Hot Water Extraction and Subsequent Kraft Pulping of Pine Wood Chips 
 
by 
 
Allen J. Smith 
 
 
 
 
A dissertation submitted to the Graduate Faculty of 
Auburn University 
in partial fulfillment of the 
requirements for the Degree of 
Doctor of Philosophy 
 
Auburn, Alabama 
May 9, 2011 
 
 
 
 
Keywords: prehydrolysis, biorefinery, pulp, 
hemicellulose, sugar, 
 
 
Copyright 2011 by Allen Joseph Smith 
 
 
Approved by 
 
Gopal Krishnagopalan, Chair, Professor Emeritus of Chemical Engineering 
Harry Cullinan, Director of Alabama Center for Paper and Bioresource Engineering 
Steve Duke, Associate Professor of Chemical Engineering 
Oladiran Fasina, Associate Professor of Biosystems Engineering 
 
 
 
 
ii 
 
 
 
Abstract 
This research investigated the effects of pre-extraction time, temperature, and pH 
on the rate and quantity of sugars recovered from Loblolly pine wood chips and the 
impact this treatment has on pulp kappa number and yield.  Composition analysis of raw 
chips, extracted chips, pulp, and hydrolyzate were performed to assess the hemicellulose 
extraction efficiency and subsequent loss during kraft pulping of Loblolly pine.  A 
pseudo kinetic model of prehydrolysis extraction was developed from the hydrolyzate 
sugar concentration data.   
Preliminary experiments demonstrated the potential for influencing pulp 
properties and sugar recovery in hydrolyzate through on-line control of prehydrolysis pH, 
reaction temperature, and time.  A second set of experiments examined these factors and 
added a presoak period to the design matrix.  A third set of experiments used a best case 
of pre-extraction conditions to test five potential pulping additives. 
 It was concluded that the extraction rate for all sugars was increased with either 
increasing temperature from 140 to 170?C or decreasing pH from 4.5 to 3.0.  The 
hydrolysis was selective for hemicellulose as opposed to cellulose by using temperature 
at 140?C.  Both pH and temperature also impacted the degradation rate of sugars in 
solution.  The 24 hour presoak at 25?C with various pH levels had no measurable effect 
on hydrolysis rate or pulp yield.  
iii 
 
The additives tested in this research:  anthraquinone, acetaldehyde, ethanolamine, 
lithium aluminium hydride, and hydroxylamine, were not successful in recovering pulp 
yield from extracted chips to that of a standard kraft cook with the conditions tested.  
More work could be justified with hydroxylamine or to test the use of anthraquinone in 
conjunction with hydroxylamine or another successful additive.   
A pseudokinetic model of the extraction was calculated using an activation energy 
of 27 kcal/mole for hemicellulose hydrolysis and a term for the acid concentration.  This 
modified H-factor model described the sugar extraction data except when significant 
degradation of sugars was observed.  The data for chip weight loss and pulp yield also fit 
a smooth curve when plotted against the modified H-factor.   
 
 
 
 
 
 
 
 
 
 
iv 
 
 
 
Acknowledgments 
Dr. Gopal has been not just my academic advisor but also a friend from the time 
of my undergraduate experience at Auburn University from 1985-89 through my 
professional experience in the pulp and paper industry.  I have not always heeded his 
advice, but I typically have seen its wisdom in hindsight.   
Dr. Harry Cullinan was particularly helpful in the latter stages of this project as I 
was exploring different potential forms for the model.  The work seemed to get more and 
more complex each week until he reminded me of Occam?s razor. 
I had many valuable discussions with Dr. Sung Hoon Yoon in the time that we 
were both working in Wilmore 156.  Some of them involved prehydrolysis experiments.   
Dr. Duke and Dr. Fasina both provided valuable comments about the dissertation 
proposal and this document. 
Funding for this project was provided by the Auburn University Natural 
Resources Management and Development Institute and the Masada Resource Group, 
LLC through the Alabama Center for Paper and Bioresource Engineering. 
The Rock ?Tenn Company supplied the pine chips. 
Dr. Y. Y. Lee?s group at Auburn University assisted with the carbohydrate 
composition analysis.  In particular, Rajesh Gupta and Kang Li trained this investigator in 
preparation of samples for analysis on the HPLC and in reading the chromatograms.   
v 
 
Dr. Saeed Maghsoodloo assisted with interpretation of statistical data and basic 
instruction in the use of the Minitab statistical software program. 
Many people have encouraged me during this process and I will not attempt to 
name all that have kept me going.  In particular Martha Evans has expressed confidence 
in me for the 30 years that I have known her as a teacher and friend.  She has been 
particularly diligent in this over the past five years.   Jim Killian and Bill Josephson have 
had exactly the right words at times when I felt that I would not have the will to finish.   
Janine and Elizabeth were children when I started this project and are now 
beautiful young women.  They have reversed roles and been an example to me of getting 
their homework done and performing well in school. 
Tony and Mary Ruth Smith have been just who parents should be for a child 
working through graduate studies even if that child has children of his own.  Thanks 
Mom for refocusing me when I got distracted.  Dad, your insightful and irreplaceable 
comments on this document completely reshaped the tone into what it should be. 
Penny has stood beside me, in front of me, behind me? all the time showing me 
that she loves me just for who I am even though she doesn?t always agree with what I do. 
I hope that my love in return is as apparent.
vi 
 
 
 
Table of Contents 
Abstract ............................................................................................................................... ii 
Acknowledgments.............................................................................................................. iv 
List of Tables ...................................................................................................................... x 
List of Figures .................................................................................................................... xi 
List of Abbreviations ..................................................................................................... xviii 
Chapter 1  Introduction ....................................................................................................... 1 
1.1 Research Synopsis .................................................................................................... 1 
1.2 Perspective ................................................................................................................ 2 
1.3 Dissertation Organization ......................................................................................... 4 
1.4 Biorefineries .............................................................................................................. 5 
1.5 Biomass Feedstock ................................................................................................. 10 
1.6 Pulping Methods ..................................................................................................... 18 
Chapter 2  Review of Previous Work ............................................................................... 28 
2.1 Dissolving Pulp ....................................................................................................... 28 
2.2 Alcohol Production from Wood Hydrolysis ........................................................... 34 
2.3 Value Prior to Pulping ............................................................................................ 36 
Chapter 3  Experimental Apparatus and Procedures ........................................................ 41 
3.1 Materials and Equipment ........................................................................................ 41 
3.2 Experimental Procedures ........................................................................................ 46 
vii 
 
3.3 Analytical Procedures ............................................................................................. 50 
Chapter 4  Prehydrolysis of Pine Wood Chips ................................................................. 51 
4.1 Preliminary Work ................................................................................................... 51 
4.2 Second Preliminary Work ....................................................................................... 53 
4.3 pH Controlled Prehydrolysis of Pine Wood Chips ................................................. 55 
4.4 Material Balances ................................................................................................... 61 
4.5 Prehydrolysis Conclusions ...................................................................................... 66 
Chapter 5  Prehydrolysis Designed Study ........................................................................ 67 
5.1 Designed Experiment Factors ................................................................................. 68 
5.1.1 Variation of Prehydrolysis Time ...................................................................... 68 
5.1.2 Variation of Temperature ................................................................................. 68 
5.1.3 Variation of pH ................................................................................................ 69 
5.1.4 Variation of Presoak Conditions ...................................................................... 70 
5.2 Hydrolysis Target ................................................................................................... 71 
5.3 Experimental Design ............................................................................................... 72 
5.4 Results ..................................................................................................................... 77 
5.4.1 Prehydrolyzate Composition ............................................................................ 77 
5.4.2 Chip Weight Loss ............................................................................................ 94 
5.4.3 Pulp Yield ........................................................................................................ 96 
5.4.4 Additional Cooks ........................................................................................... 101 
Chapter 6  Pulp Yield Recovery ..................................................................................... 108 
6.1 Anthraquinone ...................................................................................................... 111 
6.2 Acetaldehyde ........................................................................................................ 117 
viii 
 
6.3 Ethanolamine ........................................................................................................ 120 
6.4 Lithium Aluminium Hydride ................................................................................ 125 
6.5 Hydroxylamine ..................................................................................................... 128 
6.6 General Yield Improvement Conclusions ............................................................. 140 
Chapter 7  Value Prior to Pulping Model ....................................................................... 141 
7.1 Historical Development ........................................................................................ 143 
7.1.1 Wood hydrolysis or prehydrolysis models .................................................... 143 
7.1.2Kraft pulping models ...................................................................................... 147 
7.2 Model Development ............................................................................................. 150 
7.2.1 The Modified H-factor ................................................................................... 150 
7.2.2 Validation ....................................................................................................... 168 
7.2.3 Additional Improvements Required ............................................................... 169 
Chapter 8  Conclusions and Suggestions for Future Research ....................................... 170 
8.1 Conclusions from this Research ........................................................................... 170 
8.1.1 Prehydrolysis Experiments ............................................................................ 172 
8.1.2 Yield Improvement ........................................................................................ 173 
8.1.3 Modeling Results ........................................................................................... 173 
8.2 Suggestions for Further Investigation ................................................................... 174 
8.2.1 Prehydrolysis Variables ................................................................................. 174 
8.2.2 Additional Yield Improvement Research ...................................................... 176 
8.2.3 Model Enhancement ...................................................................................... 178 
8.2.4 Improvements to Testing or Experimental Procedures .................................. 181 
8.3 Final Comments .................................................................................................... 182 
ix 
 
Bibliography ................................................................................................................... 184 
Appendix 1  Minitab 15 Statistical Results..................................................................... 193 
Appendix 2  Non-Standard Test Methods ...................................................................... 195 
A2.1 ABC Liquor test ................................................................................................. 195 
A2.2 Black Liquor Residual Alkali Test .................................................................... 196 
Appendix 3  Cook Worksheets and Calculations ........................................................... 197 
 
  
x 
 
 
 
List of Tables 
Table 1  Loblolly pine wood composition measurements. ............................................... 42 
Table 2  Macromolecule composition estimation matrix. ................................................ 43 
Table 3  Preliminary study, selected conditions and results. ............................................ 53 
Table 4  Latin square experimental plan. .......................................................................... 67 
Table 5  Latin square conditions for prehydrolysis screening. ......................................... 73 
Table 6  Latin square experiment order and labeling. ...................................................... 74 
Table 7  Conditions and results for additional prehydrolysis cooks. .............................. 102 
Table 8  Treatment conditions and results for ethanolamine treatment. ......................... 123 
Table 9  Hydroxylamine treatment conditions and results. ............................................ 136 
Table 10  Hydrolysis activation energy values from literature. ...................................... 147 
 
  
xi 
 
 
 
List of Figures 
Figure 1  2007 U.S. energy flow diagram (Department of Energy 2008). ......................... 3 
Figure 2  Potential biorefinery pathways (Thorp, Thorp and Thorp 2008). ..................... 10 
Figure 3  Cellulose structure (Sjostrom 1993, 54). ........................................................... 12 
Figure 4  Galactoglucomannan structure (Sjostrom 1993, 65). ........................................ 13 
Figure 5  Arabinoglucuronoxylan structure (Sjostrom 1993, 65). .................................... 14 
Figure 6  Segment of softwood lignin structure (Sjostrom 1993, 84). ............................. 15 
Figure 7  Kraft recovery cycle (Sjostrom 1993, 161). ...................................................... 25 
Figure 8  M/K Digester configuration. ............................................................................. 44 
Figure 9  Extracted sugar concentration as a function of time for water hydrolysis. ....... 51 
Figure 10  Pulp kappa number as a function of cooking H-factor.  Legend indicates pre-
treatment time. ...................................................................................................... 54 
Figure 11  Prehydrolysis pH and temperature profiles.  Legend indicates initial chemical 
addition. ................................................................................................................ 56 
Figure 12  Pulp yield as a function of kappa number for all prehydrolysis runs (Legend 
identifies prehydrolysis time and initial chemical charge if present). .................. 57 
Figure 13  Sugar recovery as a function of H-factor for water hydrolysis. ...................... 58 
Figure 14  Sugar recovery as a function of H-factor for prehydrolysis with an initial 
charge of 0.1% NaOH followed by control with 0.5 N NaOH to pH 5. ............... 59 
xii 
 
Figure 15  Sugar recovery as a function of H-factor for prehydrolysis with an initial 
charge of 0.3% NaOH followed by control with 0.5 N NaOH to pH 5. ............... 60 
Figure 16  Sugar recovery as a function of H-factor for prehydrolysis with an initial 
charge of 3% acetic acid followed by control with 0.5 N NaOH to pH 5. ........... 60 
Figure 17  Glucan disposition for water hydrolysis. ......................................................... 61 
Figure 18  Glucan disposition for pH controlled hydrolysis. ........................................... 62 
Figure 19  Xylan disposition for water hydrolysis............................................................ 63 
Figure 20  Xylan disposition for pH controlled hydrolysis. ............................................. 63 
Figure 21  Mannan disposition for water hydrolysis. ....................................................... 64 
Figure 22  Mannan disposition for pH controlled hydrolysis. .......................................... 64 
Figure 23  Total disposition for water hydrolysis. ............................................................ 65 
Figure 24  Total disposition for pH controlled hydrolysis. ............................................... 65 
Figure 25  Hydrolyzate sugar vs. chip weight loss. .......................................................... 72 
Figure 26  Kappa number as a function of H-factor for raw chips. .................................. 75 
Figure 27  Pulp yield as a function of kappa number for raw chips. ................................ 76 
Figure 28  Sugar recovered as a function of time for prehydrolysis cook "H" pH 4.5, 
140?C, 10.5 pH Soak. ........................................................................................... 79 
Figure 29  Sugar recovered as a function of time for prehydrolysis cook "D" pH 3.5, 
140?C, DIW Soak. ................................................................................................ 80 
Figure 30  Sugar recovered as a function of time for prehydrolysis cook "B" pH 4.5, 
155?C, DIW Soak. ................................................................................................ 81 
Figure 31  Sugar recovered as a function of time for prehydrolysis cook "I" pH 3.0, 
140?C, pH 3.1 Soak. ............................................................................................. 81 
xiii 
 
Figure 32  Sugar recovered as a function of time for prehydrolysis cook "C" pH 4.5, 
170?C, pH 3.1 Soak. ............................................................................................. 82 
Figure 33  Sugar recovered as a function of time for prehydrolysis cook "A" pH 3.0, 
170?C, DIW Soak. ................................................................................................ 83 
Figure 34  Sugar recovered as a function of time for prehydrolysis cook "F" pH 3.5, 
170?C, pH 10.5 Soak. ........................................................................................... 83 
Figure 35  Sugar recovered as a function of time for prehydrolysis cook "E", pH 3.5, 
155?C, pH 3.1 Soak. ............................................................................................. 84 
Figure 36  Sugar recovered as a function of time for prehydrolysis cook "G", pH 3.0, 
155?C, pH 10.5 Soak. ........................................................................................... 85 
Figure 37  Total sugar recovered as a function of time for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). ......................... 86 
Figure 38  Arabinose recovered as a function of time for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ........................................ 87 
Figure 39  Galactose recovered as a function of time for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ........................................ 88 
Figure 40  Glucose recovered as a function of time for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ........................................ 89 
Figure 41  Mannose recovered as a function of time for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ........................................ 90 
Figure 42  Xylose recovered as a function of time for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ........................................ 91 
xiv 
 
Figure 43  Glucose and mannose recovered as a function of time compared for four 
conditions (Legend indicates trial label, temperature, reaction pH, and presoak 
pH.  Glucose has solid connections.  Mannose connections are dashed.). ........... 92 
Figure 44  Chip weight loss in prehydrolysis as a function of reaction temperature for all 
conditions (Legend indicates trial label, temperature, reaction pH, and presoak 
pH). ....................................................................................................................... 94 
Figure 45  Recovered sugar as a function of chip weight loss for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). ......................... 95 
Figure 46  Pulp lignin content as a function of H-factor for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). ......................... 98 
Figure 47  Yield as a function of kappa number for all conditions (Legend indicates trial 
label, temperature, reaction pH, and presoak pH). ............................................. 100 
Figure 48  Pulp yield as a function of recovered sugar for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 101 
Figure 49  pH and temperature for prehydrolysis cooks "E" and "J". ............................ 103 
Figure 50  Recovered sugar measurements as a function of time for prehydrolysis cooks 
"P", "Q", and "R".   Individual sugars from top to bottom based on final data 
points are mannose, xylose, galactose, arabinose, and glucose. ......................... 107 
Figure 51  Kappa number as a function of H-factor for AQ cooks (Legend indicates trial 
label, temperature, reaction pH, and presoak pH). ............................................. 114 
Figure 52  Pulp yield as a function of kappa number for AQ cooks (Legend indicates trial 
label, temperature, reaction pH, and presoak pH). ............................................. 115 
xv 
 
Figure 53  Pulp yield as a function of kappa number for additional cooks with and 
without AQ.  Legend indicates trial label with the modifiers ?AQ? for inclusion of 
AQ in the kraft cook and ?AQ4? for modification of the temperature control 
profile. ................................................................................................................. 116 
Figure 54  Yield as a function of kappa number for acetaldehyde treated and untreated 
prehydrolyzed chips. ........................................................................................... 119 
Figure 55  Yield as a function of kappa number for ethanolamine treatment. ............... 124 
Figure 56  Yield as a function of kappa number for lithium aluminium hydride treatment.
............................................................................................................................. 127 
Figure 57  Reaction pathways with hydroxylamine (Clayton and Jones 1970). ............ 132 
Figure 58  Pulp yield as a function of kappa number with hydroxylamine treatment. ... 137 
Figure 59  Chip weight loss from prehydrolysis as a function of H-factor. ................... 151 
Figure 60  Total sugar recovered as a function of H-factor (Legend indicates trial label, 
temperature, reaction pH, and presoak pH). ....................................................... 152 
Figure 61  Total sugar recovered at pH 3.0 as a function of H-factor with EA = 27 
kcal/mole (Legend indicates trial label, temperature, reaction pH, and presoak 
pH). ..................................................................................................................... 153 
Figure 62  Sugar recovered at pH 3.5 as a function of H-factor with EA = 27 kcal/mole 
(Legend indicates trial label, temperature, reaction pH, and presoak pH). ........ 154 
Figure 63  Total sugar recovered at pH 4.5 as a function of H-factor with EA = 27 
kcal/mole (Legend indicates trial label, temperature, reaction pH, and presoak 
pH). ..................................................................................................................... 154 
xvi 
 
Figure 64  Total sugar recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 156 
Figure 65  Glucose recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 158 
Figure 66  Xylose recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 159 
Figure 67  Galactose recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 160 
Figure 68  Arabinose recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 161 
Figure 69  Mannose recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). ...................................... 161 
Figure 70  Chip weight loss from prehydrolysis as a function of modified H-factor. .... 162 
Figure 71  Prehydrolyzed kraft pulp yield at 30 kappa as a function of modified H-factor.
............................................................................................................................. 163 
Figure 72  Pulp yield after "standard" cook as a function of modified H-factor. ........... 164 
Figure 73  Kappa number after "standard" cook as a function of modified H-factor. .... 164 
Figure 74  Modified H-factor model equations for Polymath 6.10. ............................... 165 
Figure 75  Total recovered sugars compared to predicted values as a function of time for 
selected conditions (Legend indicates trial label, temperature, reaction pH, and 
presoak pH). ........................................................................................................ 166 
Figure 76  Model predictions and experimental values as a function of modified H-factor 
(Legend indicates trial label, temperature, reaction pH, and presoak pH). ........ 167 
xvii 
 
Figure 77  Total recovered sugar as a function of modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH. Data labeled ?wtr 51208? is 
validation data).................................................................................................... 169 
 
  
xviii 
 
 
 
 
List of Abbreviations 
AA  Active Alkali, NaOH + Na2S 
AQ  Anthraquinone 
AHQ  Anthrahydroquinone 
ARA  Arabinose 
ASABE American Society of Agricultural and Biological Engineers 
AU  Auburn University 
bd   Bone dry 
CTO  Chief Technology Officers 
DF  Degrees of Freedom 
DIW  Distilled Deionized Water 
Dp  Degree of Polymerization 
EA  Effective Alkali, NaOH + Na2S/2 
EISA  Energy Independence and Security Act of 2007 
FPL  Forest Products Laboratory 
FUR  Furfural 
GAL  Galactose 
GHG  Greenhouse Gas 
GLU  Glucose 
HMF  Hydroxymethyl furfural 
xix 
 
MAN  Mannose 
REA  Residual Effective Alkali 
SS  Sum of Squares 
SUNY ? ESF State University of New York College of Environmental Science and  
Forestry 
 
tpd  Tons per day 
TTA  Total Titratable Alkali, NaOH + Na2S + Na2CO3 
VPP  Value Prior to Pulping 
XYL  Xylose 
Mathematical Abbreviations: 
    Carbohydrate or hemicellulose concentration 
     Initial carbohydrate or hemicellulose concentration 
    Acid concentration 
Ea  Activation energy 
R  Ideal gas constant 
    Rate constant 
     Combined rate constant at constant pH 
    Rate constant at reference conditions of temperature and pH 
     Rate constant at reference pH and constant temperature 
T  Process temperature 
    Reference temperature 
 
 
 
1 
 
 
 
Chapter 1  Introduction 
1.1 Research Synopsis 
The purpose of this research was to investigate the effects of varying hot water 
extraction time, temperature, and pH on the rate and quantity of hemicellulose sugars 
recovered from Loblolly pine wood chips and the effects this prehydrolysis treatment has 
on kraft pulp yield and the severity of kraft cooking required to reach a pulp kappa 
number target.  The target sugars are currently chemically degraded and/or burned at a 
low heating value as a component of black liquor in the industrial kraft pulping and 
chemical recovery process.  Recovery of the sugars prior to pulping would allow them to 
be redirected to a better end use.  Composition analysis of raw chips, extracted chips, 
pulp, and hydrolyzate were performed to assess the hemicellulose extraction efficiency 
and subsequent hemicellulose yield loss during kraft pulping of Loblolly pine.  On line 
control of the extraction pH and temperature were used to influence the quantity of 
hemicellulose recovered in the hydrolyzate and the yield and properties of the final pulp.  
Selected conditions of extraction and cooking were then chosen for evaluation of pulp 
yield improvement additives. 
Reports of previous efforts by other researchers and preliminary work 
accomplished in this study identified significant reductions in pulp yield and strength 
following kraft cooking of extracted chips.  The reduction in strength is particularly 
problematic with southern pine pulp since it is included in paper and board products 
2 
 
primarily for its strength contribution.  Additionally, pine lignin is more sensitive to 
deactivation (condensation reactions) in acid conditions than hardwood lignin.  
Therefore, severe extraction conditions will force stronger cooking conditions further 
reducing yield.  Economics of the process are also expected to be more difficult with pine 
than hardwoods since there is a lower proportion of hemicellulose in the native pine 
wood.  Successful conversion to ethanol of 50% of hemicellulose (12% of the wood 
weight) from pine wood that is currently processed to kraft pulp represents approximately 
280 million equivalent gallons of gasoline per year.   This could replace 0.2% of the 2010 
U.S. consumption of 138,500 million gallons of gasoline per year (EIA U.S. Petroleum 
Product Supplied 2010).  
Preliminary experiments illustrated the potential for influencing pulp yield and 
sugar recovery in hydrolyzate liquids by manipulation and control of pH, reaction 
temperature, and time of prehydrolysis.  A second set of experiments contained these 
influences and added a presoak period to the design matrix.  A third set of experiments 
used a chosen best case of pre-extraction conditions to test the potential advantage of 
pulping additives in increasing pulp yield. 
1.2 Perspective 
The United States? dependence on fossil fuel for energy and, in particular, on 
petroleum for transportation fuels has led to instability in economics, potentially 
devastating environmental effects, and difficult foreign policy situations.   These difficult 
circumstances could all be improved by substituting domestic renewable energy sources 
in the place of imported fossil fuels.  In 2007, the USA consumed 102 quadrillion Btu 
(Quads) of energy with 85% of this being derived from fossil sources.  Transportation 
3 
 
consumed 29 Quads of energy with most of this in the form of liquid fuels derived from 
crude oil or petroleum (Department of Energy 2008).  Figure 1 shows an energy flow 
diagram for the United States including energy supply by source and consumption by 
category (Department of Energy 2008).  Renewable sources including solar, wind, 
biomass, hydropower, and geothermal made up less than 7% of the energy consumed 
(Energy Information Administration 2008).   
Figure 1  2007 U.S. energy flow diagram (Department of Energy 2008). 
 
Any of these renewables can fuel electricity production at stationary sources with 
existing technology, but biomass is uniquely suited to replace petroleum transportation 
fuel needs at least until battery and fuel cell technologies are fully matured.  Biomass can 
be refined into energy dense alcohols or alkanes for substitution into liquid fuels markets 
4 
 
or even gasified to syngas and burned directly in specially modified engines (Whitley 
2008).  The U. S. Departments of Energy and Agriculture published a report now known 
as ?The Billion Ton Report? in 2005 with the aim of estimating the capability of 
sustainably harvested biomass to replace 30% of current U.S. petroleum consumption.  
The report concludes that this can be achieved with fairly conservative assumptions but 
not without considerable effort (Perlack, et al. 2005).  This research contributes to that 
effort by investigating a potential method of improving the utilization of an industrial 
biomass byproduct.   
1.3 Dissertation Organization 
This report is organized into five sections: 
Chapter 1 introduces biorefineries as they exist in 2010 and briefly discusses how 
they are expected to progress over the next twenty years.  The various biomass feedstock 
materials that are available for conversion to energy are discussed with an emphasis on 
wood and the chemistry of wood that is pertinent to this research.  This is followed by a 
brief discussion of the history of processing wood into paper and the current state of the 
art of kraft pulping.  Chapter 2 is a review of publications related to the specific area of 
biorefining that is the topic of this research, Value Prior to Pulping (VPP). 
Chapters 3, 4, and 5 present laboratory methods, experiments, and results that 
were performed to investigate prehydrolysis pulping of loblolly pine with regard to VPP. 
Chapter 6 includes a review of literature regarding chemical additives for pulp 
yield improvement and lab tests that were performed to screen several potential additives 
when pulping prehydrolyzed chips. 
5 
 
The development of mathematical models to predict sugar recovery from the 
prehydrolyzate is presented in Chapter 7 along with some review of literature in this area. 
Overall conclusions and recommendations for future work are in Chapter 8. 
1.4 Biorefineries 
A biorefinery is ?A facility that uses mechanical, thermal, chemical, and/or 
biochemical processes to convert biomass into value-added biobased products or key 
intermediates for the production of chemicals and other materials.? (ASABE S593 2006)  
Common use of the term often implies that fuel or energy must be a primary and 
potentially salable product analogous to an oil refinery and this convention is followed 
here.  A biorefinery is then viewed as a biomass processing facility that will help the 
USA reduce its dependence on fossil fuels and in particular imported petroleum.  Current 
production of biofuels consists primarily of ethanol from starch or sugar and biodiesel 
from vegetable oils or animal fats.  Energy efficiency and economics can be improved in 
all of these processes by proper utilization of byproduct materials and rejected heat.   
Ethanol is produced from sugar cane or sugar beets by fermentation with yeast to 
form ?beer? followed by distillation with heat to concentrate the ethanol.  Further 
purification with a molecular sieve can reduce water content if necessary.  When the 
starting material is starch from grain, the additional step of hydrolysis of starch to 
dextrose with enzymes or weak acid is required.  Additional pretreatments are required to 
prepare other plant materials for fermentation.  There are presently 204 ethanol refineries 
either operating or under construction in the USA with an operating capacity of 13,823 
million gallons per year (mgy) and total capacity of 14,295 mgy (Renewable Fuels 
Association 2011).  Most of these plants use corn grain as the feedstock, but milo, barley, 
6 
 
wheat starch, and alternative sources including cheese whey, potato waste, wood waste, 
waste beer, beverage waste, and sugar cane bagasse are also minimally represented.     
Biodiesel is typically produced by lye-catalyzed transesterification of the oil with 
methanol.  In April 2009, there were 137 biodiesel plants with capacity of 2285 mgy that 
used mostly vegetable oil feedstock but some could use animal fats or multiple 
feedstocks. (Biodiesel Magazine 2009)  Estimated actual production was much lower 
than capacity due to shortage of feedstock and poor economics.  In addition to the 
capacity listed above, 46 biodiesel plants were idle with a potential capacity of 594 mgy.   
The direction of biofuel production in the USA over the next 10 years will likely 
be influenced heavily by the Energy Independence and Security Act of 2007 (EISA).  
This act mandates the production of renewable fuels to increase from 9 billion gallons of 
conventional renewable biofuels in 2008 to a total of 36 billion gallons of renewable fuels 
in 2022 with a maximum of 15 billion gallons from conventional biofuels and the rest 
from advanced biofuels.  ?Conventional biofuel? includes ethanol produced from corn 
starch, but plants constructed after 2008 must also demonstrate greenhouse gas (GHG) 
emission reductions of at least 20% when compared to a baseline.  ?Advanced biofuel? 
must be made from renewable biomass other than corn starch and must demonstrate a 
50% reduction in GHG emissions.  ?Cellulosic biofuel? is an advanced biofuel made 
from renewable lignocellulosic feedstock that meets the higher standard of 60% GHG 
emission reduction (Renewable Fuels Association 2008).   
Production processes for cellulosic biofuels can generally be divided into two 
major categories:  the biochemical technologies or ?sugar platform? and thermochemical 
technologies.  Processes that fall under the sugar platform separate the sugar monomers 
7 
 
from the rest of the biomass material and ferment the sugars into ethanol.  The non-sugar 
biomass may be burned to fuel the plant or put to some other use, but the maximum yield 
of biofuel depends on the amount of sugar in the original feedstock and is typically about 
90 gallons of ethanol (equivalent to 60 gallons of gasoline) per ton of biomass.  
Thermochemical processes include gasification and other processes that use heat to break 
down all of the organic material in the biomass to small, simple molecules and then either 
burn this mixture directly for fuel or process it further into fuels or chemicals.  Biofuel 
yield in this case depends on the process but could be upward of 100 gallons of gasoline 
equivalent fuel per ton of biomass.   
There has been much written recently about converting ?pulp mills? into 
?biorefineries?.  A series of articles in Paper 360? discussed the reasons for integrating 
biorefineries with existing pulp mills (Thorp, Thorp and Murdock-Thorp 2008), the 
pathway options (Thorp, Thorp and Thorp 2008), the status of current projects (Thorp, 
Thorp and Murdock-Thorp 2008), and summarizes the economics associated with a 
hypothetical gasifier and gas-to-liquids plant (B. A. Thorp 2008). The paper and timber 
industries can be viewed as biorefineries using the ASABE S593 definition above, but 
they have really been thought of as single product lumber mills or paper mills.  The 
facilities are well suited for transition from mills to biorefineries since they have a long 
history of experience effectively managing large quantities of biomass including planting, 
harvesting, transportation, storage, and processing into finished products often with fuel 
or energy as a byproduct.  This reinvention of the pulp mill will hopefully lead to higher 
material and energy efficiencies, reduced impact on the environment and a sustainable 
future for the pulp industry in North America even though they already have a history of 
8 
 
efficient and complete use of their raw material.  Another pair of articles discussed the 
technological options (Agenda 2020 CTO Committee Working Group 2008) and business 
case (Agenda 2020 CTO Committee Working Group 2008) for a stepwise 
implementation of a thermochemical biorefinery at a pulp mill.  The final mill would be 
fossil fuel energy independent and carbon neutral in addition to producing pulp and paper 
products and renewable energy. The business case shows the plant to be profitable 
although the capital intensity is high relative to an analogous fossil fuel plant and the high 
costs of transporting low energy density biomass to the mill limits the size of the facility.    
There are several potential ways to implement a biorefinery concept at a pulp mill 
that vary from stepwise modification of existing processes to wholesale replacement of 
sections of the mill.  Installation of a hog fuel gasifier to produce syngas to fuel the lime 
kiln is an example of a small project while modification of the pulp mill or recovery 
island to produce byproduct ethanol, acetic acid, or Fischer-Tropsch liquids currently 
have more risk technologically and financially. Wood residue gasifiers have been 
installed to supply kiln fuel since the 1980s but have struggled with low reliability.  Most 
of the problems are associated with the feed systems and dryers that are required to raise 
the heating value of the wood residue (Francey, Tran and Jones 2009). Syngas has a 
lower heating value than the typical kiln fuels natural gas and fuel oil.   
The Flambeau River Biofuels project that is expected to be complete in 2013 but 
has some funding problems (Brochu 2010) is an example of a bigger biorefinery project.  
The AVAPtm process, a proprietary ethanol solvent pulping and recovery system, was 
originally proposed for this project (Retsina and Pylkkanen 2007).  If constructed as 
planned in 2009, however, the project will gasify 1000 bone dry tons per day (bd tpd) of 
9 
 
forest residuals and agricultural waste.  The syngas will be used to produce renewable 
biodiesel and paraffin via the Fisher-Tropsch process while excess process heat will be 
used in the adjacent pulp and paper mill (Patrick 2009).  Gasification of black liquor is 
now ready to move past the demonstration stage and be applied at commercial scale 
(Barry and Leblanc 2009).  Several projects at various stages of development were briefly 
described in Paper 360? (Thorp, Thorp and Murdock-Thorp 2008).   
Twelve different potential pathways for implementation of the biorefinery 
concept that are in various states of conceptualization to commercialization are illustrated 
in Figure 2 (Thorp, Thorp and Thorp 2008). These pathways each have advantages or 
disadvantages based on how close they are to commercialization, expected capital 
requirement, and potential energy transformation efficiency.  As an example, paths that 
follow the sugar platform to produce ethanol are in general further along technologically 
and have received a majority of research funding, but have lower potential for energy 
conversion than paths that can make use of lignin and other organics in addition to 
fermentable sugars.  Pathway #8 of this biorefinery concept is extracting ?Value Prior to 
Pulping? (VPP) which involves pretreating chips to remove valuable chemicals like 
acetic acid or fermentable sugars before they would be degraded by alkali in cooking the 
chips to produce pulp.  An ideal implementation of VPP would produce no changes in 
pulp yield or properties from a standard pulp.  The only change would be that some of the 
wood chemicals that are not retained in the pulp would be intentionally directed out of 
the black liquor to some higher economic end use.  Kraft pulping presently removes and 
degrades a significant percentage of hemicellulose and small amounts of cellulose from 
10 
 
the wood.  These sugar degradation products are burned in the chemical recovery unit 
along with lignin to produce steam for mill operations.  
 
 
1.5 Biomass Feedstock 
Most of the biorefineries that currently produce liquid fuel as the primary product 
make ethanol from starch or sugar or biodiesel from vegetable oils.  These processes may 
be close to their production limit politically due to conflicts over using food and farmland 
to produce fuel.  Biofuel production has been blamed for higher food costs worldwide 
and for deforestation in South America and South East Asia as palm oil plantations 
replace peat bogs and soybeans replace rain forest (Grunwald 2008).  Producing biomass 
 
S u gar  1  
B i o m a s s  o r  s o r t e d M S W  
A c i d H y dr o l y s i s  
F e r m e n t a t i o n  
( B l ue  F i r e  E t h a n o l )  
S u gar  2  
B i o m a s s  
E n z ym e  H y dr o l y s i s  
F e r m e n t a t i o n  
( P o e t &  DuP o n t )  
1  &  2  3  &  4  5  6  
T h e r m a l  1  
B i o m a s s  
Ga s i f i c a t i o n  
GT L + E  
( F R B E )  
F e r m e n t a t i o n  
( A L I C O)  
T h e r m a l  2  
B l a c k L i quo r  
Ga s i f i c a t i o n  ( No r a m p a c )  
GT L  F e r m e n t a t i o n  
S u gar  3 ( He m ice ll u l o s e )  
P ul pwoo d  
P ul p ing  
E t h a n o l  F e r m e n ta t i o n  
( A V A P  pr oc e s s )  
P ul p  
S e pa r a t i o n  L igni n  
7  8  
S u gar  4 ( He m ice ll u l o s e )  
E x tr a c t io n  
F e r m e n ta t i o n  
E t h a n o l  
( VPP )  
P ul p  
P ul p ing  
B l a c k  
L iq u o r  
an d  
9  
T h e r m al 1  
B i o m a s s  
Ga s if i c a t i o n  
GT L + E  
( F R B F )  
F e r m e n ta t i o n  
( AL I C O)  
S u gar  4 ( He m ice ll u l o s e )  
E x tr a c t io n  
F e r m e n ta t i o n  
E t h a n o l  
( VPP )  
P ul p  
P ul p ing  
B l a c k  
L iq u o r  
an d  
10  
S u gar  4 ( Hem ice ll u l os e )  
E x t r a c t i o n  
F e r m e n t a t i o n  
E t h a n o l  
( VPP )  
P u l p  
P u l p i ng  
B l a c k  
L i quo r  
T h e r m a l  2  
B l a c k L i quo r  
Ga s i f i c a t i o n  ( No r a m p a c )  
GT L +  
e n e r g y  
F e r m e n t a t i o n  
an d  
11  
T h e r m a l  1  
B i o m a s s  
Ga s i f i c a t i o n  
GT L + E  
( F R B F )  
F e r m e n t a t i o n  
( A L I C O)  
T h e r m a l  2  
B l a c k L i quo r  
Ga s i f i c a t i o n  ( No r a m p a c )  
GT L + E  F e r m e n t a t i o n  
or  or  
12  
S u gar  5 ( Hem ice ll u l os e )  
W oo d y  B i o m a s s  
S e pa r a t i o n  
P ur e  L i g ni n  F e r m e n t a t i o n  
an d  
Figure 2  Potential biorefinery pathways (Thorp, Thorp and Thorp 2008). 
11 
 
for fuel on existing crop land can cause a net increase in greenhouse gas emissions due to 
carbon emissions associated with land use change.  This increase in emissions is greatest 
if using croplands for energy requires clearing new land areas for displaced crops, but 
there can be a detrimental effect even when no new land is cleared if the new crop 
sequesters less carbon than the old one did (Searchinger 2008).  To minimize the ?carbon 
debt? associated with land use change, an emphasis should be placed on using waste 
biomass or perennials planted on unused land (Fargione, et al. 2008).   
These concerns and others were addressed in the ?billion ton report? (Perlack, et 
al. 2005) in an effort to determine a feasible path for sustainably replacing 30% of 2005 
petroleum use with biofuels by 2030.  The target for this report of approximately one 
billion dry tons of feedstock per year came from the assumption that 60 gallons of 
gasoline equivalent liquid fuel or 90 gallons of ethanol can be produced per ton of 
biomass.  Potential sources of biomass are divided into two groups for this report, 
agricultural resources and forest resources.  Agricultural resources include:  corn and 
soybeans, crop residues, perennial grasses and woody crops, animal manures, food/feed 
processing residues, and municipal solid waste.  Forest resources include logging 
residues, removal of excess biomass from forest lands, fuelwood, wood processing mill 
residues, pulping liquors, and urban wood residues.  Forest residuals contribute 
368million dry tons annually in this estimate or about one third of the total.   
Biomass resources can also be categorized by type.  The billion ton report lists 87 
million tons of grain for ethanol production and 87 million tons of process residues 
including manure and municipal solid waste.  The rest of the material can be called 
?lignocelluloses?: biomass derived from plant cell walls consisting primarily of cellulose, 
12 
 
hemicellulose, and lignin.  Lignocelluloses include crop residues like corn stover and 
wheat straw; energy crops like switchgrass, giant reed, and miscanthus; sugar crops like 
energy cane and sorghum; and wood.  Considerations on which would be the best energy 
feedstock in a particular geographical area include:  dry matter yield per acre per year, 
soil considerations, water requirement, fertilizer requirement, issues of harvest, storage 
concerns, ease of drying, invasivity, and potential for competition with food crops.  Each 
plant also has chemical differences that can be concerns including ash composition, 
hemicellulose type, and extractives composition.  Examples of these concerns are high 
silicate ash content of corn stover and wheat straw and the potential for pine resin and 
some five carbon sugars to reduce the effectiveness of or stop fermentation.  
Lignocelluloses have some common chemical structures and some that are species 
specific.  Cellulose is the most abundant biopolymer in the world and is made up of a 
long, linear chain of ?-D- glucopyranose linked by 1-4 glycosidic bonds (Stenius 2000, 
34).  A short section of cellulose is illustrated in Figure 3. It forms intra and 
intermolecular hydrogen bonds that cause it to have crystalline and amorphous regions in 
the plant cell wall.  The crystalline regions cause cellulose to be relatively inert and 
insoluble in most solvents.  The length, or degree of polymerization (Dp), of cellulose 
 
Figure 3  Cellulose structure (Sjostrom 1993, 54). 
13 
 
varies with cell type and species.  Wood cellulose normally has around 10,000 units 
while cotton has about 15,000.   
Hemicellulose is also a polymer of sugars, but it is shorter, branched, made up of 
multiple components and in some cases can be water soluble.  The branched form inhibits 
formation of crystalline regions in native hemicellulose, but some types can be crystalline 
after pulping if the branches are selectively removed (Rydholm 1985, 158).  The structure 
and composition of hemicellulose varies by species and in some cases by cell type and 
location (Stenius 2000, 36).  Examples of hemicelluloses common in softwood are 
illustrated in Figure 4 and Figure 5.   
 
Figure 4  Galactoglucomannan structure (Sjostrom 1993, 65). 
 
Lignin is an amorphous aromatic polymer of phenyl propane units joined mostly 
with ether linkages or carbon-carbon bonds.  Lignin is often thought of as the glue that 
holds the cell wall fibers together.  There is some evidence for bonding between lignin 
14 
 
and carbohydrates, especially hemicelluloses, but the nature of this bonding is not well 
understood.  The structure of lignin has been approximated as shown in Figure 6.   
 
 
Figure 5  Arabinoglucuronoxylan structure (Sjostrom 1993, 65). 
 
Extractives are generally relatively small, nonstructural and species specific 
components of lignocelluloses.  They are named because they can be extracted from the 
plant material with organic solvents or water.  They typically make up a small fraction of 
the plant by mass, but may give the plant some of its unique character.  For example, the 
scent and rot resistance of cedar comes from thujaplicin, an aromatic compound with a 
seven-member ring that is toxic to most wood decay organisms (Sjostrom 1993, 97). 
 
15 
 
 
Figure 6  Segment of softwood lignin structure (Sjostrom 1993, 84). 
 
