future studies.

In this study, a number of different enzyme mixtures were studied.
Unfortunately, most commercially available enzyme mixtures contain
monosaccharides. This means that the enzyme mixtures must be purified
with respect to monosaccharides prior to use in case that high dosages of
enzymes are applied for the hydrolysis. However, additional mono-
saccharides are often released during the enzymatic hydrolysis of the pulps,
and in such cases a “blank” sample must be analysed and the monosaccharide
content of this “blank” sample subtracted from the carbohydrate analysis
result. Details about the amount of monosaccharides present in the different
enzyme preparations in this study are presented in Appendix 2.

In the choice of the best mixture of enzymes for a certain application, the
following aspects should be taken into account:

In order to achieve complete hydrolysis of pulp, it is important that the
enzyme mixture contains enzymes capable of hydrolysing all the different
glycosidic linkages present in pulp polysaccharides. One way to ensure that
the enzymatic activity is sufficient to depolymerise the pulp sample to the
corresponding monosaccharides is to add an excess of enzymes. This is the
case in the methods used by STFI and MoRe. Consequently, addition of an
excess of enzymes also increases the amount of monosaccharides, mainly
mannose and galactose, in the “blank”. These results show in somewhat
poor repeatability for determination of mannose in pulps with low
glucomannan content (e.g. fully bleached hardwood kraft pulp).

In order to avoid contamination of the pulp samples with monosaccharides
from the enzyme preparations, some laboratories use much lower enzyme
dosages. In these methods, no purification of the enzyme preparations prior
to hydrolysis is necessary. The small amounts of monosaccharides can be
     According to STFI's Confidentiality Policy this report is assigned category 1

26   Results and recommendations from the NSP Network group Carbohydrate analysis




     accounted for by using a “blank” sample. This is the case in the method used
     by KCL. This procedure has to be optimised carefully to ensure complete
     hydrolysis of pulps.

     A way of increasing the carbohydrate yield is the combination of enzymatic
     and mild acidic hydrolysis (by trifluoroacetic acid) used by PFI. This method
     is, however, not applicable for analysis of hexenuronic acid, which is
     destroyed in the acid hydrolysis step.


     5.3                      Carbohydrate determination
     Carbohydrate determination can be performed by a number of different
     techniques, for example gas chromatography (after derivatization to
     alditols) (Sjöström et al. 1966; TAPPI 1985; Theander & Westerlund 1986),
     ion-exchange chromatography (Edwards et al. 1987; Hausalo et al. 1995;
     Wright & Wallis 1996), high-performance liquid chromatography (Syverud
     et al. 1999) or capillary zone electrophoresis (after derivatization with a UV-
     absorbing tag) (El Rassi 1994; El Rassi 1999; Dahlman et al. 2000). The
     benefits and drawbacks of the analysis techniques were not included in the
     present study. However, during the work of the NSP Network group Carbo-
     hydrate analysis, it became clear that the variation in results had its origin
     in both the hydrolysis step and the end determination of the carbohydrate
     compositions.

     Calibration
     A correct calibration of the instrument is very important in order to obtain
     reliable results from the determination of absolute carbohydrate contents.
     The need for calibration differs to a large extent between the various
     determination techniques used. For example, an ion-exchange chromato-
     graphy instrument requires a new calibration curve for every analysis
     occasion, whereas an updated calibration curve every year is sufficient for
     capillary zone electrophoresis.

     As mentioned earlier, it is well known that acid hydrolysis destroys some
     monosaccharides to various extent whereas others are released only to a
     limited extent. In the TAPPI standard method (T249 cm-85), the loss of
     carbohydrates due to degradation during hydrolysis is compensated by using
     a calibration curve prepared from monosaccharides standards that have been
     subjected to an acidic pre-treatment. The theory behind the use of such pre-
     hydrolysis of the standards is that the carbohydrate content determined by
     this procedure better reflects the “true” carbohydrate content of the pulp
     sample, since some monosaccharides are undoubtedly degraded during acid
     hydrolysis. Critical voices, however, claim that the pre-treatment of
     standards will always introduce additional uncertainties to the analysis
     results. Monosaccharides in acidic solution behave differently from poly-
     saccharides present in pulp during the course of acid hydrolysis. The com-
     pensation may thus “overcorrect” the monosaccharide losses.
                                           According to STFI's Confidentiality Policy this report is assigned category 1

                   Results and recommendations from the NSP Network group Carbohydrate analysis                            27



From the intercalibrations, it seems to be difficult to obtain accurate values
for the absolute carbohydrate content. In fact, the value for the total carbo-
hydrate content varied quite extensively (i.e. between 82% and 102 % of the
dry weight) between the different determinations even though the average
value for the total carbohydrate content for all determinations in one inter-
calibration was quite similar throughout all the intercalibrations. In contrast,
determination of the relative contents of glucose, xylose and mannose showed
much smaller variations.

As expected, the variations were reduced when the analyses were performed
5-6 times (Intercalibration IIb) compared to the initial intercalibration (I) in
which two parallel analyses were performed.