Wood has several advantages over other biomass feedstocks in terms of the 
handling characteristics during a continuously operating process.  It can be sustainably 
harvested during most of the year and stored for relatively long periods of time as whole 
logs or chips.  The inorganic ash content of wood is typically less than 1% but the 
inorganic content of bark can be substantially higher at 2-5% (Sjostrom 1993, 107,113) 
which is one reason bark is typically separated from wood at an early stage of  
processing.  The dry matter yield for wood is similar to other potential energy crops like 
switchgrass at 4 to 8 dry tons per acre year (Graham, et al. 1995) but harvesting logistics 
16 
 
would be different since it would be performed less often.  There is already an 
infrastructure in place for harvest, transport, and handling of large quantities of wood, but 
this will need to be modified somewhat to enable the use of harvesting residuals and 
other materials that are not already sent to the mills.   
Wood species are generally classified as hardwood or softwood.  These groups 
vary from each other distinctly not only in physical appearance but also in chemical 
structure.  Most woods have about 40-45% cellulose and less than 5% extractives on a 
dry basis, but softwoods typically have 25-30% hemicellulose and 25-30% lignin while 
hardwoods typically have 30-35% hemicellulose and only 20-25% lignin (Stenius, 29).   
Hemicelluloses in softwood consist mostly of galactoglucomannan and 
arabinoglucuronoxylan.  Hardwood hemicellulose content is made up of mostly 
glucuronoxylan with some glucomannan.  Lignin in softwood is generally classified as 
?guaicyl? lignin and is made of mostly trans-coniferyl alcohol precursors with some 
trans-p-coumaryl alcohol.  Hardwood lignin is called ?guaicyl-syringyl? lignin and is 
made up of about 50% each of trans-coniferyl and trans-sinapyl alcohols.  Wood 
chemical composition also varies between heartwood, sapwood, and reaction wood 
within a single tree.   
Two sets of carbohydrate reactions were of interest in this research:  those that 
happen in the acid environment of prehydrolysis and those that occur in the alkaline 
environment of kraft cooking.  Hydrolysis of bonds between sugar monomers takes place 
as long as the carbohydrates are exposed to hot cooking liquor in either case.  The rate of 
hydrolysis increases with temperature above 100?C and as pH is either increased or 
decreased from near neutral where a minimum rate occurs.  Hemicelluloses respond 
17 
 
differently at either end of the pH spectrum.  Galactoglucomannan is more active than 
xylan on the alkaline side while the xylans are more easily hydrolyzed on the acid side 
(Ingruber, Kocurek and Wong 1985, 34-38).  Suggested means of increasing the yield of 
these two components are also different.  A neutral or alkaline treatment prior to acid 
treatment can increase the yield of glucomannan while xylan loss can be minimized by 
reducing the residual alkalinity at the end of the alkaline cook.  Hydrolysis of various 
carbohydrate components occurs at different rates with the five-carbon sugars reacting 
faster than the six-carbon and galactose faster than mannose which is faster than glucose.  
Cellulose is fairly resistant to acid hydrolysis due to its ordered structure and crystalline 
regions but is reactive in more severe environments (e.g. mineral acid at temperatures 
above 120?C) and in areas damaged by mechanical action.  The hydrolysis does not cause 
substantial yield loss in an acidic environment, but it can reduce chain length and 
increase yield loss in subsequent alkaline treatments.  Acid hydrolysis is much faster than 
the similar alkaline reaction and also has a higher tendency to occur at the end of the 
polymer chain (Sjostrom 1993, 46).   
 In addition to random chain scission by hydrolysis, the most important reactions 
in strong alkali for both cellulose and hemicellulose are the endwise ?peeling? reaction 
that shortens the chain from the reducing end one sugar monomer at a time and the 
competing ?stopping? reaction that produces an alkali stable end group.  Alkaline 
scission has only a minimal direct effect on yield, but a dramatic effect on degree of 
polymerization.  A large percentage of pulp yield is lost to peeling of reducing sugars in 
alkaline pulping even though these represent only about 0.2 % by weight of the native 
wood (Procter and Apelt 1969) because peeling and scission each produces a new 
18 
 
reducing end group that can allow peeling to continue or start again.  The stopping 
reaction kinetically limits peeling to about 60 monosaccharide units (one study found the 
?zip length? to be 84 for cotton cellulose and 53 for Douglas fir derived holocellulose 
(Agarwal, McKean and Gustafson 1992)), but peeling will also be terminated by physical 
inaccessibility in crystalline areas or by dissolution of a shorter chain.  Destruction of 
glucomannan is by peeling and dissolution of smaller fractions.  Cellulose yield loss is 
also primarily by peeling; ?primary? during initial heat up and then ?secondary peeling? 
after cleavage at higher temperatures.  Xylan removal is by dissolution of soluble 
fractions rather than peeling (Grace 1989, 48).  These mechanisms were determined by 
using a borohydride to prevent peeling.  Sugar monomers are subject to rearrangement to 
other sugars or conversion to carboxylic acids in strong alkali.   
1.6 Pulping Methods 
Processes for turning wood into pulp for paper and board manufacture can be 
divided into methods that preserve wood yield by producing pulp through primarily 
mechanical forces and methods that chemically purify fibers.  Mechanical pulping 
methods use mechanical force, heat, and or pressure, and sometimes mild chemical 
treatments to physically disrupt wood structure into pulp.  This pulp is suitable for grades 
of paper that do not require great physical strength or longevity of quality like newsprint.  
The advantages of these methods are low manufacturing cost with the exception of high 
energy usage and high yield of pulp that gives good print quality.  Chemical pulping 
methods, on the other hand can produce pulp of controlled chemical purity ranging from 
removing just enough extractives and lignin to separate fibers to almost pure cellulose 
depending on product requirements.  All pulping methods have continuously evolved to 
19 
 
improve economic performance and to meet new requirements of improved product 
quality and cleaner environmental performance.   
The soda pulping process was first used commercially to manufacture pulp from 
poplar wood in 1854 (Smook 1982).  Modifications to the process included regeneration 
of alkali for economic purposes through combustion of waste liquor and causticization of 
sodium carbonate.  The kraft process almost completely replaced the soda process after 
sodium sulfate was substituted as a less expensive replacement for the sodium carbonate 
make-up chemical and the resulting pulp was found to be stronger.  The soda process is 
similar to the kraft process which will be described later except that sulfur is kept out of 
the liquor cycle.  The process is used only rarely today in cases of local sensitivity to the 
unpleasant odor of kraft recovery and waste treatment.  In these cases, delignification is 
usually enhanced by the addition of anthraquinone (AQ) to the cook to make the ?soda-
AQ? process.  There is more discussion of AQ in Chapter 6.  The only soda AQ mill 
currently operating in the USA is in Tennessee. 
Patents for a pulping process using bisulfites and sulfur dioxide were issued to 
Tilghman in 1866 and 1867 (Rydholm 1985).  Sulfite cooking liquor consists of water, 
sulfur dioxide, and a base for the resulting bisulfite that depends on the target pH.  
Commercial processes followed later using magnesium, calcium, sodium and ammonium 
as the base for the bisulfite.  The process grew in popularity because it produced a 
relatively bright unbleached pulp at good yield on wood that was easy to bleach to higher 
brightness.  The process could also be modified to produce high cellulose dissolving 
pulps.  The sulfite process is very sensitive to bark contamination and extractives, though, 
which restricts the process to only a few wood species:  spruce, fir, hemlock, and a few 
20 
 
hardwoods.  Chemical recovery systems were later developed and the sulfite process was 
the dominant chemical pulping process in the first half of the 20th century.  The acid 
sulfite process has since mostly disappeared due to advancement in bleaching kraft pulp, 
continuing environmental problems with biological oxygen demand (BOD) in liquid 
effluent, and atmospheric sulfur dioxide losses (Gullichsen and Fogelholm 1999).  Only 
four acid sulfite mills remained open in the USA in 2006 with a total production of about 
450,000 tons per year of bleached pulp.  The newest of these mills was constructed in 
1969 (Matussek 2006).   
Discussion of the sulfite process is included here because the process allowed, in 
some cases, for efficient manufacture of products from the spent liquor in addition to 
burning the waste organics for fuel.  Ethanol, proteins, vanillin, and lignosulfonates were 
all produced either directly in the process or from components of the liquor at mills like 
the one closed by Georgia Pacific in 2001 at Bellingham, WA.  The sulfite process is now 
represented mostly by Neutral Sulfite Semichemical (NSSC) mills that manufacture high-
yield high strength fibers from hardwood for corrugating medium although there has been 
some expression of optimism in the literature for mills that combine alkaline sulfite with 
AQ catalyst and for mills that mix sulfite with kraft in a combined recovery system 
(Ingruber, Kocurek and Wong 1985). 
Kraft pulping of wood is currently the dominant method of producing chemical 
pulp for paper and paperboard products.  This process uses ?white liquor?, a solution of 
sodium hydroxide and sodium sulfide, to delignify the wood at a very alkaline pH of 12 
to 14 and temperature of about 170?C.  Industrial white liquor also contains sodium 
carbonate and other components resulting from the recovery cycle.  The kraft process has 
21 
 
mostly replaced the non-sulfur soda process and the sulfite processes because it produces 
a stronger pulp and it has an efficient process for recovering cooking chemicals.   The 
recovery process also produces energy from burning the organic material in the ?black 
liquor? that is washed from the pulp after cooking.  Kraft pulping can be accomplished 
with most wood species and can handle contaminants like bark and dirt to a certain level.  
Some contaminants like excess calcium or silica can cause operational difficulties or 
reduced pulp quality.  The primary advantage of kraft pulping over soda is that the time 
required for delignification is reduced by 50%.  This effectively doubles the production 
rate while giving less time for destructive carbohydrate hydrolysis reactions (Rydholm 
1985, 311).  Kraft cooking cycles are also shorter than sulfite due to shorter heat-up time 
and typically higher final cooking temperature.  Anthraquinone is sometimes used with 
kraft pulping but the benefits are lower than that observed with soda cooking (Sjostrom 
1993, 156).  The justification for using expensive AQ in kraft pulping is usually to relax a 
bottleneck in pulp production that can be improved by increasing pulp yield.   
The kraft cook is performed in a special pressure vessel or digester.  The 
temperature of the cooking reaction controls the rate of reaction and the type of 
equipment required.  The 165 to 170?C typically used in industrial kraft pulping 
corresponds to about 100 psi since the solution is mainly water.  Traditionally, and in 
many current operations, the cook was a batch operation and the digester was a tall tank 
that was designed to withstand the pressure, temperature, and corrosive nature of the 
cooking liquors.  For batch digesters, the chips are fed into the vessel followed by steam 
to remove air followed by a combination of white liquor and spent ?black liquor? to make 
up the required volume. The digester is capped and heated by direct or indirect steaming.  
22 
 
After the appropriate time, a valve at the bottom of the digester is opened and the chips 
and liquor are ?blown? to a separate tank.  The release of pressure can help physically 
break the cooked chips into pulp, but it can also damage the pulp.  Since the first Kamyr 
continuous flow digester was built in 1950, many installations have had continuous 
digesters where the presteamed chips and liquor slurry feed in the top of the digester and 
the pulp is released from the bottom in a continuous controlled manner (Smook 1982, 
81).  There are many different innovations in the ways the liquor and heat are added and 
recovered from all types of digester systems to improve the yield, strength and chemical 
makeup of the pulp and the energy economics of the process (Gullichsen and Fogelholm 
1999, 52-56).   
The aim of chemical pulping is typically to solubilize and remove most of the 
lignin while leaving the carbohydrate holocellulose, hemicellulose and cellulose, as intact 
as possible.  A standard test of lignin remaining in pulp is the ?kappa number?.  In this 
test a measured quantity of pulp is reacted with potassium permanganate for ten minutes 
before quenching with iodide.  The solution is then titrated with thiosulfate.  A typical 
brown stock kappa number target for softwood pulp that will be bleached is about 30 
which corresponds to about 4.5% lignin remaining in the pulp.  Since some of the lignin 
is bound to carbohydrates and carbohydrates are sensitive to reaction under the same 
conditions as lignin, the carbohydrate fraction normally receives some damage.  It has 
been reported that under the strong alkaline conditions with pH above 12 in the heating 
stage of a kraft cook galactoglucomannan is 70% dissolved by 130?C, xylan begins to 
dissolve after 140?C and is 60% removed by 170?C.   Cellulose begins peeling after 
120?C followed by alkaline hydrolysis and secondary peeling at higher temperature 
23 
 
(Grace 1989, 46-49). By the end of a conventional cook, all of the pectin, starch, and 
arabinose, 70% of the galactoglucomannan, about 50% of the xylan, and about 10% of 
the cellulose would be expected to be removed resulting in a total sugar loss of just over 
20% of the original wood (Grace 1989, 26-29) in addition to the roughly 30% of wood 
that is lignin.  These sugars represent the opportunity for VPP.   
The pulp or ?brown stock? is suspended in black liquor following the cook.  
Black liquor includes the residual alkali from white liquor and salts of the organic 
materials removed from the chips.  This slurry is typically fed through a screen or 
?knotter? to remove uncooked chips before washing.  Brown stock washing separates the 
pulp from the black liquor and is typically a multi-stage counter current filtering process.  
Clean hot water that is typically reused from somewhere else in the mill is used as wash 
water on the last stage.  The filtrate from the last stage is used as wash on the next to last 
stage.  Some of the filtrate from the first stage is reused for digester dilution; the rest is 
concentrated in the evaporators and then burned in the chemical recovery boiler.   
One of the biggest hindrances to expansion of kraft pulping historically was the 
dark color of the unbleached pulp and difficulty in bleaching without severe strength loss.  
The development of multistage bleaching followed by the introduction of bleaching with 
chlorine gas in around 1930 (Rydholm 1985, 280) produced the first fully bleached kraft 
pulp using a three stage chlorination, caustic extraction, hypochlorite (CEH) bleach plant 
with washing between stages.  Chlorine dioxide (D) was introduced as an industrial 
bleach chemical in 1946 and by 1957 there were 43 D stages in operation (Dence and 
Reeve 1996, 81).  This allowed southern pine pulp to be bleached to full market 
brightness while maintaining a strength advantage over sulfite pulps.  Bleach plants 
24 
 
mostly stopped using chlorine or hypochlorite in the late 1990s due to environmental 
regulations and pulp quality concerns.  Modern bleach plants typically have oxygen 
delignification (O) followed by three stage DED, four stage DEDD, or five stage DEDED 
to reach high brightness with kraft pulp.   
Unfortunately, though understandably, most mills are divided up into the ?paper? 
side, the ?pulp? side, and the ?recovery island? as main production areas for management 
with the woodyard, waste treatment, converting, and chemical preparation areas included 
in one of these or considered apart.  These areas are often physically set apart from each 
other by a road or rail line as if to enforce the separation.  This is unfortunate because a 
pulp and paper mill consists of not just a process that transforms wood from logs to chips 
to pulp to paper, but also several water and chemical process cycles that tie the areas 
together and often make discovery of the root cause of a disturbance in any area difficult.  
The most important of the cycles in the kraft mill is the liquor cycle.  The 
chemistry of the liquor recovery process actually suggests two cycles, the sodium cycle 
and the calcium cycle.  A previously published elegantly simplified diagram of this 
process is presented as Figure 7 (Sjostrom 1993, 161).  In this diagram, the sodium cycle 
is the top rectangle and the calcium cycle is the triangle below it with the two cycles 
interacting in causticizing.  Starting with ?Pulping?, white liquor and wood are reacted 
then separated to pulp and black liquor in washing.  The black liquor is burned to make 
energy and to regenerate the molten sodium carbonate and sodium sulfide ?smelt? that is 
dissolved to form green liquor.  The efficiency of this part of the recovery cycle improved 
dramatically following the invention by Tomlinson in 1927 (Rydholm 1985, 279) and 
subsequent commercialization in 1934 of the modern recovery boiler (Smook 1982, 130).  
25 
 
The sodium carbonate in green liquor reacts with quicklime to reform sodium hydroxide 
for white liquor.  It is important that the white liquor and lime are effectively separated 
from each other as calcium in white liquor will cause scaling in the digester equipment  
and potentially give the pulp a blue shade.  Sodium in the lime cycle will make the lime 
difficult to dry and can cause operational and environmental problems in the lime kiln.   
Multistage pulping describes processes where the pulping chemistry is changed 
deliberately from outside the digester.  There are many possible combinations of changes 
that are possible and there may or may not be washing between stages.  Implementation 
of VPP will require some sort of multistage process.  One published example used a mild 
alkaline extraction before kraft pulping (Al-Dajani and Tschirner 2008).  Most work to 
 
Figure 7  Kraft recovery cycle (Sjostrom 1993, 161). 
26 
 
date has included an acidic prehydrolysis followed by kraft pulping (T. E. Amidon 2006) 
(T. Amidon 2008) (Yoon, MacEwan and van Heiningen 2006).  If done properly, the acid 
prehydrolysis can selectively remove sugars from hemicellulose while leaving lignin and 
cellulose mostly intact.   Acid-Alkali processes are typically only recommended for 
producing dissolving pulps since the acid treatment makes the remaining carbohydrates 
more sensitive to alkaline degradation, but they can also be used to produce specialty 
paper pulps.  These pulps are softer and have higher tear strength and opacity but lower 
burst and tensile index (Rydholm, 661).  There are also many process options that finish 
cooking with an acidic stage and or contain sulfite in one or more stages. 
Choice of pulping process depends on economics, but the economics will often 
depend on product quality.  Hemicellulose content of pulp in particular effects not only 
the yield of the pulp but also the strength and other paper making properties.  A recent 
study considered strength and yield relationships among ten spruce pulp samples 
produced from six different laboratory kraft pulping processes (Molin and Teder 2002).  
The processes included conventional kraft, prehydrolysis kraft, two polysulfide strategies, 
and two cooks with modified liquor profiles.  The resulting pulps all had a klason lignin 
content of 3.3?1% of wood, but hemicellulose content that varied from a low of 9.8% 
with the prehydrolysis pulp to a high of 25.4% with the vapor phase polysulfide pulp.  
Zero span tensile strength was unaffected by the various pulping methods indicating that 
changes in paper properties attributed to the different pulping methods are not due to 
changes in individual fiber tensile strength.  The specific bond strength between fibers as 
measured by z-directional tensile and Scott bond also was unaffected by changes in pulp 
composition but did trend with sheet density.  Tensile index decreased while tear index 
27 
 
increased in direct proportion to the cellulose/hemicellulose ratio.  Folding endurance and 
fracture toughness followed a similar trend to the tear index, increasing as the 
hemicellulose content decreased.  These changes were attributed to a decrease in stiffness 
of fibers as hemicelluloses are removed caused by a change in shear strength and/or 
bending strength of the fiber wall or a change in shrinkage of the fibers during drying 
with all of these possibly caused by a change in the aggregation of cellulose microfibrils.    
28 
 
 
 
Chapter 2  Review of Previous Work 
This section includes previously published work that lends insight to the 
processing of wood to chemicals and pulp.  Previous work that relates to VPP can be 
roughly divided three ways:   
1. Efforts to produce chemically pure ?dissolving? pulp grades 
2. Efforts to break down wood completely to monomers or ?wood hydrolysis? 
3. Recent laboratory work with VPP 
2.1 Dissolving Pulp 
Most pulp is produced to make paper and paperboard.  Some is produced for 
specialty applications like absorbent fillers for diapers or as chemical feedstock for 
regenerated cellulose like rayon, cellulose acetate, or even food additives.  Pulp that is 
produced as chemical feedstock usually has very specific quality constraints that are 
based less on physical characteristics (e.g. strength) than they are on chemical 
characteristics such as pentosan content or degree of polymerization.  These products are 
called ?dissolving pulp? and are produced either from bleached sulfite or by a method 
called prehydrolysis kraft (Rydholm 1985, 281).  The prehydrolysis kraft process has the 
advantage of working with many different wood species.  This multistage process begins 
with acid or water prehydrolysis followed by kraft pulping and then bleaching.  The 
29 
 
conditions used in each stage and the stages used in bleaching depend on the intended 
product.   
The hemicelluloses that are dissolved in prehydrolysis are mostly in polymer 
form.  Some are converted to monoses and some of those are decomposed to furfural, 
hydroxymethyl furfural (HMF), or other acids.  Much of the hemicellulose remains in the 
chips, but is degraded and more easily dissolved in kraft pulping through peeling at the 
additional reducing end groups.  The acidic conditions can also cause lignin condensation 
making the chips more difficult to delignify in kraft pulping.  This tendency can be 
neutralized by adding metal bisulfite (Rydholm 1985, 657), but this addition may lead to 
substantial lignin removal in the prehydrolysis.  Prehydrolysis of pine reduced xylan 
degree of polymerization (Dp) from around 200 in the native wood to between 20 and 60 
in the pulp.  Kraft pulping alone reduced xylan Dp to between 76 and 90 (Grace 1989, 
33-35).  Xylans are retained in kraft pulp due to relative stability in alkali, reduced 
solubility as alkalinity drops at the end of the cook, and mass transfer limitations due to 
branching in the polymer.  Prehydrolysis removes arabinose and methyl glucuronic acid 
side chains from xylans making them more susceptible to degradation and mass transport 
in subsequent pulping. 
Much of the published work in the area of prehydrolysis pulping focused on 
producing high alpha cellulose chemical pulps from western US wood species hemlock, 
spruce, and birch, but some of the learning can be extended to the VPP situation with 
pine.  In particular, there is some discussion of the effects of acidity, time and 
temperature on not only pulp composition and properties, but also on pulping 
requirements (Richter 1956).  One difficulty with this approach relative to VPP is that 
30 
 
yield of sugar or hemicellulose in the hydrolysis extract is less important than reduction 
of hemicellulose residuals in the final pulp.  Therefore the conditions for prehydrolysis 
and pulping in dissolving pulp manufacture can be selected so that the hemicellulose is 
removed in any stage of pulping and bleaching while with VPP it is critical that the sugar 
removal in the first stage be enhanced while removal in subsequent stages is minimized.  
Water hydrolysis is noted to cause pentosans to be more soluble in subsequent alkaline 
pulping.  The severity of prehydrolysis is determined by acidity, time, and temperature 
and this severity determines the amount of hemicellulose that is solubilized in 
prehydrolysis and the amount that is degraded and extracted during subsequent pulping.   
There is much discussion of two stage pulping with bisulfite in the first stage 
since this limits lignin condensation and represents a commercial process competitive 
with prehydrolysis kraft (Richter 1956).  One drawback with the use of bisulfite in the 
VPP application would be the extraction of considerable amounts of lignin that would 
then need to be separated from the sugar.  Prehydrolysis pulping can be used to optimize 
pulp properties not just for high alpha cellulose dissolving pulps, but also for improved 
bleachability or superior tear strength and fold endurance.  Tear resistance and fold 
endurance are initially improved with prehydrolysis, but Mullen burst and tensile index 
are typically lowered.  Extraction of pentosans can be controlled by adjusting 
prehydrolysis conditions and different wood species will respond differently to 
hydrolysis.  Deactivation (condensation reactions) of lignin must be limited in the 
prehydrolysis by limiting the combination of extended time, high temperature, and low 
pH.  Softwood lignin is much more sensitive to deactivation than hardwood lignin.  Nitric 
acid and sulfur dioxide increase the amount of deactivation even at application of only 
31 
 
0.5% on wood.  Sulfuric acid has no observed effect on the lignin at this same application 
rate as long as the temperature of prehydrolysis is kept below 120?C.   
A study to look at fundamentals of polysaccharide hydrolysis was performed at 
the Institute of Paper Chemistry (Bernardin 1958).  The study used a twin digester system 
that allowed preheating the liquor and could bring the wood ? liquor mixture to 160?1?C 
in 7 to 9 minutes and be rapidly cooled below 100?C at the proper time.  The wood was 
air-dried shavings of black gumwood and the times studied were 0, 15, 30, 60, and 120 
minutes at temperature.  The composition of the extracted chips was determined by 
extraction with alkali and precipitation in alcohol.  The total yield of wood was found to 
be a linear function of the logarithm of hydrolysis time and reduced nearly 30% over two 
hours.  About 55% of the hydrolyzate was recovered as hemicellulose; the remainder was 
a mixture of extractives, lignin, and degradation products.  The study found a rapid 
reduction in alkali stable hemicellulose in the first 15 minutes of hydrolysis.  The average 
Dp of hemicellulose was reduced by half from 130 to 60 and of cellulose from 1600 to 
1200 in the first 30 minutes.  Hemicellulose Dp was reduced to 45 and cellulose Dp to 
1000 after two hours.  Removal rate of polyose from the chips was fairly constant for the 
first hour and then dropped significantly in the second.  The first hour is said to indicate 
either a zero-order reaction or a diffusion controlled surface reaction which would be 
consistent with an understanding that hemicellulose was concentrated on the outer layers 
of the fiber wall.  Slow removal in the second hour would then be diffusion controlled 
removal of dispersed fractions from deeper within the fiber wall.  The composition of 
sugars removed did not change over time, so there is not a selective removal of any 
hemicellulose fraction in this study.   
32 
 
No pulping was performed in this study with black gum, but the results can help 
explain the enhanced removal of hemicellulose and potential for cellulose damage in 
subsequent alkaline pulping.  The decrease in average chain length of hemicellulose 
would make the carbohydrate both more soluble and easier to move by diffusion out of 
the fiber and less likely to reprecipitate on the fiber at the end of the alkaline cook.  The 
solubility of lignin and pitch in alkali that would hinder movement in an acidic 
environment would further allow already hydrolyzed fractions to move out of the fiber.  
Although very little cellulose was removed in this two hour treatment, there was 
significant reduction in average Dp of cellulose also.  This indicates the potential for 
greater cellulose damage and yield loss in subsequent alkaline pulping.   
An example of efforts to produce dissolving grade pulp using prehydrolysis kraft 
describes 16 water prehydrolysis cooks with cooking time from 2.5 to 240 minutes, 
maximum temperature from 140? to 180?C, and heating time to temperature of 30 to 120 
minutes (D. J. Brasch 1964).  A measured amount of chips were contained in a small 
basket that was removed following prehydrolysis and replaced with an empty basket prior 
to the cook.  They noted in their procedure that all cooks were started within 30 minutes 
of completion of the prehydrolysis and that the pulping chemicals were added to give the 
same active alkali charge and sulfidity for every cook.  The composition of the liquors, 
chips, prehydrolyzed chips and pulp were measured.  Final prehydrolysis liquor pH 
varied from 3.0 for higher time and temperature (150 minutes at 180?C) to 4.2 for milder 
conditions (60 minutes at 140?C).  Final pulp yield and pentosan fraction dropped as the 
time and temperature of prehydrolysis increased, but the data indicated that it would be 
possible to optimize the conditions to control the lignin and alpha cellulose content.  
33 
 
Lignin content of pulp was highest following the most severe prehydrolysis and lowest in 
conditions in the middle range of time and temperature.   
In contrast with the results from the study of prehydrolysis of black gum, the 
various components of the hemicelluloses in pine are expected to be dissolved at different 
temperatures due to the varied chemistry and solubility of the sugars present.  The first 
stage of a prehydrolysis kraft dissolving pulp process for southern pine was considered in 
detail through measurement of sugar concentrations in chips, prehydrolyzed chips, and 
hydrolyzate as a function of time and temperature (Casebier 1969).  These studies 
indicate that acetyl and uronic acid in the wood can be solubilized by room temperature 
extraction.  Arabinose and some galactose were removed in 100?C water.  Xylose, 
mannose, and some glucose were present in the liquid following extractions at 170?C.  
The ratio of sugars indicates that cellulose was not significantly damaged by this 
extraction.  Significant degradation of the sugars to furfural or hydroxymethylfurfural 
was observed at higher temperatures and longer reaction times leading to a maximum 
extracted yield of about 16% from 90 minutes at 175?C or 45 minutes at 180?C.  One 
interesting result was that the analysis of molecular weights in the hydrolyzate showed a 
maximum of 3100.  This indicates that the hemicellulose molecule Dp must be reduced to 
below about 20 for successful diffusion out of the fiber. 
A study of prehydrolysis kraft pulping of Loblolly pine and sweetgum, both 
southern woods, was performed to investigate how the various carbohydrates behaved 
during the purification process (Mitchell, et al. 1956).  Chips were prehydrolyzed with 
water at 160?C for three different times up to 75 minutes and then cooked with three 
different alkali charges.  The nine samples of each species were then divided three ways 
34 
 
for three different purification methods and then nitration.  Samples were taken at each 
step for carbohydrate analysis and percentages of the remaining fraction for each 
component was reported but total yield was not reported.  All five carbohydrate 
components, glucosan, mannan, xylan, galactan, and arabinan, remained in the pine chips 
following prehydrolysis with the percentage of glucosan increasing while the others 
decreased slightly.  Galactan and arabinan were not detectable in the pulp following 
cooking while the xylan fraction was reduced by half and mannan reduced by 75% 
relative to the original wood.  Variation in alkali concentration during the cook had no 
measurable impact on pulp composition.  Alpha-cellulose from the pulps had roughly 
half the mannan and xylan as the pulp at all tested conditions.   
2.2 Alcohol Production from Wood Hydrolysis 
Manufacturing ethanol and methanol from wood has been considered as an 
alternative source of liquid fuels for emergency purposes since at least World War I 
(Zerbe 1985).  Two plants to make fuel ethanol were built during this war using a dilute 
sulfuric acid hydrolysis but they were shut down after the war.  Another plant was built 
during World War II in Oregon that was never operated successfully.  Also built during 
the war was a plant to produce ethanol from the sugar in the spent sulfite pulping liquor 
from the pulp mill in Bellingham, WA.  This plant produced 25 gallons of ethanol per ton 
of pulp until it was shut down in 2000.  Other processes include concentrated acid and 
enzyme hydrolysis processes, each with their own variations, advantages, and 
disadvantages.  Methanol, or ?wood alcohol?, was made as a byproduct of charcoal 
manufacture and is present as a non-condensable gas that must be dealt with in pulp 
35 
 
mills.  Methanol can also be produced from syngas that is a result of gasification of wood 
or coal.   
Development of these processes has continued with large efforts in the late 1970s 
in response to the oil crisis of that time and again in the last ten years as oil prices have 
continued to rise and there is renewed recognition that substitute fuels must be found.  
Continuous improvements to the dilute acid process led to the Forest Products 
Laboratory?s ?Madison process? that can produce 64 gallons of 95% ethanol from one 
ton of dry wood waste (Zerbe 1985).  A pilot plant was built at TVA in Florence, AL to 
test a modification to this process.  The fungus Trichoderma was identified for further 
work from over 14,000 tested organisms as a source of cellulose degrading enzymes.  
This identification has encouraged an entire industry with new applications for enzymes 
in textiles, food manufacturing, and pulp and paper in addition to continued efforts 
toward an enzyme based production platform for ethanol from biomass.  
Research has also examined methods for the pretreatment of biomass to enhance 
enzymatic hydrolysis of cellulose.  One of the process difficulties in wood hydrolysis is 
degradation of sugars to furfural or hydroxymethyl furfural (HMF) that can poison the 
fermentation organisms.  Degradation of sugars prompted development of a method of 
measuring the pH of the reaction liquid during laboratory pretreatment of lignocelluloses 
for subsequent fermentation to ethanol (Weil 1998). The measurement allowed alkali 
addition on a pH target during pretreatment to minimize the acid catalyzed degradation of 
cellulose to glucose and glucose to HMF or other acids.  The system sampled the liquid 
through a cooling coil and a pressure reducing valve before the pH meter.  The cooled 
36 
 
liquid was then pumped back in to the reactor.  Liquid residence time in the system 
before the pH probe flow cell was about 1.5 minutes.   
2.3 Value Prior to Pulping 
Successful application of VPP requires that only the desired components be 
removed from the wood and that those components and the final pulp are not degraded in 
the process.  For production of ethanol feed sugars this means selected sugars are 
removed and lignin is left in the wood unchanged during prehydrolysis.  Hot water 
extraction of wood chips is an effective method of separating fermentable sugars from 
wood prior to pulping.  Potential difficulties with this practice include a reduction in yield 
(Yoon, MacEwan and van Heiningen 2006) and strength (Martinez 2006) of the final 
pulp.  The conditions of extraction also must be optimized for the wood species to 
maximize recovered sugar and minimize degradation of pulp quality.   
Work with pine has special issues that are of lesser concern with hardwoods due 
to relative importance of pulp strength and differences in lignin and hemicellulose 
composition.  Different species and grades of pulp are included in finished products to 
impart particular characteristics to those products.  Hardwood pulp is included in fine 
paper to improve optical characteristics like brightness, opacity, smoothness and print 
quality.  These fibers are in tissues to add softness and absorptivity.  Softwood fibers are 
used primarily to give the products strength to meet the physical demands put on the 
sheet or board but they are also used to make absorbent fluff pulp for hygiene products.  
Hardwood and softwood also have different amounts and forms of both lignin and 
hemicellulose.  Softwood lignin has a tendency for condensation of ether bonds to 
stronger carbon ?carbon bonds in acidic environments that hardwoods do not have.  
37 
 
Therefore, conditions that work for hardwood may not be appropriate for softwood.  
Several studies of VPP for hardwood have been performed with some apparent success 
based in part on the relatively high fraction of hemicellulose in hardwood (Al-Dajani and 
Tschirner 2008) (T. E. Amidon 2006).  Less work has been reported for pine (Yoon, 
MacEwan and van Heiningen 2006), even though pine pulp is approximately 30% of U.S. 
production (estimated at 29 million tons/year (Matussek 2006)).   
Recent prehydrolysis work with hardwood has produced pre-extracted chips with 
8, 12, and 23% mass removal (T. Amidon 2008).  Increasing the mass removal to 23% 
gave a higher quantity and concentration of sugar in the extract but produced pulp with 
lower handsheet strength relative to a control Kraft pulp at equal freeness.  Higher 
refining energy was required to reach the target freeness with this pulp.  Dr. Amidon 
emphasized a lower pulping requirement, higher ?digester yield? to reach the target of 18 
kappa number and easier bleaching thus potentially increasing production rates in the 
digester, but cautioned that this comes at the cost of 12% lower overall yield (40% versus 
52% for kraft) and that more experiments could more closely evaluate the costs and 
benefits.  The extraction was performed with water at 160?C for two hours with a 4/1 
liquid/solid ratio and was followed by two 15 minute washes with 80?C water.  About 
70% of the original xylose was removed and this accounted for about 50% of the mass 
removed.   An attempt was made to use wet end starch to reduce the strength loss.   
Another paper on the topic of VPP with hardwood had several preliminary 
observations regarding the effects of hot water preextraction on Sugar maple chips (T. E. 
Amidon 2006).  The most favorable observation was that a pulp could be produced by a 
Soda-anthraquinone (AQ) cook following pre-extraction that had similar yield and kappa 
38 
 
number to a kraft control pulp.  Further, this pulp was easier to bleach and therefore the 
yield could be improved by reducing the cook time without negatively impacting final 
pulp properties.  The result with similar yield was with a mild extraction that removed the 
acetyl groups and about 5% of the chip weight as hemicellulose.    
This line of research with hardwood was extended in the direction of improving 
delignification (Bolton 2007).  In the case of a very mild prehydrolysis of 15 minutes at 
150?C with 1.5% acetic acid, the H-factor for a Soda AQ cook could be reduced 37.5% 
while the overall yield was slightly improved relative to a kraft control.  The explanation 
proposed for accelerating the cook was that the hydrolysis formed more end groups 
capable of reducing AQ to AHQ (see Chapter 6) and that diffusion was enhanced through 
increased void volume in the chips.  Very little sugar was extracted under these mild 
conditions.  The use of a mild acid prehydrolysis of 20 minutes at 140?C followed by a 
carbonate wash for 20 minutes at 160?C was also mentioned to dramatically reduce alkali 
requirement in SAQ pulping to only 8% on chips.  Severity of hydrolysis was measured 
by kinetics and assumed to be independent of acid concentration while strongly 
dependent on time and temperature.   
A large fraction of the softwood hemicelluloses also appear relatively easy to 
remove with hot water or acid prehydrolysis.  However, a reduction in both strength and 
yield in the kraft pulp produced from prehydrolyzed southern pine chips have been 
observed (Yoon, MacEwan and van Heiningen 2006) (Martinez 2006).  Speculations 
about the cause of these losses include:   
1. Cellulose damage caused by prehydrolysis reduces individual fiber strength. 
2. Bulk hemicellulose removal limits yield and bonding potential. 
39 
 
3. A key hemicellulose fraction is removed that inhibits fiber-fiber bonding. 
4. Prehydrolysis makes all or select pulp carbohydrates more susceptible to 
alkaline degradation. 
Yoon, et al reported 3% pulp yield loss after kraft pulping relative to a control 
pulp when 5% of wood weight was extracted with hot water before pulping and 6% yield 
loss when 8% was pre-extracted (Yoon and van Heiningen 2008).  They also reported 
difficult refining and reduced tensile strength in the extracted pulp but similar viscosity, 
zero-span wet tensile, and tear resistance.  They conclude that there is little change to the 
cellulose in pine pulp due to pre-extraction but there is reduced hemicellulose content.  
One of the goals for the future in Dr. Yoon?s work was to return 1/3 of the hemicellulose 
sugars to the pulp for yield improvement, but a plan was not presented for how this 
would be accomplished.   
Dr. Yoon has continued his work in Dr. Krishnagopolan?s group at Auburn 
University.  He has found that pine yield can be preserved following prehydrolysis if 
sodium borohydride is added to the pulp following the prehydrolysis treatment (Yoon 
2009).  Sodium borohydride has long been known as a yield preservative in pulping and 
bleaching but is not used in industry due to high cost.   
Potential for VPP is not limited to kraft mills.  There has also been work done 
with thermo-mechanical pulping (TMP) where the pretreatment accomplishes a reduction 
in refining energy of about 25% in addition to extracting sugar for potential ethanol 
production.  Recent work that compared different acids for this process showed yield loss 
of up to 30% in a one hour treatment with .01 molar sulfurous acid at 150?C (Rudie, et al. 
40 
 
2007).  Very little lignin was removed in this treatment.  The primary drawback to this 
process is a significant reduction in brightness.   
In summary, sugar can be removed from hardwood and softwood by heating the 
wood in water.  The rate and quantity of sugar removal can be controlled by controlling 
the temperature and time of the reaction.  Discovery of a financially advantageous 
method of recovering sugar from softwood is complicated by a lower percentage of 
hemicellulose in the wood and lignin that is sensitive to the conditions of extraction.  It 
may be possible to further control the type of hemicellulose impacted by the extraction 
and the impact on pulp properties by addition of chemical additives for pH control or 
other purposes.   
  
41 
 
 
 
Chapter 3  Experimental Apparatus and Procedures 
The experimental portion of this work was conducted in the laboratories of the 
Chemical Engineering department at Auburn University.  The laboratory equipment and 
chemicals were provided by the Alabama Center for Paper and Bioresource Engineering. 
3.1 Materials and Equipment 
 Loblolly pine chips (Pinus Taeda) produced from two individual 16 year old trees 
by Rock -Tenn Company in Demopolis, Alabama were screened on a CHIP CLASS TM 
laboratory screen using a stack from top to bottom of 45 mm holes, 8 mm bars, and 2 mm 
bars over the pan.  Accepts were taken from the 2 mm bar tray.  The screened chips were 
air dried and stored in plastic zipper bags which were sealed in a 55 gallon drum until 
needed to maintain uniform moisture content.   Bark and knots were manually removed 
before use. This chip screening method was chosen because it closely simulates the 
accepted chips that are fed to industrial digesters.  Chips that are smaller than 2 mm 
thickness may pass through internal digester screens and plug downstream equipment.  
Chips that are larger than 8 mm thickness may not cook completely.  Smaller particle size 
may give more accurate information on the chemical kinetics of the reactions involved 
but will not give relevant information on mass transfer effects.   
Composition analysis of the chips was performed at the USDA Forest Products 
Lab (FPL) in Madison, WI and at Auburn University (AU).  The carbohydrate analyses 
from the two labs were in very good agreement with each other, but the klason lignin 
42 
 
determination was 2% of wood higher in the AU results.  Table 1 shows the wood 
composition analysis results including the averages of the AU results that were used for 
further calculation.   
 
The approximate hemicellulose composition was determined using the 
carbohydrate composition analysis and literature values for molar ratios of the sugar 
monomers and typical values for softwood composition (Sjostrom 1993, 64).  Table 2 
illustrates the method used to estimate the macromolecular composition of the wood.  
The estimated wood fraction of each polymer was input in the ?Wood? column.  These 
estimates were adjusted until monomer percentages calculated along the bottom of the 
table were close to those determined in the composition analysis of the wood chips.  The 
glucomannan fraction was estimated to be 1% lower than typical for softwood while the 
arabinoglucuronoxylan fraction was higher.   
White liquor for kraft cooks was prepared from sodium sulfide hydrate and 
sodium hydroxide pellets and tested by the ?ABC? titration with hydrochloric acid.  
Reagents for pH control were prepared from glacial acetic acid, 10 N sodium hydroxide, 
and granular sodium carbonate.  Other reagents were used as required by analytical 
Table 1  Loblolly pine wood composition measurements. 
R a w  C h i p  C o m p o s i t i o n ,  %  o f  B o n e  D r y  W o o d
AU AU AU AU FP L FP L AU FP L
G l u ca n     4 0 . 0 7      4 0 . 2 5      4 0 . 8 3      4 1 . 1 5      4 1 . 1 0      3 9 . 6 0      4 0 . 5 8      4 0 . 3 5  M a c r o m o l e c u l e
X y l a n       7 . 3 9        7 . 4 3        8 . 2 0        8 . 1 5        7 . 1 1        7 . 8 3        7 . 7 9        7 . 4 7  G a l a c t o g l u c o m a n n a n
G a l a ct a n       2 . 3 8        2 . 0 5        2 . 0 6        2 . 2 3        1 . 9 9        2 . 2 5        2 . 1 8        2 . 1 2  G l u c o m a n n a n
A ra b i n a n       1 . 5 8        1 . 5 0        1 . 5 1        1 . 5 7        1 . 3 6        1 . 4 5        1 . 5 4        1 . 4 1  A r a b i n o g l u c u r o n o x y l a n
M a n n a n       8 . 9 0        8 . 9 1        9 . 2 3        9 . 3 5        9 . 7 7        8 . 8 4        9 . 1 0        9 . 3 1  C e l l u l o s e
T o t a l  C a rb o h y d ra t e s     6 1 . 1 9      6 0 . 6 5  G a l a c t a n
K -L i g n i n  %     3 1 . 5 4      3 1 . 5 4      3 3 . 2 4      3 3 . 2 4      3 0 . 6 0      2 9 . 6 0      3 2 . 3 9      3 0 . 1 0  A r a b i n a n
A ci d -S o l u b l e  L i g n i n       0 . 4 8        0 . 4 8        0 . 3 3        0 . 3 3        0 . 4 1  
A s h  co n t e n t       0 . 3 1        0 . 3 1        0 . 3 2        0 . 3 2        0 . 4 0        0 . 5 0        0 . 3 2        0 . 4 5  
T o t a l     9 2 . 6 5      9 2 . 4 7      9 5 . 7 2      9 6 . 3 4      9 2 . 3 3      9 0 . 0 7      9 4 . 3 0      9 1 . 2 0  
A v e r a g e
43 
 
methods.  Chemical additives for pulp yield improvement trials are discussed in each 
section of Chapter 6.  
 
 
The extractions and cooks were performed in a M/K Systems Inc. twin laboratory 
digester with liquor circulation as illustrated in Figure 8.  Some of the cooks were 
performed in 500 ml stainless steel ?bomb? containers that were placed in water in the 
M/K digester.  The digester system was modified over the course of the experiments.  
Modifications to the equipment included a crossover line to allow for transfer of liquor 
(wash) from one digester to the other, systems for online pH measurement and 
adjustment, and a back-up temperature sensor.  The modifications added approximately 
250 ml to the volume of the circulation piping.   
The back-up temperature sensor was installed in the circulation loop as the 
reference point for all temperature measurements and H-factor calculations beginning 
with the prehydrolysis designed experiment.  The original sensor was found to read two 
degrees higher than other sensors at room temperature and approximately three degrees 
higher at cook temperature.  The original sensor was retained so an estimate of the back-
Table 2  Macromolecule composition estimation matrix. 
R a t i o s
G L U X Y L G A L A R A M A N A c e G L U p A U n i t
M a c r o m o l e c u l e T y p i c a l *  W o o d C a r b  162 132 162 132 162 45 177 MW
G a l a c t o g l u c o m a n n a n 5 - 8    5 . 0 0       7 . 7 6  1 0 1 0 3 1 0 855
G l u c o m a n n a n 1 0 - 1 5    9 . 0 0     1 3 . 9 8  1 0 0 . 1 0 4 1 0 871
A r a b i n o g l u c u r o n o x y l a n 7 - 1 0  1 0 . 8 0     1 6 . 7 7  0 10 0 1 . 3 0 0 2 1846
C e l l u l o s e 3 7 - 4 1  3 8 . 0 0     5 9 . 0 1  1 0 0 0 0 0 0 162
G a l a c t a n    1 . 2 0       1 . 8 6  0 0 6 1 0 0 0 1104
A r a b i n a n    0 . 4 0       0 . 6 2  0 0 0 1 0 0 0 132
 6 4 . 4 0   1 0 0 . 0 0   4 0 . 6 2    7 . 7 2   2 . 1 7   1 . 5 5   9 . 5 4   0 . 7 3     2 . 0 7  6 4 . 4 0  
s u g a r s 6 1 . 6 0  
*  T y p i c a l  v a l u e s  t a k e n  f r o m  S j o s t r o m  1 9 9 3 ,  p .  6 4 .
M a s s  P e r c e n t
44 
 
up temperature measurement could be used to recalculate the temperature readings when 
results from earlier tests were compared to those of the designed study and later. 
Figure 8  M/K Digester configuration. 
 
Several options were considered for pH control including manual sampling, 
cooling a side stream to room temperature for external measurement, and two different 
systems of online measurement.  A system of buffers was briefly considered, but was not 
attempted since the pH target and the potential effects of the buffer chemical on the 
prehydrolysis were not known.  Samples of the prehydrolyzate were taken at various 
 
Fl o w  
M et er  
m ET E
R  
pH  
M et er  
Sa m p le  
L in e  
pH C o n tr o l  
C h em ical  
H
E
A
T
E
R  
D IGES TE R  1  
(WASH  
L IQUOR  
STORAGE)  
H
E
A
T
E
R  
D IGES TE R  2  
( REACTION  
VES SEL )  
Co o li n g  Wat er  
Te m p  
Sen so r  
Te m p  
Sen so r  
Back - up 
Te m p  Senso r  
45 
 
times and the pH was measured in the early experiments, but the limited frequency of 
samples, reduction of circulation volume and time lag of cooling the samples was not 
acceptable for control.  The system for cooling a side stream for pH measurement 
discussed earlier was also considered, but the time lag in the system was fairly long and 
the system was complicated (Weil 1998).  A commercial apparatus for measuring pH at 
digester temperature and pressure in sulfite cooks was patented in 1968.  This system had 
separate glass, reference and temperature probes with pressure and temperature control 
on the reference (Ingruber, Kocurek and Wong 1985, 52-53).   This system would have 
been impractical on the laboratory digester due to the liquid volume required.   
A Rosemount Model 3300HT pH probe was selected for this research and 
installed in a cooled side stream bypass line to monitor pH continuously.  This probe is 
simple to maintain and can be used up to 145?C @ 100 psi.  The online pH measurement 
then provided timely information and required a minimum of cooling from the maximum 
circulation temperature of 170?C used in the studies.  A large cooling requirement would 
slow the response time of the digester temperature control system.   
The system for alkali or acid addition to the primary circulation loop consists of a 
LMI diaphragm pump and a couple of valves.  The required reagent is pumped from a 
graduated cylinder so the volume can be measured.   When the control requirement 
changes from acid to alkali as the extraction proceeds, the change is a simple switch of 
the cylinder and then to flush the feed line using the digester pressure and the feed pump 
from either end of the line to the drain valve.  Acetic acid was used to lower the pH.  
Sodium hydroxide was used to raise pH for the earliest experiments, but was replaced 
with sodium carbonate after sugar degradation was observed. 
46 
 
A room temperature (24?C) soak period at various conditions was performed in 
plastic zipper bags prior to the prehydrolysis step.  A vacuum oven was used to remove 
air from the dry chips for the pre soak and in the case of bomb cooks with raw chips.   
3.2 Experimental Procedures 
The moisture content of the air-dried chips was checked by drying samples in an 
oven at 105?C before they were bagged, a few weeks after they were bagged, and again a 
year later prior to a cook.  The results were all within 8?0.2% moisture.  91.82% solids 
was used to calculate chip weight for all experiments.  In the preliminary experiments, 
the calculated amount of dried chips for each prehydrolysis was weighed and put in the 
chip basket.  The dry chips were weighed into beakers that were placed in the vacuum 
oven covered with soak solution for the designed experiment and later tests.  The oven 
vacuum was run up to 26 inches of mercury vacuum until air bubbles stopped appearing 
on the chips.  Two or three runs in the oven were sometimes required to make sure all 
chips were covered with solution after the vacuum was released.  The wet chips were 
then placed in plastic bags for the pre-soak period in the designed experiment.  The soak 
solution was added to the bags to saturate the chips at a liquor-to-wood ratio of 2.  The 
bags were then closed and left at room temperature (24?C) for 24 hours.  After the soak 
period, the wet chips were placed in the chip basket for the prehydrolysis.   
The prehydrolysis extractions and the kraft cooks completed before the designed 
experiment were performed in the digester with liquor circulation.  The kraft cooks for 
the designed experiment and later tests were performed in the 500 ml ?bombs? so only 70 
grams of chips would be required and up to three cooks could be performed at once with 
the same time/temperature profile.  The liquor circulation allowed the flexibility of 
47 
 
sampling the hydrolyzate and controlling the pH of the circulation.  A generalized 
method for a digester extraction followed by digester cook is described here followed by 
a generalized description of a bomb cook.  Modifications to these methods and key 
measurements were recorded for each cook in the lab notebook and a Microsoft Excel 
spreadsheet that was then used to perform calculations to allow comparison of the cooks.  
Examples of these spreadsheets are included in Appendix 3.  
Water was added to the digester to just cover the chips and the liquor circulation 
was started.  Any chemical for pH adjustment prior to the start of the extraction was 
added at this point.  After the water had circulated for a few minutes, additional water 
was added to just cover the chips to make up for the volume in the circulation piping and 
any water absorbed by the chips.  Additional water was then added to account for the 
volume of samples to be taken.  The top of the digester was bolted down and the control 
profile was started.  The digester vent was closed when the circulation temperature 
reached 90?C.  The digester was vented to remove the remaining air when the 
temperature reached 100?C and again at 105?C.  All liquid additions or removals from 
the digester were measured so the sugar concentration measurements in the hydrolyzate 
samples could be related to the chip mass.  The programmed temperature ramp was 45 
minutes to temperature target for all runs before the implementation of pH monitoring.  
The ramp was extended to one hour after the first two runs with pH measurement due to 
heat loss in the side stream loop.  This heat loss was reduced by insulating the circulation 
piping and carefully adjusting the pH meter temperature control during the ramp.  Small 
samples of the circulating prehydrolyzate were taken at set temperatures and regular time 
intervals after the digester reached target temperature.  These samples were then tested 
48 
 
for carbohydrate composition.  The prehydrolysis liquid was drained through a cooling 
coil and collected at the end of the designated time period.  
A wash sequence was followed after the prehydrolysis.  Various methods were 
used on different preliminary runs including using heated water and heating the wash 
circulation to 105?C. Room temperature water (25?C) was added for wash in the 
designed experiment runs to aid in cooling the chips and digester.  The resulting wash 
temperature was about 60?C and was recorded for each cook.  Cooling the chips reduced 
evaporation during weighing following extraction.  After the wash, the digester was 
drained under pressure before opening.  A sample of wash was taken for pH and 
composition analysis.  The chip basket was removed from the digester and the chips were 
transferred to a zipper bag for weighing.  A sample of 10% of the prehydrolyzed chips 
was removed for composition analysis before the chips were returned to the digester for 
immediate cooking or separated into 70 g dry chip equivalent samples in individual 
zipper bags and put in the refrigerator for later ?bomb? cooking.   
Several different cooking conditions were used in the preliminary cooks.  All later 
cooks used synthetic white liquor made up of sodium hydroxide and sodium sulfide with 
30% sulfidity.  The white liquor was added to the chips for 20% active alkali on dry raw 
chips followed by sufficient water to cover the chips.  This caused relatively high liquor 
to wood ratios of around 7.  Following a digester cook, the black liquor was slowly 
drained through a coil in an ice bath.  A sample of black liquor was collected for pH 
and/or residual alkali testing.  The digester was then opened and fresh water circulated to 
wash and cool the pulp.  The pulp was then removed and washed further in a filter sock 
before it was defiberized in a laboratory blender, squeezed dry, and homogenized.  The 
49 
 
homogenized pulp was spread in a tray to air dry to improve uniformity before weighing 
for yield and kappa number analysis.   
The bombs were used for the kraft cooks to determine the standard yield versus 
kappa number curve, the kraft portion of the designed experiments, and the yield 
improvement treatments.  In each case, the equivalent of 70 grams of dry raw chips was 
processed in the bomb.  The weight of extracted chips was determined by ratio of the 
total weight of wet chips to the original weight of dry chips charged to the digester.  The 
chips were placed in the bombs and the bombs were shaken down so the chips would fit 
closely.  The white liquor or other reagents were mixed in labeled graduated cylinders so 
they would be premixed before they were contacted with the chips.  All reagents were 
charged based on a liquor to wood ratio of 7.0 for 70 g dry chip weight regardless of 
pretreatment.  In the case of dry chips, the bombs were placed in a vacuum oven to 
remove air.  About 10% of the liquor was reserved until after this treatment so the bombs 
would not foam over.  The bombs were capped and shaken vigorously for two minutes 
before they were placed in the M/K digester.  The digester was filled with water to just 
cover the bombs and then sealed for the cook.  The digester was not vented during the 
bomb cook.   
The time for the cook was calculated based on an H-factor target and the 
programmed temperature profile.  The H-factor was calculated as the cook progressed to 
account for differences in the temperature ramp and transition to steady temperature 
operation of each cook.  Examples of these determinations are included in Appendix 3.  
When the time for reaction of the bomb cooks was complete, the circulation water was 
directed through the in-line cooling coil to cool the system before the bombs were 
50 
 
removed from the digester and placed in an ice water bath for further cooling.  The 
bombs were then opened and the cooked chips dumped into the filter sock for washing. 
After washing, the pulp was blended, homogenized and spread on a pan for drying.   
When the bombs were used for intermediate yield additive treatments, the chips 
were typically not dumped from the bombs before kraft cooking.  Instead, liquid was 
poured from the bombs as needed to be replaced by the concentrated white liquor to 
maintain the original liquor to wood target.  A sample of this intermediate liquid was 
retained for further analysis.  Some other specific method changes were required for 
application of the yield additives.  These changes are discussed in Chapter 6.   
3.3 Analytical Procedures 
Tappi standard methods and methods published by the National Renewable 
Energy Laboratory in 2004 were followed for testing the composition of chips, pulp, and 
hydrolyzate.   These methods include: 
Tappi T236 cm-85 Kappa Number of Pulp 
NREL Preparation of Samples for Compositional Analysis 
NREL Determination of Total Solids in Biomass 
NREL Determination of Structural Carbohydrates and Lignin in Biomass 
NREL Determination of Sugars, Byproducts, and Degradation Products in Liquid 
Fraction Process Samples 
Customary methods were used for testing white liquor strength and black liquor 
residual alkali.  These methods are included in Appendix 2.   
51 
 
 
 
Chapter 4  Prehydrolysis of Pine Wood Chips 
4.1 Preliminary Work  
An undergraduate lab group performed the preliminary work discussed in this 
section in spring 2006 as a class project under the direction of Dr. Krishnagopalan and 
this researcher.  Samples of the prehydrolyzate were taken every 15 minutes up to a 3 
hour reaction time for water extraction at 170?C.  The measured sugar concentration 
trends shown in Figure 9 illustrate that reaction time beyond two hours had little net 
improvement in sugar recovery.  Similar experiments were performed with acetic acid 
 Figure 9  Extracted sugar concentration as a function of time for water hydrolysis. 
0
5
10
15
20
25
0 30 60 90 120 150 180
Sugar
conce
nt
ra
tion
, g
/l
Time, minutes
Glucose
Xylose
Galactose
Arabinose
Mannose
Total
52 
 
added at a charge of 3% on wood to drop the initial pH to 3.5.  The composition analysis 
results were similar, but the time required for reaction was reduced with the sugar 
concentration in the hydrolyzate approaching a maximum in one hour instead of two.  
The maximum concentration of sugar recovered in the extractions corresponds to 
approximately 50% of the hemicellulose in Loblolly pine.  The ratio of galactose: 
glucose: mannose is approximately 1: 1: 3 indicating that cellulose was minimally 
affected even by these extreme treatments.  The concentrations of xylose, galactose, and 
arabinose peaked and then fell over the course of the treatment indicating that these 
components may be degraded by the severity of the treatment.  Many of the arabinose 
and some of the mannose measurements reported are estimates due to interference 
between these two components in the HPLC Column.  The lab group did not attempt to 
measure yield of the pulp.   
The prehydrolyzed chips were pulped to a target kappa number of 30 ?2.   The 
pulp was processed in a valley beater and handsheets were prepared for property testing.  
Selected pulp properties for three of these cooks are listed in Table 3.  These tests 
indicated reductions of 25% and 15% in burst and tensile index and a dramatic decrease 
in air permeability when compared to a control kraft pulp.  The cook numbers correspond 
to different conditions:  Cook 1 was a kraft cook without prehydrolysis.  Cook 2 had a 2 
hour prehydrolysis treatment with water at 170?C  before kraft cooking.  Cook 3 had a 1 
hour prehydrolysis treatment at 170?C with an acetic acid charge of 3% on wood added at 
the beginning of the prehydrolysis followed by kraft cooking. 
 