Recommendations regarding instrument carbohydrate determination/instrument
calibration:
- The need for (and effect of) pre-hydrolysis of calibration substances in
  connection with acid hydrolysis must be carefully evaluated in each
  laboratory.
- With each set of analyses a standard pulp sample should be included, and
  a control card should be used to keep track of the variations in carbo-
  hydrate composition obtained upon analysis of this standard pulp. The
  oxygen-delignified softwood kraft pulp studied within the present work is
  suitable as such a standard pulp.
- Intercalibrations should be repeated with a certain periodicity. It might be
  possible to include carbohydrate analysis in the round-robin analyses
  routinely performed within the pulp and paper industry.
- Series of samples, the absolute carbohydrate contents of which are going to
  be compared, should be analysed at the same time (i.e., using the same
  calibration).
- In order to obtain accurate determination of the absolute carbohydrate
  content, the number of parallel samples to be analysed could be increased.
- The relative carbohydrate composition values are, as expected, much less
  sensitive to variations in instrument calibration than are the corresponding
  absolute compositions. Therefore, if the absolute carbohydrate composition
  is uncertain, the relative carbohydrate compositions could still be studied.
     According to STFI's Confidentiality Policy this report is assigned category 1

28   Results and recommendations from the NSP Network group Carbohydrate analysis




     6                        References
     Biermann, C.J. (1988) Adv. Carboh. Chem. Biochem. 46: 251-271.

     Dahlman, O., et al. (2000). J. Chromatogr. A 891: 157-174.

     Edwards, W. T., Pohl, C. A. and Rubin, R. (1987). Tappi J. 70: 138-140.

     El Rassi, Z. (1994). Adv. Chromatogr. 34: 177-250.

     El Rassi, Z. (1999). Electrophoresis 20: 3134-3144.

     Hausalo, T. et al. (1995) Proceedings of the 8th International Symposium on
     Wood and Pulping Chemistry, Helsinki, Finland; June 6-9, Vol. III, pp. 131-
     136.

     Lindberg, B., Lönngren, J. and Thompson, J. L. (1973) Carbohydr. Res. 28:
     351-357.

     Saeman, J., et al. (1954) Tappi 37: 336-343.

     Sjöström, E., Haglund, P. and Janson, J. (1966). Sv. Papperstidn. 69: 381-
     385.

     Syverud, K., Moe, S.T. and Minja, R.J.A. Proceedings of the 10th
     International Symposium on Wood and Pulping Chemistry, Yokohama,
     Japan; June 7-10, Vol. II, pp. 232-237.

     Teleman, A., et al. (1996) Carbohydr. Res. 280: 197-208.

     Theander, O., Westerlund, E. A. (1986) J. Agric. Food Chem 34:330-336.

     Wright, P. J. and Wallis, A. F. A. (1996). Holzforschung 50: 518-524.

     Carbohydrate composition of extractive-free wood and wood pulp by gas
     chromatography. Atlanta GA, Tappi Press, 1985.
        Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                         Appendix 1    1




                  Appendix 1

Report from intercalibration studies
         010221 – 011002



       Anna Jacobs, Asha Ismail Olsson
            Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                             Appendix 1    1




       Contents
                                                                                 Page


1      Background                                                                      3

2      Methods                                                                         4

3      Pulp samples                                                                    5

4      Results                                                          6
4.1    Oxygen delignified hardwood kraft pulp                           6
4.2    Bleached hardwood kraft pulp                                     7
4.3    Oxygen delignified softwood kraft pulp                           8
4.4    Softwood TMP                                                     9
4.5    Hydrolyzate (of oxygen delignified softwood kraft pulp) produced
       at STFI                                                         10

5      Conclusions/further work                                                      11

6      References                                                                    13

7      Appendix; hydrolysis parameters                                               14
7.1    Method 1                                                                      14
7.2    Method 2                                                                      14
7.3    Method 3                                                                      14
7.4    Method 4                                                                      14
7.5    Method 5                                                                      15
7.6    Method 6                                                                      15
7.7    Method 7                                                                      15
7.8    Method 8                                                                      15
7.9    Method 9                                                                      16
7.10   Method 10                                                                     16
7.11   Method 11                                                                     16
7.12   Method 12                                                                     16
                  Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                   Appendix 1    3




1            Background
The network group for carbohydrate analysis consists of participants from
the three institutes (KCL, PFI and STFI) together with participants from
five companies (Kappa, MoRe, SCA, Stora Enso, Södra).

The main objectives of this network group are:

i) to evaluate the methods currently used for carbohydrate analysis in the
   pulp and paper industry in Finland, Norway and Sweden, and
ii) to investigate the possibilities to develop standardized carbohydrate
    analysis methods.

During the first period of collaboration (010221–011002), the network
group has performed an initial interlaboratory study in order to compare
the methods currently used by the institutes and the industry. In this
study, the carbohydrate compositions of four pulp samples were analysed at
each laboratory using their in-house methods. In addition, each laboratory
has analysed the carbohydrate composition of a pulp hydrolyzate prepared
at STFI, in order to compare the techniques used for the final carbohydrate
determination.

The results from this initial study are summarized in the present report.
    Results and recommendations from the NSP Network group Carbohydrate analysis

4   Appendix 1




    2               Methods
    Carbohydrate analysis of wood pulps is normally accomplished in two
    steps. The pulp polysaccharides are first hydrolyzed to mono- and oligo-
    saccharides, which are subsequently determined by some carbohydrate
    analysis technique, for example gas chromatography (after derivatization
    to alditols) (Sjöström et al. 1966; TAPPI 1985; Theander & Westerlund
    1986), ion-exchange chromatography (Edwards et al. 1987; Hausalo et al.
    1995; Wright & Wallis 1996) or capillary zone electrophoresis (after
    derivatization with a UV-absorbing tag) (El Rassi 1994; Dahlman et al.
    1997; El Rassi 1999).

    The combinations of methods used in this interlaboratory study are listed
    in Table 1. Details concerning the parameters used for hydrolysis of the
    pulp samples are presented in the Appendix.

        Table 1. Methods used in the interlaboratory study.
    Method        Laboratory        Hydrolysis             Determination technique
    number
    1             STFI              Acid                   Gas Chromatography
    2             Kappa             Acid                   Ion Exchange Chromatography
    3             KCL               Acid                   Ion Exchange Chromatography
    4             MoRe              Acid                   Ion Exchange Chromatography
    5             PFI               Acid                   Gas Chromatography
    6             SCA               Acid                   Gas Chromatography
    7             Stora Enso        Acid                   Ion Exchange Chromatography
    8             Södra             Acid                   Gas Chromatography
    9             PFI               Acid /enzymatic        High-Performance Liquid
                                                           Chromatography
    10            STFI              Enzymatic              Capillary Zone Electrophoresis
    11            KCL               Enzymatic              Ion Exchange Chromatography
    12            MoRe              Enzymatic              Ion Exchange Chromatography
                 Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                  Appendix 1    5




3           Pulp samples
Four pulp samples were distributed to the participating laboratories. These
four pulps were:

 Oxygen delignified hardwood kraft pulp with kappa number 8 from Södra.
 Bleached (TCF) hardwood kraft pulp with kappa number 4 from Södra.
 Oxygen delignified softwood kraft pulp with kappa number 9 from
 Östrand pulp mill (SCA).
 Softwood thermomechanical pulp (TMP) with kappa number 136 from
 Ortviken (SCA).

The pulp samples were washed until the outlet water was clear, dried at
40°C overnight, and finally ground using an analytical mill.
    Results and recommendations from the NSP Network group Carbohydrate analysis

6   Appendix 1




    4               Results
    Results from carbohydrate analysis of the four pulp samples, using in-
    house methods (listed in Table 1) are presented in sections 4.1 to 4.4.
    Values below the quantification limits are denoted as “<x%”, where x is the
    limit obtained in the corresponding laboratory. The relative standard
    deviation for each carbohydrate component was calculated in the cases
    when at least five laboratories had detected values above their quantifica-
    tion limit.

    In order to evaluate the techniques used for the final carbohydrate determi-
    nation, a larger amount of the softwood kraft pulp was subjected to acid
    hydrolysis at STFI. Aliquots of the resulting hydrolyzate was also distri-
    buted to the participants. The results from carbohydrate analysis of the
    hydrolyzate are presented in section 4.5.

    In this report we have used comma (not a dot) as decimal sign as it is the
    directives within international standardization.


    4.1             Oxygen delignified hardwood kraft pulp
        Table 2. Absolute carbohydrate composition
                   Carbohydrate composition (% of dry weight)
    Method # Arabinose Galactose Mannose       Xylose     Glucose                  Total
    1            0,2        0,1       0,4        21,0       74,4                   96,1
    2           <0,1       <0,1      <0,2        19,1       64,2                   83,4
    3          <0,03      <0,03     <0,05        19,8       65,3                   85,1
    6            0,3      <0,05       0,4        18,0       72,0                   90,7
    7          <0,05      <0,05       0,3        24,8       72,3                   97,4
    8            0,2        0,3       0,4        22,5       73,1                   96,5
    9            0,4       <0,1      <0,4        21,9       69,1                   91,4
    10         <0,05      <0,05      0,7         24,5       69,9                   95,1
    11         <0,03       0,1        1,3        20,2       68,2                   89,8
    12         <0,02      <0,02      1,0         22,3       71,3                   94,6
    Average      0,3                  0,6        21,4       70,0                   92,0
    Std. dev.    0,1        not       0,4         2,2        3,3                    4,8
    RSD (%)      35     applicable    60          10          5                      5
                  Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                   Appendix 1    7



 Table 3. Relative carbohydrate composition
                    Relative carbohydrate composition
                        (% of total carbohydrates)
Method # Arabinose Galactose Mannose          Xylose Glucose
1            0,1          0,1         0,4      21,9    77,4
2           <0,1         <0,1         0,1      22,9    76,9
3           <0,03       <0,03        <0,05     23,3    76,7
6           <0,05       <0,05         0,4      19,8    79,4
7           <0,05       <0,05         0,3      25,4    74,3
8            0,3          0,3         0,4      23,3    75,8
9           <0,1         <0,1        <0,4      24,0    75,6
10          <0,05       <0,05         0,7      25,8    73,5
11          <0,03         0,1         1,5      22,5    75,9
12          <0,02       <0,02         1,1      23,6    75,3
Average                               0,6      23,3    76,1
Std. dev.    not          not         0,4       1,6     1,6
RSD (%)   applicable applicable       68         7       2