 
53 
 
Table 3  Preliminary study, selected conditions and results. 
Cook 1 2 3 
Prehydrolysis time, hours 0 2 1 
Prehydrolysis solution - Water Acetic acid, 
3% of wood 
Kappa number 30.0 29.3 30.2 
Burst Index, kpa 6.8 5.0 5.1 
Tensile Index, kN/M 190 162 162 
Porosity, seconds 37.5 99.6 69.9 
 
The preliminary studies suggest that hemicelluloses are easily removed from pine 
chips with mild acid or water at typical kraft cooking temperature and that cellulose yield 
and quality are relatively unaffected by this treatment.  However, degradation of the 
extracted sugars during the time at temperature and reduction in strength of the pulp after 
kraft cooking the prehydrolyzed chips hint that successful VPP extraction may be 
significantly different than conventional kraft pulping.  A procedure that will extract the 
sugars without degradation and without reducing the strength of pulp will require at a 
minimum that the time and temperature of the reaction be controlled. 
4.2 Second Preliminary Work 
Further preliminary work was done to test modifications to laboratory equipment 
and investigate potential variables for further testing.  Much of this work was presented at 
the August 2008 TAPPI Engineering Pulping and Environmental (EPE) Conference in 
Portland, Oregon (Smith, Cullinan and Krishnagopalan 2008). 
Several digester cooks produced control kraft pulps and kraft pulps from chips 
that had been water extracted for 30 minutes, 60 minutes, or 2 hours at 170?C.   Kraft 
cooking was to a kappa number target of 30. The temperature target and cook time were 
both varied to seek this target, so the H-factor is used to facilitate comparison of these 
cooks.  The H-factor is a severity parameter that combines the effects of temperature and 
54 
 
time into a single variable.  It is discussed further in Chapter 7 under kraft pulping 
models.  The plot of kappa vs. H-factor in Figure 10 shows that the chips with a 30 
minute prehydrolysis treatment required lower H-factor to cook to a target kappa number 
while the chips from the two hour pre-treatment required more severe conditions to cook 
to 30 kappa than raw chips.  White liquor was charged at 20% active alkali on bone dry 
raw chips, so the prehydrolyzed chips saw a slightly higher alkali charge on chip weight.  
The higher difficulty in cooking the two hour chips is in accordance with previous results 
wherein higher temperature, alkali, or both were required to reach the same kappa target 
following extended prehydrolysis of softwood at 170?C (Richter 1956).  Further 
experimentation could determine the optimum pretreatment time and temperature as it 
relates to cook time and yield at a given kappa.   
Figure 10  Pulp kappa number as a function of cooking H-factor.  Legend indicates 
pre-treatment time. 
 
20
22
24
26
28
30
32
34
36
38
2000 2100 2200 2300 2400 2500 2600 2700
Kappa
 number
H-factor
Raw chips
30 min
60 min
120 min
55 
 
4.3 pH Controlled Prehydrolysis of Pine Wood Chips 
Six preliminary prehydrolysis runs were also performed with the pH measurement 
and pH adjustment chemical addition systems in place.  Addition of the pH measurement 
loop increased the circulation volume and changed the energy balance on the digester, so 
the time-to-temperature ramps for these cooks took up to 67 minutes compared to 45 
minutes for previous runs.  The composition analysis results are indexed using H-factor 
as a reference that combines time and temperature following the example of Yoon (Yoon, 
MacEwan and van Heiningen 2006) to facilitate comparison between the pH adjusted 
prehydrolyzate samples and those from a run with no pH adjustment.  Each of the first 
three runs was held at 170?C for two hours following the ramp.  A second set of three 
batches was then run with the time at temperature to maximize sugar recovery 
determined by the prehydrolyzate sugar concentration curve. The pH profiles for the first 
set are shown in Figure 11 along with a representative pH profile for a run without pH 
control and a representative temperature profile taken from the acid run.  Two runs had 
an initial charge of NaOH and are designated ?0.1% NaOH? and ?0.3% NaOH?.  The 
third had an initial charge of acetic acid and is designated ?3% Acetic?.  The pH was 
controlled with 0.5 N NaOH at pH 5 after it dropped to that level in the cases of high 
initial pH.  The chips were allowed to react for 30 minutes before the caustic feed was 
started to control the pH at 5 in the acid case.  A sample of prehydrolyzed chips was 
taken following the wash in each case for moisture determination and composition 
analysis; the rest were stored in the refrigerator to be divided later for ?bomb? cooking.   
 
56 
 
Figure 11  Prehydrolysis pH and temperature profiles.  Legend indicates initial 
chemical addition. 
 
The pulp yield results for the bomb cooks from all of the prehydrolysis cooks are 
plotted against the kappa number in Figure 12.  There is an apparent trend of reduced 
yield with increased severity of prehydrolysis.  The comparison of pulp yield should be 
made at the same kappa number or over the same range since the kappa number itself is a 
measurement of lignin yield.  Since one cannot make that comparison with all of this 
data, this limited data set should be interpreted as if each point sat on a line with slope 
similar to the line through the ?Kraft? data. The conventional kraft pulp has the highest 
yield with the two hour treated pulp without pH control having the lowest yield.  The two 
hour pulps that received pH adjustment fall closer to the 30 minute treated pulp.  There 
remains a significant gap between the conventional pulp and all of the pretreated samples.   
 
0
20
40
60
80
100
120
140
160
180
0
2
4
6
8
10
12
0 50 100 150 200
Te
mpe
ra
tu
re
, ?
C
Liquor 
pH
Time, minutes
Water
0.1% NaOH
0.3% NaOH
3% Acetic
Temperature
57 
 
Figure 12  Pulp yield as a function of kappa number for all prehydrolysis runs 
(Legend identifies prehydrolysis time and initial chemical charge if present). 
 
The samples of the prehydrolyzate liquid were tested for sugar composition 
following a secondary hydrolysis.  The results of this testing are shown in Figure 13 
through Figure 16 which plot the extract carbohydrate composition as a percent of 
original dry wood as a function of time.   The quantity of arabinose and xylose and total 
sugar that is detected reaches a maximum at some point and then begins to decline in all 
of the runs.  This is likely due to degradation of the pentoses.  Furfural is detected as a 
potential degradation product.  The first two data points for arabinose in each plot are 
actual measurements; the rest are estimates since the arabinose peak was at least partially 
obscured by the mannose peak in the HPLC chromatograms.  It was assumed for the 
estimate that degradation of arabinose would be at a rate similar to that of xylose.  The 
y = 0.1798x + 38.561
32
34
36
38
40
42
44
46
15 20 25 30 35 40 45
Pulp 
yie
ld, 
% 
bd w
ood
Kappa number
Kraft
30 min
60 min
120 min
120 min 0.1% NaOH
120 min 0.3% NaOH
120 min 3% Acetic
60 min 0.1% NaOH
30 min 3% Acetic
58 
 
quantity of glucose and mannose detected increases slowly in all runs except for the run 
with initial acid addition.     
Figure 13  Sugar recovery as a function of H-factor for water hydrolysis. 
 
The quantities of galactose and glucose trend together for at least the first hour 
after reaching temperature target (approximately 1000 H factor) of the runs with no pH 
control and caustic added shown in Figure 13, Figure 14, and Figure 15.  The quantity of 
mannose is roughly triple that of galactose and glucose indicating that the source of these 
three carbohydrates is likely galactoglucomannan with a ratio of 1:1:3 for the galactose, 
glucose, and mannose monomers.  After this time, however, the quantity of galactose in 
solution begins to decline while the glucose and mannose continue to rise.   
0
1
2
3
4
5
6
7
8
9
0 500 1000 1500 2000 2500
Suga
rs
 re
cov
ere
d, 
%bd 
chi
ps
H-factor
Glucose
Xylose
Galactose
Arabinose
Mannose
Total
59 
 
Figure 14  Sugar recovery as a function of H-factor for prehydrolysis with an initial 
charge of 0.1% NaOH followed by control with 0.5 N NaOH to pH 5. 
 
Addition of caustic prior to the beginning of heating and then later to control pH 
at 5 has an inhibitory effect on the extraction of all of the sugars except arabinose as 
shown in Figure 14 and Figure 15 where total sugar recovery is reduced by 25% and 33% 
respectively.  Arabinose extraction is not reduced, but this is expected since it has been 
shown to be soluble even under the very mild conditions of room temperature and 100?C 
extractions (Casebier 1969).   
The plot of sugar extracts in Figure 16 shows a very different profile than the 
others.  The maximum in glucose, xylose, galactose, mannose and total sugar all occur 
much earlier than in the other runs.  The maximums are also much higher with the initial 
acid addition than those with caustic addition, but the final quantities are similar with the 
exception of xylose, which drops very quickly after addition of caustic begins.   
0
1
2
3
4
5
6
7
8
9
0 500 1000 1500 2000 2500
Sugar
s re
cov
ere
d, %
 bd c
hips
H-factor
Glucose
Xylose
Galactose
Arabinose
Mannose
Total
60 
 
Figure 15  Sugar recovery as a function of H-factor for prehydrolysis with an initial 
charge of 0.3% NaOH followed by control with 0.5 N NaOH to pH 5. 
Figure 16  Sugar recovery as a function of H-factor for prehydrolysis with an initial 
charge of 3% acetic acid followed by control with 0.5 N NaOH to pH 5. 
 
0
1
2
3
4
5
6
7
8
9
0 500 1000 1500 2000 2500
Suga
rs
 re
cov
ere
d, 
% 
bd 
chi
ps
H-factor
Glucose
Xylose
Galactose
Arabinose
Mannose
Total
0
1
2
3
4
5
6
7
8
9
0 500 1000 1500 2000 2500
Sugar
s re
cov
ere
d, %
 bd c
hips
H-factor
Glucose
Xylose
Galactose
Arabinose
Mannose
Total
61 
 
4.4 Material Balances 
The aim of VPP is to extract sugars from pine chips that would be removed in 
kraft pulping while maintaining yield and strength of the pulp after kraft cooking.  From 
an overall material balance perspective, the goal is to move sugar from the black liquor to 
the VPP extract in order to divert that material to higher value products.  Raw chips, 
prehydrolyzed chips, prehydrolyzate liquid and pulps were tested for yield and sugar 
composition for comparison of the overall material balance for the water prehydrolysis of 
kraft pulp.  The results of this testing are displayed in Figure 17 through Figure 23.  
?Black Liquor? is a calculated value that includes everything not accounted for in the 
pulp or prehydrolysis extract.  The value represented by ?Black Liquor? includes all 
degradation products from the prehydrolysis and any other components of the extract that 
are not specifically measured.   
Figure 17  Glucan disposition for water hydrolysis. 
 
The glucan dispositions shown in Figure 17 and Figure 18 demonstrate that only a 
small fraction of glucan is measured in the extract.  As the more severe 2 hour extraction 
0
5
10
15
20
25
30
35
40
45
Raw 
Chips
Kraft Pulp 30 min 60 min 120 min
Gl
uc
an 
yie
ld,
 %
 bd 
chi
ps
Black Liquor
Extract
Pulp
Chips
62 
 
treatment is done, the glucan in the extract increases to 1.6% of wood, but the glucan in 
black liquor increases by over 3%.  The pretreatment effectively moved glucan from 
finished pulp to extract and black liquor which was counter to the goal of moving glucan 
from black liquor to extract.  The most favorable tested conditions relative to the goals of 
VPP were either one hour water hydrolysis or 30 minutes with acid added.  The more 
severe treatments affected the glucan especially in two ways.  First, more time was 
available for prehydrolysis.  Second, the deactivation of lignin forced a more severe cook 
that gave more time for alkaline hydrolysis and secondary peeling. 
Figure 18  Glucan disposition for pH controlled hydrolysis. 
 
The xylan disposition change shows a trend similar to glucan.  Figure 19 and 
Figure 20 illustrate that the extraction reduces xylan yield in the pulp rather than reducing 
the quantity in black liquor.  Xylan remaining with the pulp is cut roughly in half by a 30 
minute water hydrolysis and roughly in half again by continuing the treatment to two 
hours.  More than half of the xylan remains in the kraft pulp while only about one eighth 
remains in the pulp pretreated for two hours.  The most promising extraction condition 
0
5
10
15
20
25
30
35
40
45
Kraft 
Pulp
0.1% 
NaOH   
60 min
0.1% 
NaOH 
120 min
0.3% 
NaOH 
120 min
3%  
acetic 
30min
3%  
acetic 
120 min
Gl
uca
n y
ield, 
%bd 
chips
Pulp Extract Black Liquor
63 
 
from the standpoint of pulp yield was the 30 minute 3% acetic acid treatment, but even 
this treatment sent more xylan into the black liquor than conventional kraft pulping. 
Figure 19  Xylan disposition for water hydrolysis. 
 
Figure 20  Xylan disposition for pH controlled hydrolysis. 
 
Mannan disposition shown in Figure 21 and Figure 22 illustrates some progress 
toward the goals of VPP.  Mannan remaining in the pulp is reduced from about 25% of 
0
1
2
3
4
5
6
7
8
9
10
Raw 
Chips
Kraft Pulp 30 min 60 min 120 min
Xy
lan 
yie
ld, 
%bd 
chips
Black Liquor
Extract
Pulp
Chips
0
1
2
3
4
5
6
7
8
9
10
Kraft 
Pulp
0.1% 
NaOH   
60 min
0.1% 
NaOH 
120 min
0.3% 
NaOH 
120 min
3%  
acetic 
30min
3%  
acetic 
120 min
Xy
lan 
yie
ld, 
% 
bd c
hips
Pulp Extract Black Liquor
64 
 
original wood mannan in kraft pulp to less than 10% remaining in the pretreated pulps, 
but mannan in the black liquor is reduced by half in the two hour treatment.   
Figure 21  Mannan disposition for water hydrolysis. 
 
Figure 22  Mannan disposition for pH controlled hydrolysis. 
 
The plots of total chip weight disposition in Figure 23 and Figure 24 illustrate that 
hydrolysis of pine chips by water alone followed by kraft pulping does not accomplish 
the goals of VPP.  The net movement of sugar is only from the high value pulp to the 
0
1
2
3
4
5
6
7
8
9
10
Raw 
Chips
Kraft Pulp 30 min 60 min 120 min
Ma
nnan 
yie
ld, 
%bd 
chips
Black Liquor
Extract
Pulp
Chips
0
1
2
3
4
5
6
7
8
9
10
Kraft 
Pulp
0.1% 
NaOH   
60 min
0.1% 
NaOH 
120 min
0.3% 
NaOH 
120 min
3%  
acetic 
30min
3%  
acetic 
120 min
Ma
nnan 
yie
ld, 
%bd 
chips
Pulp Extract Black Liquor
65 
 
extract.  The calculated mass of material in the black liquor has not been changed by the 
treatment, but this calculation does not include acetic acid, lignin or carbohydrate 
degradation products in the extract.   
Figure 23  Total disposition for water hydrolysis. 
 
Figure 24  Total disposition for pH controlled hydrolysis. 
 
 
 
0
10
20
30
40
50
60
70
80
90
100
Raw ChipsKraft Pulp 30 min 60 min 120 min
Tot
al 
yie
ld, 
%bd
 chips
Black Liquor
Extract
Pulp
Chips
0
10
20
30
40
50
60
70
80
90
100
Kraft 
Pulp
0.1% 
NaOH   
60 min
0.1% 
NaOH 
120 min
0.3% 
NaOH 
120 min
3%  
acetic 
30min
3%  
acetic 
120 min
Tot
al 
yie
ld, 
%bd 
chips Pulp Extract Black Liquor
66 
 
4.5 Prehydrolysis Conclusions 
Kraft pulps were produced following prehydrolysis performed under various 
conditions to explore the disposition of hemicelluloses.  It is possible to manipulate the 
yield of final pulp and extract by changing the conditions of prehydrolysis.  Online 
monitoring and manipulating pH of the extracting solution is a useful tool in this study as 
it provides detailed information about the reaction conditions and allows them to be more 
closely controlled.   
It can be concluded from these experiments that careful control of the 
pretreatment/ extraction conditions and the cooking conditions is required for value prior 
to pulping from southern pine to be an economically viable process.  The most favorable 
condition tested at 170?C from the standpoint of moving sugar from black liquor to 
hydrolyzate is a 30 minute pretreatment with an initial charge of 3% acetic acid.  Further 
work should concentrate on maintaining pulp yield relative to extracted sugar yield.  
67 
 
 
 
Chapter 5  Prehydrolysis Designed Study 
The purpose of the designed experiments was to look further at the effects of 
changing prehydrolysis time, temperature, pH, and room-temperature-presoak pH with 
the intent of maximizing sugar removal while limiting any negative impact on pulp 
properties. An ideal result would be to remove as much of the sugar as is typically 
removed in kraft pulping. The previous results narrow the list of causes of yield and 
strength losses in pine to primarily hemicellulose effects.   Understanding of 
prehydrolysis effects on kraft pulping has been extended in this study through a detailed 
look at the material balance in the process.  The use of pH monitoring and control in the 
prehydrolysis step allows further investigation into some of these effects.  The Latin 
square experimental design illustrated in Table 4 allows the combined effects of the four 
factors to be looked at together with the relative impact of each factor in fewer 
experiments.  This design assumes that interactions among the factors pH, temperature, 
and presoak condition are negligible or that the impact of one or two of the factors is 
expected to be much greater than the other(s) (Ostle 1963).   
Table 4  Latin square experimental plan. 
A1B1C1 A1B2C3 A1B3C2 
A2B1C2 A2B2C1 A2B3C3 
A3B1C3 A3B2C2 A3B3C1 
 
68 
 
5.1 Designed Experiment Factors 
5.1.1 Variation of Prehydrolysis Time 
The experimental digester was configured to allow sampling of the hydrolyzate 
while the reaction proceeds.  Extractions run under various conditions with sampling on a 
15 minute interval for up to 3 hours were discussed in Chapter 4.  This data determined 
the optimal time for extraction at 170?C to be between 30 minutes and two hours from 
the standpoint of maximum sugar recovery.  In the designed experiments, the extraction 
was run for two hours at temperature.  Following the designed experiment, three 
conditions were repeated but the extractions were terminated at an earlier time based on 
hydrolyzate sugar concentration measured in the two hour run.  The sugar concentrations 
measured in composition analysis are related to the original chip weight by careful 
measurement of the volume of liquids added and removed from the digester.   
5.1.2 Variation of Temperature 
A proper understanding of the effects of temperature of prehydrolysis on sugar 
recovery rates and pulp yield after extraction will improve technical knowledge of VPP 
and aid a design engineer in evaluation of process options.  If the prehydrolysis is to be 
performed in the digester, then a high temperature will not impact the capital cost of the 
equipment, but it will slow production due to residence time used for prehydrolysis and 
could increase operating costs if additional heating is required.  If the prehydrolysis is to 
be performed in a separate vessel, then the construction cost will depend on temperature 
according to the pressure rating required.   
The preliminary studies were performed in the laboratory digester at 170?C.  
Since prehydrolysis at a lower temperature is slower and therefore would require a larger 
69 
 
reactor, there should be an optimum temperature taking into account sugar release, 
hydrolyzate end-use compatibility, pulp strength effects, if any, and a capital cost factor.  
Previous work suggests 150?C to 160?C as optimal for water hydrolysis in production of 
dissolving grades (Bernardin 1958) (Mitchell, et al. 1956) (Richter 1956) so the designed 
experiment included a range centered on 155?C. 
5.1.3 Variation of pH 
In previous work, the maximum sugar yield for water hydrolysis was observed 
when the extract had a pH of about 3.5 (Yoon, MacEwan and van Heiningen 2006). The 
pH was dependent on the integrated time and temperature of extraction in these 
experiments, however and was not controlled independently.  In the preliminary studies 
in this research, application of NaOH for pH control appeared to have an inhibitive or 
even destructive effect on extracted sugars.  Initial application of acetic acid greatly 
accelerated the rate of hydrolysis and extraction.  In other work, the use of a mineral acid, 
0.5% sulfuric, in the hydrolysis produced similar hydrolysis effects at 120?C to those 
observed with water in the same time at 170?C (Richter 1956).  The additional constraint 
to not introduce chemicals with sulfur outside of the conventional kraft cook was initially 
placed on this research.  This constraint limited the choices of pH control agents and 
potential additives.  Acetic acid was chosen for pH reduction since this could, 
theoretically, be a product of the VPP extraction and possibly available from on-site 
production.  The use of a buffer at a pH range of 3.5 to 4.5 has also been suggested 
(Francis 2006). A buffer could be an option after an appropriate pH target is identified.  
Economics will be an important factor in determining an industrial pH control method if 
one is recommended.   
70 
 
Lower pH and higher temperature are each expected to accelerate the hydrolysis 
of hemicellulose, the degradation of dissolved sugar monomers, and condensation of 
lignin.  Observation of each of these effects at various pH levels should aid in the choice 
of an optimal pH depending on which effect is most important to success of the process.  
Highly alkaline environments also quickly degrade dissolved sugar monomers.  Sodium 
carbonate was used in place of sodium hydroxide to raise the pH in the designed 
experiment to reduce the possibility of a locally high pH. 
5.1.4 Variation of Presoak Conditions 
The intention of the presoak was to rehydrate the dried chips and test any effect of 
changing the pH of the rehydrating liquor on prehydrolysis and on subsequent kraft 
pulping.  The chips were air- dried when received to limit degradation of the chips due to 
biological growth and to reduce variability over time in measuring the quantity of wood 
for each test.  Liquor penetration in air-dried chips is much slower than that in wet chips, 
however, and can lead to increased pulp quality variability attributable to chip size 
(Jimenez, McKean and Gustafson 1990).  This delay is attributed to the required 
dissolution and diffusion of air trapped in the chips.  To optimize movement of reactants 
in and out of the chips, air should be removed from the chips.  This is accomplished 
industrially by replacing the air with water vapor in presteaming.   
Soaking the chips in liquid to remove the air does take longer than presteaming 
them, but the capillaries between tracheids in the chips facilitate this happening quickly.  
The chip voids should be completely full of the presoak when the chip weight is 67% 
water (Rydholm, 295-297) or at a liquor-to-wood ratio of 2.  The presoak was accelerated 
71 
 
by placing the chips, covered in the presoak solution, in a vacuum oven until bubbling 
stopped at 26 inches of Mercury.   
Prehydrolysis kraft is a multistage process, but adding a presoak stage provided 
an opportunity to test more effects.  Other forms of multistaging could follow the lead of 
some sulfite mills by beginning with buffered liquor at near neutral and then adding acid 
to meet the pH target for hydrolysis (Ingruber, Kocurek and Wong 1985, 303).  In this 
manner, presoak at high pH may preserve some hemicellulose through both the 
prehydrolysis and kraft cook.  Presoak at low pH may improve hydrolysis rate and yield.  
Previous studies indicate that acetyl and uronic acid along with small amounts of 
arabinogalactan in the wood can be solubilized by room temperature extraction with 
water (Casebier 1969).  Samples of the different soak solutions were retained for 
composition analysis.   
5.2 Hydrolysis Target 
Approximately 20% of the wood weight as sugars is lost to the black liquor in the 
laboratory kraft process at Auburn University.  Ideally, then the goal in VPP is to recover 
20 percent of the wood as hemicellulose in the hydrolyzate.  The difficulty is that the 
kraft cook after pretreatment also removes carbohydrates and this removal appears to be 
enhanced by the prehydrolysis.  The question is then expanded to be not only how much 
can be removed by extraction, but also what characteristics of the process should be 
observed to demonstrate sufficient extraction for value while minimizing damage to the 
primary product, the wood pulp. 
Wood weight loss in hydrolysis and sugar recovered from the hydrolyzate do not 
have a simple relationship, therefore both chip weight loss before pulping and 
72 
 
composition of the hydrolyzate liquid as a function of time were measured.  The need to 
measure both was demonstrated in the preliminary work as illustrated in Figure 25.  One 
might expect that some non-sugar material is removed and that some of the removed 
sugars are degraded by the extraction process.  An example of the latter is the ?Acetic? 
data.  The point labeled ?Max? is the total sugar analyzed after 30 minutes of hydrolysis.  
The point below it in the chart is the result after two hours in the same prehydrolysis run.  
The chip weight loss after two hours is used for both points.  Sodium hydroxide was 
added to raise the pH to 5.0 beginning at 30 minutes in this run.   
Figure 25  Hydrolyzate sugar vs. chip weight loss. 
 
5.3 Experimental Design 
The prehydrolysis screening experiments consist of a Latin square of three factors 
at three conditions (Hendrix 2002).  A fourth factor, prehydrolysis time-at-temperature, 
(Max)
0
2
4
6
8
10
12
14
0 5 10 15 20 25
Rec
ov
ere
d s
ugar
, %
bd 
wood
Chip weight loss, %bd wood
Acetic
Caustic
Water
73 
 
was measured by way of sugar recovery in the hydrolyzate for all conditions.  The factors 
and conditions are listed in Table 5 along with the letter and number corresponding to 
each factor.  As an example, ?A1? refers to prehydrolysis pH 3.0. 
The experimental plan is represented by a square with nine experiments illustrated 
in Table 4.  The design tests each level of each factor at all three levels of the other 
factors.  The experiments were performed in random order. 
Table 5  Latin square conditions for prehydrolysis screening. 
 
The randomized prehydrolysis runs were assigned a letter label for reference in 
the order the cooks were performed.  Subsequent prehydrolysis cooks that followed the 
initial nine were coded with letters in the order the cooks were performed for ease of 
reference.  The initial nine cooks and the conditions they represent are listed in Table 6. 
Each of the first nine runs was programmed to circulate for two hours after the 
ramp from 30?C to target temperature.  The ramp to target temperature was programmed 
to take 45 minute for cooks ?A? and ?B?.  The digester came to temperature late relative 
to the program, however and control was difficult at the transition from ramp to steady 
operation.  The ramp was changed to one hour for run ?C? and all subsequent 
prehydrolysis runs.  Samples of the prehydrolyzate liquid were taken via a cooling coil 
Factor: Prehydrolysis pH Prehydrolysis 
Temperature, ?C 
Pre-Soak pH 
Condition A B C 
1 3.0 140 3.5 
2 3.5 155 7( DI Water) 
3 4.5 170 10 
74 
 
and stored in a refrigerator for secondary hydrolysis and sugar composition analysis.  
Each prehydrolyzate sample was given a time assignment relative to first reaching the 
temperature target and relative to the start of the ramp.   
Table 6  Latin square experiment order and labeling. 
Label pH Temperature, ?C Pre Soak pH 
A 3.0 170 7 
B 4.5 155 7 
C 4.5 170 3.5 
D 3.5 140 7 
E 3.5 155 3.5 
F 3.5 170 10 
G 3.0 155 10 
H 4.5 140 10 
I 3.0 140 3.5 
 
Two sets of two and six sets of three bomb kraft cooks were performed using dry 
chips to develop standard information for comparison of the prehydrolyzed chips.  The 
results of the first set of two were not used as some of the chips remained undercooked 
after the bombs were opened.  The liquor to wood ratio was increased slightly after this 
by using the vacuum oven to remove air from the chips. The remaining seven sets of 
cooks used the same chip mass:  70 g bone dry, liquor charge: 20% active alkali at 30% 
sulfidity, liquor to wood ratio:  6.0, temperature target: 165?C, and ramp to temperature 
of one hour.  The time at temperature was changed to vary the H-factor.  Pulp kappa 
number and total yield were measured after the pulp had been washed and dried.  These 
75 
 
data were then used to plot kappa number as a function of H-factor and yield as a 
function of kappa number.  These plots are illustrated in Figure 26 and Figure 27.   
Figure 26  Kappa number as a function of H-factor for raw chips. 
 
The standard condition of 1850 H-factor for producing a lignin content of 30 
kappa number was interpolated from the results of cooks 2-5 and was also tested with 
resulting averages for the three bombs of  kappa number 29.0 and 45.0% yield.  Those 
data are also included in Figure 26 and Figure 27.  The curve displayed in Figure 26 is a 
third order polynomial fit by Microsoft Excel ? that approximates the curve that was 
drawn manually to chose the standard H-factor.  The yield calculated at kappa 30 for raw 
chips, 44.63%, was used as the standard for comparison to the prehydrolyzed chip cooks.   
Each sample of prehydrolyzed chips was divided into six subsamples.  Five of the 
subsamples represented 70 g dry chips.  Part of the remaining subsample was used to 
measure the moisture content of the chips for calculation of weight loss during 
20
25
30
35
40
45
50
1000 1500 2000 2500
Pulp 
ka
ppa 
number
H-factor
76 
 
Figure 27  Pulp yield as a function of kappa number for raw chips. 
 
 
prehydrolysis and the rest was reserved for composition analysis.  The first subsample of 
chips from each prehydrolysis run was cooked to 1850 H-factor; the conditions that were 
calculated to produce a kappa number of 30 from the dry chips.  The second subsample 
was cooked with the same conditions except that 0.1% AQ was added.  The chips for the 
third cook were grouped together based on the kappa number of the pulp from the first 
cook.  The H factor was then changed for each group in an attempt to produce a pulp on 
the opposite side of kappa number 30 from that achieved in the first cook.  Chips for the 
fourth and fifth cooks were again grouped together based on target H factor.  The fourth 
cook was used to add a bracketing condition or a central condition around kappa 30.  The 
aim of the fifth cook was similar to the third except that it also used AQ.  The results of 
the cooks with AQ are discussed in Chapter 6. 
y = 0.1624x + 39.761
R? = 0.9567
40
41
42
43
44
45
46
47
48
49
50
20 25 30 35 40 45 50
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
77 
 
5.4 Results 
5.4.1 Prehydrolyzate Composition 
Five different sugars were measured in the composition analysis:  arabinose, 
galactose, glucose, mannose, and xylose.  A secondary hydrolysis was part of the test 
method since the samples contained oligomers and monomers.  The absolute value of 
arabinose and mannose for samples taken at higher times from tests at 155?C and 170?C 
were estimates due to the peaks for these two sugars blending on the chromatogram.  The 
mannose peak covered the arabinose peak in most cases since it was larger.  The 
arabinose concentration was estimated and the measured concentration of mannose was 
reduced by 10% for these samples per Auburn University Chemical Engineering 
Department laboratory standard practice.  The sugar concentration measured in the 
samples was related to the original wood weight using the wood weight charged to the 
digester and the calculated volume of solution circulating in the digester and contained in 
the wet chips.  The mass transfer time between the chip voids and the free liquor was 
assumed to be short relative to the time between samples.  The dissolved sugar 
concentration and temperature were then considered to be uniform since the 
prehydrolyzate liquor was completely turned over by the digester circulation 
approximately once per minute.   
Other components of the hydrolyzate potentially include acid-soluble lignin, 
acetic acid, uronic acid, and degradation products, but these were not quantified in this 
study.  The acid-soluble lignin concentration was measured for some of the preliminary 
hydrolyzate samples discussed in Chapter 4, but these measurements and material 
balances both consistently indicated that this component was less than 1% of wood and 
78 
 
consequently it was not measured for this portion of the research.  The acetyl component 
of the galactoglucomannan and the methyl-glucopyranosyluronic acid component of the 
arabinoglucuronoxylan were also not measured even though collectively they were 
estimated to make up about 3% of the dry wood (0.7% and 2.1% respectively based on 
typical softwood hemicellulose compositions (Sjostrom 1993, 63-66)).  Acetic acid 
measurements would have been confounded by the use of acetic acid for pH control.  
Uronic acid measurements are not a standard test in this lab at Auburn University and 
would be complicated by the lack of available uronic acid standards.  Peaks for the 
primary degradation products for pentoses and hexoses, furfural (FUR) and 
hydroxymethylfurfural (HMF) respectively, were observed on some of the 
chromatograms and notes of a qualitative nature were made regarding their presence or 
absence, but no attempt was made to quantify these or any other degradation products.   
The results of the prehydrolyzate composition analysis reveal a strong dependence 
of sugar in solution on reaction temperature and pH.  The effect of the presoak condition 
was less obvious and is discussed later in this chapter.  Higher temperature and lower pH 
both increased the rate of sugar dissolution.  The highest temperature and lowest pH 
condition did not yield the highest recovered sugar quantity, however, due to degradation 
of the monomers after their release into solution.  The hydrolyzate sugar composition 
data as a function of time are presented in Figure 28 through Figure 36 and discussed in 
rough order of increasing maximum total recovered sugar quantity. 
The slowest sugar removal was at pH 4.5 and 140?C.  Figure 28 shows an almost 
linear increase in total sugar concentration over the time at temperature.  This is the 
combination of relatively fast removal of arabinose in the first 45 minutes that then 
79 
 
slowed and initially slow dissolution of the other four sugars that then accelerated over 
the last hour.  The total sugar removed after two hours represents less than 3% of the dry 
wood and only about 12% of the hemicellulose.  Arabinose is a disproportionate 37% of 
the total recovered since 69% of the original arabinose was found in the hydrolyzate.  No 
FUR or HMF peaks were observed in the hydrolyzate sample chromatograms for this run.   
Figure 28  Sugar recovered as a function of time for prehydrolysis cook "H" pH 4.5, 
140?C, 10.5 pH Soak. 
 
Reducing the pH from 4.5 to 3.5 at the same temperature nearly doubled the total 
sugar removed but was still only 5.5% of wood or 23% of hemicellulose.  Figure 29 
shows again an almost linear increase over the two hours at 140?C.  The sugar recovered 
was doubled with the exception of arabinose.  The individual sugar trends were similar to 
the higher pH condition.  A very small FUR peak was observed in the final sample 
indicating that degradation of pentoses was beginning even under these mild conditions.   
0
1
2
3
4
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
80 
 
Figure 29  Sugar recovered as a function of time for prehydrolysis cook "D" pH 3.5, 
140?C, DIW Soak. 
  
 
Conditions B and I yielded results very similar to each other, showing that 
changes in temperature and pH can be used to control the extent of hydrolysis.  In this 
case, lowering the pH from 4.5 to 3.0 countered the effect of lowering the temperature 
from 155?C to 140?C.  As shown in Figure 30 and Figure 31, the total yield of sugars was 
moderate in these cases at between 7 and 8% of wood and the rate of appearance of 
sugars in the hydrolyzate only slightly decreased in the second hour.  In these two tests 
and all remaining tests in the designed study, essentially all of the arabinose is removed 
from the chips.  The responses of glucose, xylose, galactose, arabinose, and mannose 
were very similar to their counterparts in the two tests.  Very small FUR peaks were 
observed in the chromatograms beginning after one hour for 155?C and pH 4.5 and after 
90 minutes for 140? and pH 3.0 
0
1
2
3
4
5
6
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
81 
 
Figure 30  Sugar recovered as a function of time for prehydrolysis cook "B" pH 4.5, 
155?C, DIW Soak. 
 
Figure 31  Sugar recovered as a function of time for prehydrolysis cook "I" pH 3.0, 
140?C, pH 3.1 Soak. 
0
1
2
3
4
5
6
7
8
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
0
1
2
3
4
5
6
7
8
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
82 
 
Conditions C, A, and F, the tests at 170?C shown in Figure 32, Figure 33, and 
Figure 34, were similar in the very fast initial appearance of sugars followed by a 
reduction in total sugar quantity due to degradation of primarily the pentoses:  xylose and 
arabinose.  Both the initial rate of recovery and the rate of degradation were increased at 
lower pH.  This phenomenon produces optimums relative to sugar recovery in time and 
in pH for operation at 170?C.  The optimum time was 90 minutes for pH 4.5 and 3.5 but 
only 60 minutes for pH 3.0.  The best pH of the three tested at 170?C is 3.5 where 56% of 
the wood hemicellulose sugars are found in the hydrolyzate.  FUR was observed after 
only ten minutes at 170?C in all three tests and the peak was large after 45 minutes at pH 
3.0.  HMF was observed after 90, 45, and 20 minutes respectively as pH decreased from 
4.5 to 3.5 and finally to 3.0.  These observations confirm significant degradation of both 
pentose and hexose at 170?C and especially as the pH dropped below 4.5.   
Figure 32  Sugar recovered as a function of time for prehydrolysis cook "C" pH 4.5, 
170?C, pH 3.1 Soak. 
0
2
4
6
8
10
12
14
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
83 
 
Figure 33  Sugar recovered as a function of time for prehydrolysis cook "A" pH 3.0, 
170?C, DIW Soak. 
 
Figure 34  Sugar recovered as a function of time for prehydrolysis cook "F" pH 3.5, 
170?C, pH 10.5 Soak. 
0
2
4
6
8
10
12
14
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
0
2
4
6
8
10
12
14
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
84 
 
The lower two pH levels tested at 155?C potentially gave the best results overall.  
The results of these tests, illustrated in Figure 35 and Figure 36, show a continual 
increase in concentration of all sugars over the test period with the exception of arabinose 
which was completely removed from the chips within the first hour.  Galactose was also 
completely removed by the end of the test at pH 3.0.  The arabinose data at higher times 
was obscured by the mannose data, so it was not possible to tell if there was significant 
degradation of arabinose under these conditions.  The maximum percent hemicellulose 
removed in these two conditions, 50% and 57% are very similar to the maximum 
removed at the same pH at 170?C, but the trends indicate that more sugars could be 
recovered at 155?C if the prehydrolysis were allowed to continue for a longer time 
period.  A small FUR peak was observed after 30 minutes at pH 3.5 and after only 15 
Figure 35  Sugar recovered as a function of time for prehydrolysis cook "E", pH 3.5, 
155?C, pH 3.1 Soak. 
0
2
4
6
8
10
12
14
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
85 
 
minutes at pH 3.0.  Pentose degradation appeared significant at pH 3.0 after 45 minutes 
and a small HMF peak appeared in the 90 minute sample of this run.  
Figure 36  Sugar recovered as a function of time for prehydrolysis cook "G", pH 3.0, 
155?C, pH 10.5 Soak. 
 
Several conclusions can be made from this group of data by examination of 
Figure 28 to Figure 36.   For convenience, Figure 37  shows the total sugar curves from 
all nine conditions.  The colored lines for each curve are based on temperature:  Yellow is 
140?C.  Brown is 155?C.  Blue is 170?C.  It is clear that higher temperature increases the 
rate and total amount of sugar recovered for all conditions except for sugar recovered 
after more than one hour for pH 3.  Similarly, reducing pH in the range 4.5 to 3.0 
increases the rate and total amount of sugar for all temperatures except for 170?C after 30 
minutes of reaction time where the response at pH 3.5 began to exceed the response at pH 
3.0 due to degradation of sugars.  The solution pH is represented in the chart by the shape 
0
2
4
6
8
10
12
14
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Total
Mannose
Xylose
Galactose
Arabinose
Glucose
86 
 
of the data points:  Square is 3.0.  Triangle is 3.5.  Circle is 4.5.  The presoak pH is 
represented by the color of the data points:  White is 10.  Black is 7.  Green is 3.5.  There 
is no clear trend for presoak pH among these results.  For example, presoak at pH 10 has 
the two highest results, but also the lowest.   
Figure 37  Total sugar recovered as a function of time for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
Comparing the behavior of each sugar for all nine conditions is also valuable.   
Arabinose was removed from the chips quickly under all test conditions.  Even at 
140?C and pH 4.5 (Condition H) 69% of the arabinose in the raw chips was removed 
after two hours and the trend indicates that the remainder would be in solution in less 
than three hours.  Arabinose is expected to significantly degrade after it enters solution at 
170?C, but the arabinose data at higher times was obscured by the growing mannose peak 
on the chromatogram.  As can be seen in Figure 38, the arabinose concentration reaches a 
0
2
4
6
8
10
12
14
0 50 100 150 200
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
87 
 
maximum in the 170?C tests after about 10 to 30 minutes at temperature.  After this time, 
the quantity in solution drops as either the degradation rate exceeds the dissolution rate or 
all of the arabinose has entered solution and begins to degrade.  The degradation rate is 
much slower at the lower temperatures and, therefore most of the arabinose remains in 
solution at the conclusion of the test.  Arabinose is a small fraction of the wood and 
hemicellulose, 1.5% and 7% respectively, but is a disproportionately large part of the 
final hydrolyzate, 12-37% depending on extraction conditions, since it enters solution so 
readily.  Arabinose would be an even higher proportion of the dissolved sugar under any 
of these conditions if the prehydrolysis treatment was stopped early.   
Figure 38  Arabinose recovered as a function of time for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
Galactose removal from the chips was much more widely varied than arabinose 
with only 20% of the original galactose removed at 140?C and pH 4.5 and 100% 
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0 50 100 150 200
Ara
binos
e e
xtrac
ted, 
%bd 
chips
Time, minutes
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
88 
 
recovered at 155? and pH 3.0.  As shown in Figure 39, the rate of removal and quantity 
recovered increased with time, increasing temperature and dropping pH for all tests at 
140 and 155?C.  The observed behavior of galactose at 170?C was more complicated 
with the rate of removal and the rate of degradation both appearing to be pH dependent.  
The reduced rate of removal at pH 4.5 followed by significant degradation at the high 
temperature forced the recovered galactose to reach a maximum at 1.5% of wood or 
about 70% of the galactose available.  The high rate of removal and degradation at pH 3.0 
allowed a maximum of 100% recovery after 60 minutes at temperature but only 91% 
after 105 minutes.  Both rates were slowed slightly at pH 3.5 as 92% of the galactose was 
recovered after 90 minutes at temperature.   
Figure 39  Galactose recovered as a function of time for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
0.0
0.5
1.0
1.5
2.0
2.5
0 50 100 150 200
Ga
lac
to
se
 ex
trac
ted, 
%bd 
chips
Time, minutes
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
89 
 
The glucose recovery trends illustrated in Figure 40 were the simplest of the five 
sugars with recovery increasing with time under all conditions.  Additionally, recovery 
increased with temperature at all pH levels and increased with acidity at all temperatures.  
The rate of recovery did begin to decrease at longer times for the tests at 170?C and for 
the test at pH 3 and 155?C.  The final recovered quantity varied dramatically from less 
than 0.2% of wood up to slightly higher than 2%.  The lower quantity represented only 
7% of the glucose attributed to hemicellulose, while the highest was 78%.   
Figure 40  Glucose recovered as a function of time for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
Mannose is the most plentiful sugar recovered at the end of the prehydrolysis for 
all conditions except for 140? and pH 4.5.  It is also the most plentiful during the course 
of the prehydrolysis with the exception of arabinose during the early part of the 
extractions.  This is due not to efficiency of extraction, however, but to the fact that 
0.0
0.5
1.0
1.5
2.0
2.5
0 50 100 150 200
Gl
ucos
e e
xtrac
ted, 
%bd 
chips
Time, minutes
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
90 
 
mannose is 40% of the initial wood hemicellulose.  The efficiency of extraction varies 
from 7% up to a maximum of 57%.  The shapes of the mannose trends illustrated in 
Figure 41 are very similar to the glucose trends with the exception of the two conditions 
at 170? and lower pH.  This should be expected since the two sugars are found on the 
same hemicellulose molecule.  In the milder extractions, the initial rate is slow before 
accelerating toward the end of the reaction.  In the stronger conditions, the initial rate is 
fast before slowing as the extraction nears 50% of the mannose in the wood.  Evidence of 
degradation is observed at 170?C where the rate slows dramatically after about one hour 
and the concentration even decreases at pH 3.0.  One potential indicator of an 
improvement due to the high pH presoak would have been a lower recovery of mannose 
from the chips pretreated at alkaline pH.  This does not appear to be significant.  
Figure 41  Mannose recovered as a function of time for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
0
1
2
3
4
5
6
0 50 100 150 200
Ma
nnose
 ex
trac
ted, 
%bd 
chips
Time, minutes
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
91 
 
Xylose is the second most plentiful sugar in the hemicellulose and, in most cases, 
it is the second most plentiful sugar recovered in the hydrolyzate.  The trends shown in 
Figure 42 and relationship to pH and temperature are similar to the other sugars with the 
exception of the pronounced degradation observed at 170?C.  The rate of degradation is 
equal to or greater than the rate of removal from the chips after 30 minutes of reaction for 
all conditions at this high temperature.  The recovery rate of xylose was the lowest of all 
the sugars and varied from 7% to a maximum of 41% of the sugar in the original wood 
while it consistently represented about 25% of the sugar found in the hydrolyzate.   
Figure 42  Xylose recovered as a function of time for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). 
 