4.2         Bleached hardwood kraft pulp
 Table 4. Absolute carbohydrate composition
               Carbohydrate composition (% of dry weight)
Method # Arabinose Galactose Mannose       Xylose     Glucose                   Total
1            0,1        0,1       0,5        21,3       76,9                    98,9
2           <0,1      <0,1       <0,2        19,4       65,2                    84,7
3          <0,03      <0,03       0,3        20,2       66,0                    86,5
6            0,3      <0,05       0,5        19,0       80,0                    99,8
7          <0,05      <0,05       0,5        24,9       72,9                    98,3
8            0,3        0,4       0,5        23,0       75,2                    99,4
9            0,5      <0,1        0,4        20,8       66,5                    88,2
10           0,1      <0,05       0,5        24,4       69,3                    94,3
11          0,04       0,05       1,3        19,4       66,7                    87,5
12         <0,02     <0,02       0,7         22,8       72,3                    95,8
Average      0,2                  0,6        21,5       71,1                    93,3
Std. dev.    0,2        not       0,3         2,1        5,2                     6,0
RSD (%)      77     applicable    49          10          7                       6
    Results and recommendations from the NSP Network group Carbohydrate analysis

8   Appendix 1



      Table 5. Relative carbohydrate composition
                      Relative carbohydrate composition
                          (% of total carbohydrates)
    Method # Arabinose Galactose Mannose        Xylose Glucose
    1            0,1        0,1         0,5      21,5    77,8
    2           <0,1       <0,1        <0,2      23,0    76,9
    3          <0,03      <0,03         0,3      23,4    76,3
    6            0,3      <0,05         0,5      19,0    80,1
    7          <0,05      <0,05         0,5      25,3    74,2
    8            0,3        0,4         0,5      23,1    75,7
    9            0,6       <0,1         0,5      23,6    75,4
    10           0,1      <0,05         0,5      25,9    73,5
    11           0,1        0,1         1,4      22,1    76,3
    12         <0,02      <0,02         0,8      23,8    75,5
    Average      0,2                    0,5      22,9    76,4
    Std. dev.    0,2        not         0,1       2,1     2,0
    RSD (%)      81     applicable      19         9       3

    4.3             Oxygen delignified softwood kraft pulp
      Table 6. Absolute carbohydrate composition

                     Carbohydrate composition (% of dry weight)
    Method # Arabinose Galactose Mannose                      Xylose       Glucose   Total
    1            0,8      0,4       6,1                         7,4          76,5    91,2
    2            0,6      0,3       5,9                         6,5          72,6    85,9
    3            0,6      0,3       5,3                         6,6          69,2    82,1
    6            0,9      0,5       6,9                         6,6          87,0    101,9
    7            0,7      0,3       6,6                         8,5          79,9    96,0
    8            0,8      0,7       6,5                         7,9          83,4    99,3
    9            1,0     <0,1       5,3                         8,3          76,2    90,8
    10           1,0      0,3       7,1                         8,9          76,6    93,9
    11           0,7      0,3       5,4                         8,0          76,2    90,6
    12           0,7      0,3       7,0                         7,8          80,2    95,9
    Average      0,8      0,4       6,2                         7,7          77,8    92,8
    Std. dev.    0,1      0,2       0,7                         0,8           5,1     6,0
    RSD (%)     18,9     41,3      11,4                        11,0           6,6     6,4
                  Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                   Appendix 1    9



 Table 7. Relative carbohydrate composition
                  Relative carbohydrate composition
                      (% of total carbohydrates)
Method # Arabinose Galactose Mannose        Xylose Glucose
1            0,8        0,4         6,7       8,1    83,9
2            0,7        0,3         6,9       7,6    84,5
3            0,8        0,4         6,5       8,1    84,3
6            0,9        0,5         6,8       6,5    85,3
7            0,7        0,3         6,9       8,9    83,2
8            0,8        0,7         6,5       8,0    84,0
9            1,1       <0,1         5,8       9,1    83,9
10           1,1        0,3         7,6       9,5    81,6
11           0,8        0,3         6,0       8,8    84,1
12           0,7        0,3         7,3       8,1    83,6
Average      0,8        0,4         6,7       8,3    83,8
Std. dev.    0,1        0,1         0,5       0,9     1,0
RSD (%)     17,1       36,3         7,9      10,5     1,2

4.4         Softwood TMP
 Table 8. Absolute carbohydrate composition
               Carbohydrate composition (% of dry weight)
Method # Arabinose Galactose Mannose       Xylose     Glucose                   Total
1            1,1       2,2       12,6         5,4       46,7                    68,0
2            1,0       1,7       10,6         4,4       41,2                    58,9
3            1,0       1,7        9,6         4,4       37,2                    53,8
4            1,1       1,9       10,8         5,1       39,6                    58,4
5            1,0       1,8       11,1         5,5       61,0                    80,4
6            1,2       2,4       13,0         5,2       52,0                    73,8
7            1,1       1,9       12,7         5,8       44,9                    66,4
8            1,3       2,5       12,0         6,0       46,2                    68,0
Average      1,1       2,0       11,5         5,2       46,1                    66,0
Std. dev.    0,1       0,3        1,2         0,6        7,6                     8,7
RSD (%)     10,4      15,3       10,4        11,5       16,5                    13,2


 Table 9. Relative carbohydrate composition
                  Relative carbohydrate composition
                      (% of total carbohydrates)
Method # Arabinose Galactose Mannose        Xylose Glucose
1            1,6        3,2        18,5       7,9    68,7
2            1,6        3,0        18,0       7,4    70,0
3            1,9        3,2        17,8       8,1    69,0
4            1,8        3,2        18,5       8,7    67,8
5            1,2        2,2        13,8       6,8    75,9
6            1,6        3,3        17,6       7,0    70,5
7            1,7        2,9        19,1       8,8    67,6
8            1,9        3,7        17,6       8,8    67,9
Average      1,6        3,0        17,6       7,8    69,9
Std. dev.    0,2        0,4         1,8       0,8     2,8
RSD (%)     12,7       12,0         9,9       9,7     4,0
     Results and recommendations from the NSP Network group Carbohydrate analysis