Glucose is unique among the sugars because it is also available for removal from 
the cellulose component.  One method of considering the impact of the prehydrolysis on 
cellulose is to consider the ratio of glucose to mannose recovered relative to the amount 
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0 50 100 150 200
Xy
los
e e
xtrac
ted, 
%bd c
hips
Time, minutes
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
92 
 
present in the hemicellulose since these sugars are together in the backbone chain of the 
glucomannans and therefore would be expected to be removed at about the same rate and 
degraded at a similar rate since they both have six carbons.  A ratio close to 0.275 would 
indicate very little impact on cellulose while a higher ratio would indicate that some of 
the glucose likely came from cellulose.  This ratio can be compared visually by putting 
the two sets of sugar data on the same plot with separate axes and adjusting the scales of 
the dependent axes so their magnitude is close to the ratio of the two sugars calculated to 
be in the original wood hemicellulose, 0.275.  As can be seen in Figure 43 where the 
glucose data from four tests is connected by solid lines and the mannose data from the  
Figure 43  Glucose and mannose recovered as a function of time compared for four 
conditions (Legend indicates trial label, temperature, reaction pH, and presoak pH.  
Glucose has solid connections.  Mannose connections are dashed.). 
0
1
2
3
4
5
6
7
8
9
0.0
0.5
1.0
1.5
2.0
2.5
0 50 100 150 200
Ma
nnose
 ex
trac
ted, 
%bd 
chips
Gl
ucos
e e
xtrac
ted, 
%bd 
chips
Time, minutes
F 170,3.5,10
G155,3.0, 10
B 155,4.5, 7
H 140,4.5,10
F 170,3.5,10
G155,3.0, 10
B 155,4.5, 7
H 140,4.5,10
93 
 
same tests is illustrated with dashed lines, the plots of the mildest condition, test H, 
almost exactly overlay each other while the stronger conditions, tests G and F, yield 
much more glucose in relation to mannose.   
The impact on cellulose appears to increase with time, as pH drops, and especially 
as temperature increases in the range studied.  The impact illustrated in this chart likely 
understates the real damage done to the cellulose since only relatively small oligomers 
are removed from the chips.  For glucose from cellulose to be recovered in solution the 
hydrolysis must occur very close to an end of the molecule or there must be two 
hydrolytic cleavages very close together.   
Since the experiments were designed in a Latin square, the three tested factors can 
be compared to each other by averaging the results of the three tests at each level 
(Hendrix 2002).  The average results are comparable because each contains the 
variability associated with each other factor at all three levels.   
An analysis of variance can also be performed to test for the significance of each 
factor in changing the results.  Minitab 15, a statistical software package published by 
Minitab, Inc., was used to test a general linear regression model of the three factors for 
several results.  The sugar recovery quantities were considered at the maximum level as 
well as the amounts measured in the final (2 hours) and 45 minute samples.  Temperature 
was identified as the strongest factor at all three positions with percentage contribution 
ranging from 68% for the ?Final? samples to 87% for the samples at 45 minutes.  The 
second largest contributor was pH which ranged from a 12% contribution at 45 minutes 
to 28% at maximum sugar.  The presoak condition was attributed from 3 to 4% 
contribution for ?final? and ?max? samples but was not significant at the 5% level of 
94 
 
confidence.  Presoak was only attributed to 1% of the variation in the data for the 45 
minute sample, but this contribution was determined to be significant statistically.  The 
adjusted R-squared for the models was very good at 99.1, 99.6, and 99.9 respectively for 
Final, Max, and 45 minutes samples.  The residual error was 1% or less for all three.  The 
results of the Minitab 15 analysis for these and other variables are shown in Appendix 1. 
5.4.2 Chip Weight Loss  
Chip weight loss is plotted as a function of temperature for the nine study 
conditions in Figure 44.  The same color and shape scheme is used as in the sugar plots 
so all three factors can be observed in one figure.  The data marker for condition ?D? is 
partially obscured by the marker for condition ?H?.  Weight loss is higher at lower pH at 
all temperatures, if only slightly at 140?C.  Loss is also higher at higher temperature at all 
Figure 44  Chip weight loss in prehydrolysis as a function of reaction temperature for 
all conditions (Legend indicates trial label, temperature, reaction pH, and presoak pH). 
0
5
10
15
20
25
30
130 140 150 160 170 180
W
eight 
los
s, 
%bd 
chips
Temperature, ?C
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
95 
 
pH levels.  Examination of Figure 44 suggests the presoak at pH 3.5, indicated by green 
shape fill, may have increased weight loss slightly, but the ANOVA indicated that 
presoak was not significant for chip weight loss.  The only statistically significant 
variable was temperature which was attributed 76% of the variation.   
Maximum recovered sugar is plotted as a function of weight loss in Figure 45.  
The reference line is a least squares linear fit that was forced to intercept at 0.  The 
marker for condition ?F? is partially obscured by the marker for Condition ?G?.  The 
final recovered sugar values for the high temperature conditions were slightly lower than 
the maximum values used here.  All the plotted values are close to the plotted line except 
for two.  The most severe prehydrolysis condition, test A, had high degradation of sugars 
resulting in high weight loss relative to sugar recovery.  Condition D shows higher sugar 
recovery than weight loss, an impossibility.  The weight loss measurement in this case is 
Figure 45  Recovered sugar as a function of chip weight loss for all conditions 
(Legend indicates trial label, temperature, reaction pH, and presoak pH). 
0
2
4
6
8
10
12
14
16
0 5 10 15 20 25 30
Rec
ov
ere
d s
ugar
, %
 bd c
hips
Weight loss, % bd chips
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
ref
Ideal
96 
 
probably in error.  The prehydrolyzed chip weight measurement is very sensitive to the 
solids content measurement in the wet chips.  Condition D had the highest measured dry 
solids content.  The method was changed later to duplicate this measurement.   
5.4.3 Pulp Yield 
The H-factor for the bomb cooks with raw chips was varied from a low of 1200 to 
a high of 2600 to allow estimation of the required H-factor to reach a lignin content of 
kappa number 30 that was the chosen standard pulp.  The broad range also demonstrates 
for this set of chips the manner in which kappa number varies with cook time and the 
way that pulp yield varies with kappa number.  These data are plotted in Figure 26 and 
Figure 27.  Each point in the plots represents the average measurement from two or three 
bombs that were cooked at the same time.   The error bars indicate one standard deviation 
in either direction from the average.  The kappa number standard deviation decreased 
from 1.84 to 0.15 kappa units as H-factor increased in the study range.  The average 
standard deviation was 0.82, but the pooled standard deviation that takes into account the 
limited number of degrees of freedom in each test was 1.02.  The standard deviation of 
yield did not trend with H-factor and varied from 0.16% to 0.49% of wood with average 
and pooled values of 0.35% and 0.37% respectively.  Variability accounted for in these 
numbers includes that attributed to the raw chips and to each individual bomb cook from 
measurement and handling, i.e. measurement of the dry chips, liquor, water, dry pulp, 
moisture content, and kappa number.  Handling included the choice of bomb, positioning 
in the digester, mixing of chips and liquor, washing and drying of pulp.   
Figure 26 and Figure 27 illustrate for raw chips the ?ease? of cooking to a given 
kappa number and the expected yield at a given kappa number.  Important numbers that 
97 
 
come from these plots are that a cook of approximately 1800 H-factor is required to reach 
kappa 30 and the expected yield from a conventional kraft cook of the Loblolly pine 
chips utilized for this research in the laboratory digester at Auburn University at kappa 30 
is about 44.5%.  The required H-factor was estimated to be 1850 before the final two 
cooks were performed and 1850 was used to target the ?standard? cook for the 
prehydrolyzed chips.  The plots also are important for the shapes of the curves.  The pulp 
kappa number vs. H-factor curve is approximated by a third order polynomial for this 
plot, but the observation that reducing the kappa number below about 25 requires a much 
more extreme cook is important.  The linear nature of the yield vs. kappa curve and the 
slope, 0.16, are important guides taken from Figure 27 for interpreting other data.   
The two hour prehydrolysis treatment had a dramatic impact on the required 
cooking severity in some cases but had very little effect in others.  As shown in Figure 
46, the chips prehydrolyzed at 170?C with pH below 4.5 required 25 to 75% longer cooks 
than the standard for raw chips to reach a target of 30 kappa number, while the chips 
from the test at 155? and pH 4.5 reached the target in 25% less time than the standard.  
The chips from all but the two tests at high temperature and low pH had lower than 30 
kappa number after the standard cook, but in three cases had higher lignin content than 
the raw chips after shorter cooks.  The chips from the mildest treatment had slightly 
lower lignin content than the raw chips over the whole tested range of conditions.  These 
results are similar to those shown in Figure 10 where the shorter length of pretreatment at 
170?C cooked to a lower kappa number at a given H-factor while a longer cook was 
required to process the chips pretreated for longer than one hour.  The H-factor required 
to reach the target of 30 kappa with each set of prehydrolyzed chips was estimated from 
98 
 
this chart using manually drawn curves.  The curves shown are ?power? trendlines from 
Microsoft Excel ? and approximate the manually drawn curves that were used.   
Figure 46  Pulp lignin content as a function of H-factor for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
There appear to be at least two effects that are working counter to each other 
regarding the prehydrolyzed chips response to kraft cooking.  The pretreatment opens up 
the chip structure by breaking bonds and removing some carbohydrates thus reducing 
mass transfer limitations to lignin removal in cooking.  Extended time at high 
temperature and low pH causes condensation reactions in the lignin structure that make 
this lignin resistant to alkaline attack in cooking.   At low temperature and high pH in the 
test conditions H, D, and B, kraft cooking becomes easier as the conditions become 
stronger.  For moderate conditions E, I, C, and G, kraft cooking becomes harder as 
prehydrolysis conditions become stronger and lignin condensation becomes more 
15
20
25
30
35
40
45
50
55
60
65
1000 1500 2000 2500 3000 3500
Kappa
 number
H-factor
Raw
A 170, 3.0,   7
F 170, 3.5, 10
C 170, 4.5,  3.5
G 155, 3.0, 10
E 155,  3.5,  3.5
B 155,  4.5,  7
I  140,  3.0,  3.5
D 140, 3.5,  7
H 140, 4.5,10
99 
 
important, but reaching kappa 30 or below remains easier than for a standard process.  
Finally, at the high temperature and low pH conditions F and A, lignin condensation 
effects dominate and reaching the target lignin content requires a much longer cook.   
A precise interpretation of the impact of the prehydrolysis on cooking severity 
required to reach a kappa number target and on pulp yield cannot be made because the 
alkali charge was not optimized for each condition.  The liquor charge was calculated 
based on the original dry chip weight, not the prehydrolyzed chip weight.  The white 
liquor chemical and water charged to each bomb were held constant while the weight of 
wood remaining after prehydrolysis varied.  The liquor to wood ratio varied from a high 
of 8.3 for condition ?A? to a low of 6.4 for condition ?H? compared to the 6.0 used for 
the raw chips.  Black liquor residual alkali was not titrated for these tests, but the pH of 
the black liquor was measured instead as verification that there was not significant 
leakage in or out of the bombs and that the alkali was not completely depleted before the 
end of the cook.   
Yield loss relative to the standard process is significant in even the mildest of the 
tested prehydrolysis conditions as seen in Figure 47.  The effects of the two main factors, 
pH and temperature, were also consistent within the range of this research:  pulp yield 
loss increased with temperature at every pH level and with acidity at every temperature.  
Kraft pulp yield at 30 kappa was calculated using the best fit linear regression of the three 
cooks.  The slopes of these lines averaged 0.10% per kappa unit, but varied from 0.025 to 
0.25.  The slope is slightly lower on average for the prehydrolyzed chips than the raw 
chips since there is less carbohydrate relative to lignin at the beginning of the cook.  
There is higher variability in the yield data for the prehydrolyzed chips than with the raw 
100 
 
chips since they were weighed when they were wet from the prehydrolysis.  The presoak 
was added to test the possibility of reducing yield loss, but the soak condition was not 
statistically significant according to ANOVA of this data.  Temperature was attributed 
70% of the variance and pH 29% with an adjusted R-squared of 98.4.   
Figure 47  Yield as a function of kappa number for all conditions (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). 
 
Yield loss relative to sugar recovered is a key comparison in evaluating the 
financial impact of prehydrolysis.  Figure 48 shows the calculated pulp yield at 30 kappa 
number as a function of the measured maximum recovered sugar expressed as a 
percentage of the original bone dry chip weight.  The trend line is a linear regression of 
the nine data points forced to pass through the calculated pulp yield for untreated chips.  
The data markers for the high temperature tests at 170?C all fall below the trend line even 
using the maximum values.  The actual final values would be slightly lower.  The data 
y = 0.1604x + 39.819
R? = 0.8882
25
30
35
40
45
50
10 20 30 40 50 60 70
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Raw
A 170, 3.0,   7
F 170, 3.5, 10
C 170, 4.5,  3.5
G 155, 3.0, 10
E 155,  3.5,  3.5
B 155,  4.5,  7
I  140,  3.0,  3.5
D 140, 3.5,  7
H 140, 4.5,10
101 
 
markers for 155?C all fall above the trend line while the low temperature values are very 
close to the line.  The pH and presoak pH conditions appear randomly relative to the line.  
Statistically, all three factors were significant in ANOVA of this ratio, even though pH 
was only attributed 2% of the variance.  Temperature accounted for 68% and presoak 
30%.  This was the only tested situation where the presoak was deemed significant by 
ANOVA.  The R-squared was 99.9.  The average values for the three sets of three 
conditions for this ratio are 1.253, 1.106, and 1.272 for presoak pH levels of 3.5, 7, and 
10 respectively.  The data shown in Figure 48 indicate that 155?C and presoak conditions 
away from neutral can improve pulp yield relative to recovered sugar.   
Figure 48  Pulp yield as a function of recovered sugar for all conditions (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
5.4.4 Additional Cooks 
Ten additional prehydrolysis tests were performed at conditions that appeared 
interesting after the first nine tests.  Tests J and K were intended to test stronger presoak 
25
27
29
31
33
35
37
39
41
43
45
0 2 4 6 8 10 12 14
Unsc
re
ened 
pulp y
ield 
at 30
 ka
ppa
Recovered sugar, % bd chips
Raw
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
102 
 
conditions.  Tests L, M, N, and O did not use a presoak, but the vacuum oven was used to 
remove air from the chips.  Preparing the chips in this manner required about one hour 
from the time the chips were first wet with solution to the start of the temperature ramp.  
The intention of test L was to limit degradation by reducing the temperature after a set 
time period.  Tests M, N, and O were designed to extract a target amount of the wood as 
sugar by repeating conditions E, G, and F at shorter time periods.  Table 7 compares 
operating conditions and results for condition E with these first six additional tests. 
Table 7  Conditions and results for additional prehydrolysis cooks. 
 
The remaining four prehydrolysis tests were all the same and were used to 
produce chips for tests of potential yield improving additives.  These tests each processed 
enough chips to test 12 different conditions in bombs.  The results of the additive tests are 
discussed in Chapter 6.   
Acetic acid was used at a charge of 2% on wood for the presoak in test J which 
was otherwise intended to have the same conditions as test E.  This only dropped the pH 
of the presoak to 2.97 compared to the 3.16 in runs C, E, and I.  The results were similar 
to but not exactly the same as those for test E.   Both chip weight loss and sugar recovery 
were higher in E while pulp yield and cooking severity required were better in J.  The 
Cook E J K L M N O
P re s o a k p H 3 . 1 6 2 . 9 7 11 none none none none
p H  t a rg e t 3 . 5 3 . 5 3 . 5 3 . 5 3 . 5 3 3 . 5
T e m p e ra t u re 155 155 155 1 7 0 ,  1 5 5 155 155 170
M i n u t e s  a t  T e m p e ra t u re 120 120 120 3 0 ,  9 0 60 45 20
H  f a ct o r 450 452 458 717 252 201 392
% W e i g h t  l o s s 1 9 . 8 1 5 . 7 1 8 . 5 1 6 . 4 8 . 4 1 8 . 6 1 9 . 9
M a x  S u g a r 1 1 . 5 1 1 0 . 3 4 7 . 7 7 1 1 . 4 4 7 . 6 9 9 . 6 1 9 . 6 3
% R e c /  %  W t  L o s s 0 . 5 8 0 . 6 6 0 . 4 2 0 . 7 0 0 . 9 2 0 . 5 2 0 . 4 8
K ra f t  y i e l d  a t  3 0 ka p p a 3 6 . 3 3 7 . 3 3 8 . 7 3 5 . 2 3 7 . 8 3 7 . 0 3 3 6 . 4 1
Y i e l d  L o s s 8 . 4 7 . 3 6 . 0 9 . 4 6 . 8 7 . 6 8 . 2
% R e c /  % Y  L o s s 1 . 3 7 1 . 4 1 1 . 3 0 1 . 2 1 1 . 1 3 1 . 2 6 1 . 1 7
H  f a ct o r t o  3 0 1480 1380 1700 1450 1360 1520 1320
103 
 
operating pH and temperature targets for both runs were the same, but the actual achieved 
pH and temperature were slightly different.  The actual measurements are plotted in 
Figure 49.  The pH for test E is slightly lower than that for test J over most of the time 
frame of the experiment.  Acetic acid was added slowly to reduce the pH until about 15 
minutes past the end of the ramp in both cases.  During test E, the operator decided the 
pH remained close enough to the target of 3.5 that it did not need to be adjusted upward.  
Sodium Carbonate was used in test J beginning one hour after the ramp to adjust the pH 
back to target of 3.5.  The variation in pH during cooling of the reactor is typical of all 
the tests as is the temperature variation at the transition from ramp to steady operation at 
the target.  The temperature data displayed is the data that was used for calculation of H 
factor with steady operation at about 152?C.  The reactor temperature was controlled 
automatically with the digester temperature sensor at 155?C.   
Figure 49  pH and temperature for prehydrolysis cooks "E" and "J". 
80
90
100
110
120
130
140
150
160
3.30
3.35
3.40
3.45
3.50
3.55
3.60
3.65
3.70
0 50 100 150 200
Te
mpe
ra
tu
re
, ?
C
Hy
droly
za
te 
pH
Time, minutes
J pH
E pH
J Temp
E Temp
104 
 
The maximum sugar recovery data for E and J presented in Table 7 are slightly 
different than what is reported elsewhere in this document.  Sugar recovery standards are 
analyzed along with the samples in composition analysis and are used to adjust the results 
for degradation during the analysis.  When the total sugar recovery for runs E and J were 
compared and found to be substantially different, 11.85% and 9.8% respectively, the test 
records were compared and it was noticed that the recovery ratios for glucose and 
mannose were unusually low at 91% for E and unusually high at 103% for J.  When a 
typical standard recovery ratio of 98% was used in both cases, the sugar recovery 
measurements for the two runs were much closer as listed in the table.  The remaining 
difference in sugar recovery measured in the samples can be attributed to the slight 
difference in pH over the course of the reaction.  Lowering the presoak pH was intended 
to increase the rate of sugar removal from the chips, but did not appear to make a 
difference in this test.   
Sodium carbonate was added to the presoak solution at a charge of 2% on wood 
for test K.  The solution pH was 11.0 at the beginning of the presoak and dropped to 9.22 
after 24 hours.  The intention of this test was to attempt to maintain higher pulp yield.  
Dilute acetic acid was used to control pH during the prehydrolysis.  The pulp yield was 
slightly higher, but the quantity of sugar recovered in the hydrolyzate was lower, so the 
ratio of sugar recovered to yield lost was not improved.  The chips produced in this 
treatment also required a slightly stronger cook to reach 30 kappa.  The conclusion 
following test K was that the 24 hour presoak at 25?C in the range of pH 3 and 11 had a 
very small, if any impact on either sugar recovered in hydrolyzate or pulp yield, so this 
treatment was discontinued for the remaining tests.   
105 
 
A two level temperature control was attempted for condition L to test the 
possibility of increasing sugar recovery by limiting the degradation of sugars seen at the 
highest temperature in this study.  The temperature was held for 30 minutes at 170?C and 
then dropped to 155?C for 90 minutes.  Dilute acetic acid was used to control pH initially 
and 0.2 N sodium carbonate was used after 30 minutes at temperature.  Sugar recovery 
increased in this test relative to condition J, but so did yield loss.  The glucose: mannose 
ratio for the mixed temperature run was between those for the trials that used just 155?C 
or just 170?C for the entire time, but much closer to that of the 170?C run.  This suggests 
that cellulose may have been more strongly affected by the use of high temperature for 
part of the test.  Lignin does not seem to have been significantly deactivated by this 
treatment, however as the H-factor required to reach 30 kappa was only 1450.   
Conditions M, N, and O were replicates of conditions E, G, and F but held at 
shorter time periods with the intention of extracting 8 - 10% of the wood as sugar.  The 
ratio of recovered sugar to yield loss was improved by the shorter treatment at 170?C, but 
was not as good for the two conditions at 155?C.  Thus the optimal reaction time and 
extraction target with reference to this ratio is different for each given set of reaction 
conditions.  Both are higher for 155?C than for 170?C.   
Limited results from the last four prehydrolysis tests are included in this chapter 
because they can be used to help understand the level of variability associated with the 
experimental process and the analysis methods.  Prehydrolysis runs P, Q, R, and S were 
intended to be identical with prehydrolysis liquor added to the chips in the vacuum oven 
at pH 3.5 followed by a prehydrolysis treatment with the pH controlled at 3.5 with dilute 
acetic acid.  The temperature profile included a one hour ramp to 155?C that was held for 
106 
 
68 minutes before cooling.  The chips were divided into 12 different samples equivalent 
to 70 g bone dry untreated chips based on a ratio of wet treated chip weight to dry 
untreated chips.  Duplicate samples were also taken to measure chip dry fraction.  A 
careful procedure was followed to mix the treated chips before they were divided and to 
portion the chips equally into each sample from the main bag.   
Variability in the chip weight loss measurement was identified as a potential 
problem in the designed experiment when the sugar recovered was higher than the weight 
loss measured for condition D.  Duplicate samples of chips were taken for the chip dry 
fraction measurement in an attempt to reduce this problem.  The pooled standard 
deviation of the dry fraction measurement for the eight chip samples is 0.0127.  This 
suggests that the weight loss for prehydrolysis Q is 13.6% ?3.4% or somewhere between 
10.2% and 17% given one standard deviation.   
Composition analysis was performed on the liquid samples from runs P, Q, and R 
to determine sugar recovery.   The data from these runs are illustrated in Figure 50.  This 
figure has been modified in an attempt to show how close the data from the three runs 
are.  The ?Total Sugar? data has been plotted against a secondary Y axis to expand the 
primary axis.  Run P is illustrated with ?Plus? symbols.  Run Q is illustrated with 
triangles.  Run R is illustrated with square markers and lines.  The data from the three 
runs are basically identical with the exception of the 120 minute data set from run P that 
is slightly higher than the other two.  This lack of variability among these three runs 
implies that the method of producing and analyzing these samples is very consistent.  The 
liquid samples from these three runs were prepared for analysis on the HPLC and run in 
107 
 
sequence with standards checked before and after each set of seven samples.  The wash 
samples are not shown on the plot.   
Figure 50  Recovered sugar measurements as a function of time for prehydrolysis 
cooks "P", "Q", and "R".   Individual sugars from top to bottom based on final data points 
are mannose, xylose, galactose, arabinose, and glucose.   
 
Six of the chip samples from run S were bomb cooked without any additive using 
the standard cook profile to estimate the variability in pulp kappa number and yield 
measurements of the bomb cooks with wet chips.  The resulting pulps had an average 
kappa number of 26.6 with a standard deviation of 0.92 and average yield of 37.36% with 
a standard deviation of 0.79.  The kappa number standard deviation is similar to that of 
the cooks with raw chips, but the yield standard deviation is more than doubled.  This 
increase can probably be attributed to dividing the wet prehydrolyzed chips into smaller 
samples for bomb cooking.   
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
50 70 90 110 130 150
Tot
al 
ca
rbohy
dra
tes
, %
bd 
chips
Carbohy
dra
tes
 ex
trac
ted, 
%bd 
chips
Time, minutes
Mannose
Xylose
Galactose
Arabinose
Glucose
Total P
Total Q
Total R
108 
 
 
 
Chapter 6  Pulp Yield Recovery 
Pulp yield on wood is a huge financial driver in the industry.  A 1% change in 
yield can mean millions of dollars per year to a pulp mill.  Typical brown stock kraft pulp 
yield of Pine is only 47% compared to an ideal yield of 68% if pulping was purely 
selective to lignin and extractives.  The extra yield loss can be attributed to cellulose 
(4%), glucomannan (13%), and xylan (3%) (McDonough 1998).  Glucomannan losses in 
alkaline pulping are about equally from dissolution and peeling and not strongly 
dependent on alkali concentration or temperature.  Xylan losses are due to dissolution and 
are highly dependent on both temperature and alkali concentration.  Cellulose loss is due 
to peeling and secondary peeling at higher temperature.   
Prehydrolysis that is expected to be part of a VPP strategy increases yield loss 
beyond what is observed in conventional kraft pulping.  Acid prehydrolysis removes 
sugar from wood through random cleavage of polysaccharides.  This treatment produces 
oligomers and monomers that are dissolved and recovered in the prehydrolyzate, but it 
also damages the polymers that remain in the chips.  These damaged polymers are then 
more susceptible to peeling and dissolution typical of kraft pulping.   
Several methods have been proposed and attempted for increasing pulp yield from 
the conventional kraft process.  These include modifications to the raw material, 
modifications to the cooking process, and the use of chemical additives.  Investigations of 
how modifications to raw materials or the cooking process may relate to VPP are outside 
109 
 
the scope of this research and therefore are only introduced briefly for context.  This 
research is concentrated on the effect of extraction of hemicellulose prior to pulping on 
pulp yield.  A few potential methods of improving pulp yield in the context of chemical 
addition to the digester or additional pretreatment steps prior to cooking were tested.  
Discussion of these additives and the laboratory work performed to test them are the 
focus of this chapter.   
Pulp yield can be improved by changing the raw material: the wood chip.  It has 
been shown by modeling and mill studies that chip thickness screening and optimization 
can improve yield by reducing rejects and reducing delignification variability from chip 
to chip (Jimenez, McKean and Gustafson 1990).  Optimization of chipping technique can 
also improve pulp yield and strength while minimizing the need for screening (Wang and 
Gullichsen 1998). The quality of the wood itself can also be changed through selective 
hybridization and eventually through genetic modification to control wood composition 
(MacRae and Cotterill 1998).   
Modified cooking or extended delignification methods were developed in the 
1970s and 80s with the intention of reducing the kappa number before bleaching without 
reducing strength or yield.  Several different commercial methods for both continuous 
and batch cooking were developed based on the principles of maintaining alkali 
concentration low and even, having a high sulfide concentration in the initial part of the 
cook, and maintaining a low dissolved lignin concentration.  Innovation in these methods 
has continued by adding white liquor addition points and changing flows in continuous 
cooking and changing composition and order of addition of liquors in batch cooking.  
The result is a potential yield increase of about 2% at the same kappa number.  Addition 
110 
 
of polysulfide and/or AQ to a modified cook can result in an additional 1 to 2% yield 
improvement (Courchene 1998).   
Chemical additives can be used in pulping to reduce reactions of carbohydrates or 
increase reactivity of lignin.  One method of increasing pulp yield is to stabilize the 
carbohydrates with regard to peeling by treating the carbonyl end group with an oxidizing 
or reducing agent.  Often, however, the concentration of additive required is not 
economical or the effect is difficult to measure at mill scale.  Additives that have been 
tried previously include sodium dithionite, sodium borohydride, polysulfide, hydrogen 
sulfide, Anthraquinone (AQ), and surfactants (Courchene 1998).  Sodium borohydride 
and polysulfide are both effective at improving glucomannan yield but can actually lead 
to increased xylan losses if the alkali in the cook is not adjusted.  They also are most 
effective at lower cook temperatures.  AQ functions catalytically on both primary wood 
components by oxidizing carbonyl groups for increased yield and reducing lignin for 
improved delignification.  AQ and polysulfide can also be used in combination for a 
synergistic effect resulting in up to 4% yield improvement.  Surfactants can also help by 
improving chemical penetration therefore reducing rejects at the same total yield.   
Five additives were screened as to their potential to improve pulp yield.  Sodium 
borohydride and sodium polysulfide were not included because they were being tested by 
another member of Auburn University?s research group (Yoon 2009).  Hydrazine looked 
promising based on a literature search (Gillespie 1964) but was not tested following a 
recommendation from Dr. Susanne Striegler of the Auburn University Chemistry 
Department that personal exposure to toxic quantities of the carcinogen would be too 
111 
 
difficult to avoid in the lab or industrial environment.  Acetaldehyde was substituted for 
the recommended formaldehyde based on similar logic.   
Since this was a screening study, it was not feasible to attempt to hit a target 
kappa number or optimize the alkali charge so the yield at each condition could be 
compared on an equivalent basis.  It was assumed instead that pulp yield is a linear 
function of kappa number in the kappa number range observed in this group of studies 
(van Donkelaar 1998) and that the slope of that function is approximately 0.1% per kappa 
unit as observed in the designed experiment discussed in Chapter 5 of this research.  No 
correction was made for variation in residual alkali.  The temperature target for all the 
kraft cooks was 165?C.  Reducing the temperature may increase pulp yield by reducing 
alkaline cleavage and secondary peeling.  This was not attempted due to the increase in 
time of cook that would be required.   
Discussion of each additive begins with a description of the chemical and any 
potential handling concerns since these products may not be typical to the industry.  This 
is followed by literature examples of the use of the product in similar situations if it is 
available and discussion of expected reactions.  Laboratory tests with the product are then 
described followed by results and conclusions.  General conclusions regarding this 
research are given at the end of the chapter.   
6.1 Anthraquinone 
Anthraquinone (AQ) is typically delivered to a pulp mill in slurry form of 50-70% 
solids with a surfactant to aid in dispersal.  The mixture must be agitated to avoid settling 
in storage tanks.  9, 10- Anthraquinone powder, 97% grade was purchased from Alfa 
Aesar for this research.  AQ is considered harmful with limited evidence of carcinogenic 
112 
 
effect.  It can enter the body through skin absorption, breathing of dust, or consumption 
of dust (Alfa Aesar MSDS 2007).  It is included in the California Proposition 65 list of 
chemicals known to cause cancer or reproductive toxicity (California Office of 
Environmental Health Hazard Assessment 2010).  AQ is only barely soluble in water or 
white liquor, but can be dissolved by hot organic solvents. 
Anthraquinone has been such an important factor in industrial pulp yield 
improvement efforts since the 1970s that it was discussed in nine of the 25 presentations 
at TAPPI?s ?Breaking the Pulp Yield Barrier Symposium? in 1998.  A collection of 
papers discussing AQ pulping has also been published (Goyal 1997).  AQ accelerates 
delignification and protects polysaccharides from degradation, so the benefits can be 
utilized by a mill in several ways.  Application of AQ is typically justified based on 
increased pulp production.  Production increases are a result of increased yield, increased 
digester throughput, and decreased black liquor solids firing in the chemical recovery 
boiler.  An industry survey reported yield increase of 0.5 to 5% of wood with most results 
from 1 to 1.5% with a typical dosage of 0.05% on wood (Laubach 1998).  Other potential 
justifications for AQ could include reducing sulfidity or reducing bleaching requirement.   
This investigator has personal experience with the use of AQ in a mill 
environment.  The additive was used to reduce alkali required in the digester and improve 
yield.  Together these effects allowed the mill to increase pulp production under a 
chemical recovery limitation.  A yield increase of 1% was estimated based on a reduction 
of black liquor solids fired per ton of pulp produced and composition analysis of the pulp.    
The principle reactions of interest for AQ in pulping are oxidation of the reducing 
end groups of polysaccharides and reduction of lignin through several potential paths 
113 
 
(Lindenfors 1980).  The oxidized polysaccharides are stabilized relative to the peeling 
reaction by conversion to carboxylic acids.  One of the potential reactions with lignin 
inhibits condensation reactions.  AQ itself functions catalytically in both sets of reactions.  
It is reduced by the polysaccharides to anthrahydroquinone which is oxidized back to 
anthraquinone by reaction with lignin (Sjostrom 1993, 157).   
Two of the five chip subsamples from each of the nine tests in the designed 
experiment were cooked with the addition of 0.1% AQ on raw wood weight.  The first 
cook with AQ used the standard digester control profile to reach 1850 H-factor.  The H-
factor for the second cook with AQ for each sample was chosen so 30 kappa number 
would be bounded by the kappa number of the two cooks.  Only sample A had a kappa 
number higher than 30 from the first run, so the H-factor for the second cook of chips 
from sample A was much higher than standard at 3400.  The other second cook H factor 
targets were lower than standard.  The pattern in cook severity required to reach 30 kappa 
number for the nine sets of subsamples with AQ was similar to the pattern for cooks 
without AQ.  The required cook time decreased as the prehydrolysis conditions became 
more severe and reached a minimum with condition E at 155?C and pH 3.5.   The 
required cook time then increased with prehydrolysis severity finally becoming longer 
than standard with sample A.   
The H-factor required to reach 30 kappa number was estimated by reading from 
the plot of this data shown as Figure 51.  The curves shown are ?power? trendlines from 
Microsoft Excel ? and approximate the manually drawn curves that were used to 
estimate the H-factor.  The average H-factor required to reach 30 kappa number with AQ 
for the nine sets of chip samples was 1531; 283 points less than the 1814 required without 
114 
 
AQ.  This is a 27 minute shorter cook at 165?C.  Cook time can also be a significant 
consideration when determining an optimal prehydrolysis.  Two of the better conditions 
from the standpoint of pulp yield and sugar recovery were conditions E and G, 155?C at 
pH 3.5 and 3.0 respectively.  The H-factor required to reach 30 kappa number was 290 
points lower for E than G without AQ and 400 points lower with AQ.   
Figure 51  Kappa number as a function of H-factor for AQ cooks (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). 
 
The pulp yield measurements from each subsample were plotted as a function of 
kappa number in Figure 52 along with data from the cooks with raw chips for reference.  
The yield at 30 kappa number was calculated using the equation of the line between the 
two data points.  The improvement in yield varied from 0 to 1.8% of wood with an 
average of 0.74%.  The three highest yield improvements with AQ occurred with the 
15
20
25
30
35
40
45
50
55
60
65
1000 1500 2000 2500 3000 3500
Kappa
 number
H-factor
Raw
A 170, 3.0,   7
F 170, 3.5, 10
C 170, 4.5,  3.5
G 155, 3.0, 10
E 155,  3.5,  3.5
B 155,  4.5,  7
I  140,  3.0,  3.5
D 140, 3.5,  7
H 140, 4.5,10
115 
 
three prehydrolysis runs that were performed at pH 3.5.  The two runs with no 
improvement were prehydrolyzed at pH 4.5.    
Figure 52  Pulp yield as a function of kappa number for AQ cooks (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). 
 
Chips from prehydrolysis runs J, K, and L were also cooked with and without 
AQ.  Subsamples 1 and 3 were cooked without AQ.  Subsamples 2, 4, and 5 were cooked 
with AQ.  The temperature control profile for subsample 4 was changed so the 
temperature was ramped up to 150?C and held for 75 minutes and then ramped to 165?C 
and held for 80 minutes.  This profile had the same cumulative H-factor as cook #5 but 
the time was extended to be the same as cook #1.  The H-factor required to reach 30 
kappa number was reduced for the AQ cooks, but the yield was only improved by the AQ 
y = 0.1604x + 39.819
R? = 0.8882
25
30
35
40
45
50
10 20 30 40 50 60 70
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Raw
A 170, 3.0,   7
F 170, 3.5, 10
C 170, 4.5,  3.5
G 155, 3.0, 10
E 155,  3.5,  3.5
B 155,  4.5,  7
I  140,  3.0,  3.5
D 140, 3.5,  7
H 140, 4.5,10
116 
 
addition for condition L.  The pulp yield measurements from all subsamples are plotted 
vs. kappa number for conditions J, K, and L in Figure 53.  The yield to kappa relationship 
was not changed by the modified temperature profile for conditions J and L.  This 
subsample had a lower yield for condition K.   
Figure 53  Pulp yield as a function of kappa number for additional cooks with and 
without AQ.  Legend indicates trial label with the modifiers ?AQ? for inclusion of AQ in 
the kraft cook and ?AQ4? for modification of the temperature control profile. 
 
The average increase in yield of about 1% from application of 0.1% AQ was 
expected from literature and the investigator?s personal experience.  The indication from 
the design study samples that pH of prehydrolysis could influence effectiveness of AQ 
was interesting, but not confirmed with the follow up samples J, K, and L.  No further 
work with AQ alone as an additive is recommended as the results observed in this 
research closely match what has already been documented.  The beneficial effects of AQ 
y = 0.1604x + 39.819
R? = 0.8882
32
34
36
38
40
42
44
46
48
15 20 25 30 35 40 45 50
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Raw Chips
J
J AQ
J AQ4
K
K AQ
K AQ4
L
L AQ
L AQ4
117 
 
on pulp yield and delignification rate have been shown to be additive when it is applied 
along with polysulfide (Courchene 1998).  It might prove beneficial to try AQ along with 
other chemicals or process changes that improve yield to determine if these effects are 
also additive.  Attempts in this research to improve the effect of AQ by dissolving it in 
acetaldehyde or ethanolamine before addition to the white liquor are discussed in those 
sections.   
6.2 Acetaldehyde 
Acetaldehyde, 99.5% was purchased from Acros Organics for this research.  It is 
a volatile, flammable clear liquid that can be irritating to the eyes and respiratory system.  
There is also limited evidence of acetaldehyde acting as a carcinogen.  As such, it is 
included on the California Proposition 65 list (California Office of Environmental Health 
Hazard Assessment 2010).  The boiling point for acetaldehyde is 21?C, but the flash 
point is -27?C.  Acetaldehyde should be stored in a refrigerator approved for flammables.  
It is very soluble in water (Acros Organics MSDS 2010).   
No literature references were found for using acetaldehyde for pulp yield 
improvement.  A patent was issued in 2001 for the addition of an aldehyde to a chlorine 
dioxide delignification bleaching stage to improve delignification (Jiang, van Lierop and 
Berry 2001).  The kappa number was about one point lower and the final brightness about 
one point higher with the addition of 2% formaldehyde.  In another test, the kappa 
number was 0.6 lower after the addition of 1% glucose to the D0 stage.  The aldehyde 
was mixed with the pulp just prior to the addition of chlorine dioxide and the D0 stage 
was pH adjusted to 3.0 with sulfuric acid.  Treatment was at 50?C for 30 minutes.   
118 
 
A possible reaction of acetaldehyde with the carbonyl end group of the 
carbohydrates is an aldol condensation that could then limit peeling of this end group.  
Aldol condensation can occur between any two aldehydes and is base catalyzed.  To 
avoid condensation of acetaldehyde with itself, the additive was added with water to the 
chips and given time to diffuse into the wood chips at room temperature for 60 minutes 
before 1% sodium hydroxide on chips was added dropwise to raise the pH.  The bombs 
were then closed and shaken before being left to rest for 90 minutes.  After 90 minutes, a 
measured amount of soak solution was removed and replaced with white liquor for the 
standard cook.  Three bombs were prepared in this manner.  One was a blank that 
underwent the same treatment without acetaldehyde.  The other two had 1% or 5% 
acetaldehyde by weight on chips. 
Acetaldehyde was tested in four bomb cooks from prehydrolysis ?S?.   Pulp yield 
as a function of kappa number for these four samples, the blank, and the six samples used 
to determine the ?Standard? kappa and yield results are shown in Figure 54.  The blank 
sample had the lowest kappa number, the 5% sample the highest.   The 5% sample also 
had a higher yield of pulp than the other two, but all five were very close to the 
?Standard? line.   
The red ?Standard? line and point and green ?Upper? and ?Lower? lines in Figure 
54 are generated from the measurements from cooks S7 to S12 that were all performed 
without additives or additional treatment.  The red line was generated by plotting a line of 
slope 0.1 through the average yield and kappa for these six cooks.  The green lines 
indicate one standard deviation or 0.8% in yield from the standard line.  The standard 
deviation in kappa number was 0.9 units.  All of the yield data from the cooks with 
119 
 
acetaldehyde was within one standard deviation in yield from standard at the measured 
kappa number.  None of the tests were within one standard deviation of the standard 
kappa measurement.  The closest point to standard kappa number was the ?blank? test 
that did not have any acetaldehyde added.   
Figure 54  Yield as a function of kappa number for acetaldehyde treated and untreated 
prehydrolyzed chips. 
 
One difficulty in working with acetaldehyde is the low boiling point of 21?C.  The 
additive was kept in the refrigerator until just before it was measured and a pipette was 
used to quantitatively transfer the measured amount into the water immediately before it 
was added to the chips.  The bombs were immediately closed and shaken before the rest 
time.  An odor was present even with these precautions.   
34
35
36
37
38
39
40
20 22 24 26 28 30 32 34 36 38 40
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Standard
Upper
Lower
Standard
Blank
Acetaldehyde
AA with AQ
120 
 
The two attempts to use acetaldehyde to dissolve AQ were more eventful.  
Initially acetaldehyde was added directly to the dry AQ powder.  A paste formed, but 
quickly dried back to powder.  Some water was then added to the AQ with no effect on 
the dried paste now stuck on the beaker side.  Acetaldehyde was added to the water and 
the AQ paste released from the beaker.  The entire mixture was then added to the chips 
before the white liquor was added.  The beaker with AQ inside was placed in an ice water 
bath for the second attempt.  The paste remained after the acetaldehyde was added.  
When white liquor was added directly to the paste, there was a crackling sound as the 
solution turned orange. A dispersed particulate formed when water was added to this 
solution before the entire mix was added to the chips for cooking.  The kappa number 
was reduced below the standard in these two cases, but the yield remained proportional to 
the standard line. 
The conclusion from these four experiments is that acetaldehyde addition 
increased the yield and kappa number without changing the standard relationship 
between the two.  The use of acetaldehyde to aid in dissolution of AQ had no significant 
effect on pulp yield.  No further work is recommended with acetaldehyde as an agent to 
increase yield of pulp from prehydrolyzed chips either with or without AQ.   
6.3 Ethanolamine 
Ethanolamine is both a primary amine and a primary alcohol that is also known as 
monoethanolamine.  It reacts as a weak base and is produced by reacting ethylene oxide 
with aqueous ammonia (Morrison and Boyd 1983, 547).  Ethanolamine is a colorless, 
viscous liquid with an ammonia-like odor.  It is classified as moderately flammable.  
Exposure should be kept below the odor threshold (2 to 4 ppm) to avoid irritation to eyes 
121 
 
and lungs.  Ethanolamine should be stored in a cool, dry, well ventilated area in properly 
labeled tightly sealed containers.  Small spills should be soaked up with noncombustible 
absorbent material (Occupational Safety and Health Administration n.d.).  Ethanolamine, 
99% was purchased from Acros Organics for this research.   
The closest literature example for the use of ethanolamine for yield improvement 
is a patent for a pretreatment of wood prior to soda cooking that would increase yield and 
improve strength properties of the resulting pulp (Procter and Chow 1978).  The primary 
purpose of the process was to produce a pulp with properties comparable to kraft pulp 
without the use of sulfur, but the pretreatment was also claimed to increase the yield of 
kraft pulp by about 2%.  The method called for pretreatment with 5 to10% amine and 0.5 
to 3% sodium hydroxide based on the weight of wood at a liquor to wood ratio of 5 to 1in 
a pressure vessel with nitrogen over pressure of about 25 psi.  The treatment would be 
held at 120? to160?C for 30 to 60 minutes.  The solution would be drained from the chips 
after treatment for recovery and reuse of the remaining amine.  Cooking liquor would 
then be added for the soda or kraft cook.   
Ethanolamine has also been referenced for use in preserving strength and yield of 
cellulose as an additive in bleaching, as a co-solvent with water and soda for solvent 
pulping, and for improving decay resistance of wood.  None of these additional 
applications specifically relate to the use of ethanolamine for yield improvement after 
prehydrolysis, but they do give a range of conditions that were used as a starting point for 
screening the chemical for this purpose.    
A patent was issued in 1981 for the use of amines to inhibit cellulose degradation 
in a hypochlorite bleaching stage (Breslin and Cosper 1981).  The invention called for the 
122 
 
use of 0.05 to 0.2% of monoethanolamine on pulp added with the bleaching chemicals, 
0.5% sodium hydroxide, and water to bring the pulp to 10% consistency.  The bleaching 
reaction was then allowed to proceed for 60 minutes at 110?F.  The additive increased 
pulp viscosity after bleaching from 14.7 to 16.6 cps.  Pulp strength properties were also 
improved.   
Ethanolamine has also been suggested for use in solvent pulping.  In one example, 
the ethanolamine concentration varied from 5 to 15% and was used along with soda at 2.5 
to 7.5% concentration.  The wood was cooked at 165 to 195?C for 30 to 90 minutes.  The 
maximum yield was attained with the lowest temperature, longest time, lowest soda 
concentration and highest ethanolamine concentration (Jimenez, et al. 2004).   
Upgrading of wood is a chemical modification that is performed to improve 
biological resistance or dimensional stability of wood without increasing flammability.  A 
study of upgrading wood with ethanolamine used solutions ranging from 1.6 to 20% 
ethanolamine in water (Humar, et al. 2003).  The treatment was performed by soaking 
wood in the solution at room temperature for times from 7 hours to 2 weeks.  Vacuum 
was used to improve uptake of the solution in the wood blocks.  Measurements indicated 
that the treatment improved resistance to fungal rot due to chemical reaction of the 
ethanolamine with both hemicellulose and lignin.  The improvement in rot resistance was 
attributed to an increase in the pH of the wood from 5.1 to 9 following the treatment.   
Ethanolamine reacts with the carbonyl end of carbohydrates and benzene ring 
groups in lignin (Humar, et al. 2003).  The reaction with carbohydrates could limit 
degradation during pulping due to peeling of this end group.  Reactions with lignin could 
restrict condensation reactions (Procter and Chow 1978). 
123 
 
Ethanolamine was tested with the chips from prehydrolysis run ?P?.  Ten different 
bomb cooks were performed with various chemical addition rates, temperatures, and time 
of intermediate treatment.  General conditions for each treatment and results are shown in 
Table 8.  For treatments P1, P2, and P3, a 30 minute ramp was used and then held for one 
hour at temperature.  After this time the bombs were removed from the digester and 
opened.  129 ml of intermediate solution was poured from the bombs and replaced with 
white liquor.  The bombs were then closed, shaken well and put back in the digester for 
the cook to 1400 H-factor.  The pH of the intermediate solution was measured and 
recorded.  The procedure for P4, P5, and P6 was the same except the ramp time was 
increased to 40 minutes since the temperature was higher.  The ramp time was increased 
to one hour for P7 and P8.  P9 was left at room temperature for the two hours required to 
process P7 and P8 and then cooked with P7 and P8 after the solutions had been changed.  
P10 had AQ mixed with ethanolamine as discussed earlier.  The AQ appeared to dissolve 
in the ethanolamine initially, but then some precipitate formed when white liquor and 
water were added to the beaker.   
Table 8  Treatment conditions and results for ethanolamine treatment. 
 