10   Appendix 1




     4.5             Hydrolyzate (of oxygen delignified softwood kraft
                     pulp) produced at STFI
       Table 10. Absolute carbohydrate composition
                       Carbohydrate composition (% of dry weight)
     Lab          Arabinose Galactose Mannose      Xylose     Glucose                 Total
     STFI            0,8       0,4       6,2          7,4       76,5                  91,3
     Kappa           0,6       0,3       6,2          6,9       74,7                  88,7
     KCL             0,6       0,3       5,9          6,0       68,9                  81,7
     MoRe            0,6       0,3       5,8          6,6       73,9                  87,2
     PFI             0,5        -        5,3          6,5       69,6                  81,9
     SCA             0,6       0,5       5,8          5,6       74,0                  86,5
     Stora
     Enso              0,6           0,3           6,3            7,2          79,2   93,6
     Södra
     Average           0,6           0,3           5,9            6,6          73,8   87,3
     Std. dev.         0,1           0,1           0,4            0,6           3,6    4,5
     RSD (%)          14,8          28,2           6,0            9,7           4,9    5,1



       Table 11. Relative carbohydrate composition
                       Relative carbohydrate composition
                           (% of total carbohydrates)
     Method # Arabinose Galactose Mannose        Xylose Glucose
     STFI         0,9        0,4         6,8       8,1    83,8
     Kappa        0,7        0,3         7,0       7,8    84,2
     KCL          0,8        0,3         7,2       7,3    84,4
     MoRe         0,7        0,3         6,7       7,6    84,7
     PFI          0,6         -          6,5       7,9    85,0
     SCA          0,6        0,6         6,7       6,5    85,6
     Stora
     Enso         0,7        0,3         6,8       7,7    84,6
     Södra
     Average      0,7        0,4         6,8       7,6    84,6
     Std. dev.    0,1        0,1         0,2       0,5     0,6
     RSD (%)     12,2       28,6         3,3       7,1     0,7
                     Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                       Appendix 1   11




5               Conclusions/further work
We have tried to perform a deeper evaluation of the results from the
interlaboratory test. When the relative standard deviations in the
sugar determinations of the five samples are compared (Table 12 and
13), it can be clearly seen that the relative standard deviations for
components of higher concentration (i.e. glucose and xylose) is much
lower than for components with low abundance in the pulps (i.e.
arabinose and galactose).

    Table 12. Comparison of standard deviations for determination of the absolute
    carbohydrate compositions
Pulp                                   Relative standard deviation (%)
                            Arabinose Galactose Mannose Xylose Glucose Total
O2-del hardwood kraft           35             -            60         10          5          5
Bleached hardwood kraft         77             -            49         10          7          6
O2-del softwood kraft           19            41            11         11          7          6
TMP                             10            15            10         12         17         13
O2-del softwood kraft           15            28             6         10          5          5
Hydrolyzed at STFI


    Table 13. Comparison of standard deviations for determination of the relative
    carbohydrate compositions
Pulp                                   Relative standard deviation (%)
                             Arabinose Galactose Mannose Xylose                     Glucose
O2-del hardwood kraft              -             -               68          7           2
Bleached hardwood kraft           81             -               19          9           3
O2-del softwood kraft             17            36                8         11           1
TMP                               13            12               10         10           4
O2-del softwood kraft             12            29                3          7           1
Hydrolyzed at STFI


Furthermore, the relative standard deviations for determination of the
carbohydrate composition of the oxygen delignified kraft pulp are
consistently higher than the corresponding values for the same pulp
hydrolysed at STFI. This means that there must be differences
caused by the hydrolysis procedures at the different laboratories.
However, we were not able to find any correlation between the
results from the carbohydrate analyses and the analysis methods
employed. In contrast, none of the evaluations showed any clear
trends or tendencies. The lack of such correlation is most likely due
to the fact that the differences in the results from the participating
laboratories are of the same magnitude as the standard deviation of
the individual analysis methods. We believe that, in order to perform
     Results and recommendations from the NSP Network group Carbohydrate analysis

12   Appendix 1



     a more thorough evaluation of the correlation between the analysis
     methods employed and the corresponding results, more statistical
     data is needed.
     To further investigate these correlations, we suggest that a few extra
     experiments must be included. In that case, each laboratory should
     perform the whole analysis procedure (pretreatment, hydrolysis and
     carbohydrate determination) for one pulp in a series of parallel
     experiments (i.e.; 6-8 replicates). This way we will get
     i) carbohydrate analysis results less sensitive to random variations
        and
     ii) a measure of the standard deviation for each laboratory. We think
       this would give us the necessary statistical background for further
       evaluations.
     Finally, it must be mentioned that the acidic carbohydrates (i.e. 4-O-
     methylglucuronic acid and hexenuronic acid) were not taken into
     account in this study. The hexenuronic acid is destroyed by the
     conditions employed during acid hydrolysis. Thus, hexenuronic acid
     was analysed only by the laboratories utilizing enzymatic hydrolysis.
     In the future work, more emphasis will be laid on the enzymatic
     hydrolysis.
                  Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                   Appendix 1    13




6            References
Dahlman, O., et al. (2000). J. Chromatogr. A 891: 157-174.

Edwards, W. T., Pohl, C. A. and Rubin, R. (1987). Tappi J. 70: 138-140.

El Rassi, Z. (1994). Adv. Chromatogr. 34: 177-250.

El Rassi, Z. (1999). Electrophoresis 20: 3134-3144.

Hausalo, T. et al. (1995) Proceedings of the 8th International Symposium on
Wood and Pulping Chemistry, Helsinki, Finland; June 6-9, Vol. III, pp. 131-
136.

Sjöström, E., Haglund, P. and Janson, J. (1966). Sv. Papperstidn. 69: 381-
385.

Theander, O., Westerlund, E. A. (1986). J. Agric. Food Chem: 330-336.

Wright, P. J. and Wallis, A. F. A. (1996). Holzforschung 50: 518-524.

Carbohydrate composition of extractive-free wood and wood pulp by gas
chromatography. Atlanta GA, Tappi Press, 1985.
     Results and recommendations from the NSP Network group Carbohydrate analysis

14   Appendix 1




     7               Appendix; hydrolysis parameters

     7.1             Method 1
     Laboratory: STFI
     Acid hydrolysis
     Step 1: 72% sulphuric acid, 30°C, 60 minutes
     Step 2: 4% sulphuric acid, 125°C (autoclave), 60 minutes
     Derivatization: acetylation
     Internal standard: 2-deoxygalactose
     Monosaccharide standards subjected to pre-hydrolysis


     7.2             Method 2
     Laboratory: Kappa
     Acid hydrolysis
     Step 1: 72% sulphuric acid, 30°C, 60 minutes
     Step 2: 4% sulphuric acid, 125°C (autoclave), 60 minutes
     Derivatization: none
     Internal standard: none
     No pre-treatment of the monosaccharide standards


     7.3             Method 3
     Laboratory: KCL
     Acid hydrolysis
     Step 1: 72 % sulphuric acid, 30°C, 60 minutes
     Step 2: 4 % sulphuric acid, 120°C (autoclave), 60 minutes
     Derivatization: none
     Internal standard: D-fucose
     No pre-treatment of the monosaccharide standards


     7.4             Method 4
     Laboratory: MoRe
     Acid hydrolysis
     Step 1: 72% sulphuric acid, 30°C, 60 minutes
     Step 2: 4% sulphuric acid, 120°C (microwave oven), 40 minutes
     Derivatization: none
     Internal standard: none
     No pre-treatment of the monosaccharide standards
                   Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                    Appendix 1    15




7.5         Method 5
Laboratory: PFI
Acid hydrolysis
Step 1: 72% sulphuric acid, 30°C, 60 minutes
Step 2: 4% sulphuric acid, 125°C (autoclave), 60 minutes
Derivatization: acetylation
Internal standard: myo-inositol
Monosaccharide standards subjected to pre-hydrolysis


7.6         Method 6
Laboratory: SCA
Acid hydrolysis:
Step 1: 72% sulphuric acid, 30°C, 60 minutes
Step 2: 4% sulphuric acid, 125°C (autoclave), 60 minutes
Derivatization: acetylation
Internal standard: 2-deoxygalactose
Monosaccharide standards subjected to pre-hydrolysis


7.7         Method 7
Laboratory: Stora Enso
Acid hydrolysis
Step 1: 72% sulphuric acid, 30°C, 60 minutes
Step 2: 4% sulphuric acid, 120°C (autoclave), 20 minutes
Derivatization: none
Internal standard: D-fucose
Monosaccharide standards subjected to pre-hydrolysis


7.8         Method 8
Laboratory: Södra
Acid hydrolysis
Step 1: 72% sulphuric acid, 30°C, 60 minutes
Step 2: 4% sulphuric acid, 125°C (autoclave), 60 minutes
Derivatization: acetylation
Internal standard: 2-deoxygalactose
Monosaccharide standards subjected to pre-hydrolysis
     Results and recommendations from the NSP Network group Carbohydrate analysis

16   Appendix 1




     7.9             Method 9
     Laboratory: PFI
     Combination of enzymatic and acid hydrolysis
     Step 1; enzymatic hydrolysis:
     Enzymes: Ecopulp C-15 (Röhm), Ecopulp TX-200C (Röhm), Mannanase
     ABM (Röhm). Volume ratio= 1:1:1.
     Amount of enzyme used: 0,6 ml / g pulp added at the beginning of
     hydrolysis, after 3 hours and after 6 hours of hydrolysis (e.g. 1,8 ml / g
     pulp).
     Conditions: 48 h, 55°C, pH: 5
     Step 2; acid hydrolysis:
     Conditions: 0,68 M trifluoroacetic acid, 100°C, 60 min.
     Internal standard: mannitol


     7.10            Method 10
     Laboratory: STFI
     Enzymatic hydrolysis
     Enzymes: Celluclast 1.5L (Novozymes), Novozym 188 (Novozymes),
     Econase HC-400 (Röhm). Volume ratio=1:1:1.
     Amount of enzyme used: 0,05 ml / mg pulp
     Internal standard: ribose