 
Bo m b EA % T e m p N a O H % pH ka p p a Y i e l d %
P1 0 8 0 ? C  1% 8 . 9 1 2 5 . 1 3 7 . 2 5
P2 0 . 5 8 0 ? C 1% 1 0 . 0 4 2 8 . 0 3 7 . 4 3
P3 5 8 0 ? C 1% 1 0 . 8 5 2 5 . 2 3 7 . 1 6
P4 0 1 1 0 ? C  1% 6 . 5 9 2 5 . 9 3 7 . 3 6
P5 0 . 5 1 1 0 ? C 1% 8 . 1 6 2 7 . 5 3 8 . 1 1
P6 5 1 1 0 ? C 1% 1 0 . 3 0 2 9 . 7 3 8 . 2 1
P7 0 . 5 1 4 0 ? C 1% 5 . 6 4 3 0 . 6 3 8 . 5 9
P8 2 1 4 0 ? C 1% 7 . 5 2 3 6 . 4 3 7 . 4 5
P9 5 2 5 ? C 1% 1 1 . 4 5 2 3 . 4 3 7 . 0 9
P1 0  a q 2 1 1 0 ? C 2 %  W L 9 . 8 0 2 5 . 1 3 7 . 1 4
124 
 
The yield is plotted as a function of kappa number for all the tests in Figure 55.  
The red ?Standard? line and point and green ?Upper? and ?Lower? lines are generated 
from the measurements from cooks S7 to S12 that were all performed without additives 
or additional treatment.  All of the yield data was very close to being within one standard 
deviation in yield from standard at the measured kappa number.  Only two of the tests 
with ethanolamine treatment were within one standard deviation of the standard kappa 
measurement.  Two of the closest points were the ?blank? tests, P1 and P4, which did not 
have any ethanolamine added.  The test with AQ had results almost identical to the blank.   
Figure 55  Yield as a function of kappa number for ethanolamine treatment. 
 
The ethanolamine treatment had very little, if any, effect on the yield of pulp from 
prehydrolyzed chips.  The treatments did increase or decrease the kappa number of the 
pulp in certain cases.  The highest yield was generally observed at higher temperature and 
higher percentage of ethanolamine, but the differences from standard were small and 
34
35
36
37
38
39
40
20 25 30 35 40
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Standard
Upper
Lower
Blank
EtohAmine
EA with AQ
125 
 
possibly not significant.  The lowest kappa number resulted from the run with the highest 
pH following intermediate treatment, P9.  The high pH prior to addition of white liquor 
may have increased swelling in the chips and therefore increased the rate the white liquor 
was transported into the chips, resulting in the three point reduction in kappa number.  
The two highest kappa numbers were measured following runs with low pH and high 
temperature in the intermediate treatment.  In these cases, the depletion of alkali at higher 
temperature may have caused condensation of lignin resulting in higher kappa number.   
Even though the ethanolamine appeared to dissolve the AQ, this did not improve 
pulp yield.   
If additional studies are attempted with ethanolamine as an agent to increase yield 
of pulp from prehydrolyzed chips, higher dosage of ethanolamine and sodium hydroxide 
should be tested in the higher temperature range of 120? to 160?C.  It also might be 
instructive to apply an overpressure of nitrogen as suggested in the patent for 
pretreatment prior to soda cooking (Procter and Chow 1978).   
6.4 Lithium Aluminium Hydride 
Lithium aluminium hydride (LiAlH4) was purchased from Acros Organics for this 
research.  The primary difficulty working with this product is that it reacts violently with 
water to release hydrogen gas that may ignite spontaneously.  This can cause severe 
burns if the powder comes in contact with skin or eyes.  The melting point of LiAlH4 is 
125?C which is also the decomposition temperature (Acros Organics MSDS 2010).   
Lithium aluminium hydride is a versatile reducing agent for organic synthesis, but 
example reactions were performed in organic solvents (Morrison and Boyd 1983, 473).  
LiAlH4 can be used to reduce aldehydes to alcohols (Morrison and Boyd 1983, 754).  
126 
 
This reaction is similar to the sodium borohydride reduction of the end groups of 
carbohydrates to prevent peeling.  The reaction of LiAlH4 with water is very fast.  
Five chip samples from prehydrolysis ?R? were used to test the effect of LiAlH4 
on pulp yield.  The hydride was added to the pulp and liquor in different ways for each 
sample in an attempt to find a method to contain the fast reaction.  For R1, 0.2% on wood 
of the hydride was placed in the bottom of a graduated cylinder.  The white liquor and 
water were poured over it and this solution was then poured into the bomb reactor over 
the chips.  The chips and liquor were placed in the bomb first for R2 before 0.5% of the 
hydride was added to the top of the reactor.  A gas was immediately released followed by 
a flame and popping sound.  The hydride was measured into the bomb reactor at a dosage 
of 1.0% on wood for R3.  The chips were added on top of the hydride followed by the 
white liquor and water.  All three bombs were capped quickly and shaken before they 
were placed in the digester for cooking.  There was a concern, especially for  R1 and R2, 
that the hydrogen off gassed prior to capping the bombs.  Small plastic weighing dishes 
were modified for R4 and R5 so the dish would fit in the top of the bomb.  The hydride 
was measured for a 1% dose in the weighing dish.  The chips were placed in the bombs 
followed by water and white liquor.  Only enough white liquor was added to R4 to raise 
the pH to about 10.  The full dose of white liquor was added to R5.  The full weighing 
dishes were carefully floated on top of the chips and liquor and the bombs were capped.  
R4 was then shaken and placed in the digester where it was heated to 120?C and held for 
60 minutes.  R5 was shaken and set aside.  The bomb reactor for R4 was cooled in an ice 
water bath after it was removed from the digester.  It was opened and sufficient solution 
was poured from the reactor so the remainder of the white liquor could be added.  The 
127 
 
bomb was then closed, shaken again, and R4 and R5 were placed in the digester for the 
cook.  Both were cooled in ice water before they were opened.  Some pressure was 
released when R4 and R5 were opened for the first time following the addition of the 
hydride.  R6 was a standard cook with no additive and no pretreatment. 
The yield is plotted as a function of kappa number for the tests with LiAlH4 in 
Figure 56.  The standard, upper, and lower lines are generated from the S7 to S12 cooks 
without additive.  The ?blank? point is R6.  All the yield results are within one standard 
deviation of the standard line.  Samples R3 and R4 had higher kappa numbers than 
standard.  Samples R4 and R5 had higher yield than the other samples and the standard.   
Figure 56  Yield as a function of kappa number for lithium aluminium hydride 
treatment. 
 
34
35
36
37
38
39
40
20 22 24 26 28 30 32 34 36 38 40
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Standard
Upper
Lower
Blank
LiAlH4
128 
 
The yield of pulp was not increased by the addition of LiAlH4 like it would have 
been expected to have been increased by the addition of sodium borohydride.  The 
improved hydride addition method utilizing the weighing dishes may have allowed the 
reaction to be contained in the bomb and contributed to the higher yield in those two 
cases, but the yield improvement remained small.  No further work is recommended with 
LiAlH4 as an agent to increase yield of pulp from prehydrolyzed chips.   
6.5 Hydroxylamine 
Hydroxylamine is typically stored and transported as either a chloride or sulfate 
salt.  Hydroxylamine hydrochloride was purchased from Acros Organics for this 
research.  It is an irritant to eyes and skin and is suspected of causing cancer, but is not on 
California?s Proposition 65 list of potential carcinogens.  Immediate medical attention is 
required following exposure to eyes or skin.  Hydroxylamine hydrochloride decomposes 
at 152?C and can produce hydrogen chloride gas.  It is very soluble in water (Acros 
Organics MSDS 2010).   
Hydroxylamine can be used to eliminate reducing sugars in various applications.  
Patents have been issued for modifying or removing reducing sugars to increase pulp 
yield (Paterson 1968), purify citric acid preparations (Borchert 1981), and as a bleaching 
aid for mechanical pulp (Beurich and Scholl 1988).  Each of these patented applications 
claims that it doesn?t matter which form of hydroxylamine is used unless there are other 
considerations that might recommend one form over the others.  Compatibility with the 
kraft liquor cycle would favor the sulfate form if one of these were used.   
Different reaction conditions are recommended for each of these three 
applications.  Hydroxylamine sulfate or bisulfate was added directly to the white liquor 
129 
 
for pulp yield improvement.  Some additional alkali was required to neutralize the 
hydroxylamine.  A cook with 6.35 parts hydroxylamine bisulfate improved unscreened 
and screened yield from 48.4% and 44.7% to 52.8% and 48.1% respectively and 
improved viscosity 30%.  The control cook was performed with the same conditions 
except for the modification of chemicals.  The inventor suggests the method to be 
particularly useful for pine (Paterson 1968).  Substantially milder conditions were 
successful in eliminating reducing sugars from trisodium citrate, possibly since it is a 
homogeneous reaction.  Recommended conditions for this treatment are pH 9 to 10, 
60?C, and 12 hours reaction time.  The charge of hydroxylamine is approximately 
stoichiometric at 0.1% for every 0.25% of reducing sugar (Borchert 1981).  
Hydroxylamine is used in the mechanical pulp bleaching application as a reductive 
bleaching agent along with various chelant solutions that are the primary subject of the 
patent.  The reaction conditions for examples that included hydroxylamine were pH 6 to 
8, 60?C for 1 to 2 hours with 0.5 to 1% hydroxylamine.  The bleaching agent could be 
added to the wood chips before pulping or to the disintegrated pulp (Beurich and Scholl 
1988).   
Other studies identified hydroxylamine as an interesting additive for pulp yield 
improvement.  A study of the effects of twenty two additives on the yield stability of 
hydrocellulose after boiling in 1 N NaOH found six that increased the yield by at least 
10%.  The most effective was sodium borohydride followed closely by hydroxylamine 
hydrochloride (Clayton and Marraccini 1966).  The screening tests were performed on 
specially prepared partially hydrolyzed cellulose samples to increase the sensitivity to the 
peeling reaction.    Hydroxylamine increased the yield of the hydrocellulose to 90% 
130 
 
compared to 52% for the control sample.  In Kraft pulping of spruce and birch with 
hydroxylamine hydrochloride or sulfate additive  the results were less dramatic, but still 
significant with a 3% increase in carbohydrate yield on wood while at the same time 
generating a 3% decrease in lignin yield on pulp with a 5% charge on wood.  Tests with 
1% sodium borohydride in this same study generated a 7% increase in carbohydrate yield 
but also a 1% increase in lignin yield.  Tests with soda pulping had higher carbohydrate 
yield improvements, but the lignin content remained very high in all pulp samples.  The 
authors suggested that the hydroxylamine reacted preferentially with lignin degradation 
products, therefore reducing the effect in kraft pulping where these products would be 
more prevalent earlier in the cook.  When compared to sodium borohydride, the 
hydroxylamine was more effective on hardwood than softwood.  The authors cite a 
previous study with sodium borohydride wherein yield increases were primarily due to 
increased glucomannan content since xylans are removed by dissolution of oligomers 
instead of peeling (Aurell and Hartler 1963).   
When the required reaction time for pretreatment with hydroxylamine was tested, 
the resistance to peeling of the hydrocellulose continued to gradually rise even after two 
hours of pretreatment with hydroxylamine in 1 N NaOH at 100?C.  The time dependence 
of peeling resistance is due to the slow conversion of oxime to nitrile.  Test cooks were 
performed by adding the hydroxylamine directly to the white liquor (Clayton and Jones 
1970).  These tests indicated that hydroxylamine addition at less than 1% would 
accelerate delignification but not improve carbohydrate yield.  Application of 6.6% 
hydroxylamine increased carbohydrate yield by 4% of wood.   
131 
 
A study that compared the end group stabilizing efficiency of hydrogen sulfide 
treatment under various conditions of time, temperature and pH to that of eighteen other 
reagents also found hydroxylamine to be one of the best (Procter and Apelt 1969).   This 
study used a hydrocellulose with number average degree of polymerization of 46 to test 
the stabilizing effect toward alkaline peeling. Molecules with stabilized end groups would 
not change in alkali while unmodified molecules would likely be completely hydrolyzed.  
The hydroxylamine performed well when added with the hydrocellulose directly to the 
refluxing 1 N sodium hydroxide test solution with 55 and 85% of the end groups 
stabilized with 0.05 M and 0.2 M hydroxylamine.  When 0.05 M hydroxylamine was 
used as a pretreatment in water or a carbonate buffer at pH 10, it only achieved 
stabilization in 40 and 45% of the molecules respectively.  The author attributes the lower 
stabilization to lower accessibility of the end groups at reduced pH, but sodium 
borohydride was actually more efficient when used as a pretreatment with 89 and 86% 
stabilization in water and pH 8 buffer compared to 79 percent when added directly to the 
alkali.  The concentrations of reagents used in this study were similar to what could be 
seen in pulping when compared on a reducing end group weight basis.   
Hydroxylamine has also been studied as an additive to increase the delignification 
rate of soda cooking.  In a study with bagasse pith, 10% hydroxylamine hydrochloride 
was added with 16% sodium hydroxide on pith.  The solution was boiled for one hour at 
atmospheric pressure before washing.  The black liquor was acidified and washed with 
water to produce a high lignin sample that was then prepared for UV and IR analysis.  
The analysis confirmed that hydroxylamine does form an oxime with the coniferyl 
aldehyde that contributes to acceleration of delignification.  Potentiometric titrations of 
132 
 
the additive solutions and the black liquor also confirmed this mechanism (Ahmed, Saad 
and Mohamed 1987).   
The protective effect of hydroxylamine on carbohydrates is due to the reversible 
formation of an oxime at the reducing end of the polysaccharide.  This oxime is subject to 
three competing reactions in the presence of alkali.  The oxime can revert to the 
aldehyde, convert to a nitrile that stabilizes the polysaccharide or it can be ?peeled? by 
alkali which produces a new reducing end group and allows continued peeling of the 
polysaccharide.  The reactions of hydroxylamine in this system are summarized in Figure 
57 (Clayton and Jones 1970).  Another reaction shown is the degradation of 
hydroxylamine by alkali to ammonia, nitrogen, and water.   
Figure 57  Reaction pathways with hydroxylamine (Clayton and Jones 1970). 
  
133 
 
Hydroxylamine appears to increase pulp yield by first reducing the rate of peeling 
through formation of the oxime and second preventing subsequent peeling if the oxime 
degrades to a nitrile.  Both depend on the reaction with the carbohydrate carbonyl 
reducing end group to form an oxime.  This oxime remains subject to peeling, but at a 
slower rate than the original reducing sugar.  The oxime must convert to a nitrile which 
then subsequently degrades in alkali to a carboxylate ion in order to permanently protect 
the rest of the polymer.  The primary reaction of the oxime in alkali is to form the nitrile 
if there is a large excess of hydroxylamine.  The peeling and stopping reactions of the 
hydrocellulose and the oxime are both suppressed in this case.  The primary reaction of 
the oxime in alkali is peeling if there is a very low concentration of hydroxylamine with 
the peeling reaction about twice as fast as the desired degradation to a nitrile (Clayton 
and Jones 1970).   
Two mechanisms for production of the oxime are proposed, one to occur 
optimally at pH 4 to 5 and the other to occur above pH 9.  The reaction of hydroxylamine 
with a carbonyl is expected to be very slow either below about pH 2 or near neutral at 
25?C (Jencks 1959).  This maximum in rate below neutral pH is a function of two 
competing rate-limiting factors:  acid concentration for catalysis of the dehydration step 
from the addition complex and free hydroxylamine base concentration for formation of 
the addition complex with the carbonyl (pKa of hydroxylamine is 6.0). The rate of 
oximation of benzaldehyde and acetone with hydroxylamine increased with hydroxide 
concentration in the range of pH 10 and 12.5.  These reactions had a minimum in rate 
between about pH 8 and 10.  The explanation given for this behavior is that the reaction 
can be catalyzed by the specific base, hydroxide, or a general acid, either oxonium or 
134 
 
water depending on the pH (Williams and Bender 1962).  In discussion of the hundred to 
thousand fold difference in rate of dehydration of the intermediate carbinolamines formed 
in the reactions with acetone and benzaldehyde, the authors argue the rate must be 
controlled by the ?relief of steric strain? that should be greater in the case of the ketone.  
This reasoning would imply that the reaction with a carbohydrate would be even slower 
since the steric relief in the aldehyde would be low and the dehydration would not have 
the advantage of resonance interaction that is present in the aromatic ring. 
There are several different opinions in literature about what the best conditions 
are for first forming the oxime.  The exact conditions of solvent, reagent, temperature, 
and pH all should be carefully controlled according to several sources, but none of these 
sources specify what all these conditions should be (Morrison and Boyd 1983, 756-757) 
(Rojas-Escudero, et al. 2004).  A study of preparation of sugar samples for gas 
chromatography found that a 5% hydroxylamine hydrochloride concentration worked 
best in several organic solvents and sodium acetate in water.   The best reaction 
conditions in the sodium acetate solution were 60?C for 30 minutes (Rojas-Escudero, et 
al. 2004).  Neither the concentration nor pH of sodium acetate solution was specified.  In 
this and other studies solvents other than water were recommended for the best reaction 
yield.  Those recommended were ethanol with pyridine (Portnoy, Reine and Arthur 1972) 
and aniline (Rojas-Escudero, et al. 2004) among others tested.  A diagnostic procedure 
for producing the oxime from cinnamaldehyde is:  ? ?0.12  g of the oil?was dissolved 
in ethanol (3 ml) and treated with a solution of hydroxylamine hydrochloride (0.15 g) and 
anhydrous sodium acetate (0.3 g) in water (5 ml)? for several hours.  The precipitate was 
135 
 
then recrystallized for melting point determination (Birkinshaw, Chaplen and Findlay 
1957).   
Catalysts have been reported that increase either the rate or yield of the 
production of oximes.  Silica gel is reported to catalyze production of oximes 
quantitatively and with a stoichiometric amount of hydroxylamine, but this reaction is 
carried out under ?dry? conditions (Hajipour, et al. 1999).  Ultrasonic radiation is also 
reported to improve the rate and yield of oximes from condensation of aldehydes or 
ketones with hydroxylamine hydrochloride (Li, Li and Li 2006), but this has only been 
demonstrated on a laboratory scale.  Methylamine can function as a nucleophilic catalyst 
in oximation with hydroxylamine by reversibly forming the intermediate ketimine 
(Williams and Bender 1962).  
Like the literature references for producing an oxime, references for producing a 
nitrile from the oxime refer to homogeneous reactions in organic solvents such as dry 
dichloromethane (Chaudhari and Akamanchi 1999) (Telvekar and Akamanchi 2004).  In 
particular, it is emphasized that for the highest efficiency, the solvent should be dry in 
order to encourage the dehydration reaction.  The use of aqueous alkali is listed as one 
option for this reaction but no process conditions are listed (Meudt, Scherer and Boehm 
2008).   
The chip samples from prehydrolysis run ?Q? were used to test hydroxylamine 
under a variety of conditions.  The chemical charges, reaction temperature, pH data 
where available, and pulp kappa and yield for each run are listed in Table 9.  All were 
held at the pretreatment temperature for 60 minutes except for Q7, Q8, and Q9 which had 
no pretreatment time.  The ramp time depended on the target: 30 minutes for 60?C, 40 
136 
 
minutes for 110?C, or 45 minutes for 120?C.  Sodium hydroxide or white liquor was 
added to all the bombs that had a pretreatment with the aim of reaching an outlet pH 
target of either 9 to 10 or 4 to 5.  Q1 was a blank sample.  The bombs were cooled 
following the pretreatment and sufficient solution was poured off to be replaced with 
white liquor for the kraft cook.  Hydroxylamine was added directly to the chips with the 
white liquor for Q7, Q8, and Q9 before cooking with the standard H-factor target of 1400 
used for all P, Q, R, and S cooks discussed in this chapter.   
Table 9  Hydroxylamine treatment conditions and results. 
 
The results listed in Table 9 are plotted in Figure 58 along with the standard red 
and green lines.  All but four of the results were either on or within the green lines 
indicating one standard deviation from the standard yield-kappa relationship while only 
three were within one standard deviation from the standard kappa number.  It is notable 
that the blank result had lower yield and kappa than the standard average and the lowest 
yield of any of the conditions tested in this yield improvement study.  The yield 
improvements observed in Q4, Q6, and Q11 are 1% of wood higher if they are measured 
against the blank for run Q instead of the standard from run S.   
N a O H %  o r
Label B o m b H y d ro x % W L ,  m l T e m p ka p p a Y i e l d % in out
B l a n k Q1 0 . 0 1 . 0 % 6 0 ? C 2 3 . 7 3 5 . 9 7 1 0 . 3 9
6 0 ? C  N a O H Q2 0 . 1 6 1 . 0 % 6 0 ? C 2 7 . 5 3 7 . 7 3 9 . 7 9
Q3 2 . 0 1 . 0 % 6 0 ? C 2 2 . 1 3 6 . 4 0 4 . 8 5
1 1 0 ? C  N a O H Q4 1 . 0 1 . 0 % 1 1 0 ? C 2 2 . 0 3 8 . 1 8 1 1 . 1 1 5 . 3 7
Q5 1 . 0 1 . 5 % 1 1 0 ? C 2 2 . 6 3 7 . 4 8 1 2 . 4 9 6 . 4 9
Q6 2 . 0 1 . 5 % 1 1 0 ? C 1 8 . 9 3 7 . 8 6 1 0 . 2 4 5 . 2 7
t o  W L Q7 0 . 1 6 1 2 9   1 6 5 ? C 2 7 . 2 3 7 . 3 5
Q8 1 . 0 1 2 9   1 6 5 ? C 2 6 . 7 3 6 . 5 5
Q9 2 . 0 1 2 9   1 6 5 ? C 3 2 . 0 3 8 . 6 3
1 2 0 ? C  W L Q 1 0 5 . 0    1 . 5 1 2 0 ? C 3 1 . 2 3 8 . 6 8 1 . 8 7
Q 1 1 5 . 0  6 1 2 0 ? C 2 5 . 1 3 8 . 5 8 5 . 0 + 4 . 4 0
Q 1 2 5 . 0 18 1 2 0 ? C 2 9 . 8 3 7 . 5 9 1 2 . 2 0 9 . 3 7
pH
137 
 
Figure 58  Pulp yield as a function of kappa number with hydroxylamine treatment. 
 
The variety of conditions was chosen to screen the impact of hydroxylamine 
dosage, alkali source and strength, temperature of pretreatment, and pretreatment pH on 
the pulp kappa number and yield at a given kappa number.  It is difficult to draw general 
conclusions from this data about the impact of any of these factors on yield improvement 
due to the small differences, but specific subsets of the samples can be compared to direct 
future work.   
For dosage rate, the two treatments that had only stoichiometric additions of 
hydroxylamine, Q2 and Q7, were the closest in both yield and kappa number to the 
standard average indicating little effect.  Increasing the dosage from 1% to 2% reduced 
the kappa number by 3.7 for Q5 and Q6 at 110?C but raised the kappa number by 5.3 for 
34
35
36
37
38
39
40
18 20 22 24 26 28 30 32 34 36 38
Pulp 
yie
ld, 
% 
of
 bd w
ood
Kappa number
Standard
Upper
Lower
Blank
to WL
60?C NaOH
110?C NaOH
120?C WL
138 
 
Q8 and Q9 when the hydroxylamine was added to the kraft cooking liquor.  The yield 
was improved in both cases by 0.38% and 2.08% of wood, respectively.  
High and low yield results were found with both sodium hydroxide and white 
liquor as the alkali source.  Addition of hydroxylamine directly to the white liquor for a 
standard kraft cook resulted in higher yield and kappa number at the highest dosage 
tested, 2% on wood, but the lower dosages of hydroxylamine gave kappa and yield very 
near the standard average.  Higher kappa number measured for sample Q9 was 
unexpected since the hydroxylamine is expected to react with both lignin and 
carbohydrate to accelerate delignification while increasing yield.  This change may have 
been due to consumption of alkali by the hydroxylamine HCl.  The yield change 
remained within one standard deviation of the standard for all three samples.  The alkali 
charge is only an independent variable if the hydroxylamine charge is held the same for 
the samples under comparison since these two together strongly influence the solution 
pH.  Comparison of Q4 with Q5 shows increasing the alkali charge increased the solution 
pH out of the preferred range of 4 to 5.  The pulp showed a slight increase of 0.6 in kappa 
number and a decrease of 0.7% in yield.  Similarly, comparison of Q10, Q11, and Q12 
shows that the charge of alkali that results in the proper final pH produces the best results 
of yield and kappa number.   
 In addition to the general observation that the four best examples of pulp yield 
improvement relative to standard were pretreated at 110 or 120?C, samples Q3 and Q6 
can be compared to look at temperature related differences.  The alkali charges are 
different, but the final pH measurements for these two samples are fairly close.  The 
kappa number for the pulp pretreated at the higher temperature is 3.2 points lower while 
139 
 
the yield is 1.46% higher.  These differences indicate that while the kappa number of Q3 
is low at 22.1, increasing the temperature improved reaction with both lignin and 
carbohydrate.    
Hitting a target final pretreatment solution pH by manipulation of alkali charge in 
response to hydroxylamine charge proved difficult.  Samples Q2 and Q12 finished in the 
higher target range of 9 to 10 pH while samples Q3 and Q11 were in the lower range of 4 
to 5 pH with Q4 and Q6 just above this range.  The results indicate the lower range is to 
be preferred where Q4, Q6, and Q11 all had a kappa number below average and yield 
significantly higher than the standard at the measured kappa number.   
Pretreatment with hydroxylamine produced the best improvement in both pulp 
yield increase and kappa number decrease of any of the additives tested in this study.    
Control of solution pH in the range of 4.0 to 5.5 was the most critical factor in success as 
long as there was at least a 1% charge of hydroxylamine.  The improvement in yield in 
three samples was significant, but not sufficient to return the yield of pulp from 
prehydrolyzed chips back to the yield from untreated chips of above 44% of wood.  
Hydroxylamine addition directly to the digester for the kraft cook did not bring an 
improvement in this study, but a much lower dosage was used than that recommended in 
the literature examples in an attempt to improve potential economics.   
Larger yield improvements were expected with hydroxylamine based on the 
literature.  Further work with hydroxylamine should be attempted to verify the proper pH 
target, temperature of pretreatment, and dosage rate.  Other options for pH control agents 
are sodium sulfide and ammonium hydrosulfide.  Discovery of the optimal temperature 
and pH sequence for nitrile formation from the oxime should also increase yield.  Since 
140 
 
hydroxylamine is reactive with both lignin and carbohydrates, it would be helpful to 
document the relative reactivity of those wood components as functions of pH and 
temperature.  Improved understanding of these relationships may allow more selective 
production of oxime and nitrile from carbohydrates without degradation of the 
hydroxylamine or side reactions with lignin.  An example of an additional experiment 
would be to attempt to produce the oxime at pH 4 where lignin would be insoluble and 
then add alkali to drive degradation of the oxime to a nitrile.   
6.6 General Yield Improvement Conclusions 
Pulp yield was difficult to measure even in a laboratory setting due to variability 
in moisture content of the extracted chips and pulp and the variability of chip 
composition and size if the chip sample is small. This screening study indicated only 
modest yield improvements from the additives and conditions tested.  Analysis of the 
results of the study indicated that division of a sample of wet chips into smaller samples 
for comparative study doubles the variability in pulp yield.  A larger set of data is then 
required to statistically demonstrate a change.  Replicates of the study conditions were 
not performed as the study was only intended to identify additives worthy of further 
future study.   
Based on a combination of literature results and results in this study 
hydroxylamine is recommended for further study as a yield additive.  AQ could also be 
included in some cooks after proper conditions for yield improvement with 
hydroxylamine are determined.   
Further attempts to improve pulp yield should include optimization of alkali 
charge to a residual target in addition to kappa number targeting.    
141 
 
 
 
Chapter 7  Value Prior to Pulping Model 
A mathematical model of a process can be used in process design to predict 
optimal conditions, test modifications to a built process or new material specifications 
(Jimenez, McKean and Gustafson 1990), and, after the process is built, to aid in process 
control.  There have been many different predictive models developed for kraft pulping.  
They include simple kinetic models, models that include mass and heat transfer 
considerations and empirical models that are accurate but limited to the boundary 
conditions under which the data were gathered for model development.  The form of the 
model will depend on the assumptions that are made to simplify the actual physical 
situation into something that can be described mathematically.  The appropriate type of 
model and development method depends on the intended end use for the model.  
Fundamental models provide understanding of a process and facilitate testing of variable 
changes on specific property values.  Empirical or input/output models may have no 
fundamental significance or applicability outside of the specific process they were 
developed for, but they are typically easier to develop and provide a suitable basis for 
process control (Saucedo and Krishnagopalan 1999).  Models published prior to 1983 
have been discussed and summarized previously including descriptions of the models and 
their individual advantages and disadvantages (Grace 1989, 45-73).  This chapter 
includes discussion of a variety of models used to describe wood hydrolysis and pulping 
to provide a context for the development of a complete VPP model.   
142 
 
An ideal VPP model would predict sugar recovered from prehydrolyzate and pulp 
characteristics following prehydrolysis and pulping of wood.  The inputs available to this 
model would include the wood composition and process conditions of the prehydrolysis 
and pulping operations.  The VPP model would be general enough that the wood 
composition data is sufficient characterization of the raw material.  In addition, ideally, it 
should have a flexible structure so that changes in pulping or prehydrolysis conditions 
could be easily incorporated into the model and still give good estimates of 
prehydrolyzate composition and quantity and pulp kappa number, yield, and composition.  
The model would require at least two different stages of calculation, one each for 
prehydrolysis and pulping.  The primary input for each stage would be the composition 
and quantity of wood chips entering the stage.  Additional pulp or chip treatments could 
then be considered by adding additional stages.  Composition information would be 
formatted to include quantity of sugar monomers and approximations as to how these 
monomers are distributed into polysaccharide molecules.   
This research does not attempt to complete the ideal VPP model.  Development of 
the ideal model will require experiments that isolate kinetic and mass transfer effects and 
composition analysis of not only sugar monomers, but also oligomers and degradation 
products.  The experiments performed in this research were intended to screen water 
extraction variables and potential yield improvement additives for process conditions that 
would facilitate industrial implementation of VPP with Loblolly pine.  This development 
looks only at the first stage of the ideal VPP model: the prehydrolysis.  The development 
attempts to produce a model that incorporates pH, temperature, and time to improve 
understanding of how these interact in the extraction of sugars from wood chips.  A 
143 
 
variety of model forms were considered in this development in hope that in addition to 
representing the data available from this research, understanding of what would be 
required to develop the ideal model would be improved.   
7.1 Historical Development  
7.1.1 Wood hydrolysis or prehydrolysis models 
The prehydrolysis factor or ?P-factor? is used to combine time and temperature in 
the first stage of a prehydrolysis-kraft cook into a single variable (Brasch and Free 1965). 
The development is compared to Vroom?s H-factor (discussed in kraft pulping models), 
but the approximation that the reaction rate trebles for each 10?C rise in temperature is 
used instead of the Arrhenius law.  They do not include liquor pH in the correlation 
because it is controlled by the release of acid from the wood in water prehydrolysis and is 
therefore already a function of time and temperature.  They state that initial pH might be 
important in the case of mineral acid prehydrolysis but specifically exclude that case 
from this work.  Another limitation to use of the P-factor is that the second stage or kraft 
conditions have to be the same for all cooks in order to make direct comparisons of the 
pulps prepared with different P-factor.   Included in the paper are plots of various 
characteristics of the prehydrolyzed chips, hydrolyzate, and final pulp as a function of 
prehydrolysis factor.  For reference, prehydrolysis with a 40 minute heat up time to 
170?C and then one hour at 170?C yielded a prehydrolysis factor of 2400.  A single stage 
kraft pulp was prepared for the ?0? P-factor sample.  Yield of chips and pulp dropped 
very quickly with rising P-factor before leveling off after about 5000 P-factor.  
Hydrolyzate pH leveled off at 3.0 at about 10000 P-factor.  Pentosans in the chips 
dropped quickly to about 5% at 2000 P-factor and then continued a slower decline while 
144 
 
pentosans in the final pulp dropped very quickly and then leveled at just under 2% after 
2000 P-factor.  Kappa number reached a minimum and alpha cellulose a maximum at 
about 1000 P-factor.  The authors? explanation of these results is that the initial 
hydrolysis is of easily accessible carbohydrates while later hydrolysis is mass transfer 
limited in the first stage.  The maximum in alpha cellulose signals a transition to 
hydrolysis of cellulose while the minimum in kappa number shows the extent of lignin 
condensation after the pH drops below 3.5.   
The P-factor is an example of a severity parameter, a single value that 
incorporates the contributions of more than one variable to allow simple comparisons of 
different conditions.  In the cases of the P-factor and H-factor, those variables are time 
and temperature.  Chum added an acidity function to previous severity measurements by 
adding a fitted parameter that is species specific and then adjusting the severity factor by 
the pH value (Chum 1990). This ?combined severity parameter? is then used to compare 
water and solvent based acid-catalyzed fractionation of Aspen.  In the case of a first order 
reaction, the fitted parameter can be derived from the activation energy or the activation 
energy can be calculated from the fit data.  Fractionation experiments on Aspen were 
performed with various concentrations of methanol in water with sulfur dioxide, sulfuric 
acid or phosphoric acid catalyst.  This data was compared to published data on water-
based dilute acid hydrolysis of Aspen.  The combined severity parameter produced a 
good correlation for xylan removal in both cases and for lignin removal in the solvent 
situation.  There was no effect of varying methanol concentration in the 40 to 70% range 
studied.    
145 
 
Species specific kinetic models have also been developed for prehydrolysis of 
hardwoods.  An effort to model autohydrolysis xylan removal from five different 
hardwoods at 170?C proposed dividing the hemicellulose into fast and slow reacting 
components that were removed by parallel first order reactions (A. H. Conner 1984).  The 
rate constants for the two reactions correlated well with the fast reaction approximately 
20 times faster than the slow reaction for all five species.  The constants varied widely 
among species with red oak approximately four times faster than American elm.  The 
authors discussed a decreased initial fast xylan removal as the proportion of uronic acid 
groups in xylan increased from one species to another, but then opined that the difference 
between fast and slow hydrolysis did not depend on the chemical nature of the polymer, 
but instead on the accessibility of the hemicellulose from within the lignin matrix.   
Further work by this same group considered hydrolysis with 0.1 M HCl at 120?C 
(Conner, Libkie and Springer 1985) and 5% acetic acid at temperatures from 170 to 
240?C (Conner and Lorenz 1986).  In the case with HCl, the faster reaction was from 10 
to 15 times faster than the slower reaction for all species, but the slow rate was almost the 
same for all.  Once again, the fast rate correlated with the uronic acid/ xylan ratio, but in 
this case the fastest species was only 1.5 times faster than the slowest.  The authors of the 
1985 paper speculate that the higher acid concentration of HCl may simply dwarf 
interspecies differences in acetyl content and that mass transfer effects are probably more 
important at the lower temperature.  The group considered the production of furfural and 
other degradation products in addition to xylan in the study at higher temperatures.  They 
again used the two component model but found the fraction of slow reacting xylan to 
decrease with increasing temperature so that at 240?C the slow reacting fraction was 
146 
 
replaced by a small non-reactive fraction.  This again suggests the fast and slow 
phenomena are due to mass transfer rather than chemical structure differences.  They 
assumed the complex reactions involved in the degradation of xylan could be modeled 
with apparent rate constants for a single reaction.   
When the best fit values for each rate constant were used to predict concentrations 
of oligomers and monomers of xylan, the model over predicted monomers in the early 
time, and under predicted them in the later times of reaction.  This was corrected by 
adding a time variance to the rate constant for the monomer production reaction: 
 k2 = k2max(1 ? exp(-Bt)) (1) 
The constants k2max and B were determined in the fitting process.  The logic in this 
form was that hydrolysis of the oligosaccharides is a random process and thus more likely 
to produce free xylose monosaccharides as the length of the oligomers is reduced over 
time.  The model then fit the data well and was used to predict time and temperature for 
maximum yield of anhydroxylose in the prehydrolyzate.  The best yield was found at 
higher temperatures, but this was a pretreatment process for wood hydrolysis, not 
pulping.  Addition of acetic acid reduced the time to maximum xylan yield, but did not 
change the maximum quantity of xylan removed.  An advantage of using water or acetic 
acid over dilute mineral acid is that the prehydrolyzate should not require neutralization 
prior to use.  A potential disadvantage is the production of more oligomers with the 
weaker acid.   
Values of the activation energy for xylan hydrolysis that are reported in the 
literature are presented in Table 10.   These values are determined from modeling of 
experimental data with feedstock including corn cobs (Nabarlatz, Farriol and Montane 
147 
 
2004) and various hardwood species including red oak (Conner and Lorenz 1986), aspen 
(Springer 1966), and paper birch (Maloney, Chapman and Baker 1985).    
Table 10  Hydrolysis activation energy values from literature. 
 
7.1.2Kraft pulping models 
There have been many mathematical models of pulping developed to predict the 
impact of changing reaction conditions on particular variables.  One of the simplest and 
most useful for kraft pulping is Vroom?s ?H-factor? that indexes lignin removal as a 
function of time and temperature (Grace 1989, 49-54).  The H-factor is used successfully 
in industry to help digester operators adjust temperature to maintain pulp uniformity 
during production rate changes.  The model gives a reaction rate relative to an assigned 
value of 1.0 at 100?C using the Arrhenius law with assumed activation energy of 134 
kJ/mol.  The relative reaction rate is then integrated as a function of time using the 
cooking schedule.  The result of the integral is the H-factor, a single number that 
describes the exposure of the chips to temperature changes along the entire cook 
kJ/ m o l e kca l / m o l e C o n d i t i o n s So u rce
X y l a n  H y d ro l y s i s
88 2 1 . 0 5 %  A ce t i c A ci d C o n n e r,  1 9 8 6
108 2 5 . 8 W a t e r A u t o h y d ro l y s i s C o n n e r,  1 9 8 6
116 2 7 . 7 5 %  A ce t i c A ci d ,  " f a s t "  re a ct i o n C o n n e r,  1 9 8 6
118 2 8 . 2 D i l u t e  H C l Sp ri n g e r,  1 9 6 6
119 2 8 . 4 X y l o s e  m o n o m e r f ro m  o l i g o m e r N a b a rl a t z ,  2 0 0 4
127 3 0 . 4 " f a s t "  re a ct i o n  i n  d i l u t e  H 2 SO 4 M a l o n e y ,  1 9 8 4
1 2 7 . 3 3 0 . 4 " R e a ct i v e "  x y l a n  p o l y m e r t o  o l i g o m e r N a b a rl a t z ,  2 0 0 4
136 3 2 . 5 W a t e r " f a s t "  re a ct i o n C o n n e r,  1 9 8 6
2 5 1 . 7 6 0 . 2 " L e s s  re a ct i v e  x y l a n  p o l y  t o  o l i g N a b a rl a t z ,  2 0 0 4
A ra b i n o s e  H y d ro l y s i s
1 0 6 . 2 2 5 . 4 A ra b i n o s e  f ro m  x y l a n  o l i g o m e r N a b a rl a t z ,  2 0 0 4
A ct i v a t i o n  En e rg y
148 
 
schedule.  The H-factor has been correlated to yield as well as lignin content of pulp 
when also given the applied effective alkali (Hatton 1973).   
H-factor has been used more recently to combine time and temperature in 
predicting weight loss during water hydrolysis of pine chips.  Weight loss was plotted vs. 
H-factor for various temperature and time of reaction combinations.  The authors 
concluded that the activation energy for water hydrolysis must be similar to that for kraft 
delignification, 32 kcal/mol (Yoon, MacEwan and van Heiningen 2006). 
A model for pulping based on a single chip evolved over several years 
(Gustafson, et al. 1983).  The original model considered wood to be an infinite slab with 
set thickness made up of lignin, carbohydrate, and acetyl groups.  The acetyl was 
considered separately because of its effect on alkali concentration.  The equations used to 
describe pulping included kinetics divided into initial, bulk, and residual periods and 
mass transfer of alkali with a diffusivity that depended on temperature, lignin remaining 
in wood, and alkali concentration.  Mass transfer of other species were not considered 
because all reactions were considered to be irreversible, sulfide concentration was 
assumed constant, and the model output concentrations were for well washed pulp.  
Another input to the model was the lignin content remaining when delignification 
changed from bulk to residual.  This model was used to make predictions about potential 
problems with chip size variation and to interpret experimental data in this area.   
Gustafson?s model was later modified so it could be used to simulate extended 
pulping methods (Pu, McKean and Gustafson 1991).  This required a more detailed 
consideration of the initial period of pulping particularly with regard to the carbohydrate 
reactions and a revised accounting of consumption of alkali.  Cellulose and hemicellulose 
149 
 
are considered separately to improve yield predictions.  The carbohydrate reactions are 
also no longer a function of lignin concentration, but of the carbohydrate reduced by an 
?unreactive %? of the species, 5% for hemicellulose and 32% for cellulose.   
The model was used to simulate Rapid Displacement Heating (RDH) batch 
pulping where the chips are heated quickly with warm and then hot black liquor before 
white liquor is added.  Each of these liquors is displaced by the following one with the 
aim of saving energy and reducing residual lignin sent to bleaching.  The results of 
experiments and modeling had two surprising results:  The yield/kappa relationship was 
similar to conventional pulping and the delignification kinetics were similar as well.  
These results were found even though RDH pulping reaches a much lower kappa number 
at the same H-factor.  The explanations for these results were that the initial pulping 
period was accomplished in the black liquor portion of the cook and the transition from 
bulk to residual delignification is reduced from 50 kappa to 20.  Yield is proposed to be 
the same because reduced carbohydrate losses in initial pulping due to the relatively low 
alkali black liquor at the beginning are balanced by reduced xylan precipitation from the 
relatively high alkali black liquor at the end of the cook.   
The geometric basis for Gustafson?s model was later changed from an infinite 
slab to a sphere with equivalent specific surface area of an average chip (Agarwal and 
Gustafson 1997).  This allows diffusion in three dimensions to be approximated while 
maintaining the mathematical simplicity of the one dimensional model.  The model 
results suggest that decreasing temperature while increasing alkali charge may reduce the 
fiber?to-fiber kappa variability.  This could allow low kappa pulping at similar 
productivity and yield.   
150 
 
7.2 Model Development 
In the current research, model development began with the simplest approach 
using the H-factor as the basis of development and the total sugar recovered as the 
primary variable of interest. Adjustment of the activation energy in the H-factor and the 
addition of a pH dependence to the relative rate constant produce a method of comparing 
extraction severities of various prehydrolysis conditions on an equivalent basis.   
7.2.1 The Modified H-factor 
The H-factor was used previously as a guide for comparing chip weight loss data 
from hot water extraction for up to 90 minutes at temperatures from 160 to 190?C (Yoon, 
MacEwan and van Heiningen 2006).  The pH of the hydrolysis solution was measured at 
the end of each extraction in that study and was found to decrease slowly with increasing 
H-factor.  The first attempt at modeling in this study was to compare chip weight loss and 
total sugar recovered from the nine tests in the Latin square designed experiment 
discussed in Chapter 5 on the basis of H-factor alone.    
The chip weight loss displayed as a function of H-factor in Figure 59 shows that 
the pH of the solution must be included in the analysis.  The weight loss is substantially 
higher at lower pH for all tested pH and H-factor levels with the exception of the low H-
factor test at pH 3.5 that may exhibit lower than expected weight loss due to an error in 
measurement of chip moisture (refer to Chapter 5).  The plot of total sugar recovered in 
the hydrolyzate as a function of H-factor shown as Figure 60 also illustrates the need to 
include pH.  The legend includes the letter indicating the test, the temperature target, 
solution pH target, and presoak pH.  The solution pH in Figure 60 is represented by the 
shape of the data marker, the temperature by line color, and the presoak pH by marker 
151 
 
color.  The data for the three different pH levels are generally grouped together on the 
plot with the lower pH showing higher recovery at a given H-factor.   
Figure 59  Chip weight loss from prehydrolysis as a function of H-factor. 
 
There is also a general trend in each pH group in Figure 60 for the higher 
temperature to show a lower sugar recovery at the same H-factor.  A possible reason for 
the trend with temperature is that the activation energy used to calculate the H-factor, 32 
kcal/mol, is higher than many of the reported values for the hydrolysis reactions that 
produce the sugars listed in Table 10.  A second possible explanation is that the measured 
quantity, recovered sugars, is an intermediate product in a series of reactions and that the 
activation energies for the reactions that degrade the sugars in solution are higher than 
those for the reactions that produce the sugars.  Reported values for dehydration of xylose 
to furfural vary from 29 to 32 kcal/mol (Nabarlatz, Farriol and Montane 2004).  
 
0
5
10
15
20
25
30
0 200 400 600 800 1000 1200 1400 1600
Chip 
we
ight 
los
s, 
% 
bd w
ood
H-factor
pH 3
pH 3.5
pH 4.5
152 
 
Figure 60  Total sugar recovered as a function of H-factor (Legend indicates trial 
label, temperature, reaction pH, and presoak pH). 
 
The potential explanations for the data taken at different temperatures not 
matching at the same H-factor were tested by recalculating the H-factor using different 
activation energies and comparing the data from each pH group graphically.  Figure 61 
illustrates the recovered sugar data collected at pH 3 for all three temperatures plotted 
against a modified H-factor calculated using 27 kcal/mol as the activation energy.  This 
value agrees with reported values shown in Table 10and allowed a good fit for most of 
the data.  The 140? and 155?C data overlay each other very well, while the data taken at 
170?C still shows reduced recovery at the same modified H-factor.  The data markers for 
the 140?C runs have been removed for clarity in the rest of the plots of this type in this 
chapter.  The data from the tests at pH 3.5 are similarly plotted in Figure 62.  As at pH 3, 
the two lower temperature tests plot on the same line while the test at 170?C shows lower 
recovery.  The recovery at higher temperature is not reduced as much at pH 3.5 as at pH 
0
2
4
6
8
10
12
14
0 200 400 600 800 1000 1200 1400 1600
Suga
rs
 re
cov
ere
d, 
%bd 
chi
ps
H-factor
A 170,3.0,  7
F 170,3.5,10
C 170,4.5, 3.5
G155,3.0, 10
E 155,3.5, 3.5
B 155,4.5, 7
I  140,3.0,  3.5
D 140,3.5, 7
H 140,4.5,10
153 
 
3.0.  Finally, the data at pH 4.5 are shown in Figure 63 and shows that the modified H-
factor is able to account for the variation due to temperature among these three tests up to 
a two hour time at temperature.  These plots suggest that both reasons for the inability of 
the conventional H-factor to account for the variation of measured sugar recovery exist.  
Adjustment of the activation energy to 27 kcal/mol properly accounted for differences 
due to temperature except in the cases where significant degradation of extracted sugars 
is suspected.   
 
Figure 61  Total sugar recovered at pH 3.0 as a function of H-factor with EA = 27 
kcal/mole (Legend indicates trial label, temperature, reaction pH, and presoak pH). 
 
The mathematical form of the pH influence can be determined by looking at an 
assumed kinetic rate law for a reaction depending on native wood carbohydrate 
concentration, CC, and acid concentration, CA: 
                (2) 
0
2
4
6
8
10
12
14
0 100 200 300 400 500 600
Suga
rs
 re
cov
ere
d, 
%bd 
chi
ps
Modified H-factor, EA=27
A 170,3,   7
G155,3,   10
I  140,3,   3.5
154 
 
 
Figure 62  Sugar recovered at pH 3.5 as a function of H-factor with EA = 27 
kcal/mole (Legend indicates trial label, temperature, reaction pH, and presoak pH). 
 
Figure 63  Total sugar recovered at pH 4.5 as a function of H-factor with EA = 27 
kcal/mole (Legend indicates trial label, temperature, reaction pH, and presoak pH). 
0
2
4
6
8
10
12
14
0 100 200 300 400 500 600
Sugar
s re
cov
ere
d, %
bd c
hips
Modified H-factor, EA=27
F 170,3.5,10
E 155,3.5,3.5
D 140,3.5,7
0
2
4
6
8
10
12
14
0 100 200 300 400 500 600
Sugar
s re
cov
ere
d, %
bd c
hips
Modified H-factor, EA=27
C 170,4.5,3.5
B 155,4.5,7
H 140,4.5,10
155 
 
 
Since the pH was maintained at a target level for these experiments and the acid is 
not consumed in the reaction, the acid concentration term can be considered to be part of 
the rate constant along with the temperature effects that have already been shown to 
follow the Arrhenius relationship:   
            
  
  
 
    
 
   (3) 
If only the three tests performed at 140?C are used for this part of the 
development, then degradation effects should be minimal and the temperature term can 
be included in the constant: 
             (4) 
Taking the natural log of both sides: 
                    (5) 
A plot of ln(kC) against ln (CA) will have a slope of the exponent, n, for the acid 
term.  The individual kC?s were calculated from the slope of a plot of ln(CC0/CC) vs time 
where CC0 is the total initial carbohydrates, 62% of wood, and CC is calculated as CC0 
minus the sugars measured in the hydrolyzate.  The exponent for acid was calculated to 
be 0.317. 
The modified H-factor references for each test in the designed experiment were 
recalculated using the new kC.  The number 25 is a proportionality constant added to 
return the new modified H-factor to approximately the same scale as the original for this 
data: 
                                     (6) 
156 
 
All nine sets of recovered sugar data are plotted as a function of the modified H-
factor including pH in Figure 64.  All of the variability in this data is covered by the 
modified H-factor except for that caused by degradation of sugar monomers in solution at 
170?C and either pH below 4.5 or time at temperature longer than 45 minutes.   
Figure 64  Total sugar recovered as a function of the modified H-factor (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
The spreadsheet that was used to calculate the modified H-factor was constructed 
so the activation energy, Ea, and the order exponent for the pH effect, n, could be varied 
to gauge the impact of changing these parameters on the appearance of the final 
composite curve of all nine prehydrolysis tests.  The constant was then calculated to force 
the modified H-factor to equal the conventional H-factor for condition ?A?, the most 
severe case.  Activation energies from 25 to 29 kcal/mole yielded acceptable appearance 
with 27 kcal/mole the best fit for the tests with little to no degradation of sugars.  
0
2
4
6
8
10
12
14
0 200 400 600 800 1000 1200 1400 1600
Sugar
s re
cov
ere
d, %
bd c
hips
Total modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
157 
 
Changing the exponent, n, on the acid term yielded acceptable fit from 0.26 to 0.33 with 
0.31 the best.  The proportionality constant, k0, was 23.5 with the ?best? parameters.   
Each sugar was also considered separately for variation of activation energy and 
pH effects on extraction from wood.  The activation energy was chosen by comparing 
plots of three sets of three tests at the same pH but different temperatures.  The best 
number for each sugar varied from a low of 23 kcal/mole for arabinose to a high of 29 
kcal/mole for glucose, but the value of 27 kcal/mole determined for total sugar would 
have worked well in all cases.  The exponent for the acid term was first calculated using 
the rate of appearance of each sugar at 140?C in a similar manner to that used for the total 
sugars discussed previously yielding exponents from 0.23 for glucose to 0.41 for xylose, 
but the calculated values did not work well when tested graphically.  The number used 
for the plots was chosen by comparing three sets of three tests at the same temperature 
but different pH levels.  These exponents varied in a much smaller range from 0.26 for 
arabinose to 0.31 for glucose and galactose.  When different values of the parameters 
appeared to be optimal in the three plots, the plot of the less severe condition was given 
priority since degradation of sugars would be a smaller consideration.  The determination 
of the two parameters separately in comparisons of three tests at a time can lead to a 
suboptimal set relative to when the data from all nine tests are observed together, but the 
differences were not large.  The modified H-factor was calculated with the parameters 
determined from the two sets of three separate plots for charts of the five sugars shown 
here. 
The plots of each sugar as a function of the modified H-factor make it obvious 
what conditions cause degradation of each sugar.  Glucose, shown in Figure 65, appears 
158 
 
to have very little degradation, but as was discussed in Chapter 5, this is at least partially 
due to recovery of sugar hydrolyzed from cellulose in addition to hemicellulose under the 
conditions that cause degradation of monomers in solution.  The activation energy for 
glucose hydrolysis was estimated to be 29 kcal/mole through visual inspection of the 
three plots at different pH levels, but 27 kcal/mole appeared better when all nine tests 
were viewed together.   
Figure 65  Glucose recovered as a function of the modified H-factor (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
The pentoses, xylose and arabinose, showed the most deviation from an imagined 
smooth extension of the composite curve beyond 500 modified H-factor.  The recovered 
xylose is plotted against a modified H-factor calculated using 26 kcal/mole for activation 
energy, 0.30 for the acid exponent, and a proportionality factor of 26.9 in Figure 66.  This 
figure shows the data from the tests at 170?C begin to deviate from the composite curve 
at about 300 modified H-factor.  There must have been sufficient xylose monomers in 
0.0
0.5
1.0
1.5
2.0
2.5
0 200 400 600 800 1000 1200 1400 1600
Gl
ucos
e re
cov
ere
d, %
bd 
chips
Glucose modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
159 
 
solution that the rate of degradation to furfural began to exceed the rate of xylose 
monomers or oligomers entering solution.   
Figure 66  Xylose recovered as a function of the modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH). 
 
The modified H-factor for galactose was calculated with an activation energy of 
25 kcal/mole and an acid exponent of 0.31 for Figure 67.  The deviation from the 
composite curve, and therefore the expected degradation rate, was greater for galactose 
than for the other hexoses, glucose and mannose.  This could be because galactose was 
more likely to enter solution as a monomer since it is derived from sidechains of the 
galactoglucomannan that are susceptible to acid hydrolysis and arabinogalactans that are 
soluble in water (Sjostrom 1993, 65).   
Arabinose is completely hydrolyzed by 200 modified H-factor, so the composite 
curve of all the data should then form a flat line at 1.55%.  The data plotted in Figure 68 
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0 200 400 600 800 1000 1200 1400 1600
Xy
los
e re
cov
ere
d, %
bd 
chips
Xylose modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
160 
 
indicates that arabinose begins to degrade from that point, but the arabinose data in that 
range are only estimates in most cases due to interference from the larger mannose peak.  
The fit parameters used for this plot were 23 kcal/mole for Ea, 0.26 for n, and 37.3 for k0, 
but the appearance of the plot was not very sensitive to these values.   
Figure 67  Galactose recovered as a function of the modified H-factor (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
The mannose recovered sugar data in Figure 69 shows deviation from the 
composite curve beginning at about 300 modified H-factor.  The modified H-factor was 
calculated using 27.5 kcal/mole for Ea, 0.29 for n, and 18.5 for k0.  The degradation of 
mannose and arabinose may be more sensitive to pH than the other sugars since the curve 
for run ?A? at 170?C and pH 3.0 was lower than that for run ?F? at 170?C and pH 3.5 at 
higher modified H-factor.   
 
0.0
0.5
1.0
1.5
2.0
2.5
0 200 400 600 800 1000 1200 1400 1600
Ga
lac
to
se
 re
cov
ere
d, %
bd c
hips
Galactose modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
161 
 
Figure 68  Arabinose recovered as a function of the modified H-factor (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
Figure 69  Mannose recovered as a function of the modified H-factor (Legend 
indicates trial label, temperature, reaction pH, and presoak pH). 
 
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0 200 400 600 800 1000 1200 1400 1600
Ara
binos
e re
cov
ere
d, %
bd c
hips
Arabinose modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
0.0
1.0
2.0
3.0
4.0
5.0
6.0
0 200 400 600 800 1000 1200 1400 1600
Ma
nnose
 re
cov
ere
d, %
bd 
chips
Mannose modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
162 
 
The modified H-factor with pH adjustment can be useful for predicting the 
behavior of more than just the sugar recovery from which it was derived.  The wood 
weight loss and pulp yield at 30 kappa number were also plotted as functions of the 
modified H-factor.  Figure 70 shows an improvement over Figure 59 as to the data being 
closer to a single relationship, but there is still a lot of scatter. The scatter in the data is 
likely due to errors in this measurement as discussed before.  The pulp yield data all align 
on one relatively smooth curve in Figure 71 with the untreated reference pulp shown as 
?Raw?.  The comparison of pulp yield can be made because all of the data is referenced 
to a standard of 30 kappa number since the modified H-factor refers only to the 
conditions of the prehydrolysis.  This plot indicates a significant pulp yield loss of 6% 
relative to the standard pulp after only 200 modified H-factor.   
Figure 70  Chip weight loss from prehydrolysis as a function of modified H-factor. 
 
0
5
10
15
20
25
30
0 200 400 600 800 1000 1200 1400 1600
Chi
p w
eight 
los
s, 
% 
bd 
wood
Modified H-factor with pH, Ea = 27
Raw
pH 3.0
pH 3.5
pH 4.5
Water
163 
 
Figure 71  Prehydrolyzed kraft pulp yield at 30 kappa as a function of modified H-
factor. 
 
Comparisons of pulp characteristics can also be made using the modified H-factor 
if the pulping conditions for the kraft cook are all the same.  The cook that was performed 
on the first subset of chips from each of the nine prehydrolysis test conditions was with 
the same liquor charge and to the same H-factor, 1850.  The yield from each of these 
cooks is plotted as a function of modified H-factor with pH in Figure 72.  Again, a fairly 
smooth curve could be drawn through this collection of data that previously would have 
had three separate lines; one for each pH level.  A final example is the plot of the kappa 
number from these cooks shown in Figure 73.  A plot of this type would make it very 
easy to pick the optimum prehydrolysis conditions to minimize lignin content of pulp.   
An additional prehydrolysis test designed to end at a modified H-factor of 300 would 
help to define the lower range of this dataset.   
 
25
27
29
31
33
35
37
39
41
43
45
0 200 400 600 800 1000 1200 1400 1600
Pulp 
yie
ld, 
%bd 
wood
Modified H-factor with pH, Ea = 27
Raw
pH 3.0
pH 3.5
pH 4.5
Water
164 
 
Figure 72  Pulp yield after "standard" cook as a function of modified H-factor. 
 
Figure 73  Kappa number after "standard" cook as a function of modified H-factor.  
 
25
27
29
31
33
35
37
39
41
43
45
0 200 400 600 800 1000 1200 1400 1600
% P
ulp 
yie
ld, 
% bd w
ood
Modified H-factor with pH, Ea = 27
Raw
pH 3.0
pH 3.5
pH 4.5
15
20
25
30
35
40
45
50
55
60
65
0 200 400 600 800 1000 1200 1400 1600
Kappa
 number
 at 18
50
 H
-fac
to
r
Modified H-factor with pH, Ea = 27
Raw
pH 3.0
pH 3.5
pH 4.5
165 
 
The total recovered sugar data displayed in Figure 37 was modeled by putting the 
differential rate equation in Polymath 6.10 with the necessary explicit equations:   
                  (7) 
The rate constant is defined here: 
                             (8) 
The initial rate constant, k0, was determined to be 0.00078 minute-1 by matching 
the curve generated in Polymath 6.10 with the data taken at 140?C and pH 3.0.  The 
Polymath program for the total sugars represented by these equations is fairly simple and 
is illustrated in Figure 74.  Cc0 is defined here as the initial hemicellulose mass fraction 
of the dry wood.   
 
This program was run for the nine conditions by changing the temperature target 
and the pH.  The time, temperature, and sugar (S) outputs were exported to Microsoft 
Dif f e re n t ial  e qu a t io ns   
1  d( Cc)/ d( t ) = - k c * Ca ^ n * Cc  
 
Exp lici t  e q uat i on s   
1  T t a rg = 1 4 0   
2  pH =  3 .0   
3  k 0  =  0 . 0 0 0 7 8   
4  n  =  0 .3 1   
5  E a  =  2 7 0 0 0   
6  Cc0  =  0 . 2 3 6 1   
7  T in i =  3 0   
8  T 1  =  if (t < 6 0 ) t h e n  (t * (T t a r g - T i n i)/ 6 0 + T in i + 2 7 3 ) e lse  ( T t a rg + 2 7 3 )  
9  T 0  =  3 7 3   
10  R  =  1 . 9 8 7   
11  Ca  =  1 0 ^ ( - p H)  
12  k c  =  k 0 * e xp(E a / R * (1/ T 0 - 1 / T 1 ))   
13  S =  1 0 0 * (C c 0 - C c )  
 
Figure 74  Modified H-factor model equations for Polymath 6.10. 
166 
 
Excel ? for calculations and plotting.  Selected examples of the predicted values of 
carbohydrates extracted are plotted with the corresponding experimental values as a 
function of time in Figure 75.  The predicted values are shown with lines in the same 
colors as before for different temperatures.  The lines for the experimental data are 
reduced in size to improve visual clarity of the plot.   
  Figure 75  Total recovered sugars compared to predicted values as a function of time 
for selected conditions (Legend indicates trial label, temperature, reaction pH, and 
presoak pH). 
 
The four experimental conditions that had very little degradation of sugars, 
represented in Figure 75 by condition I, are modeled well with this system of equations.  
The data from more severe conditions, illustrated by condition A, digress significantly 
from the predictions.  The predicted values for condition C, 170?C and pH 4.5, are so 
close to the predicted values for G, 155?C and pH 3.0, that the line for condition C is 
0
5
10
15
20
25
0 50 100 150 200
Suga
rs
 re
cov
ere
d, 
%bd 
chi
ps
Time, minutes
A 170,3.0,  7
C 170,4.5, 3.5
G155,3.0, 10
I  140,3.0,  3.5
A Model
C Model
G Model
I Model
167 
 
almost completely covered, but the data from the higher temperature run shows higher 
deviation from the model prediction due to more degradation of monomers in solution. 
The modified H-factor for the predicted values was calculated with the time and 
temperature data output from Polymath.  The predicted and experimental values are then 
plotted together as a function of modified H-factor in Figure 76.  The experimental values 
align with the predicted values up to approximately 250 modified H-factor before the 
more severe conditions digress.  These two examples of comparisons of predicted 
extraction of sugars and actual recovery of sugars show that the model can be a useful 
guide in choosing proper extraction conditions.   
Figure 76  Model predictions and experimental values as a function of modified H-
factor (Legend indicates trial label, temperature, reaction pH, and presoak pH). 
 
0
5
10
15
20
25
0 200 400 600 800 1000 1200 1400 1600 
Sugar
s re
cov
ere
d, %
bd c
hips
Modified H-factor
A Model A 170,3.0,  7
F 170,3.5,10 C 170,4.5, 3.5
G155,3.0, 10 E 155,3.5, 3.5
B 155,4.5, 7 I  140,3.0,  3.5
D 140,3.5, 7 H 140,4.5,10
168 
 
7.2.2 Validation 
The model was tested with data from a prehydrolysis run that was performed with 
the pH meter installed but without pH adjustment.  The H-factor for this run was 
recalculated using the modified H-factor parameters of 27 kcal/mole activation energy, 
pH exponent of 0.31, and a correction factor of 23.5.  The pH and temperature were 
recorded on approximately five minute intervals for this test.  The temperature for the 
calculation was estimated from the digester temperature since the second temperature 
probe had not been installed at the time of the test.  The pH started at 5.3 and gradually 
dropped to 3.67 at the end of the test after one hour at the temperature target of 170?C.  
The pH was 4.24 at the end of the temperature ramp.  Samples of the prehydrolyzate 
liquor were taken on fifteen minute intervals beginning with the end of the temperature 
ramp.  The degradation products furfural and hydroxymethylfurfural were observed in the 
chromatograms beginning with the 15 minute and 45 minute samples respectively.   
The validation data set, labeled ?wtr 51208?, is plotted as a function of the 
modified H-factor along with the nine data sets used to set the parameters in Figure 77.  
The validation data are slightly offset below the other nine sets, but are only about 10% 
lower than the other data from tests at 170?C.   The chip weight loss was measured after 
the prehydrolysis extraction.  The pulp yield at 30 kappa number was also calculated 
from four bomb cooks with chips from this run.  These two measurements are shown in 
Figure 70 and Figure 71 with the label ?Water?.  These two points fit right in line with 
the other data in these plots illustrating good agreement of this run with those used to 
calibrate the modified H-factor.   
169 
 
Figure 77  Total recovered sugar as a function of modified H-factor (Legend indicates 
trial label, temperature, reaction pH, and presoak pH. Data labeled ?wtr 51208? is 
validation data). 
 
7.2.3 Additional Improvements Required 
The modified H-factor with pH is a functional tool to compare prehydrolysis runs 
with different pH and temperature, but it was derived from a small range of pH and 
temperature.  It would be interesting to see how well data from outside these ranges are 
represented.    
0
2
4
6
8
10
12
14
0 200 400 600 800 1000 1200 1400 1600
Sugar
s re
cov
ere
d, %
bd c
hips
Modified H-factor
A 170,3,   7
F 170,3.5,10
C 170,4.5,3.5
G155,3,   10
E 155,3.5,3.5
B 155,4.5,7
I  140,3,   3.5
D 140,3.5,7
H 140,4.5,10
wtr 51208
170 
 
 
 
Chapter 8  Conclusions and Suggestions for Future Research 
This research started with a basic question:  What conditions of prehydrolysis 
extraction of Loblolly pine chips can produce a sufficient quantity of sugars in the 
prehydrolyzate solution to support production of marketable products including ethanol 
while continuing to produce high quality bleachable grade kraft pulp?  Answering this 
question, in whole or in part, significantly contributes to the body of chemical 
engineering knowledge and to the economic welfare of the pulp and paper industry.  
Extracting the sugar turns out to be relatively simple but a method for performing this 
extraction without damaging the pulp yield and strength remains to be found.  The 
process of searching for this method produced significant conclusions and elucidated 
ideas for the direction of future efforts that are discussed in this chapter.  
8.1 Conclusions from this Research 
Based on the limitations of the experimental data obtained in this research, the 
following conclusions can be drawn: 
The knowledge of prehydrolysis for value-prior-to-pulping of pine chips is 
extended to include the contribution of solution pH independent of time and 
temperature.   
The pH and temperature of the extraction solution can be used 
independently or in concert to control the extent of sugar removal and the rate of 
degradation of monomers in solution.   
171 
 
Five potential yield improvement additives were tested with two that show 
potential for future development.   
A kinetic model of the prehydrolysis was developed that accounts for pH 
and temperature variation in the form:   
                (9) 
When originally planned, the experiments for this research were intended to be 
performed in four phases: screening of prehydrolysis conditions, targeting, validation, 
and comparison.  The screening phase included the designed experiment discussed in 
Chapter 5.  Three experimental conditions identified in screening were rerun utilizing a 
reduced reaction time in an attempt to optimize time, temperature, and pH profiles for 
prehydrolysis and pulping yields.  The validation phase was intended to consist of 
repeating key conditions from prehydrolysis through pulp washing in an unbroken 
process to avoid any uncertainty introduced by multiple handling steps and storage of the 
chips between cooking stages.  The final comparison phase then would produce sufficient 
pulp with a standard kraft cook and the identified optimal multistage cook to produce 
refining curves and handsheets for strength testing of the pulp.  
The process validation and comparison steps were not completed because no 
condition was identified in the screening or targeting phases that produced pulp at an 
acceptable yield.  Experiments with chemical additives that might help recover this lost 
yield were substituted for these last two phases.  No immediately promising conditions 
were identified in these experiments, but additional work with anthraquinone and 
hydroxylamine is justified.  A simple kinetic model for sugar extraction was developed 
based on the results of the screening experiments.   
172 
 
8.1.1 Prehydrolysis Experiments 
The designed experiment was intended to show if the selective removal of 
hemicelluloses can be controlled by manipulation of three prehydrolysis variables: time, 
temperature, and pH.  The sugar extraction rate for all sugars was increased with either 
increasing temperature from 140 to 170?C or decreasing pH from 4.5 to 3.0.  The 
hydrolysis was selective for hemicellulose as opposed to cellulose by using temperature 
at 140?C or pH at 4.5 for reaction at 155?C.  The maximum in extraction yield at pH 3.5 
that was observed in previous work (Yoon, MacEwan and van Heiningen 2006) was a 
function of total severity integrating time, temperature, and pH rather than a function of 
pH alone.   
Both pH and temperature were found to have a significant impact on degradation 
of sugars in solution.  Degradation products were not observed in the chromatograms of 
samples taken at 140?C, but they were observed in all tests at 170?C.   For the tests at 
155?C, only very small amounts of furfural were observed after one hour at pH 4.5 but 
significant quantities of both furfural and hydroxymethylfurfural appeared at pH 3.0.   
Sodium hydroxide was used for pH control in the preliminary work at 170?C and 
significantly degraded sugars in solution.  Sodium carbonate was thus chosen as the agent 
for increasing pH in the designed experiment and did not appear to increase degradation 
between 140 and 170?C.   
A presoak period at room temperature with various pH levels was also tested with 
very little effect.  It is likely that the conditions of the presoak were too mild.  The low 
pH presoak was expected to accelerate extraction.  Either the experimental pH was not 
low enough or the temperature not high enough to enhance extraction.  The high pH 
173 
 
presoak was intended to stabilize the galactoglucomannan toward subsequent acid 
hydrolysis.  The curves for mannose shown in Figure 69 that represent presoak at high 
pH do not differ from the others in any greater way than in the figures representing other 
sugars as a function of the modified H-factor.  A literature reference found after the 
experiments were completed suggests that the degree of stabilization can be measured by 
the amount of acetyl removed (Ingruber, Kocurek and Wong 1985, 36).  A reference has 
not yet been found describing the recommended process conditions for this pretreatment.   
8.1.2 Yield Improvement 
Chemical pulping additives have been demonstrated to be successful in 
substantial yield recovery in other work performed in this lab at Auburn University 
(Yoon 2009), but none of the five additives tested in this research were successful in 
recovering yield close to a standard kraft cook with the conditions tested thus far.  More 
work could be justified with hydroxylamine or possibly to test the use of anthraquinone 
in conjunction with hydroxylamine or another successful additive.  No further work is 
recommended with acetaldehyde, ethanolamine, or lithium aluminium hydride.   
8.1.3 Modeling Results 
The H-factor was used as the basis for the development of a pseudokinetic model 
of prehydrolysis. The model was developed from sugar concentration data taken from 
nine prehydrolysis tests at three levels each of pH and temperature.  The modified H-
factor was calculated using a lower activation energy than that used for kraft pulping, 27 
kcal/mole, and was multiplied by the calculated acid concentration raised to the 0.31 
power.  This modified H-factor was able to represent all of the data from the nine runs on 
a single curve except when significant degradation of sugars in solution was observed.  
174 
 
The data for wood chip weight loss and pulp yield from all nine runs and a validation set 
fit a smooth curve in proportion to the error in the measurement when plotted against the 
modified H-factor.  This model is helpful in choosing prehydrolysis process conditions to 
minimize degradation of sugars and to minimize kappa number after kraft cooking. 
8.2 Suggestions for Further Investigation 
This section includes ideas that have not yet been researched and ideas that have 
been more thoroughly looked into.  Included are ideas that were thought of early in this 
research and others that were added as the report was written.  The ideas are divided into 
the three general areas of work presented in this report. 
8.2.1 Prehydrolysis Variables 
Additional experiments would further define pH and temperature relationships in 
multiple stage treatment of chips.  Potential experiments include an extension of the 
presoak used in the designed experiment to true multistage tests with higher temperature 
and higher and lower pH conditions in presoak and wash stages.  These tests would 
attempt to change the relationship between sugar extraction and pulp yield.  
Analysis of carbohydrates in the spent liquor from a batch cook with continuous 
liquor flow at low alkali concentrations revealed that carbohydrate dissolution begins 
almost immediately and that 20% is lost before any significant lignin is removed 
(Courchene 1998).  A mild alkaline treatment has also been applied for extraction of 
hemicellulose from aspen wood chips prior to pulping.  The yield of sugar in the extract 
was low, but the pulp yield was unaffected by the treatment (Al-Dajani and Tschirner 
2008).  This may indicate that a mildly alkaline wash at moderate temperature after a 
175 
 
mild prehydrolysis could increase the yield of recovered sugars in the hydrolyzate while 
limiting the negative effect of prehydrolysis on pulp yield.   
An example of this approach would be an acetic acid prehydrolysis to a modified 
H-factor of 100 followed by an alkaline soak with green liquor at 90?C before 
conventional kraft pulping.  This should increase sugar yield while improving pulp yield 
and other properties.   
Three of the tests following the designed experiment were intended to target a 
selected range of sugar extraction (8-10%) to compare pulp yield at similar sugar 
recovery or similar weight loss.  The modified H-factor should make it easier to reach the 
target sugar level with different prehydrolysis conditions.  
One of the potential problems with VPP extraction is degradation of the released 
sugars at the more severe conditions of high temperature and either low or high pH.  One 
possible method of reducing this effect would be to perform a counter-current extraction 
wherein cold (<100?C) chips are fed to one end of a continuous reactor while liquor at a 
higher temperature and low pH is fed to the other end.  In this manner, the chips would 
gradually rise to process temperature across the reactor as the hydrolyzate is cooled while 
flowing the opposite direction.  Evaluation of this process could be performed using a 
series of small batch reactors with one continuous flow of liquid between them.   
It would be interesting to find out what chemical difference contributes to strength 
loss of prehydrolyzed pulp relative to standard kraft pulp and what effect modification of 
prehydrolysis conditions has on strength.  A study of the effects of different types of 
alkaline carbohydrate degradation in holocellulose showed that pulp variables like yield 
and viscosity do not always predict pulp strength (Agarwal and Gustafson 1995).  The 
176 
 
authors wrote that dissolution, peeling, and chain scission should be tracked separately to 
accurately predict effects on pulp properties.  Direct dissolution of hemicellulose reduces 
pulp yield significantly and depends on alkali concentration.  This has the general effect 
of increasing viscosity because shorter chains are removed but decreasing strength.  
Primary peeling has a minimal effect on viscosity because longer chains are reduced 
slightly in length but shorter chains can be completely dissolved.  Chain cleavage 
followed by secondary peeling has a smaller effect on yield, but decreases both viscosity 
and fiber strength.  The prehydrolysis stage adds another group of reactions that change 
the nature of the chips before pulping.  In addition to pulp yield and the strength 
properties of tensile, tear, zero-span tensile, and burst, the prehydrolysis could change 
bulk, porosity, refining energy requirement, bleachability, and pulp softness.  These other 
properties should also be tested when a method of maintaining yield with VPP of pine 
chips is found. 
References disagree about the effect of direct steam for heating the chips for 
prehydrolysis (Rydholm 1985, 665) (Richter 1956, 199).  Potential advantages would be 
fast heat up, easy control of reaction temperature, and a high liquor-to wood ratio leading 
to more concentrated sugars in the extract.  Disadvantages of using steam alone include 
potential nonuniformity of treatment including local pH differences that can lead to lignin 
condensation and difficulty bleaching. 
8.2.2 Additional Yield Improvement Research 
The alkali charge used for the kraft cooks should be optimized to a residual target 
to minimize yield loss due to excess alkali.  The alkali requirement should be lower than 
177 
 
that for a conventional cook since much of the alkali in a conventional cook is consumed 
by degradation products of hemicelluloses.   
The acid prehydrolysis likely increased the number of reducing end groups 
available to alkaline peeling.  Reagents that function to stop peeling should also be more 
reactive.  It may be a question of finding the right conditions for reagents that we are 
already familiar with.  For example, AQ functions catalytically to preserve carbohydrates 
by oxidizing the reducing end groups, but the yield improvement observed from 
application of AQ in this research was modest while acceleration of delignification was 
pronounced.  A possible method to apply the AQ under conditions that are more 
conducive to reaction with carbohydrates would be to supply a secondary oxidizing agent 
like oxygen to regenerate the AQ.   After the AQ gets a head start with the damaged 
carbohydrates, the conditions could be shifted to those of a more conventional kraft cook.  
The AQ could then function to protect any new reducing end groups formed by the 
conditions of kraft cooking.   
More work with hydroxylamine could identify the proper pH and temperature to 
optimize prevention of peeling by selective reaction with the reducing end groups of the 
polysaccharides.  This process would probably require the successive steps of producing 
an oxime that degrades to a nitrile to stabilize the end group.  One possible test would be 
to add hydroxylamine to the prehydrolysis extraction for the first step.   
Other additives that showed promise in literature were cyanide (A. R. Procter 
1975) and ammonium hydrosulfide (Cho, Chiang and Sarkanen 1986).  Each of these 
additives would have handling concerns and would require modification to the chemical 
recovery systems.   
178 
 
Hydrogen gas was suggested as a possible supplement to polysulfide addition or 
as a reducing agent on its own under the proper conditions.   
8.2.3 Model Enhancement 
As discussed earlier, the model presented in this research is a simple and practical 
way of combining pH, temperature, and time effects into one parameter for comparing 
the effects of different levels of these variables on sugar release during extraction, chip 
weight loss, and pulp yield after kraft cooking of the prehydrolyzed chips.  However, the 
model has limitations:   
The model has only been tested within the range of variables used for its 
development:  pH from 3.0 to 4.5, temperature from 140? to 170?C, and up to two 
hours at the temperature target. 
The model does not account for degradation of sugars in the hydrolyzate. 
The model does not infer what changes have occurred within the chip due 
to the prehydrolysis. 
The model does not explicitly account for heat or mass transfer. 
The proven range of the current model can be expanded by additional tests similar 
to the nine in the design study.  Some of the tests that only reached low modified H-factor 
levels at their conclusion could be repeated for longer times.  Additional testing could be 
performed at lower temperature and pH targets to expand these ranges.  Neither higher 
temperature nor higher pH (while remaining in the acid range) appear to be fruitful areas 
of additional research at this time in a single stage extraction due to increased degradation 
at higher temperature and longer reaction time required at higher pH.   
179 
 
Removing the other limitations listed above will require a more complicated 
model and additional measurements.  Degradation of sugars can be accounted for by 
adding a second reaction rate equation to the model.  The second equation could be 
similar to the one used to develop the modified H-factor, but instead of depending on the 
concentration of hemicellulose it would depend on the concentration of sugar monomers 
in solution.  This would require a new dataset that included measurement of monomers in 
the hydrolyzate samples before the secondary hydrolysis that is performed as part of the 
analysis.  Measurement of degradation products like furfural, hydroxymethylfurfural, 
levulinic acid, and formic acid may also help with the development.   
Inference of chemical changes within the chip could be made with information 
about the extent of removal of monomers and oligomers from the native wood 
polysaccharides. The difficult part comes in determining the damage that is done to the 
polymers that remain in the chips. The prehydrolysis treatment releases sugars to the 
liquor that were measured as monomers after secondary hydrolysis.  The oligomers in 
solution can also be characterized by composition and molecular weight measurements.  
It may be possible to measure the composition of different size fractions.  The appearance 
of different oligomers and monomers would then be used to estimate rate parameters for 
the various hydrolysis reactions involved and estimate the effect on polymers that are not 
removed.  The information about the modified wood chip would then be fed to another 
module that would model the kraft pulping portion of the process.  The output of the 
pulping model would be a chemical description of the pulp including composition and 
degree of polymerization of the polysaccharides that remain.   
180 
 
An approach similar to this has been used to model the autohydrolysis of xylan 
hemicelluloses found in corncobs (Nabarlatz, Farriol and Montane 2004).  The corncob 
xylan model considered xylan hydrolysis to oligomers that entered solution and were then 
hydrolyzed to monomers that were subject to potential degradation reactions.  The kinetic 
rate parameters for all of the reactions were fit with an optimization package using 
solution composition data.  This effort was simpler than the one that would be required 
for pine chips since only arabinoxylan was involved.  The model for pine would include 
two types of galactoglucomannan, arabinoglucuronoxylan, arabinogalactan, and 
cellulose.   
Transport processes were an important consideration in several older models of 
pulping (Grace 1989, 62-68) and have been readdressed more recently (Simao, et al. 
2008), (Dang and Nguyen 2008).  Heat transport on a chip scale is likely not an important 
consideration in this process for the temperature and time ranges considered thus far.  
Without any calculations, consider an ice cube dropped into hot water.  A typical ice cube 
is roughly twice as thick as the largest screened chip and requires the latent heat of the 
phase change in addition to sensible heat to melt and yet does so quickly relative to the 
one hour given for heating the system to temperature in the experiments for this research.   
In contrast, mass transfer considerations must be included in the development of 
an accurate model of a heterogeneous reaction system like pulping. An example of why 
this is necessary is a recent study that showed that proper accounting for mass transfer of 
alkali into wet chips must include not only diffusion inside the chips but also 
consideration of the film between the free liquor and the chip (Simao, et al. 2008).  This 
study used measurements of alkali concentration in free liquor and the liquor entrapped in 
181 
 
the chip to develop a lumped parameter model.  Another experimental approach to 
separate mass transfer and kinetic effects was to cook very thin chips and standard chips 
under the same conditions with the assumption that mass transfer effects in the cooks of 
thin chips would be minimal (Agarwal and Gustafson 1997).  Mass transfer of reaction 
products out of the chips is not a concern in modeling pulping since the chips are subject 
to further physical treatment and washing after digestion, but it is a consideration in VPP 
since the product consists of only the sugars that are removed from the intact chips before 
pulping.  An example of this concern is the maximum molecular weight observed in 
prehydrolysis extract oligomers of 3100 (Casebier 1969).  Experimental methods of 
determining mass transfer effects might be to use small chips to have more ?pure? kinetic 
information or to isolate mass transfer effects by performing composition analysis on 
segments of chips that have undergone prehydrolysis.   
8.2.4 Improvements to Testing or Experimental Procedures  
Several areas were identified over the course of the experiments that should be 
investigated to improved precision or accuracy of the data that is produced: 
 Several different methods of cooking have been utilized in the lab at Auburn 
University to treat chips and produce pulp for this research and another similar project.  
Compare variability associated with multiple digester cooks to the use of bomb cooks for 
multiple cooks from the same set of chips or for a small set of chips through multiple 
stages of processing.   
Measure and reduce run-to-run variability on the HPLC.  Improve measurement 
of arabinose and mannose on the HPLC.  Dilution of the samples may help separate the 
peaks.   
182 
 
Lignin concentration was not measured in most of the liquid samples.  Methods 
for measuring both acid soluble lignin and acid insoluble lignin in these samples should 
be tested to help complete the material balance on the weight loss that is not accounted 
for in the sugar composition measurements.   
Improve chip weight loss measurement.  The problems here appear to be 
associated with moisture measurement.  A larger sample may be required to be 
representative of the cook.  Variability in pulp yield from bomb cooks may be in part 
caused by measurement of the wet chips.   
Concentrations of acetyl and uronic acids were not measured in this research.  The 
extraction rates of acetyl and glucuronic acids relative to other components of the 
hemicellulose could be instructive in understanding the behavior of the hemicelluloses.  
Removal of acetyl groups from glucomannan is said to stabilize this polymer to acid 
attack.  Glucuronic acid is considered to be resistant to acid hydrolysis (Sjostrom 1993, 
45) and the rate of xylan hydrolysis has been correlated to the ratio of uronic acid to 
xylan (A. H. Conner 1984).  This resistance may contribute to slow xylose removal in 
acid but fast further degradation in alkali.  The lab at Auburn University does test for 
acetic acid in liquid solutions, but this was not tested in this research since acetic acid 
was added to the solutions for pH control in most experiments.  This lab does not 
currently have a test for glucuronic acid.  Two methods were referred to in literature (A. 
H. Conner 1984).  
8.3 Final Comments 
This quest has not found exactly what it was looking for:  A simple set of 
industrially relevant process conditions for extracting VPP from Loblolly pine chips in a 
183 
 
modern kraft pulp mill without significantly reducing the yield or quality of the pulp.  
The quest has improved the scientific understanding of the impact of pH and temperature 
over time on the extraction of sugars from pine chips, the degradation of those sugars in 
solution, and the reduction in yield after kraft pulping.   This knowledge will aid future 
efforts toward developing an economically practical method for implementation of VPP.  
184 
 
 
 
Bibliography 
Acros Organics MSDS. "Hydroxylamine." Acros Organics. November 16, 2010. 
http://www.acros.be/portal/alias__Rainbow/lang__en/tabID__41/DesktopDefault.aspx 
(accessed November 19, 2010). 
Acros Organics MSDS. "Acetaldehyde." Acros Organics. November 10, 2010. 
http://www.acros.be/portal/alias__Rainbow/lang__en/tabID__41/DesktopDefault.aspx 
(accessed November 19, 2010). 
?. "Lithium Aluminium Hydride." Acros Organics. November 16, 2010. 
http://www.acros.be/portal/alias__Rainbow/lang__en/tabID__41/DesktopDefault.aspx 
(accessed November 19, 2010). 
Agarwal, Niraj, and Rick Gustafson. "A Contribution to the Modeling of Kraft 
Pulping." The Canadian Journal of Chemical Engineering 75 (February 1997): 8-15. 
Agarwal, Niraj, and Rick Gustafson. "Effect of carbohydrate degradation on zero-
span tensile strength." Tappi Journal 78, no. 1 (January 1995): 97-100. 
Agarwal, Niraj, William T. McKean, and Richard R. Gustafson. "Cellulose 
degradation kinetics in alkaline pulping." Appita 45, no. 3 (May 1992): 165-169. 
Agenda 2020 CTO Committee Working Group. "Adding biofuel/bioproduct 
capacity to existing U.S. mills Part 1: Options." Paper 360?, April 2008: 33-37. 
?. "Adding biofuel/bioproduct capacity to existing U.S. mills Part 2: The 
Business Case." Paper 360?, June/July 2008: 24-28. 
Ahmed, Y. Z., S. M. Saad, and H. A. Mohamed. "Some Aspects on the Effect of 
Nitrogen Containing Additives in Soda Pulping of Egyptian Bagasse Pith." 
Holzforschung 41, no. 1 (1987): 47-50. 
Al-Dajani, Waleed Wafa, and Ulrike W. Tschirner. "Pre-extraction of 
hemicelluloses and subsequent kraft pulping Part I: alkaline extraction." TAPPI Journal, 
June 2008: 3-8. 
Alfa Aesar MSDS. "9,10- Anthraquinone, 97%." February 20, 2007. 
Amidon, T. E., Bolton, T. S., Francis, R. C., and Gratien, K. " Effect of Hot Water 
Pre-Extraction on Alkaline Pulping of Hardwoods." TAPPI Environ. Pulping and 
Engineering Conference. Atlanta, 2006. Session 57. 
185 
 
Amidon, Thomas. "Extraction of Hemicellulose from Hardwood Chips with 23% 
Total Mass Reduction." Presentation at the TAPPI International Bioenergy and 
Bioproducts Conference. Portland, OR, August 21, 2008. 
"ANSI/ASABE S593 Terminology and Definitions for Biomass Production, 
Harvesting and Collection, Storage, Processing, Conversion and Utilization." In ASABE 
Standards, 1-3. St. Joseph, MI: ASABE, 2006. 
Aurell, Ronnie, and Nils Hartler. "Sulfate cooking the the Addition of Reducing 
Agents Part III: The Effect of Added Sodium Borohydride." Tappi 46, no. 4 (1963): 209-
215. 
Barry, Daniel A., and Richard J. Leblanc. "Entrained flow gasification 
biorefinery." Paper 360?, August 2009: 34-35. 
Bernardin, Leo J. "The Nature of the Polysaccharide Hydrolysis in Black 
Gumwood Treated with Water at 160?C." Tappi 41, no. 9 (September 1958): 491-499. 
Beurich, Harald, and Margarete Scholl. Process and Composition for Bleaching 
Wood Pulp. United Staes of America Patent 4,752,354. June 21, 1988. 
Biodiesel Magazine. Plant List. April 6, 2009. 
http://www.biodieselmagazine.com/plant-list.jsp (accessed April 14, 2009). 
Birkinshaw, J. H., P Chaplen, and W.P.K. Findlay. "Biochemistry of the Wood-
rotting Fungi." Biochem. J. 66 (1957): 188-192. 
Bolton, T. S., Francis, R. C., and Amidon, T. E. " A Biorefinery Aproach: Non-
Sulfur Pulping with Partial Lignin Recovery and Conversion." Environmental. Pulping, 
and Engineering Conference. Jacksonville,FL: TAPPI , 2007. Session 35. 
Borchert, Peter J. Elimination of Reducing Sugars Present in Trisodium Citrate. 
United States of America Patent 4,308,398. December 29, 1981. 
Brasch, D. J. and Free, K. W. "The Chemistry of new Zealand Grown Pinus 
radiate III. Prehydrolysis-Sulfate Pulps." Tappi 47, no. 4 (April 1964): 186-189. 
Brasch, D. J., and K. W. Free. "Prehydrolysis-Kraft Pulping of Pinus radiata 
Grown in New Zealand." Tappi 48, no. 4 (April 1965): 245-248. 
Breslin, Michael D., and David R. Cosper. Use of Additives in Pulp Bleaching 
Processes to Preserve Pulp Strength. USA Patent 4298428. November 3, 1981. 
Brochu, Ron. "Northern Wisconsin News." October 12, 2010. 
http://www.wisconsin-north.com/2010/10/12/park-falls-biofuel-project-hit-a-
bureaucratic-snag (accessed November 9, 2010). 
California Office of Environmental Health Hazard Assessment. "Proposition 65." 
California Government. November 12, 2010. 
http://oehha.ca.gov/prop65/prop65_list/Newlist.html (accessed November 19, 2010). 
186 
 
Casebier, R.L., Hamilton, J.K., and Hergert, H.L. "Chemistry and Mechanism of 
Water Prehydrolysis on Southern Pine Wood." Tappi 52, no. 12 (December 1969): 2369-
2377. 
Chaudhari, Sachin S., and Krishnacharya G. Akamanchi. "Thionyl Chloride-
Benzotriazole: An Efficient System for Transformation of Aldoximes to Nitriles." 
Synthetic Communications 29, no. 10 (1999): 1741-1745. 
Cho, H. J., V.L. Chiang, and K. V. Sarkanen. "Stabilization of W. hemlock 
hemicelluloses with ammonium hydrosulfide for kraft pulping." Paperi ja Puu - Papper 
och Tra, 1986: 94-96,98. 
Chum, Helena L., Johnson, David K. Black, Stuart K., and Ralph P. Overend Vol. 
"Pretreatment ? Catalyst Effects and the Combined Severity Parameter." APPLIED 
BIOCHEMISTRY AND BIOTECHNOLOGY (The Humana Press) 24/25 (1990): 1-14. 
Clayton, D. W., and H. G. Jones. "The Effect of Hydroxylamine on the Stability 
of Polysaccharides in Hot Alkali." Svensk Papperstidning 73, no. 23 (December 1970): 
753-760. 
Clayton, D.W., and L. M. Marraccini. "The Effect of Additives on the Stability of 
Polysaccharides in Hot Alkali." Svensk Papperstidning, 1966: 311-321. 
Conner, Anthony H. "Kinetic Modeling of Hardwod Prehydrolysis. Part I. Xylan 
Removal by Water Prehydrolysis." Wood and Fiber Science 16, no. 2 (April 1984): 268-
277. 
Conner, Anthony H., and Linda F. Lorenz. "Kinetic Modeling of Hardwod 
Prehydrolysis. Part III. Water and Dilute Acetic Acid Prehydrolysis of Southern Red 
Oak." Wood and Fiber Science 18, no. 2 (April 1986): 248-263. 
Conner, Anthony H., Kimball Libkie, and Edward L. Springer. "Kinetic Modeling 
of Hardwod Prehydrolysis. Part II. Xylan Removal by Dilute Hydrochloric Acid 
Prehydrolysis." Wood and Fiber Science 17, no. 4 (October 1985): 540-548. 
Courchene, Charles E. "The tried, the true, and the new - getting more pulp from 
chips - modification to the kraft process for increased yield." Breaking the Pulp Yield 
Barrier Symposium. Atlanta: Tappi, 1998. 11-20. 
Dang, Vinh Q., and Kien L. Nguyen. "A universal kinetic model for 
characterisation of the effect of chip thickness on kraft pulping." Bioresource Technology 
99 (2008): 1486-1490. 
Dence, Carlton W., and Douglas W. Reeve. Pulp Bleaching- Principles and 
Practice. Atlanta: TAPPI Press, 1996. 
Department of Energy. Annual Energy Review 2007. June 23, 2008. 
http://www.eia.doe.gov/emeu/aer/pdf/pages/sec1_3.pdf (accessed April 16, 2009). 
187 
 
EIA U.S. Petroleum Product Supplied. July 29, 2010. 
http://tonto.eia.doe.gov/dnav/pet/pet_cons_psup_dc_nus_mbblpd_a.htm (accessed April 
18, 2011). 
Energy Information Administration. Energy in Brief. August 21, 2008. 
http://tonto.eia.doe.gov/energy_in_brief/renewable_energy.cfm (accessed October 24, 
2008). 
Fargione, Joseph, Jason Hill, David Tilman, Stephen Polasky, and Peter 
Hawthorne. "Land Clearing and the Biofuel Carbon Debt." Sciencexpress, February 7, 
2008: 1-4. 
Francey, Sabrina, Honghi Tran, and Andrew Jones. "Current status of alternative 
fuel use in lime kilns." TAPPI JOURNAL, October 2009: 33-39. 
Francis, Raymond, interview by Allen Smith. Discussion following Dr. Yoon's 
presentation at the 2006 TAPPI EPE conference (November 11, 2006). 
Gillespie, Athey G. Pulping Cellulose Material with an Alkaline Cooking Liquor 
Containing Hydrazine. USA Patent 3161562. December 15, 1964. 
Goyal, G. C. Anthraquinone Pulping: A TAPPI Press Anthonlogy of Published 
Papers, 1977 - 1996. Atlanta: TAPPI Press, 1997. 
Grace, T. M. and Malcolm, E. W. ed. PULP AND PAPER MANUFACTURE 
VOLUME 5, ALKALINE PULPING. Atlanta: TAPPI Press, 1989. 
Graham, Robin L., Erik Lichtenberg, Vernon O Roningen, Hossein Shapouri, and 
Marie E Walsh. "The Economics of Biomass Production in the United States." Bioenergy 
Feedstock Information Network. 1995. 
http://bioenergy.ornl.gov/papers/bioam95/graham3.html (accessed April 19, 2010). 
Grunwald, Michael. "The Clean Energy Scam." TIME, April 7, 2008: 40-45. 
Gullichsen, Johan, and Carl-Johan Fogelholm. Chemical Pulping . Helsinki, 
Finland: Fapet Oy, 1999. 
Gustafson, Richard R., Charles A. Sleicher, William T. McKean, and Bruce A. 
Finlayson. "Theoretical Model of the Kraft Pulping Process." Ind. Eng. Chem. Process 
Des. Dev. 22, no. 1 (1983): 87-96. 
Hajipour, A. R., I Mohammadpoor-Baltork, K Nigbaghat, and G Imanzadeh. 
"Solid-Phase Synthesis of Oximes." Synthetic Communications 29, no. 10 (1999): 1697-
1701. 
Hatton, J. V. "Development of Yield Prediction Equations in Kraft Pulping." 
TAPPI 56, no. 7 (1973): 97-100. 
Hendrix, Charlie. Experiments That Get Results. New York: America Institute of 
Chemical Engineers, 2002. 
188 
 