     7.11            Method 11
     Laboratory: KCL
     Enzymatic hydrolysis
     Enzymes: Novozym 188 (Novozymes), Econase CEP (Röhm). Volume
     ratio=1:1.
     Amount of enzyme used: 0,72 ml + 0,2 g / g pulp
     Conditions: 48 h, 50°C, pH 5
     Internal standard: D-fucose


     7.12            Method 12
     Laboratory: MoRe
     Enzymatic hydrolysis
     Enzymes: Celluclast 1.5L (Novozymes), Novozym 188 (Novozymes),
     Ecopulp C15 (Röhm), Ecopulp TX-200C (Röhm). Volume ratio=1:1:1:1.
     Amount of enzyme used: 0,05 ml / mg pulp
     Conditions: 30 h, 40°C, pH 4
     Internal standard: none
      Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                       Appendix 2




                Appendix 2

Carbohydrate content in enzyme
         preparates


                 Christiane Laine
                     Mia Tehomaa

                         10.5.2002
According to STFI's Confidentiality Policy this report is assigned category 1

Appendices
CW 000




Contents
Appendix                                                                              Page


1                        Background                                                      1

2                        Analysis method                                                 2

3                        Enzyme preparates                                               3

4                        Results                                                         4
4.1                      Monosaccharides in the enzyme preparates                        4
4.2                      Monosaccharides in enzyme preparates after acid hydrolysis      4

5                        References                                                      7
                    Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                     Appendix 2    1




1            Background
The network group for carbohydrate analysis consists of participants from
the three Nordic pulp and paper research institutes (KCL, PFI and STFI)
together with participants from five companies (Kappa, MoRe, SCA,
StoraEnso, Södra).

Enzymes are nowadays used in many laboratories to conduct the hydrolysis
step of pulp polysaccharides for carbohydrate analysis.i,ii,iii,iv,v The liberated
monosaccharides in the hydrolyzate are then quantified using different
methods e.g. liquid chromatography. Applied enzymes are available as
commercial enzyme preparates. Still, the activity and the purity of the
commercial enzyme preparates are not always satisfying. Next to the
enzymes itself, the preparates contain small amounts of carbohydrates in
the form of monosaccharides, oligosaccharides and/or polysaccharides,
partly bound to the enzymes. These carbohydrates lead to a background
concentration of monosaccharides in the hydrolyzates.

This study was performed to find out, how much carbohydrates are present
in commercial enzyme preparations applied for enzymatic hydrolysis at
partners of the Network Group Carbohydrate analysis. The analysis was
performed directly from the diluted enzyme preparates and after mild acid
hydrolysis. The direct analysis delivers information on the free
monosaccharides. The analysis after mild hydrolysis shows the total amount
of present carbohydrates including oligo- and polysaccharides bound to the
enzymes themselves.

It has to be remembered that the analysis results are only valuable for the
studied batches. Different batches of the same preparate differ from each
other for example in respect to their carbohydrate composition.
    Results and recommendations from the NSP Network group Carbohydrate analysis

2   Appendix 2




    2                Analysis method
    Mild acid hydrolysis was performed at a concentration of 4 weight-% of
    sulphuric acid at 120ºC in an autoclave for 60 min.v

    The free or liberated monosaccharides were analyzed using high performance
    anion exchange chromatography with pulse amperometric detection (HPAEC-
    PAD, equipment of Dionex).
                      Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                       Appendix 2    3




3               Enzyme preparates
The studied enzyme preparates are shown in Table 1. It is indicated, which of
the laboratories uses the different enzyme preparates.

    Table 1. Overview of the studied enzyme preparates and their use at different
    laboratories.
Name of enzyme           STFI            KCL            MoRe            Stora            PFI
preparate
Novozyme 188               *               *               *
Celluclast                 *                               *
Econase HC-400             *
Ecopulp TX-200C                                            *              *               *
Ecopulp C-15                                               *              *               *
Econase CEP                                *
Gammanase                                                                 *
Mannanase AMB                                                                             *
    Results and recommendations from the NSP Network group Carbohydrate analysis

4   Appendix 2




    4                Results
    In this report we have used comma (not a dot) as decimal sign as it is the
    directives within international standardization.

    4.1              Monosaccharides in the enzyme preparates
    The content of monosaccharides in the enzyme preparates is shown in
    Table 2. The content of monosaccharides varied between 3-27 mg/l for the
    liquid preparates and was 12,6 mg/100 mg for the solid Econase CEP.
    Glucose was the major monosaccharide in Novozym 188, Celluclast, Ecopulp
    TX-200 C and Econase CEP. Econase HC-400, Ecopulp C15, Mannanase
    AMB and Gammanase contained mannose as the major monosaccharide.
    Gammanase contained galactose as second major monosaccharide.

    This part of carbohydrates can be separated from the enzymes using for
    example size exclusion chromatography.