Humar, M., M. Petric, F. Pohleven, and R. Despot. "Upgrading of spruce wood 
with ethanolamine treatment." Holz als Roh- und Werkstoff, 2003: 29-34. 
Ingruber, O. V., M. J. Kocurek, and A. Wong. Pulp and Paper Manufacture 
Voume 4: Sulfite Science and Technology. Atlanta: The Joint Textbook committee of the 
Paper Industry, 1985. 
Jencks, William P. "Studies on the Mechanism of Oxime and Semicarbazone 
Formation." Journal of the American Chemical Society 81 (January 1959): 475-481. 
Jiang, Zhi-Hua, Barbara van Lierop, and Richard McK. Berry. Chlorine Dioxide 
Bleaching in the Presence of an Aldehyde. USA Patent 6235154 B1. May 22, 2001. 
Jimenez, Graciela, William T. McKean, and Richard R. Gustafson. "Using a kraft 
pulping model to improve pulp uniformity." Tappi Journal 73, no. 7 (July 1990): 173-
176. 
Jimenez, L., A. Rodriquez, A. M. Calero, and M. E. Eugenio. "Use of 
Ethanolamine-Soda-Water Mixtures for Pulping Olive Wood Trimmings." Chemical 
Engineering Research and Design 82, no. A8 (August 2004): 1037-1042. 
Laubach, George D. "1997 TAPPI Survey of Pulping Additives- AQ and Chip 
Penetrants." 1998 Breaking the Pulp Yield Barrier Symposium. Atlanta: TAPPI Press, 
1998. 103-111. 
Li, Ji-Tai, Xiao-Liang Li, and Tong-Shuang Li. "Synthesis of oximes under 
ultrasound irradiation." Ultrasonics Sonochemistry 13 (2006): 200-202. 
Lindenfors, Sven. "Additives in alkaline pulping - what reduces what?" Svensk 
Papperstidning, 1980: 165-173. 
MacRae, Sharmane, and Paul P. Cotterill. "Genetically Engineering High Yield, 
High Quality Eucalyptus Wood Fiber." Breaking the Pulp Yield Barrier Symposium. 
Atlanta: TAPPI Press, 1998. 45-58. 
Maloney, Mark T., Thomas W. Chapman, and Andrew J. Baker. "Dilute Acid 
Hydrolysis of Paper Birch: Kinetics Studies of Xylan and Acetyl-Group Hydrolysis." 
Biotechnology and Bioengineering 27 (March 1985): 355-361. 
Martinez, C., Webster, V., and Mugg, B. Previously Unpublished undergraduate 
research Data, Auburn University, 2006. 
Matussek, Heide. Lockwood-Post Directory of Pulp and Paper Mills, Global 
Edition. 2006. 
McDonough, Thomas J. "Kraft Pulp Yield Basics." Breaking the Pulp Yield 
Barrier Symposium. Atlanta: Tappi, 1998. 1-9. 
189 
 
Meudt, Andreas, Stefan Scherer, and Claudius Boehm. Method for Producing 
Nitriles by Elimination of Water from Aldehyde Oximes with Alkylphosphonic 
Anhydrides. USA Patent 7405318 B2. July 29, 2008. 
Mitchell, R L, M A Millett, W E Moore, and J F Saeman. "The Persistence of 
Wood Xylan and Mannan Through Successive Prehydrolysis, Pulping, Purification, and 
Nitration." Tappi 39, no. 8 (August 1956): 571-5785. 
Molin, Ulrika, and Ants Teder. "Importance of cellulose/hemicellulose-ratio for 
pulp strength." NORDIC PULP AND PAPER RESEARCH JOURNAL 17, no. 1 (2002): 
14-19,28. 
Morrison, Robert Thornton, and Robert Neilson Boyd. Organic Chemistry, 
Fourth Ed. Boston: Allyn and Bacon, Inc., 1983. 
Nabarlatz, Debora, Xavier Farriol, and Daniel Montane. "Kinetic Modeling of the 
Autohydrolysis of Lignocellulosic Biomass for the Production of Hemicellulose-Derived 
Oligosaccharides." Ind. Eng. Chem. Res. 43 (June 2004): 4124-4131. 
Occupational Safety and Health Administration. Ethanolamine. 
http://www.osha.gov/SLTC/healthguidelines/ethanolamine/recognition.html (accessed 
November 12, 2010). 
Ostle, Bernard. Statistics in Research. Second Edition. Ames, Iowa: The Iowa 
State University Press, 1963. 
Paterson, Richard M.L. Increasing Kraft Pulp Yield with Hydroxylamine. United 
States Patent 3,401,076. September 10, 1968. 
Patrick, Ken. "Becoming fossil fuel free." Paper 360?, September 2009: 32-33. 
Perlack, Robert D, Lynn L Wright, Anthony F Turhollow, Robin L Graham, 
Bryce J Stokes, and Donald C Erbach. Biomass as Feedstock for a Bioenergy and 
bioproducts Industry: The Technical Feasibility of a Billion-Ton Annual Supply. 
Department of Energy, Department of Agriculture, Oak Ridge, Tennessee: Oak Ridge 
National Laboratory, 2005. 
Portnoy, Norman A., Alden H. Reine, and Jr, Jett C. Arthur. "Synthesis of 
nitrogenous derivatives of cellulose-poly(methyl vinyl ketone) copolymers. Oximes, 
semicarbazones, and hydantoins." Carbohydrate Research 24 (1972): 198-201. 
Procter, A. R., and Heidi M. Apelt. "Reactions of Wood Components with 
Hydrogen Sulfide III. The Efficiency of Hydrogen Sulfide Pretreatment Compared to 
Other Methods for Stabilizing Cellulose to Alkaline Degradation." Tappi 52, no. 8 
(August 1969): 1518 - 1522. 
Procter, Alan R. Stabilization of Cellulosic Material by Treatment with 
Ferricyanide Ion. United States of America Patent 3929559. December 30, 1975. 
190 
 
Procter, Alan Robert, and Wayne Ming Chow. Alkaline pulping of lignocellulosic 
material with amine pretreatment. USA Patent 4091749. May 30, 1978. 
Pu, Qiusheng, William McKean, and Richard Gustafson. "Kinetic model of 
softwood kraft pulping and simulation of RDH process." Appita 44, no. 6 (1991): 399-
404. 
Renewable Fuels Association. Biorefinery Locations. April 11, 2011. 
http://www.ethanolrfa.org/industry/statistics/ (accessed April 18, 2011). 
?. Renewable Fuels Standard. January 2008. 
http://www.ethanolrfa.org/resource/standard/ (accessed April 14, 2009). 
Retsina, Theodora, and Vesa Pylkkanen. "Back to the Biorefinery." Paper 360?, 
February 2007: 18-19. 
Richter, George A. "Some Aspects of Prehydrolysis Pulping." TAPPI 39, no. 4 
(April 1956): 193-210. 
Rojas-Escudero, Elba, Ana Luisa Alarcon-Jimenes, Patricia Elizalde-galvan, and 
Francisco Rojo-Callejas. "Optimization of carbohydrate silylation for gas 
chromatography." Journal of Chromatography A, 2004: 117-120. 
Rudie, Alan, Richard Reiner, Nancy Ross-Sutherland, and William Kenealy. 
"Acid Prehydrolysis of Wood." Engineering, Pulping & Environmental Conference. 
Jacksonville, FL: Tappi, 2007. 
Rydholm, Sven A. PULPING PROCESSES. Malabar, FL: Robert E. Krieger 
Publishing Company, 1985. 
Saucedo, Victor M., and Gopal A. Krishnagopalan. "Kinetics and State Space 
Modeling of Modified Kraft Pulping." AIChE Syposium Series No. 322, Vol. 95. 
American Institute of Chemical Engineers, 1999. 7-13. 
Searchinger, Timothy, et.al. "Use of U.S. Croplands for Biofuels Increases 
Greenhouse Gases Through Emissions from Land Use Change." Sciencexpress, February 
7, 2008: 1-6. 
Simao, Joao P. F., Ana P.V. Egas, M. Graca Carvalho, Cristina M.S.G. Baptista, 
and Jose Almiro A.M. Castro. "Heterogeneous studies in pulping of wood: Modelling 
mass transfer of alkali." Chemical Engineering Journal 139 (2008): 615-621. 
Sjostrom, Eero. WOOD CHEMISTRY. New york: Academic Press , 1993. 
Smith, Allen J., Harry T. Cullinan, and Gopal A. Krishnagopalan. "Effect of Hot 
Water Extraction Conditions on the Subsequent Kraft Pulping of Pine Chips." TAPPI 
Engineering, Pulping and Environmental Conference. Portland, OR: TAPPI Press, 2008. 
33.2. 
191 
 
Smook, Gary A. HANDBOOK FOR PULP & PAPER TECHNOLOGISTS . 
Montreal: Canadian Pulp and Paper Association, 1982. 
Springer, Edward L. "Hydrolysis of Aspenwood Xylan with Aqueous Solutions of 
Hydrochloric Acid." Tappi 49, no. 3 (March 1966): 102-106. 
Stenius, Per. Forest Products Chemistry. Edited by Johan Gullichsen and Hannu 
Paulapuro. Vol. 3. 19 vols. Helsinki: Fapet Oy, 2000. 
Telvekar, Vikas N., and Krishnacharya G. Akamanchi. "A Simple System for 
Preparation of Protected Glycosidic Carbohydrate Nitriles from Corresponding Oximes." 
Synthetic Communications 34, no. 13 (2004): 2331-2336. 
Thorp, B A, Benjamin A IV Thorp, and Diane Murdock Thorp. "A Compelling 
Case for Integrated Biorefineries (Part II)." Paper 360?, April 2008: 20-22. 
Thorp, B A, Benjamin A IV Thorp, and Diane Murdock Thorp. "A Compelling 
Case for Integrated Biorefineries (Part IV)." Paper 360?, June/July 2008: 16-17. 
Thorp, B. A., Benjamin A. Thorp, and L. Diane Murdock-Thorp. "A Compelling 
Case for Integrated Biorefineries (Part III)." Paper 360?, May 2008: 12-14. 
Thorp, B.A., Benjamin A Thorp, and L. Diane Murdock-Thorp. "A Compelling 
Case for Integrated Biorefineries (Part I)." Paper 360?, March 2008: 14-15. 
van Donkelaar, Arie. "Kraft SWD Yield: Break the Barriers." Breaking the Pulp 
Yield Barrier Symposium. Atlanta: Tappi, 1998. 22. 
Wang, Zhong, and Johan Gullichsen. "New Chipping technique for Kraft 
Pulping." Breaking the Pulp Yield Barrier Symposium. Atlanta: TAPPI Press, 1998. 37-
44. 
Weil, Joseph, Mark Brewer, Richard Hendrickson, Ayda Sarikaya, and Michael 
R. Ladischl. "Continuous pH Monitoring During Pretreatment of Yellow Poplar Wood 
Sawdust by Pressure Cooking in Water." APPLIED BIOCHEMISTRY AND 
BIOTECHNOLOGY 70-72 (1998): 99-111. 
Whitley, Brittany. "Truck begins cross-country trip fueled mainly by wood." 
Opelika-Auburn News, September 27, 2008: 3A. 
Williams, Andrew, and Myron L. Bender. "Studies on the Mechanism of Oxime 
and Ketimine Formation." Journal of the American Chemical Society 88, no. 11 (June 
1962): 2508-2513. 
Yoon, Sung-Hoon, interview by Allen J. Smith. Pulp yield recovery with additive 
treatment following prehydrolysis (June 5, 2009). 
Yoon, Sung-Hoon, and Adriaan van Heiningen. "Kraft Pulping and Papermaking 
Properties of Hot-Water Pre-Extracted Loblolly Pine for Integrated Forest Products Bio-
refinery." Tappi Journal, July 2008: 22-27. 
192 
 
Yoon, Sung-Hoon, Kimberly MacEwan, and Adriaan van Heiningen. "Pre-
Extraction of Southern Pine Chips with Hot Water followed by Kraft Pulping." TAPPI 
Engineering, Pulping and Environmental Conference Proceedings. Atlanta, GA, 2006. 
Zerbe, John I. "Alcohol from wood as an alternative transportation fuel." 
Alternative energy in the midwest . Schaumberg, IL: Illinois Department of Energy and 
Natural Resources, 1985. 3-3 - 3-7. 
 
  
193 
 
 
 
Appendix 1  Minitab 15 Statistical Results  
General Linear Model: Max Sugar, Final, ... versus pH, Temp, ...  
Factor   Type    Levels   Values 
pH       fixed        3   0, 1, 2 
Temp    fixed        3   0, 1, 2 
Soak     fixed        3   0, 1, 2 
 
 
Analysis of Variance for Max Sugar, using Adjusted SS for Tests 
Source   DF    Seq SS   Adj SS   Adj MS     F     P 
pH        2    31.673   31.673   15.837    83.00   0.012 
Temp    2    77.196   77.196   38.598   202.28   0.005 
Soak      2     2.653    2.653    1.326     6.95   0.126 
Error     2     0.382    0.382    0.191 
Total     8   111.903 
S = 0.436819   R-Sq = 99.66%   R-Sq(adj) = 98.64% 
 
Analysis of Variance for Final, using Adjusted SS for Tests 
Source   DF    Seq SS   Adj SS   Adj MS       F       P 
pH        2    28.251   28.251   14.125   29.55   0.033 
Temp      2    70.441   70.441   35.220   73.68   0.013 
Soak      2     3.940    3.940    1.970    4.12   0.195 
Error     2     0.956    0.956    0.478 
Total     8   103.587 
S = 0.691408   R-Sq = 99.08%   R-Sq(adj) = 96.31% 
 
Analysis of Variance for 45min, using Adjusted SS for Tests 
Source   DF    Seq SS    Adj SS Adj MS        F       P 
pH       2    15.111   15.111    7.556    200.24  0.005 
Temp      2   115.151   115.151   57.576   1525.86   0.001 
Soak      2     1.346     1.346   0.673     17.83   0.053 
Error     2     0.075     0.075    0.038 
Total     8   131.684 
S = 0.194251   R-Sq = 99.94%   R-Sq(adj) = 99.77% 
 
 
 
 
 
194 
 
Analysis of Variance for Wt loss%, using Adjusted SS for Tests 
Source   DF    Seq SS    Adj SS    Adj MS       F       P 
pH        2    96.816    96.816    48.408   10.98   0.083 
Temp      2   339.616   339.616   169.808   38.52   0.025 
Soak      2     4.616     4.616     2.308    0.52   0.656 
Error     2     8.816    8.816     4.408 
Total     8   449.862 
S = 2.09947   R-Sq = 98.04%   R-Sq(adj) = 92.16% 
 
Analysis of Variance for 30k Yield, using Adjusted SS for Tests 
Source   DF  Seq SS Adj SS  Adj MS F P 
pH        2    31.307   31.307   15.653    70.32   0.014 
Temp      2    75.977   75.977   37.989   170.66   0.006 
Soak      2     0.577    0.577    0.288     1.29   0.436 
Error     2     0.445    0.445    0.223 
Total     8   108.305 
S = 0.471809   R-Sq = 99.59%   R-Sq(adj) = 98.36% 
 
Analysis of Variance for Yield Loss, using Adjusted SS for Tests 
Source   DF    Seq SS   Adj SS   Adj MS        F       P 
pH        2    31.307   31.307   15.653    70.32   0.014 
Temp      2    75.977   75.977   37.989   170.66   0.006 
Soak      2     0.577    0.577    0.288     1.29   0.436 
Error     2     0.445    0.445    0.223 
Total     8   108.305 
S = 0.471809   R-Sq = 99.59%   R-Sq(adj) = 98.36% 
 
Analysis of Variance for %Rec/%Yloss, using Adjusted SS for Tests 
Source   DF     Seq SS     Adj SS     Adj MS        F       P 
pH        2   0.013902   0.013902   0.006951   48.84   0.020 
Temp      2   0.175915   0.175915   0.087957  617.97   0.002 
Soak      2   0.049565   0.049565   0.024782  174.11   0.006 
Error     2   0.000285   0.000285   0.000142 
Total     8   0.239666 
S = 0.0119304   R-Sq = 99.88%   R-Sq(adj) = 99.52% 
 
Analysis of Variance for H factor to 30, using Adjusted SS for Tests 
Source   DF    Seq SS    Adj SS   Adj MS      F       P 
pH        2    598822    598822   299411   2.47   0.288 
Temp      2   1633089   1633089   816544   6.74   0.129 
Soak      2    299089    299089   149544   1.23   0.448 
Error     2    242422    242422   121211 
Total     8   2773422 
S = 348.154   R-Sq = 91.26%   R-Sq(adj) = 65.04% 
  
195 
 
 
 
Appendix 2  Non-Standard Test Methods 
A2.1 ABC Liquor test 
A Test: 
1. pipette 5 ml clear liquor to 250 ml erlenmeyer 
2. Add 50 ml water 
3. Add 25 ml 10% BaCl2 
4. Add phenolphthalein (thymol phthalein) 
5. Titrate with 0.5 N HCl until pink(blue) disappears 
6. Record A, do not refill buret 
 
B Test: 
1. Add 5 ml 40% formaldehyde 
2. wait 30 seconds 
3. Titrate with 0.5 N HCl until pink(blue) disappears 
4. Record B, do not refill buret 
C Test: 
1. Add methyl orange 
2. Titrate with 0.5 N HCl until red first appears 
3. Record C 
Calculations: 
Na2S = 6.4*0.5/0.5167*(B-A) 
NaOH =3.2*0.5/0.5167*(2*A-B) 
Na2CO3 =3.2*0.5/0.5167*(C-B) 
Sulfidity% = Na2S/( Na2S + NaOH) 
TTA = Na2S + NaOH + Na2CO3 
AA =  Na2S + NaOH 
EA = NaOH + Na2S/2 
196 
 
A2.2 Black Liquor Residual Alkali Test 
1. Pipette 10ml Black Liquor into 90ml DI water 
2. Titrate with 0.5 N HCl to pH 11.3 
3. Record Volume, V 
EA Residual, g/l = N*V*31/10 
EA Residual, g  = (N*V*31/10)*Volume of liquor charged,l 
EA Residual, %on wood = EA Residual, g/ Chip Mass, g 
REA 2, g/l =31*((0.95*V/10)-0.012)*N 
REA 2 %on wood  = REA 2, g/l * Volume of liquor charged ,l / Chip Mass, g 
 
 
  
197 
 
 
 
Appendix 3  Cook Worksheets and Calculations 
The cook worksheets were used to record data during the cook and also to make 
calculations with this data.  A summary listing of all the worksheets and four examples of 
cook worksheets are included in this document.  The worksheets are each different 
depending on what was actually done in the cook.  The simplest worksheets like ?A4? 
contain only limited information about the time/temperature profiles after the cook plan.  
Later worksheets like ?A9? have more detailed temperature and pH listings with 
calculations of H-factor.  The red numbers on the cook plan pages are calculated values.   
The worksheets were modified somewhat to improve the view for printing.  Each 
worksheet starts with the prehydrolysis followed by the kraft cook data.  The listings of 
pH and temperature then follow with recalculations of this data to improve the H-factor 
and modified H-factor integrations as illustrated in ?A19?.  Kraft cooks like ?A27? do not 
have prehydrolysis data.  When the kraft cook was performed in bombs, each was 
detailed separately.  When chips from separate prehydrolysis runs were cooked together 
in bombs, the bomb cook data was only listed in one worksheet and the other worksheets 
refer to this location.   
Copies of all the cook worksheets are in the pulp and paper lab (presently 
Wilmore 156) and are retained by the author. 
      Worksheet Summary
Ref Date Cook Objective Notes
10/03/07 1003 Prehydrolysis test Temp control high overshoot
A1 10/10/07 1010 2 hr Prehydrolysis test good T control
A2 10/17/07 1017 30 minutes preH at 170
A3 10/23/07 1023 kraft cook control profile manual errors
A4 10/24/07 1024 30 minutes preH at 170 replicate
A5 11/05/07 1105 kraft cook
A6 11/07/07 1107 2 hr Prehydrolysis cook
A7 12/15/07 1215 2 hr Prehydrolysis cook target 30 kappa
A8 02/28/08 22808 preH cook to test pH mods
A9 03/06/08 0.1%NaOH preH pH 5 0.2N NaOH pH control
A10 03/25/08 0.3%NaOH test initial NaOH charge 0.5N NaOH pH control
A11 03/27/08 3% Acetic initial Acetic charge 0.5N NaOH pH control
04/04/08 BombCook1 target 30 kappa
04/08/08 BC2     Chips from 3/06,25,27/08
04/11/08 BC3  
05/08/08 BC4
A12 05/12/08 51208 Water preH to 1400H factor leaking at pH probe
A13 05/16/08 51608 0.1% NaOH for 1 hour at pH5
A14 05/17/08 51708 3% acetic for 30minutes started NaOH after 15 minutes 
05/19/08 BC1a target 30 kappa
05/22/08 BC2a    Chips from 5/12,16,17/08
06/28/08 BC3a
07/12/08 BC4a
A15 05/23/09 BK1 Bomb Kraft Cook 1 uncooked in each, Bomb 4 leak
A16 05/25/09 PreH A pH 3.0, 170, pH 7 presoak ramp long, flow problems
A17 05/26/09 PreH A2 ramp long, cook time short
A18 05/27/09 PreH B 4.5, 155, 7 ramp long, cook time short
A19 05/28/09 PreH C 4.5, 170, 3.5 ramp extended to one hour
A20 05/29/09 BK2
A21 06/01/09 PreH D 3.5, 140, 7
A22 06/02/09 PreH E 3.5, 155, 3.5
A23 06/03/09 PreH F 3.5, 170, 10.5
A24 06/08/09 PreH G 3.0, 155, 10.6
A25 06/09/09 PreH H 4.5, 140, 10.6
A26 06/10/09 PreH I 3.0, 140, 3.5
A27 06/11/09 BK3
A28 06/19/09 BK4
A29 06/20/09 BK5
06/24/09 BC D1F1G1 Stand conds for Kappa30 Random Order
06/25/09 BC A1B1C1
BC E1H1I1
06/26/09 BC CFG2 Stand cond w/ AQ Random Order
06/29/09 BC BDH2
BC AEI2
A30 07/01/09 BK6
07/02/09 BC BD3
07/03/09 BC EI3
07/04/09 BC CGH3
07/05/09 BC AF3
A31 07/06/09 BK7 Low kappa cook
A32 07/07/09 BK8 "standard" cook test
07/18/09 BC CEG4
07/19/09 BC A4A5
           198
      Worksheet Summary
07/20/09 BC DF4
BC BE5
07/21/09 BC HI4
BC CDG5
07/23/09 BC HI5
BC B4F5
A33 08/05/09 PreH J lower pH presoak
A34 08/06/09 PreH K
A35 08/07/09 PreH L Step down T from 170 to 155
08/11/09 BC JKL1
08/12/09 BC JKL2
08/20/09 BC JKL3
08/21/09 BC JKL5
09/10/09 BC JKL4
A36 09/22/09 PreH M 3.5, 155 no soak, 1hr at T
A37 09/23/09 PreH N 3.0, 155, no S,  45min at T
A38 09/24/09 PreH O 3.5, 170, 20min at T
09/25/09 BC MNO1 Stand conds for Kappa30
BC MNO2 Higher Kappa
09/28/09 BC MNO3
10/06/09 BC MNO4
A39 10/12/09 PreH P
10/15/09 BC P123 Ethanolamine
10/16/09 BC P456
10/23/09 BC P789
A40 10/26/09 PreH Q
A41 10/30/09 PreH R
11/04/09 BC Q123 Hydroxylamine
11/06/09 BC Q456
11/11/09 BC Q789
11/12/09 BC R123 LiAlH4
A42 11/13/09 PreH S
11/14/09 BC S123 Acetaldehyde
11/18/09 BC Q10 R45
11/20/09 BC R6 Q1112
11/23/09 BC P10 S45
11/24/09 BC MNO5
11/25/09 BC S101112 Standard Cook
11/29/09 BC S789
           199
      A4
Prehydrolysis Cook Plan 10/24/07
Time Activity Data Calc 250 g BD Chips to Cook
PreH Clean Digester, drain, close valve 302.52 g AD Chips 91.82 %Chip Moisture
Measure mass of chips, put in basket, g 302.67 24.76 10 %preH Chip Sample
measure volume of water added to just cover chips, ml 1670 277.912 BD Chips
Add volume for samples 150 1844.76 total water
831 Start Pump, start Profile 22 6.64 preH wtr/wood
Close lid, leave vent cracked open Profile 22 Prehydrolysis
Close vent at 100?C target Asurd Soak Off
856 Open vent briefly at 100?C measurement Controls Active
Open vent briefly at 105?C measurement Time Seg 0:00:00 300
Sample liquid at appropriate times 150 Time Seg 0:44:20 1680
Manually help controller at end of ramp Controls Off
Turnaround Delay 0:00:10
946 Vent digester at end of prehydrolysis Controls Active
open drain, blow through ice water, Collect hydrolyzate Time Seg 0:00:30 1700
turn pump off Time Seg 0:30:00 1700
Purge coils with short pulse of air pressure Controls Off
Measure volume of hydrolyzate 1310 End of Profile
Verify all pressure released 384.76 water in chips
Open lid Profile 1 Wash 
Close drain valve Controls Active
Add measured volume of wash water to digester 1560 Time Seg 0:00:00 770
turn pump on Time Seg 0:05:00 1050
close lid, close vent Time Seg 0:05:00 1050
1001 run wash Profile 1 Controls Off
open drain valve to blow through ice water End of Profile
Measure volume of wash 1320 624.76 water in chips
Verify all pressure released
Remove chip basket White Liquor
Dump chips into plastic beaker 10/02/07 test date
Weigh total chips 685.5 902.67 est WL Sample
Remove chip sample by weight 68.55 2 Volume,l
Return remaining chips to basket 616.95 ml 0.5N HCl
Return basket to digester 20 %AA 26.30 A
Close drain valve 50 AA g 29.7 B
Add measured volume of white liquor to digester 544 544 liq Vol ml 30.5 C
measure volume of water added to just cover chips 1025 366.95 water in chips
Start pump 1935.95 total water gpl Na2O
7.74 Cook liq/wood 21.1 Na2S
Cook Start Cook profile 23 70.9 NaOH
1026 Close lid 2.5 Na2CO3
Vent digester at 100? 22.90% Sulf%
Vent digester at 105? 94.4 TTA
1115 Manually adjust controller at end of ramp 171 92.0 AA
1348 Vent digester at end of cook 81.4 EA
open drain, blow through ice water, Collect black liquor
turn pump off
Purge coils with short pulse of air pressure
Measure volume of black liquor 1510 Profile 23 Kraft Cook
Verify all pressure released Controls Active
Open lid Time Seg 0:00:00 600
add two liters of water Time Seg 0:46:20 1680
circulate Controls Off
drain Delay 0:00:10
Remove chip basket 100.05 AD Pulp Controls Active
30 Dump chips into wash bag, wash well 93.64 %Pulp Dry Time Seg 0:00:30 1700
Defibrate pulp in blender 93.69 BD Pulp Time Seg 2:30:00 1700
45 Squeeze dry 37.46 %Yield Controls Off
homogenize in mixer, spread to dry 28.5 Kappa End of Profile
Time Minutes Dig Temp2 kr avg H Factor
1026 0 60 0.01         -            -             
1042 16 100 1.02         0.51          0.1             
Max T reduced 2? for temp offset 1113 47 168 792.35     396.69      205.1         
1348 197 168 792.35     792.35      2,186.0      
1353 202 168 792.35     792.35      2,252.0      
1358 207 59.9 0.01         396.18      2,285.0      
1363 212 59.9 0.01         0.01          2,285.0      
           200
      A9
Prehydrolysis with bomb Cook Plan 3/6/08 "0.1% NaOH"
Time Activity Data Calc 300 g BD Chips to Cook
PreH Clean Digester, drain, close valve 363.03 g AD targ 91.82 %Chip dry matter
Measure mass of chips, put in basket, g 363.05 29.70 10 %preH Chip Sample
measure volume of water added to just cover chips, ml 2300 333.353 BD Chips
Add volume for samples(50ml 0.2N NaOH) 150 3362.70 total water 0.1% NaOH
1001 Start Pump, start Profile 22 10.09 preH wtr/wood
Close lid, leave vent cracked open Profile 22 Prehydrolysis
1023 Close vent at 100?C target Asurd SoakOff
1025 Open vent briefly at 100?C measurement Controls Active
1029 Open vent briefly at 105?C measurement Time Seg 0:00:00 300
Sample liquid at appropriate times 186 Time Seg 0:44:20 1680
824 Manually help controller at end of ramp Controls Off
Turnaround Delay 0:00:10
1308 Vent digester at end of prehydrolysis no Controls Active
3 open drain, blow through ice water, Collect hydrolyzate Time Seg 0:00:30 1700
turn pump off Time Seg 2:00:00 1700
Purge coils with short pulse of air pressure 883 Controls Off
Measure volume of hydrolyzate, NaOH used 2640 End of Profile
Verify all pressure released 536.70 water in chips
10 Open lid Profile 1 Wash 
Close drain valve Controls Active
Add measured volume of wash water to digester 3000 Time Seg 0:00:00 250
turn pump on Time Seg 0:05:00 250
close lid, close vent Time Seg 0:05:00 250
1350 run wash Profile 1 Controls Off
1400 open drain valve to blow through ice water End of Profile
Measure volume of wash 2900 636.70 water in chips
30 Verify all pressure released
Remove chip basket
Dump chips into plastic beaker preH chips
Weigh total chips 899.67 970.05 est         w 30.23
BD preH chips 275.29 17.4% loss       d 9.25
wet chips remaining 869.44 dry frac 0.305987
BD preH chips remaining 266.0
Bomb Cook 1 4 70 BD Chips White Lqr 4/2/08 test date
measure wet chips needed 228.84 228.77 targetml 0.5N HCl 29.20 A
42 Add measured volume of white liquor to bomb 132 20 %AA       g 14 34.2 B
Add volume of water to reach 5.8 l/w 115 132 liq Vol ml 34.4 C
Seal bombs and shake well 158.818 water in chipsgpl Na2O 31.0 Na2S
Cook place bombs and water in Digester 405.818 total water 74.9 NaOH
Start pump 2:38 5.80 Cook liq/wood 0.6 Na2CO3
1004 Start Cook profile 23 29.24% Sulf%
Close lid 171 106.5 TTA
1042 Manually adjust controller at end of ramp 105.9 AA
1342 Vent digester at end of cook and run cooling coil 640.39 640.6 wet chips remain 90.4 EA
Verify all pressure released Profile 23 Kraft Cook
1402 Open lid Controls Active
30 drain Time Seg 0:00:00 300
1422 Cool bombs in water before opening Time Seg 0:59:20 1690
Take sample of black liquor 34.13 AD Pulp Controls Off
45 Dump chips into wash bag, wash well 92.619 %Pulp Dry Delay 0:00:10
1625 Defibrate pulp in blender 31.61 BD Pulp Controls Active
Squeeze dry 37.28 %Yield Time Seg 0:00:30 1710
homogenize in mixer, spread to dry 41.3 Kappa Time Seg 2:38:00 1710
Controls Off
End of Profile
Bomb Cook 2 6 70 BD Chips White Lqr 4/2/08 test date
measure wet chips needed 228.76 228.77 target ml 0.5N HCl 29.20 A
42 Add measured volume of white liquor to bomb 139 21 %AA       g 14.7 34.2 B
Add volume of water to reach 5.8 l/w 171 139 liq Vol ml 34.4 C
Seal bombs and shake well                                   158.762 wtr in chips gpl Na2O 31.0 Na2S
Cook place bombs and water in Digester 468.762 total water 74.9 NaOH
Start pump 2:38 6.70 Cook liq/wood 0.6 Na2CO3
940 Start Cook profile 23 29.24% Sulf%
Close lid 172 106.5 TTA
1104 Manually adjust controller at end of ramp 105.9 AA
1318 Vent digester at end of cook and run cooling coil 411.52 411.84 wet chips remain 90.4 EA
Verify all pressure released 125.92 BD Chips Profile 23 Kraft Cook
1402 Open lid Controls Active
           201
      A9
30 drain Time Seg 0:00:00 600
1422 Cool bombs in water before opening Time Seg 0:59:20 1700
Take sample of black liquor 31.71 AD Pulp Controls Off
45 Dump chips into wash bag, wash well 92.547 %Pulp Dry Delay 0:00:10
1625 Defibrate pulp in blender 29.35 BD Pulp Controls Active
Squeeze dry 34.62 %Yield Time Seg 0:00:30 1720
homogenize in mixer, spread to dry 19.2 Kappa Time Seg 2:38:00 1720
Controls Off
End of Profile
Bomb Cook 3 5 60 BD Chips White Lqr 4/2/08 test date
measure wet chips needed 196.06 196.09 target ml 0.5N HCl 29.20 A
Add measured volume of white liquor to bomb 119 21 %AA       g 12.6 34.2 B
Add volume of water to reach 5.8 l/w 153 119 liq Vol ml 34.4 C
Seal bombs and shake well                                   136.068 wtr in chips gpl Na2O 31.0 Na2S
Cook place bombs and water in Digester 408.068 total water 74.9 NaOH
Start pump 2:38 6.80 Cook liq/wood 0.6 Na2CO3
845 Start Cook profile 23 29.24% Sulf%
Close lid 171 106.5 TTA
945 Manually adjust controller at end of ramp 105.9 AA
1223 Vent digester at end of cook and run cooling coil 215.2 215.78 wet chips remain 90.4 EA
Verify all pressure released 65.8485 BD Chips Profile 23 Kraft Cook
1243 Open lid Controls Active
drain Time Seg 0:00:00 300
1250 Cool bombs in water before opening Time Seg 0:59:20 1700
Take sample of black liquor 28.04 AD Pulp Controls Off
45 Dump chips into wash bag, wash well 95.15 %Pulp Dry Delay 0:00:10
1625 Defibrate pulp in blender 26.68 BD Pulp Controls Active
Squeeze dry 36.73 %Yield Time Seg 0:00:30 1715
homogenize in mixer, spread to dry 25.2 Kappa Time Seg 2:38:00 1715
Controls Off
End of Profile
Time Minutes Dig Temp2 kr avg H Factor
845 0 25 0.0         -         -         
945 60 171 1,014.1   507.0     507.0     
1153 47 171 1,014.1   1,014.1   287.3     
1431 215 171 1,014.1   1,014.1   3,126.7   
1436 220 171 1,014.1   1,014.1   3,211.2   
1441 225 59.9 0.0         507.0     3,253.4   
1446 230 59.9 0.0         0.0         3,253.4   
72.9976 BD Raw Chips
Bomb Cook 4 6 60 BD Chips White Lqr 4/2/08 test date
measure wet chips needed 197.01 196.09 target ml 0.5N HCl 29.20 A
Add measured volume of white liquor to bomb 138 20 %AA       g 14.59953 34.2 B
Add volume of water to cover chips 210 138 liq Vol ml 34.4 C
Seal bombs and shake well                                   136.727 wtr in chips gpl Na2O 31.0 Na2S
Cook place bombs and water in Digester 484.727 total water 74.9 NaOH
Start pump 8.04 Cook liq/wood 0.6 Na2CO3
802 Start Cook profile 23 802 29.24% Sulf%
Close lid 106.5 TTA
906 end of ramp 4 minutes late 105.9 AA
1130 Vent digester at end of cook and run cooling coil 17.51 18.77 wet chips remain 90.4 EA
Verify all pressure released 5.35784 BD Chips Profile 23 Kraft Cook
Open lid Controls Active
drain Time Seg 0:00:00 300
Cool bombs in water before opening Time Seg 0:59:20 1700
Take sample of black liquor 28.09 AD Pulp Controls Off
Dump chips into wash bag, wash well 94.19 %Pulp Dry Delay 0:00:10
Defibrate pulp in blender 26.46 BD Pulp Controls Active
Squeeze dry 36.24 %Yield Time Seg 0:00:30 1715
1328 homogenize in mixer, spread to dry 23.2 Kappa Time Seg 2:28:00 1715
Controls Off
End of Profile
Time Minutes Dig Temp2 kr avg H Factor
802 0 25 0.0         -         -         
902 60 171.5 1,056.3   528.1     528.1     
1130 208 171.5 1,056.3   1,056.3   3,133.6   
1135 215 171.5 1,056.3   1,056.3   3,256.9   
1140 220 171.5 1,056.3   1,056.3   3,344.9   
1145 225 59.9 0.0         528.1     3,388.9   
1150 230 59.9 0.0         0.0         3,388.9   
           202
      A9
Time minutes pH 1 Dig Temp kr Avg H Factor NaOH Vol Total Water Sample
1001 0 11.16 25.6 0.000        0 0
1005 4 10.81 41.3 0.000        0.00        0.00          
1010 9 10.21 57.6 0.004        0.00        0.00          
1015 14 9.58 73.1 0.036        0.02        0.00          
1020 19 8.78 88.5 0.258        0.15        0.01          
1025 24 7.42 98.4 0.846        0.55        0.06          
1030 29 6.66 110.4 3.287        2.07        0.23          
1035 34 6.11 125.4 15.974      9.63        1.03          
1040 39 5.74 138.6 58.383      37.18      4.13          
1045 44 5.36 151 183.29      120.84    14.20        
1050 49 5.05 158.1 342.59      262.94    36.11        
1055 54 5.04 164.1 572.05      457.32    74.22        
1100 59 4.81 167.8 779.34      675.69    130.53      
1105 64 4.81 168.5 825.81      802.57    197.41      
0 1108 67 170 934.35      880.08    241.42      2500
1110 69 4.88 170.2 949.80      942.08    272.82      
1115 74 4.94 170.6 981.43      965.62    353.29      
1120 79 4.97 170.3 957.62      969.53    434.08      
15 1123 82 170.3 957.62      957.62    481.96      2625
1125 84 4.99 170.3 957.62      957.62    513.88      
1130 89 4.98 169.9 926.71      942.17    592.40      
1135 94 5 170.1 942.05      934.38    670.26      
30 1138 97 170 934.35      938.20    717.17      2750
1140 99 5.02 169.9 926.71      930.53    748.19      
1145 104 5 170.2 949.80      938.26    826.38      
1150 109 4.99 170.1 942.05      945.92    905.20      
45 1153 112 170 934.35      938.20    952.11      2875
1155 114 5.01 170 934.35      934.35    983.26      
1200 119 5.01 170.2 949.80      942.08    1,061.77   
1205 124 5.01 169.8 919.14      934.47    1,139.64   
60 1208 127 170 934.35      926.74    1,185.98   3000
1210 129 5.01 170.2 949.80      942.08    1,217.38   
1215 134 5.02 170 934.35      942.08    1,295.88   
1220 139 5 170.1 942.05      938.20    1,374.07   
1225 144 4.99 170 934.35      938.20    1,452.25   
1230 149 4.99 170 934.35      934.35    1,530.11   
1235 154 4.99 170 934.35      934.35    1,607.98   
90 1238 157 170.1 942.05      938.20    1,654.89   3200
1240 159 4.99 170.2 949.80      945.92    1,686.42   
1245 164 5 170 934.35      942.08    1,764.92   
1250 169 5.01 163.7 553.07      743.71    1,826.90   
1255 174 5.02 167.5 760.19      656.63    1,881.62   
1300 179 5.01 170.5 973.43      866.81    1,953.85   
1305 184 4.98 171 1,014.06   993.75    2,036.66   
120 1308 187 5 170 934.35      974.20    2,085.37   3360
           203
      A19
Prehydrolysis Cook Plan A3B3C1 5/28/09 pH4.5 170?C pH 3.5
Time Activity Data Calc 380 g BD Chips to Cook
Soak Measure mass of chips, put in Ziploc, g 460.03 459.84 g AD targ 91.82 %Chip Moisture
1145 Add Soak Solution 807.169 37.63 wtr in chips 10 %preH Chip Sample
1300 After Soak pH 4.18 422.40 BD Chips
PreH Clean Digester, drain, close valve 30 844.80 Soak Water
1250 Put wet chips in basket in dig, add weight 2.00 Soak l/w
1300 measure volume of water added to just cover chips, ml 2600 3601.80 PreH water 0.25% Acetic Acid
1302 add acid (ml CH3COOH) 1 8.53 preH wtr/wood
1304 Add volume for samples 100 Profile 22 Prehydrolysis
1310 Start Pump, start Profile 22 Asurd SoakOff
Close lid, leave vent cracked open Controls Active
1340 Close vent at 100?C target Time Seg 0:00:00 300
1342 Open vent briefly at 100?C measurement Time Seg 1:00:00 1700
804 Open vent briefly at 105?C measurement Time Seg 2:00:00 1700
Sample liquid at appropriate times Controls Off
824 Manually help controller at end of ramp End of Profile
Turnaround
Vent digester at end of prehydrolysis no
open drain, blow through ice water, Collect hydrolyzate 86 pH control vol
turn pump off
Purge coils with short pulse of air pressure 684.80 water in chips
Measure volume of hydrolyzate 2917 Profile 1 Wash 
Verify all pressure released Controls Active
Open lid Time Seg 0:00:00 250
Close drain valve Time Seg 0:05:00 250
Add measured volume of wash water to digester 2770 Time Seg 0:05:00 250
turn pump on 3455 total wash Controls Off
close lid, close vent End of Profile
run wash Profile 1 870.80 wtr in chips
open drain valve to blow through ice water
Measure volume of wash 2584
Verify all pressure released preH chips
Remove chip basket 1293.20 est         w 29.99
Dump chips into plastic beaker 17.0% loss       d 9.45
Weigh total chips 1112.2 dry frac 0.3151
BD preH chips 350.444
wet chips remaining 1082.16
BD preH chips remaining 340.994
Bomb Cook 1 6 70 BD Chips White Lqr 6/25/09 test date
measure wet chips needed 184.305 222.15 184.30536ml 0.5N HCl 26.5 A
Add measured volume of white liquor to bomb 145 20 %AA       g 14 31.2 B
Add volume of water to reach 5.8 l/w 159 145 liq Vol ml 32 C
900 Seal bombs and shake well 126.23 water in chipsgpl Na2O 29.1 Na2S
Cook place bombs and water in Digester 430.23 total water 67.5 NaOH
Start pump 7.41 Cook liq/wood 2.5 Na2CO3
910 Start Cook profile 4 30.13% Sulf%
Close lid 99.1 TTA
1104 Manually adjust controller at end of ramp 96.6 AA
1342 Vent digester at end of cook and run cooling coil 82.1 EA
Verify all pressure released Profile 4 Kraft Cook
1402 Open lid Controls Active
30 drain Time Seg 0:00:00 300
1422 Cool bombs in water before opening Time Seg 0:59:20 1660
Take sample of black liquor 13.4 26.55 AD Pulp Controls Off
45 Dump chips into wash bag, wash well 93.54 %Pulp Dry Delay 0:00:10
1625 Defibrate pulp in blender 24.83 BD Pulp Controls Active
Squeeze dry 35.48 %Yield Time Seg 0:00:30 1680
homogenize in mixer, spread to dry 27 Kappa Time Seg 2:50:00 1680
Controls Off
End of Profile
           204
      A19
Bomb Cook 2 4 70 BD Chips White Lqr 6/25/09 test date
measure wet chips needed 184.305 222.15 target ml 0.5N HCl 26.5 A
Add measured volume of white liquor to bomb 145 20 %AA       g 14 31.2 B
Add volume of water to reach 5.8 l/w 159 145 liq Vol ml 32 C
Seal bombs and shake well                                   126.23 wtr in chips gpl Na2O 29.1 Na2S
Cook place bombs and water in Digester 430.23 total water 67.5 NaOH
Start pump 7.41 Cook liq/wood 2.5 Na2CO3
940 Start Cook profile 4 30.13% Sulf%
Close lid 99.1 TTA
1042 Manually adjust controller at end of ramp 96.6 AA
1318 Vent digester at end of cook and run cooling coil 82.1 EA
Verify all pressure released Profile 4 Kraft Cook
1402 Open lid Controls Active
30 drain Time Seg 0:00:00 300
1422 Cool bombs in water before opening Time Seg 0:59:20 1660
Take sample of black liquor 13.45 26.35 AD Pulp Controls Off
45 Dump chips into wash bag, wash well 93.62 %Pulp Dry Delay 0:00:10
1625 Defibrate pulp in blender 24.67 BD Pulp Controls Active
Squeeze dry 35.24 %Yield Time Seg 0:00:30 1680
homogenize in mixer, spread to dry 18.8 Kappa Time Seg 2:50:00 1680
Controls Off
End of Profile
Bomb Cook 3 4 70 BD Chips White Lqr 7/3/09 test date
measure wet chips needed 184.305 222.15 target ml 0.5N HCl 27.6 A
Add measured volume of white liquor to bomb 139 20 %AA       g 14 32.5 B
Add volume of water to reach 5.8 l/w 165 139 liq Vol ml 32.8 C
Seal bombs and shake well                                   126.23 wtr in chips gpl Na2O 30.3 Na2S
Cook place bombs and water in Digester 430.23 total water 70.3 NaOH
Start pump 7.41 Cook liq/wood 0.9 Na2CO3
845 Start Cook profile 4 30.15% Sulf%
Close lid 101.6 TTA
945 Manually adjust controller at end of ramp 100.6 AA
1223 Vent digester at end of cook and run cooling coil 85.5 EA
Verify all pressure released Profile 4 Kraft Cook
Open lid Controls Active
drain Time Seg 0:00:00 300
Cool bombs in water before opening Time Seg 0:59:20 1660
Take sample of black liquor 13.53 27.39 AD Pulp Controls Off
1333 Dump chips into wash bag, wash well 92.98 %Pulp Dry Delay 0:00:10
1625 Defibrate pulp in blender 25.47 BD Pulp Controls Active
Squeeze dry 36.38 %Yield Time Seg 0:00:30 1680
homogenize in mixer, spread to dry 53.2 Kappa Time Seg 1:46:00 1680
Controls Off
End of Profile
70 BD Raw Chips
Bomb Cook 4 4 70 BD Chips White Lqr 7/3/09 test date
measure wet chips needed 184.305 222.15 target ml 0.5N HCl 27.6 A
Add measured volume of white liquor to bomb 139 20 %AA       g 14 32.5 B
Add volume of water to cover chips 165 139 liq Vol ml 32.8 C
Seal bombs and shake well                                   126.23 wtr in chips gpl Na2O 30.3 Na2S
Cook place bombs and water in Digester 430.23 total water 70.3 NaOH
Start pump 7.41 Cook liq/wood 0.9 Na2CO3
802 Start Cook profile 4 30.15% Sulf%
Close lid 101.6 TTA
906 end of ramp 4 minutes late 100.6 AA
1130 Vent digester at end of cook and run cooling coil 85.5 EA
Verify all pressure released Profile 4 Kraft Cook
Open lid Controls Active
drain Time Seg 0:00:00 300
Cool bombs in water before opening Time Seg 0:59:20 1660
Take sample of black liquor 13.57 27.35 AD Pulp Controls Off
Dump chips into wash bag, wash well 93.22 %Pulp Dry Delay 0:00:10
Defibrate pulp in blender 25.50 BD Pulp Controls Active
Squeeze dry 36.42 %Yield Time Seg 0:00:30 1680
1328 homogenize in mixer, spread to dry 33.2 Kappa Time Seg 2:20:00 1680
Controls Off
End of Profile
           205
      A19
70 BD Raw Chips
Bomb Cook 5 4 70 BD Chips White Lqr 7/3/09 test date
measure wet chips needed 184.305 222.15 target ml 0.5N HCl 27.6 A
Add measured volume of white liquor to bomb 139 20 %AA       g 14 32.5 B
Add volume of water to cover chips 158 139 liq Vol ml 32.8 C
Seal bombs and shake well                                   126.23 wtr in chips gpl Na2O 30.3 Na2S
Cook place bombs and water in Digester 423.23 total water 70.3 NaOH
Start pump 7.29 Cook liq/wood 0.9 Na2CO3
802 Start Cook profile 4 30.15% Sulf%
Close lid 101.6 TTA
906 end of ramp 4 minutes late 100.6 AA
1130 Vent digester at end of cook and run cooling coil 85.5 EA
Verify all pressure released Profile 4 Kraft Cook
Open lid Controls Active
drain Time Seg 0:00:00 300
Cool bombs in water before opening Time Seg 0:59:20 1660
Take sample of black liquor 12.75 27.06 AD Pulp Controls Off
Dump chips into wash bag, wash well 93.33 %Pulp Dry Delay 0:00:10
Defibrate pulp in blender 25.26 BD Pulp Controls Active
Squeeze dry 36.08 %Yield Time Seg 0:00:30 1680
1328 homogenize in mixer, spread to dry 34.1 Kappa Time Seg 1:47:00 1680
Controls Off
End of Profile
           206
      A19
PreH C
Time minutes pH Dig Temp kr Avg H Factor OmegaT kr avg H Factor
1310 0 4.47 25.7 0.00         -           -             23.5       0.00        -         -         
1320 10 4.52 53.3 0.00         0.0           0.00           51.6       0.00        0.0          0.0         
1330 20 4.75 76.4 0.06         0.0           0.00           74 0.04        0.0          0.0         
1340 30 4.69 94.8 0.55         0.3           0.06           92.9 0.44        0.2          0.0         
1350 40 4.68 120.2 9.36         5.0           0.88           118.9 8.17        4.3          0.8         
1400 50 4.67 145.1 107.25      58.3         10.60         143.6 93.37      50.8        9.2         
1405 55 4.62 156.4 295.49      201.4       27.38         155 261.38    177.4      24.0       
1408 58 4.57 161.2 447.28      371.4       45.95         159.5 386.61    324.0      40.2       
1410 60 4.53 164.1 572.05      509.7       62.94         162.2 487.07    436.8      54.8       
1412 62 4.48 166.6 705.37      638.7       84.23         164.7 601.68    544.4      72.9       
1414 64 4.43 169.2 874.89      790.1       110.57       167.1 735.35    668.5      95.2       
1415 65 4.38 170 934.35      904.6       125.64       167.8 779.34    757.3      107.8     
1417 67 4.4 172.2 1,118.19   1,026.3    159.85       170 934.35    856.8      136.4     
1419 69 4.48 171.7 1,073.63   1,095.9    196.38       168.8 846.51    890.4      166.1     
1421 71 4.44 169.7 911.62      992.6       229.47       167.1 735.35    790.9      192.4     
1425 75 4.49 169.7 911.62      911.6       290.25       167 729.26    732.3      241.2     
1430 80 4.49 170.3 957.62      934.6       368.13       167.6 766.52    747.9      303.6     
1435 85 4.46 170.1 942.05      949.8       447.28       167.4 753.90    760.2      366.9     
1445 95 4.49 170.1 942.05      942.0       604.29       167.3 747.67    750.8      492.1     
1500 110 4.47 169.9 926.71      934.4       837.89       167.2 741.48    744.6      678.2     
1515 125 4.5 170.3 957.62      942.2       1,073.43    167.4 753.90    747.7      865.1     
1545 155 4.49 170.1 942.05      949.8       1,548.34    167.2 741.48    747.7      1,239.0   
1610 180 4.49 170 934.35      938.2       1,939.26    167 729.26    735.4      1,545.4   
1612 182 4.46 163.5 543.81      739.1       1,963.89    159 370.31    549.8      1,563.7   
1614 184 4.45 153 219.07      381.4       1,976.61    147.9 138.57    254.4      1,572.2   
1616 186 4.45 140 66.66        142.9       1,981.37    134.8 40.55      89.6        1,575.2   
1618 188 4.45 131.1 28.25        47.5         1,982.95    125.9 16.80      28.7        1,576.1   
1620 190 4.45 124.6 14.73        21.5         1,983.67    119.1 8.34        12.6        1,576.5   
Time NaOH VolTotal WaterSample
1310 3515.8
1320 3515.8
1330 3515.8
1340 3515.8
1350 3515.8
1400 3515.8 3516
1405 3515.8 3516
1408 3515.8 3516
1410 3515.8 3516
1412 3515.8 3516
1414 3515.8 3516
1415 3489.8 -26 3516
1417 3489.8 3490
1419 3489.8 3490
1421 0 3489.8 3490
1425 22 3485.8 -26 3512
1430 3485.8 3486
1435 10 3469.8 -26 3496
1445 3 3446.8 -26 3473
1500 10 3430.8 -26 3457
1515 7 3411.8 -26 3438
1545 20 3405.8 -26 3432
1610 3405.8 3406
1612 3405.8 3406
1614 3405.8 3406
1616 3405.8 3406
1618 3405.8 3406
1620 14 3379.8 -40 3420
86 -222
see A17 (A) for cook1 details
           207
      A19
6/26/09 BC CFG2
Time Minutes Dig Temp2 kr avg H Factor OmegaT kr avg H Factor
930 0 26.8 0.00         -           -             24.8       0.00        -         -         
1000 30 98.7 0.88         0.44         0.2             97.2 0.74        0.4          0.2         
1028 58 162.6 503.90      252.39     118.0         160.5 421.28    211.0      98.7       
1030 60 167.1 735.35      619.62     138.7         165 617.03    519.2      116.0     
1032 62 167.5 760.19      747.77     163.6         165 617.03    617.0      136.5     
1034 64 167.7 772.90      766.54     189.1         165.1 622.23    619.6      157.2     
1036 66 168 792.35      782.63     215.2         165.4 638.08    630.2      178.2     
1038 68 168 792.35      792.35     241.6         165.4 638.08    638.1      199.5     
1040 70 168 792.35      792.35     268.0         165.2 627.47    632.8      220.5     
1320 230 168 792.35      792.35     2,381.0      165 617.03    622.3      1,879.9   
1322 232 163.5 543.81      668.08     2,403.3      157.2 316.84    466.9      1,895.5   
1324 234 158.2 345.57      444.69     2,418.1      150.5 175.26    246.0      1,903.7   
1326 236 150.1 169.07      257.32     2,426.7      145.2 108.25    141.8      1,908.4   
1328 238 140.9 72.56        120.81     2,430.7      136.5 47.77      78.0        1,911.0   
1330 240 131.7 29.97        51.26       2,432.4      127.5 19.74      33.8        1,912.1   
1332 242 123.6 13.30        21.63       2,433.1      119.6 8.79        14.3        1,912.6   
1334 244 116.5 6.34         9.82         2,433.4      112.4 4.09        6.4          1,912.8   
DigT old DigT New OmT OmT n
ramp 138.7       61.8           116.0     51.7        
cook 2,242.3    2,242.3      1,763.9   1,764.8   
cool 52.5         51.6           32.9       32.2        
total 2,433.4    2,355.7      1,912.8   1,848.6   
see A24 (G) for cook3 details
7/18/09 BC CEG4
Time Minutes Dig Temp2 kr avg H Factor OmegaT kr avg H Factor
1155 0 26.9 0.00         -           -             24.7       0.00        -         -         
1225 30 98.5 0.86         0.43         0.21           97.2 0.74        0.4          0.2         
1253 58 162.7 508.19      254.52     118.99       161.1 443.48    222.1      103.8     
1255 60 167.2 741.48      624.84     139.82       165.6 648.86    546.2      122.0     
1257 62 167.8 779.34      760.41     165.17       165.7 654.31    651.6      143.8     
1259 64 168 792.35      785.84     191.36       165.7 654.31    654.3      165.6     
1301 66 168 792.35      792.35     217.77       165.6 648.86    651.6      187.3     
1303 68 168 792.35      792.35     244.19       165.6 648.86    648.9      208.9     
1305 70 168 792.35      792.35     270.60       165.6 648.86    648.9      230.5     
1307 72 168 792.35      792.35     297.01       165.5 643.45    646.2      252.1     
1309 74 168 792.35      792.35     323.42       165.5 643.45    643.4      273.5     
1343 108 168 792.35      792.35     772.42       165.4 638.08    640.8      636.6     
1452 177 168 792.35      792.35     1,683.63    165.3 632.76    635.4      1,367.4   
1502 187 168 792.35      792.35     1,815.69    165.4 638.08    635.4      1,473.3   
1505 190 168 792.35      792.35     1,855.31    165.3 632.76    635.4      1,505.0   
1515 200 168 792.35      792.35     1,987.36    165.4 638.08    635.4      1,610.9   
1517 202 165.9 665.35      728.85     2,011.66    160.9 435.96    537.0      1,628.8   
1519 204 153.7 233.08      449.21     2,026.63    148.2 142.40    289.2      1,638.5   
1521 206 142.3 82.73        157.90     2,031.90    136.8 49.17      95.8        1,641.7   
1523 208 130.8 27.42        55.08       2,033.73    125.5 16.14      32.7        1,642.8   
1525 210 119.8 8.98         18.20       2,034.34    114.5 5.13        10.6        1,643.1   
1527 212 111.6 3.75         6.36         2,034.55    107.6 2.41        3.8          1,643.2   
DigT old DigT New OmT OmT n
ramp 139.8       62.2           122.0     54.2        
cook 1,847.5    1,847.5      1,488.9   1,489.0   
cool 47.2         45.8           32.3       31.2        
total 2,034.6    1,955.5      1,643.2   1,574.4   
see A24 (G) for cook 5 details
           208
      A19
b 23.54 pH 4.5
PreH C recal n 0.31 0.948
time digT kr avg H Factor OmegaT kr avg H Factor Mod kr Mod H Fact
0 25.7 0.0         -         0.0 23.5 0.00        0 0.0 0.00          0.0
10 53.3 0.0         0.0          0.0 51.6 0.00        0.00         0.0 0.00          0.0
20 76.4 0.1         0.0          0.0 74 0.04        0.02         0.0 0.06          0.0
30 94.8 0.6         0.3          0.1 92.9 0.44        0.24         0.0 0.47          0.1
31 97.34 0.7         0.7          0.1 95.5 0.60        0.52         0.1 0.62          0.1
32 99.88 1.0         0.9          0.1 98.1 0.82        0.71         0.1 0.80          0.1
33 102.4 1.3         1.2          0.1 100.7 1.10        0.96         0.1 1.03          0.1
34 105.0 1.8         1.6          0.1 103.3 1.49        1.30         0.1 1.32          0.1
35 107.5 2.4         2.1          0.2 105.9 2.00        1.74         0.1 1.70          0.1
36 110.0 3.2         2.8          0.2 108.5 2.67        2.33         0.2 2.16          0.2
37 112.6 4.2         3.7          0.3 111.1 3.55        3.11         0.2 2.75          0.2
38 115.1 5.5         4.8          0.3 113.7 4.70        4.13         0.3 3.49          0.3
39 117.7 7.2         6.3          0.5 116.3 6.21        5.46         0.4 4.42          0.3
40 120.2 9.4         8.3          0.6 118.9 8.17        7.19         0.5 5.57          0.4
41 122.7 12.1       10.7        0.8 121.37 10.57      9.37         0.7 6.92          0.5
42 125.2 15.6       13.9        1.0 123.84 13.63      12.10       0.9 8.57          0.6
43 127.7 20.1       17.9        1.3 126.31 17.51      15.57       1.1 10.59        0.8
44 130.2 25.7       22.9        1.7 128.78 22.44      19.98       1.5 13.05        1.0
45 132.7 32.9       29.3        2.2 131.25 28.67      25.56       1.9 16.04        1.2
46 135.1 41.9       37.4        2.8 133.72 36.51      32.59       2.4 19.67        1.5
47 137.6 53.2       47.6        3.6 136.19 46.37      41.44       3.1 24.06        1.9
48 140.1 67.4       60.3        4.6 138.66 58.72      52.54       4.0 29.37        2.3
49 142.6 85.2       76.3        5.9 141.13 74.14      66.43       5.1 35.75        2.9
50 145.1 107.3     96.2        7.5 143.6 93.37      83.76       6.5 43.42        3.5
51 147.4 131.9     119.6      9.5 145.88 115.23    104.30     8.2 51.85        4.3
52 149.6 161.9     146.9      11.9 148.16 141.88    128.56     10.4 61.80        5.3
53 151.9 198.3     180.1      14.9 150.44 174.31    158.10     13.0 73.51        6.4
54 154.1 242.3     220.3      18.6 152.72 213.69    194.00     16.2 87.29        7.8
55 156.4 295.5     268.9      23.1 155 261.38    237.54     20.2 103.45      9.3
56 158.0 339.6     317.6      28.4 156.5 298.09    279.74     24.9 115.58      11.2
57 159.6 390.0     364.8      34.4 158 339.63    318.86     30.2 129.02      13.2
58 161.2 447.3     418.6      41.4 159.5 386.61    363.12     36.2 143.91      15.5
59 162.7 506.0     476.7      49.4 160.85 434.10    410.36     43.1 158.68      18.0
60 164.1 572.1     539.0      58.3 162.2 487.07    460.58     50.7 174.86      20.8
61 165.4 635.4     603.7      68.4 163.45 541.51    514.29     59.3 191.21      23.8
62 166.6 705.4     670.4      79.6 164.7 601.68    571.60     68.8 208.98      27.2
63 167.9 785.8     745.6      92.0 165.9 665.35    633.51     79.4 227.48      30.8
64 169.2 874.9     830.4      105.8 167.1 735.35    700.35     91.1 247.50      34.8
65 170.0 934.3     904.6      120.9 167.8 779.34    757.34     103.7 259.93      39.0
66 171.1 1,022.4   978.4      137.2 168.9 853.52    816.43     117.3 280.64      43.5
67 172.2 1,118.2   1,070.3   155.1 170 934.35    893.94     132.2 302.89      48.4
68 172.0 1,095.7   1,106.9   173.5 169.4 889.41    911.88     147.4 290.56      53.3
69 171.7 1,073.6   1,084.7   191.6 168.8 846.51    867.96     161.9 278.70      58.1
70 170.7 989.5     1,031.6   208.8 167.95 789.08    817.80     175.5 262.66      62.6
71 169.7 911.6     950.6      224.6 167.1 735.35    735.35     176.9 247.50      66.8
75 169.7 911.6     911.6      285.4 167 729.26    732.30     225.7 245.77      83.3
80 170.3 957.6     934.6      363.3 167.6 766.52    747.89     288.0 256.32      104.2
85 170.1 942.0     949.8      442.4 167.4 753.90    760.21     351.4 252.75      125.4
95 170.1 942.0     942.0      599.4 167.3 747.67    750.79     476.5 250.99      167.4
110 169.9 926.7     934.4      833.0 167.2 741.48    744.58     662.6 249.24      229.9
125 170.3 957.6     942.2      1068.6 167.4 753.90    747.69     849.6 252.75      292.6
155 170.1 942.0     949.8      1543.5 167.2 741.48    747.69     1223.4 249.24      418.1
180 170.0 934.3     938.2      1934.4 167 729.26    735.37     1529.8 245.77      521.3
181 166.8 714.2     824.3      1948.2 163 521.28    625.27     1540.2 185.17      524.9
182 163.5 543.8     629.0      1958.6 159 370.31    445.79     1547.7 138.78      527.6
183 158.3 347.1     445.4      1966.1 153.45 227.98    299.14     1552.6 92.19        529.5
184 153.0 219.1     283.1      1970.8 147.9 138.57    183.28     1555.7 60.58        530.8
185 146.5 122.0     170.5      1973.6 141.35 75.69      107.13     1557.5 36.38        531.6
186 140.0 66.7       94.3        1975.2 134.8 40.55      58.12       1558.5 21.49        532.0
187 135.6 43.6       55.1        1976.1 130.35 26.23      33.39       1559.0 14.88        532.4
188 131.1 28.2       35.9        1976.7 125.9 16.80      21.52       1559.4 10.22        532.6
189 127.9 20.4       24.3        1977.1 122.5 11.88      14.34       1559.6 7.63          532.7
190 124.6 14.7       17.6        1977.4 119.1 8.34        10.11       1559.8 5.67          532.8
           209
      A19
BC CEG4
time digT kr avg H Factor OmegaT kr avg H Factor
0 26.9 0.00       0 0.0 24.7     0.00        0 0.0
1 29.29 0.00       0.00        0.0 27.1     0.00        0.00         0.0
2 31.67 0.00       0.00        0.0 29.5     0.00        0.00         0.0
3 34.06 0.00       0.00        0.0 32.0     0.00        0.00         0.0
4 36.45 0.00       0.00        0.0 34.4     0.00        0.00         0.0
5 38.83 0.00       0.00        0.0 36.8     0.00        0.00         0.0
6 41.22 0.00       0.00        0.0 39.2     0.00        0.00         0.0
7 43.61 0.00       0.00        0.0 41.6     0.00        0.00         0.0
8 45.99 0.00       0.00        0.0 44.0     0.00        0.00         0.0
9 48.38 0.00       0.00        0.0 46.5     0.00        0.00         0.0
10 50.77 0.00       0.00        0.0 48.9     0.00        0.00         0.0
11 53.15 0.00       0.00        0.0 51.3     0.00        0.00         0.0
12 55.54 0.00       0.00        0.0 53.7     0.00        0.00         0.0
13 57.93 0.00       0.00        0.0 56.1     0.00        0.00         0.0
14 60.31 0.01       0.01        0.0 58.5     0.00        0.00         0.0
15 62.7 0.01       0.01        0.0 60.95 0.0          0.01         0.0
16 65.09 0.01       0.01        0.0 63.3667 0.0          0.01         0.0
17 67.47 0.02       0.01        0.0 65.7833 0.0          0.01         0.0
18 69.86 0.02       0.02        0.0 68.2 0.0          0.02         0.0
19 72.25 0.03       0.03        0.0 70.6167 0.0          0.02         0.0
20 74.63 0.04       0.04        0.0 73.0333 0.0          0.03         0.0
21 77.02 0.06       0.05        0.0 75.45 0.0          0.04         0.0
22 79.41 0.08       0.07        0.0 77.8667 0.1          0.06         0.0
23 81.79 0.11       0.10        0.0 80.2833 0.1          0.08         0.0
24 84.18 0.15       0.13        0.0 82.7 0.1          0.11         0.0
25 86.57 0.20       0.18        0.0 85.1167 0.2          0.15         0.0
26 88.95 0.27       0.24        0.0 87.5333 0.2          0.20         0.0
27 91.34 0.37       0.32        0.0 89.95 0.3          0.27         0.0
28 93.73 0.49       0.43        0.0 92.3667 0.4          0.36         0.0
29 96.11 0.65       0.57        0.0 94.7833 0.6          0.48         0.0
30 98.5 0.86       0.75        0.0 97.2 0.7          0.64         0.0
31 100.8 1.12       0.99        0.1 99.4821 1.0          0.85         0.1
32 103.1 1.45       1.28        0.1 101.764 1.2          1.10         0.1
33 105.4 1.88       1.67        0.1 104.046 1.6          1.43         0.1
34 107.7 2.43       2.16        0.2 106.329 2.1          1.86         0.1
35 110 3.13       2.78        0.2 108.611 2.7          2.40         0.2
36 112.3 4.02       3.58        0.3 110.893 3.5          3.08         0.2
37 114.6 5.15       4.59        0.3 113.175 4.4          3.96         0.3
38 116.8 6.58       5.87        0.4 115.457 5.7          5.06         0.4
39 119.1 8.38       7.48        0.6 117.739 7.2          6.46         0.5
40 121.4 10.63     9.50        0.7 120.021 9.2          8.21         0.6
41 123.7 13.46     12.05      0.9 122.3   11.64      10.41       0.8
42 126 17.00     15.23      1.2 124.6   14.70      13.17       1.0
43 128.3 21.41     19.20      1.5 126.9   18.53      16.62       1.3
44 130.6 26.89     24.15      1.9 129.2   23.28      20.91       1.6
45 132.9 33.68     30.29      2.4 131.4   29.19      26.24       2.1
46 135.2 42.09     37.89      3.0 133.7   36.49      32.84       2.6
47 137.5 52.47     47.28      3.8 136.0   45.51      41.00       3.3
48 139.8 65.24     58.85      4.8 138.3   56.63      51.07       4.2
49 142.1 80.93     73.08      6.0 140.6   70.28      63.45       5.2
50 144.4 100.15    90.54      7.5 142.8   87.02      78.65       6.5
51 146.7 123.65    111.90    9.4 145.1   107.50    97.26       8.1
52 148.9 152.31    137.98    11.7 147.4   132.49    120.00     10.1
53 151.2 187.20    169.76    14.5 149.7   162.93    147.71     12.6
54 153.5 229.57    208.39    18.0 152.0   199.91    181.42     15.6
55 155.8 280.92    255.24    22.2 154.3   244.76    222.34     19.3
56 158.1 343.01    311.96    27.4 156.5   299.02    271.89     23.9
57 160.4 417.94    380.48    33.8 158.8   364.53    331.77     29.4
58 162.7 508.19    463.07    41.5 161.1   443.48    404.00     36.1
59 165 614.45    561.32    50.9 163.4   536.95    490.22     44.3
60 167.2 741.48    677.97    62.2 165.6   648.86    592.91     54.2
61 167.5 760.19    750.83    74.7 165.7   651.58    650.22     65.0
62 167.8 779.34    769.76    87.5 165.7   654.31    652.94     75.9
63 167.9 785.82    782.58    100.5 165.7   654.31    654.31     86.8
64 168 792.35    789.09    113.7 165.7 654.3      654.31     97.7
65 168 792.35    792.35    126.9 165.65 651.6      652.94     108.6
66 168 792.35    792.35    140.1 165.6 648.9      650.22     119.4
           210
      A19
67 168 792.35    792.35    153.3 165.6 648.9      648.86     130.2
68 168 792.35    792.35    166.5 165.6 648.9      648.86     141.1
69 168 792.35    792.35    179.7 165.6 648.9      648.86     151.9
70 168 792.35    792.35    192.9 165.6 648.9      648.86     162.7
71 168 792.35    792.35    206.1 165.55 646.1      647.50     173.5
72 168 792.35    792.35    219.3 165.5 643.4      644.80     184.2
73 168 792.35    792.35    232.5 165.5 643.4      643.45     194.9
74 168 792.35    792.35    245.8 165.5 643.4      643.45     205.7
108 168 792.35    792.35    694.8 165.4 638.1      640.76     568.8
177 168 792.35    792.35    1606.0 165.3 632.8      635.42     1299.5
187 168 792.35    792.35    1738.0 165.4 638.1      635.42     1405.4
188 168 792.35    792.35    1751.2 165.4 638.1      638.08     1416.0
189 168 792.35    792.35    1764.4 165.4 638.1      638.08     1426.7
190 168 792.35    792.35    1777.6 165.3 632.8      635.42     1437.3
200 168 792.35    792.35    1909.7 165.4 638.1      635.42     1543.2
201 167 726.23    759.29    1922.4 163.15 527.9      583.01     1552.9
202 165.9 665.35    695.79    1933.9 160.9 436.0      481.95     1560.9
203 159.8 396.72    531.03    1942.8 154.55 251.2      343.60     1566.6
204 153.7 233.08    314.90    1948.0 148.2 142.4      196.82     1569.9
205 148 139.84    186.46    1951.2 142.5 84.3        113.34     1571.8
206 142.3 82.73     111.28    1953.0 136.8 49.2        66.73       1572.9
207 136.6 48.00     65.36      1954.1 131.15 28.4        38.78       1573.6
208 130.8 27.42     37.71      1954.7 125.5 16.1        22.26       1573.9
209 125.3 15.81     21.62      1955.1 120 9.2          12.65       1574.2
210 119.8 8.98       12.40      1955.3 114.5 5.1          7.15         1574.3
211 115.7 5.83       7.40        1955.4 111.05 3.5          4.33         1574.3
212 111.6 3.75       4.79        1955.5 107.6 2.4          2.97         1574.4
           211
      A19
BC CFG2
time digT kr avg H Factor OmegaT kr avg H Factor
0 26.8 0.00          0 0.0 24.8 0.00            0 0.0
1 29.2 0.00          0.00            0.0 27.2 0.00            0.00       0.0
2 31.6 0.00          0.00            0.0 29.6 0.00            0.00       0.0
3 34.0 0.00          0.00            0.0 32.0 0.00            0.00       0.0
4 36.4 0.00          0.00            0.0 34.5 0.00            0.00       0.0
5 38.8 0.00          0.00            0.0 36.9 0.00            0.00       0.0
6 41.2 0.00          0.00            0.0 39.3 0.00            0.00       0.0
7 43.6 0.00          0.00            0.0 41.7 0.00            0.00       0.0
8 46.0 0.00          0.00            0.0 44.1 0.00            0.00       0.0
9 48.4 0.00          0.00            0.0 46.5 0.00            0.00       0.0
10 50.8 0.00          0.00            0.0 48.9 0.00            0.00       0.0
11 53.2 0.00          0.00            0.0 51.3 0.00            0.00       0.0
12 55.6 0.00          0.00            0.0 53.8 0.00            0.00       0.0
13 58.0 0.00          0.00            0.0 56.2 0.00            0.00       0.0
14 60.4 0.01          0.01            0.0 58.6 0.00            0.00       0.0
15 62.8 0.01          0.01            0.0 61.0 0.01            0.01       0.0
16 65.1 0.01          0.01            0.0 63.4 0.01            0.01       0.0
17 67.5 0.02          0.01            0.0 65.8 0.01            0.01       0.0
18 69.9 0.02          0.02            0.0 68.2 0.02            0.02       0.0
19 72.3 0.03          0.03            0.0 70.7 0.03            0.02       0.0
20 74.7 0.04          0.04            0.0 73.1 0.04            0.03       0.0
21 77.1 0.06          0.05            0.0 75.5 0.05            0.04       0.0
22 79.5 0.08          0.07            0.0 77.9 0.07            0.06       0.0
23 81.9 0.11          0.10            0.0 80.3 0.09            0.08       0.0
24 84.3 0.15          0.13            0.0 82.7 0.13            0.11       0.0
25 86.7 0.21          0.18            0.0 85.1 0.17            0.15       0.0
26 89.1 0.28          0.24            0.0 87.5 0.23            0.20       0.0
27 91.5 0.37          0.33            0.0 90.0 0.31            0.27       0.0
28 93.9 0.50          0.44            0.0 92.4 0.41            0.36       0.0
29 96.3 0.66          0.58            0.0 94.8 0.55            0.48       0.0
30 98.7 0.88          0.77            0.1 97.2 0.74            0.64       0.0
31 101.0 1.14          1.01            0.1 99.5 0.96            0.85       0.1
32 103.3 1.48          1.31            0.1 101.7 1.24            1.10       0.1
33 105.5 1.92          1.70            0.1 104.0 1.61            1.43       0.1
34 107.8 2.48          2.20            0.2 106.2 2.07            1.84       0.1
35 110.1 3.18          2.83            0.2 108.5 2.67            2.37       0.2
36 112.4 4.08          3.63            0.3 110.8 3.42            3.04       0.2
37 114.7 5.22          4.65            0.3 113.0 4.37            3.90       0.3
38 117.0 6.66          5.94            0.4 115.3 5.57            4.97       0.4
39 119.2 8.47          7.56            0.6 117.5 7.09            6.33       0.5
40 121.5 10.74        9.60            0.7 119.8 8.98            8.04       0.6
41 123.8 13.58        12.16          0.9 122.1 11.36          10.17     0.8
42 126.1 17.12        15.35          1.2 124.3 14.32          12.84     1.0
43 128.4 21.54        19.33          1.5 126.6 18.02          16.17     1.3
44 130.7 27.02        24.28          1.9 128.9 22.60          20.31     1.6
45 132.9 33.81        30.42          2.4 131.1 28.28          25.44     2.0
46 135.2 42.21        38.01          3.1 133.4 35.29          31.79     2.6
47 137.5 52.56        47.38          3.8 135.6 43.94          39.62     3.2
48 139.8 65.28        58.92          4.8 137.9 54.58          49.26     4.0
49 142.1 80.90        73.09          6.0 140.2 67.64          61.11     5.1
50 144.3 100.02      90.46          7.5 142.4 83.61          75.62     6.3
51 146.6 123.37      111.69        9.4 144.7 103.13        93.37     7.9
52 148.9 151.82      137.59        11.7 146.9 126.92        115.02    9.8
53 151.2 186.42      169.12        14.5 149.2 155.84        141.38    12.1
54 153.5 228.41      207.42        18.0 151.5 190.94        173.39    15.0
55 155.8 279.25      253.83        22.2 153.7 233.45        212.20    18.6
56 158.0 340.68      309.97        27.4 156.0 284.81        259.13    22.9
57 160.3 414.76      377.72        33.7 158.2 346.74        315.78    28.1
58 162.6 503.90      459.33        41.3 160.5 421.28        384.01    34.5
59 164.9 609.31      556.60        50.6 162.8 510.35        465.82    42.3
60 167.1 735.35      672.33        61.8 165.0 617.03        563.69    51.7
61 167.3 747.67      741.51        74.2 165.0 617.03        617.03    62.0
62 167.5 760.19      753.93        86.7 165.0 617.03        617.03    72.3
63 167.6 766.52      763.35        99.5 165.1 619.63        618.33    82.6
64 167.7 772.90      769.71        112.3 165.1 622.23        620.93    92.9
65 167.9 782.57      777.74        125.2 165.3 630.11        626.17    103.4
66 168.0 792.35      787.46        138.4 165.4 638.08        634.10    113.9
           212
      A19
67 167.9 785.82      789.09        151.5 165.4 638.08        638.08    124.6
68 167.8 779.34      782.58        164.6 165.4 638.08        638.08    135.2
69 167.9 785.82      782.58        177.6 165.3 632.76        635.42    145.8
70 168.0 792.35      789.09        190.8 165.2 627.47        630.12    156.3
71 168.1 795.64      794.00        204.0 165.2 627.47        627.47    166.8
72 168.1 798.94      797.29        217.3 165.2 627.47        627.47    177.2
73 168.1 798.94      798.94        230.6 165.2 627.47        627.47    187.7
74 168.1 798.94      798.94        243.9 165.2 627.47        627.47    198.1
75 168.1 795.64      797.29        257.2 165.2 627.47        627.47    208.6
76 168.0 792.35      794.00        270.4 165.2 627.47        627.47    219.0
77 168.0 792.35      792.35        283.6 165.2 627.47        627.47    229.5
78 168.0 792.35      792.35        296.8 165.2 627.47        627.47    240.0
79 168.0 792.35      792.35        310.1 165.2 624.85        626.16    250.4
80 168.0 792.35      792.35        323.3 165.2 627.47        626.16    260.8
230 168.0 792.35      792.35        2304.1 165.0 617.03        622.25    1816.5
231 165.8 657.05      724.70        2316.2 161.1 443.48        530.26    1825.3
232 163.5 543.81      600.43        2326.2 157.2 316.84        380.16    1831.6
233 160.9 434.10      488.95        2334.4 153.9 236.19        276.51    1836.2
234 158.2 345.57      389.83        2340.9 150.5 175.26        205.72    1839.7
235 154.2 242.53      294.05        2345.8 147.9 137.94        156.60    1842.3
236 150.1 169.07      205.80        2349.2 145.2 108.25        123.09    1844.3
237 145.5 111.27      140.17        2351.5 140.9 72.22          90.23     1845.8
238 140.9 72.56        91.92          2353.1 136.5 47.77          60.00     1846.8
239 136.3 46.86        59.71          2354.1 132.0 30.86          39.32     1847.5
240 131.7 29.97        38.41          2354.7 127.5 19.74          25.30     1847.9
241 127.7 20.04        25.00          2355.1 123.6 13.23          16.49     1848.2
242 123.6 13.30        16.67          2355.4 119.6 8.79            11.01     1848.4
243 120.1 9.22          11.26          2355.6 118.1 7.47            8.13       1848.5
244 116.5 6.34          7.78            2355.7 116.5 6.34            6.91       1848.6
           213
      A27
Kraft Bomb Cook Plan 6/11/09 70 g BD Chips to Cook
Time Activity Data Calc 91.82 %Chip solids
prep Clean Digester, drain, close valve 76.24 g AD Chips 0 %preH Chip Sample
1330 Measure mass of chips, put in bomb, g 76.24 6.24 wtr in chips White Liquor
70.0036 BD Chips 5/23/09 test date
1420 Add measured vol of white liquor to bomb 144 20 %AA ml 0.5N HCl
1420 measure vol of wtr added to just cover chips 270 14 AA g 27.05 A
Use vacuum oven to remove air from chips 144 liq Vol ml 31.5 B
1430 Seal bombs and shake well 420.24 total water 32.6 C
Cook place bombs and water in Digester 6.00 Cook liq/woodgpl Na2O 27.6 Na2S
13:30 Start pump 4 Bomb 70.0 NaOH
Start Cook profile 4 3.4 Na2CO3
Close lid 28.25% Sulf%
14:30 Manually adjust controller at end of ramp 168 100.9 TTA
1715 Vent digester at end of cook and run cooling coil 97.5 AA
Verify all pressure released 83.8 EA
Open lid 33.31 AD Pulp Profile 4 Kraft Cook
drain 92.86 %Pulp Dry Controls Active
30 Cool bombs in water before opening 30.93 BD Pulp Time Seg 0:00:00 300
Take sample of black liquor 44.19 %Yield Time Seg 0:59:20 1660
45 Dump chips into wash bag, wash well 26.8 Kappa Controls Off
Defibrate pulp in blender, Squeeze dry Delay 0:00:10
homogenize in mixer, spread to dry       BL pH 13.21 Controls Active
Time Seg 0:00:30 1680
BOMB #2 Time Seg 3:04:00 1680
prep Clean Digester, drain, close valve 76.24 g AD Chips Controls Off
Measure mass of chips, put in bomb, g 76.24 6.24 wtr in chips End of Profile
70.0036 BD Chips
Add measured volume of white liquor to bomb 144 20 %AA
measure volume of water added to just cover chips 270 14 AA g 70 g BD Chips to Cook
Seal bombs and shake well 144 liq Vol ml 91.82 %Chip solids
420.24 total water 0 %preH Chip Sample
Cook place bombs and water in Digester 6.00 Cook liq/wood White Liquor
13:30 Start pump 5 Bomb 5/23/09 test date
Start Cook profile 4 ml 0.5N HCl
Close lid 27.05 A
1430 Manually adjust controller at end of ramp 168 31.5 B
1715 Vent digester at end of cook and run cooling coil 32.6 C
Verify all pressure released gpl Na2O 27.6 Na2S
Open lid 33.54 AD Pulp 70.0 NaOH
drain 93.05 %Pulp Dry 3.4 Na2CO3
30 Cool bombs in water before opening 31.21 BD Pulp 28.25% Sulf%
Take sample of black liquor 44.58 %Yield 100.9 TTA
45 Dump chips into wash bag, wash well 27.2 Kappa 97.5 AA
Defibrate pulp in blender, Squeeze dry 83.8 EA
homogenize in mixer, spread to dry       BL pH 13.14
BOMB #3
prep Clean Digester, drain, close valve 76.24 g AD Chips
Measure mass of chips, put in bomb, g 76.24 6.24 wtr in chips
70.0036 BD Chips
Add measured volume of white liquor to bomb 144 20 %AA 70 g BD Chips to Cook
measure volume of water added to just cover chips 270 14 AA g 91.82 %Chip solids
Seal bombs and shake well 144 liq Vol ml 0 %preH Chip Sample
420.24 total water White Liquor
Cook place bombs and water in Digester 6.00 Cook liq/wood 5/23/09 test date
13:30 Start pump 6 Bomb ml 0.5N HCl
Start Cook profile 4 27.05 A
Close lid 31.5 B
1430 Manually adjust controller at end of ramp 168 32.6 C
1715 Vent digester at end of cook and run cooling coil gpl Na2O 27.6 Na2S
Verify all pressure released 70.0 NaOH
Open lid 32.77 AD Pulp 3.4 Na2CO3
drain 93.17 %Pulp Dry 28.25% Sulf%
30 Cool bombs in water before opening 30.53 BD Pulp 100.9 TTA
Take sample of black liquor 43.61 %Yield 97.5 AA
45 Dump chips into wash bag, wash well 26.6 Kappa 83.8 EA
Defibrate pulp in blender, Squeeze dry
homogenize in mixer, spread to dry       BL pH 13.06
           214
      A27
Time Minutes Dig Temp2kr avg H Factor OmegaT kr avg H Factor
1547 0 26.1 0.00        -             -           24.10         0.00       -         -         
1617 30 100 1.02        0.51           0.3           98.8 0.89       0.44       0.2          
1632 45 134.1 37.89      19.45         5.1           132.6 32.73     16.81     4.4          
1642 55 157.2 316.84    177.36       34.7         155.3 268.36   150.55   29.5        
1645 58 164.1 572.05    444.44       56.9         162.1 482.94   375.65   48.3        
1647 60 167.8 779.34    675.69       79.4         165.5 643.45   563.20   67.1        
1649 62 168.6 832.65    805.99       106.3       166 670.93   657.19   89.0        
1651 64 168.8 846.51    839.58       134.3       165.9 665.35   668.14   111.2      
1653 66 168.5 825.81    836.16       162.1       165.7 654.31   659.83   133.2      
1655 68 168.3 812.27    819.04       189.4       165.5 643.45   648.88   154.9      
1659 72 168.1 798.94    805.60       243.2       165.3 632.76   638.10   197.4      
1700 73 168 792.35    795.65       256.4       165.2 627.47   630.12   207.9      
1800 133 168 792.35    792.35       1,048.8    165 617.03   622.25   830.2      
1950 243 168 792.35    792.35       2,501.4    164.9 611.87   614.45   1,956.7   
1952 245 166.9 723.21    757.78       2,526.7    160 403.60   507.74   1,973.6   
1954 247 155.9 282.86    503.04       2,543.4    148.9 151.72   277.66   1,982.8   
1957 250 141.5 76.76      179.81       2,552.4    137 50.12     100.92   1,987.9   
2000 253 128.9 22.71      49.74         2,554.9    124.5 14.58     32.35     1,989.5   
2002 255 120.9 10.07      16.39         2,555.5    116.3 6.21       10.39     1,989.9   
2008 261 100 1.02        5.54           2,556.0    98.8 0.89       3.55       1,990.2   
DigT old DigT New OmT OmT n
ramp 79.4         69.2           67.1       58.5       
cook 2,422.0    2,435.3      1,889.6  1,899.9  
cool 54.6         39.9           33.5       24.0       
total 2,556.0    2,544.4      1,990.2  1,982.3  
time digT kr avg H Factor OmegaT kr avg H Factor
0 26.1 0.00        0 0.0 24.1 0.00       0 0.0
1 28.6 0.00        0.00           0.0 26.6 0.00       0.00       0.0
2 31.0 0.00        0.00           0.0 29.1 0.00       0.00       0.0
3 33.5 0.00        0.00           0.0 31.6 0.00       0.00       0.0
4 36.0 0.00        0.00           0.0 34.1 0.00       0.00       0.0
5 38.4 0.00        0.00           0.0 36.6 0.00       0.00       0.0
6 40.9 0.00        0.00           0.0 39.0 0.00       0.00       0.0
7 43.3 0.00        0.00           0.0 41.5 0.00       0.00       0.0
8 45.8 0.00        0.00           0.0 44.0 0.00       0.00       0.0
9 48.3 0.00        0.00           0.0 46.5 0.00       0.00       0.0
10 50.7 0.00        0.00           0.0 49.0 0.00       0.00       0.0
11 53.2 0.00        0.00           0.0 51.5 0.00       0.00       0.0
12 55.7 0.00        0.00           0.0 54.0 0.00       0.00       0.0
13 58.1 0.00        0.00           0.0 56.5 0.00       0.00       0.0
14 60.6 0.01        0.01           0.0 59.0 0.00       0.00       0.0
15 63.1 0.01        0.01           0.0 61.5 0.01       0.01       0.0
16 65.5 0.01        0.01           0.0 63.9 0.01       0.01       0.0
17 68.0 0.02        0.02           0.0 66.4 0.01       0.01       0.0
18 70.4 0.02        0.02           0.0 68.9 0.02       0.02       0.0
19 72.9 0.03        0.03           0.0 71.4 0.03       0.02       0.0
20 75.4 0.05        0.04           0.0 73.9 0.04       0.03       0.0
21 77.8 0.07        0.06           0.0 76.4 0.06       0.05       0.0
22 80.3 0.09        0.08           0.0 78.9 0.08       0.07       0.0
23 82.8 0.13        0.11           0.0 81.4 0.11       0.09       0.0
24 85.2 0.17        0.15           0.0 83.9 0.14       0.13       0.0
25 87.7 0.23        0.20           0.0 86.4 0.20       0.17       0.0
26 90.1 0.32        0.27           0.0 88.8 0.27       0.23       0.0
27 92.6 0.43        0.37           0.0 91.3 0.36       0.32       0.0
28 95.1 0.57        0.50           0.0 93.8 0.49       0.43       0.0
29 97.5 0.76        0.67           0.0 96.3 0.66       0.58       0.0
30 100.0 1.02        0.89           0.1 98.8 0.89       0.77       0.0
31 102.3 1.32        1.17           0.1 101.1 1.15       1.02       0.1
32 104.5 1.71        1.52           0.1 103.3 1.49       1.32       0.1
33 106.8 2.21        1.96           0.1 105.6 1.92       1.71       0.1
34 109.1 2.85        2.53           0.2 107.8 2.47       2.20       0.2
35 111.4 3.65        3.25           0.2 110.1 3.17       2.82       0.2
36 113.6 4.67        4.16           0.3 112.3 4.05       3.61       0.3
37 115.9 5.96        5.32           0.4 114.6 5.17       4.61       0.3
38 118.2 7.58        6.77           0.5 116.8 6.57       5.87       0.4
39 120.5 9.62        8.60           0.6 119.1 8.33       7.45       0.6
40 122.7 12.17      10.89         0.8 121.3 10.53     9.43       0.7
           215
      A27
41 125.0 15.35      13.76         1.1 123.6 13.28     11.90     0.9
42 127.3 19.31      17.33         1.3 125.8 16.70     14.99     1.2
43 129.6 24.24      21.78         1.7 128.1 20.95     18.83     1.5
44 131.8 30.34      27.29         2.2 130.3 26.22     23.59     1.9
45 134.1 37.89      34.11         2.7 132.6 32.73     29.48     2.4
46 136.4 47.36      42.62         3.4 134.9 40.82     36.78     3.0
47 138.7 59.05      53.21         4.3 137.1 50.79     45.81     3.7
48 141.0 73.45      66.25         5.4 139.4 63.05     56.92     4.7
49 143.3 91.14      82.30         6.8 141.7 78.07     70.56     5.9
50 145.7 112.82    101.98       8.5 144.0 96.45     87.26     7.3
51 148.0 139.33    126.07       10.6 146.2 118.88   107.66   9.1
52 150.3 171.67    155.50       13.2 148.5 146.19   132.54   11.3
53 152.6 211.04    191.36       16.4 150.8 179.39   162.79   14.0
54 154.9 258.87    234.96       20.3 153.0 219.65   199.52   17.4
55 157.2 316.84    287.85       25.1 155.3 268.36   244.01   21.4
56 159.5 386.61    351.72       31.0 157.6 327.10   297.73   26.4
57 161.8 470.77    428.69       38.1 159.8 397.86   362.48   32.4
58 164.1 572.05    521.41       46.8 162.1 482.94   440.40   39.8
59 166.0 668.13    620.09       57.1 163.8 557.76   520.35   48.4
60 167.8 779.34    723.73       69.2 165.5 643.45   600.60   58.5
61 168.2 805.58    792.46       82.4 165.8 657.05   650.25   69.3
62 168.6 832.65    819.12       96.1 166.0 670.93   663.99   80.4
63 168.7 839.56    836.10       110.0 166.0 668.13   669.53   91.5
64 168.8 846.51    843.03       124.0 165.9 665.35   666.74   102.6
65 168.7 836.10    841.31       138.1 165.8 659.81   662.58   113.7
66 168.5 825.81    830.95       151.9 165.7 654.31   657.06   124.6
67 168.4 819.01    822.41       165.6 165.6 648.86   651.58   135.5
68 168.3 812.27    815.64       179.2 165.5 643.45   646.15   146.3
69 168.3 812.27    812.27       192.7 165.5 643.45   643.45   157.0
70 168.2 805.58    808.92       206.2 165.4 638.08   640.76   167.7
71 168.2 805.58    805.58       219.7 165.4 638.08   638.08   178.3
72 168.1 798.94    802.26       233.0 165.3 632.76   635.42   188.9
73 168.0 792.35    795.65       246.3 165.2 627.47   630.12   199.4
133 168.0 792.35    792.35       1038.6 165.0 617.03   622.25   821.6
243 168.0 792.35    792.35       2491.3 164.9 611.87   614.45   1948.1
244 168.0 792.35    792.35       2504.5 164.9 611.87   611.87   1958.3
245 166.9 723.21    757.78       2517.1 160.0 403.60   507.74   1966.8
246 161.4 454.98    589.10       2526.9 154.5 249.03   326.31   1972.2
247 155.9 282.86    368.92       2533.1 148.9 151.72   200.38   1975.6
248 151.1 184.94    233.90       2537.0 144.9 105.62   128.67   1977.7
249 146.3 119.75    152.35       2539.5 141.0 73.02     89.32     1979.2
250 141.5 76.76      98.26         2541.2 137.0 50.12     61.57     1980.2
251 137.3 51.58      64.17         2542.2 132.8 33.49     41.80     1980.9
252 133.1 34.37      42.97         2543.0 128.7 22.19     27.84     1981.4
253 128.9 22.71      28.54         2543.4 124.5 14.58     18.38     1981.7
254 124.9 15.18      18.95         2543.7 120.4 9.56       12.07     1981.9
255 120.9 10.07      12.62         2544.0 116.3 6.21       7.88       1982.0
256 117.4 6.99        8.53           2544.1 113.2 4.46       5.34       1982.1
257 113.9 4.82        5.91           2544.2 110.1 3.19       3.83       1982.2
258 110.5 3.30        4.06           2544.3 107.1 2.27       2.73       1982.2
259 107.0 2.25        2.78           2544.3 104.0 1.61       1.94       1982.3
260 103.5 1.52        1.88           2544.3 100.9 1.13       1.37       1982.3
261 100.0 1.02        1.27           2544.4 98.8 0.89       1.01       1982.3
           216