        Table 2. Free monosaccharides in enzyme preparates (direct injection after
        dilution). (Rha = Rhamnose, Ara = Arabinose, Gal = Galactose, Glu = Glucose,
        Xyl = Xylose, Man = Mannose)

                                                                              Total
    Enzyme                Supplier     Batch or LOT Rha Ara Gal Glu Xyl Man monosacch.
    Novozym 188, g/l      Novo Nordisk DCN 00022     - 0,5 2,6 19,3 - 4,7      27,1
    Celluclast 1.5L, g/l  Novo Nordisk CCN 03054     -   - 0,5 6,3 - 0,9       7,6
    Econase HC-400, g/l   Röhm Enzyme 20005002;21    - 0,2 0,7 0,2 0,3 1,9      3,3
    Ecopulp TX-200 C, g/l Röhm Enzyme 10317413       - 0,2 0,9 15,6 - 0,7      17,4
    Ecopulp C-15, g/l     Röhm Enzyme 20001-1-1      - 0,1 0,4 0,3 0,2 2,1      3,2
    Econase CEP, mg/100mg Primalco*    111155032     -   -   + 11,8 - 0,8      12,6
    Gammanase, g/l        Novo Nordisk CHNO1021     1.7 0,8 7,4 0,4 1,6 12,1   24,0
    Mannanase KMB, g/l    Röhm Enzyme 100803         - 0,3 1,9 - 0,9 7,6       10,7
    * nowadays Röhm Enzyme
    - less than 0,1 g/l or 0,01 mg/100 mg, + less than 0,1 mg/100 mg


    4.2              Monosaccharides in enzyme preparates after acid
                     hydrolysis
    The total carbohydrate content in the enzyme preparates was determined
    after mild acid hydrolysis as monosaccharides (Table 3). The carbohydrate
    content increased for all preparates compared to the direct analysis. This
    means that this additional part of carbohydrates was present as oligo- or
    polysaccharides and/or bound to the enzyme. The content was between 1,5 to
    6 times as high after the mild acid hydrolysis compared to the direct analysis.
    The carbohydrate content was highest in Novozym 188 with 53,2 g/l followed
    by Ecopulp TX-200C, Mannanase AMB and Celluclast and lowest for Econase
    HC-400 with 9,5 g/l. The part of oligo- or polysaccharides was 6 g/l or higher
    (difference between Table 2 and 3). Mannose was in the oligo- or poly-
                       Results and recommendations from the NSP Network group Carbohydrate analysis

                                                                                         Appendix 2   5



saccharides of all enzyme preparates next glucose and the other mono-
saccharides.

These liberated monosaccharides may partly be liberated in the course of
the enzymatic hydrolysis causing blank values. It is most probably not
possible to remove this part of the carbohydrates, because they may follow
the enzyme in the separation step.

  Table 3. Carbohydrates of enzyme preparates determined as monosaccharides
  after acid hydrolysis. (Rha = Rhamnose, Ara = Arabinose, Gal = Galactose, Glu =
  Glucose, Xyl = Xylose, Man = Mannose)
                                                                                      Total
Enzyme                 Supplier    Batch           Rha Ara Gal Glu          Xyl Man monosacch.
                       Novo
Novozym 188, g/l       Nordisk     DCN 00022        0,6   0,8   3,4 20,7 1,0 26,7           53,2
                       Novo
Celluclast 1.5L, g/l   Nordisk     CCN 03054         -     -    2,0 15,5     -    13,4      30,9
Econase HC-400, g/l    Röhm        20005002;21
                       Enzyme                        -    0,7 1,5     0,7   0,9   5,8        9,5
                      Röhm
Ecopulp TX-200 C, g/l Enzyme       10317413          -    0,5   1,8 33,9 0,5      4,3       41,1
                       Röhm
Ecopulp C-15, g/l      Enzyme      20001-1-1         -    0,8   1,8   2,4   1,0 12,1        18,0
Econase CEP,
mg/100mg               Primalco* 111155032           -    0,5   0,4 24,0 0,6      6,0       31,5
                       Novo
Gammanase, g/l         Nordisk     CHNO1021         2,8   1,3   9,1   1,2   5,1 16,5        36,0
                   Röhm
Mannanase KMB, g/l Enzyme          100803            -    1,8   2,8   1,4   2,3 16,7        25,1
* nowadays Röhm Enzyme
- less than 0,1 g/l or 0,01 mg/100 mg
                    Results and recommendations from the NSP Network group Carbohydrate analysis

                                                           Appendix Fel! Hittar inte referenskälla.   7




5             References

i
      Tenkanen M , Hausalo T , Siika-aho M , Buchert J , Viikari L, 8th
      International Symposium on Wood and Pulping Chemistry, Helsinki,
      Finland, June 6-9, 1995, Vol. III, pp 189-194.
ii
      Dahlman O , Jacobs A , Liljenberg A , Ismail Olsson A, J. Chromatogr. A,
      Vol. 891, 2000, pp 157-174.
iii
      Syverud K , Moe S T , Minja R J A , 10th International symposium on
      wood and pulping chemistry, Yokohama, Japan, June 7-10, 1999, vol. 2, pp
      232-237.
iv
      Vaaler D , Syverud K , Moe S T, The 3rd biennial Johan Gullichsen
      colloquium, Espoo, Finland, Sep13-14, 2001, pp 87-93.
v
      Hausalo T, 8th International Symposium on Wood and Pulping Chemistry,
      Helsinki, Finland, June 6-9, 1995, Vol. III, pp 131-136.

				
DOCUMENT INFO
Shared By:
Stats:
views:27
posted:8/23/2011
language:English
pages:56
Description: Carbohydrates are the body's basic energy supply of material, energy sports training provided mainly by glycogen, so we have to add enough glycogen. The most effective way is doing fitness training before breakfast before the game or when dietary intake of carbohydrates for energy to provide adequate training to prepare. Meanwhile, in order to get a good training effect, need to pay attention to the campaign before, during and after the added sugar drinks.