Hydrothermal Treatment for Biofuels: Lignocellulosic Biomass to Bioethanol, 
Biocrude, and Biochar 
 
by 
 
Sandeep Kumar 
 
 
 
 
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 14, 2010 
 
 
 
 
Keywords: subcritical, supercritical, biocrude, liquefaction,  
pretreatment, carbonization 
 
Copyright 2010 by Sandeep Kumar 
 
 
Approved by 
 
Ram B. Gupta, Chair, Professor of Chemical Engineering 
Christopher B. Roberts, Professor of Chemical Engineering 
Yoon Y. Lee, Professor of Chemical Engineering 
Sushil Adhikari, Assistant Professor of Biosystem Engineering 
Abstract 
 
 Biofuels are viewed as an alternative, renewable fuel and an important component 
of energy security for the future. The dissertation work is directly aimed at supporting the 
commercial production of biofuels from lignocellulosic biomass. The prime objective is 
to apply chemical engineering fundamentals for capitalizing on the extraordinary solvent 
properties of water at elevated temperature and support green and sustainable chemistry. 
Hydrothermal treatment refers to processing of biomass in sub- and super-critical water 
medium.  
Microcrystalline cellulose (MCC) was pretreated (Chapter 2) using subcritical 
water in a continuous flow reactor for enhancing its enzymatic digestibility. The degree 
of polymerization of cellulose steadily decreased with an increase in the pretreatment 
temperature, with a rapid drop occurring above 300?C. The partial transformation of 
cellulose I to II polymorph was noticed in the MCC treated at ? 300 ?C in subcritical 
water.  A nearly three-fold increase in the initial enzymatic reactivity was observed for 
the sample treated at 315?C. The hydrothermal pretreatment of switchgrass (Chapter 3) 
was conducted in a flow through reactor to enhance and optimize the enzymatic 
digestibility of pretreated biomass. More than 80% of glucan digestibility could be 
achieved at 190?C in hydrothermal medium. Addition of K
2
CO
3
 (0.45-0.9 wt%) in water 
helped in enhancing the enzymatic activity of biomass. An alternate pathway for the 
utilization of liquid hydrolyzate was developed and sugar loss in the liquid fraction 
during pretreatment was recovered as high heating value solids. 
ii 
 
Liquefaction of cellulose in sub- and super-critical water was studied (Chapter 4) 
in a continuous flow reactor. The focus of this study was to maximize the yield of sugar 
products (oligomers and monomers) from cellulose hydrolysis. About 65% of cellulose 
converted to the oligomers and monomers at 335?C in 4.8 s and also at 354?C in 3.5 s. In 
the supercritical region, the produced oligomers and monomers partially degraded to 
degradation products. 
Liquefaction of switchgrass (Chapter 5) in subcritical water was studied to 
produce biocrude. The effects of reaction temperature and catalysis by K
2
CO
3
 were 
examined. Potassium carbonate significantly enhanced the hydrolysis of switchgrass 
components into water soluble products. More than 50 wt% of organic carbon available 
in switchgrass was converted to biocrude at 235?C in the presence of 0.15 wt% of 
K
2
CO
3
. Biocrude contained oxygenated hydrocarbons. The subcritical water treatment 
caused the complete breakdown of lignocellulosic structure of switchgrass.  
 Hydrothemal carbonization of switchgrass (Chapter 6) was studied to produce the 
high energy density (coal-like) biochar. The effects of temperature, residence time, and 
pressure on the yield and heating value of biochar were examined. Besides the solid fuel 
application, biochar was also studied for its sorption properties for removing heavy metal 
contaminants from ground water. The batch adsorption results (Chapter 7) with uranium 
[U(VI)] as a solute showed that biochar can be a potential low cost adsorbent for such 
application.  
Future work (Chapter 8) involves the upgrade of biocrude to liquid / gaseous fuel. 
Also pyrolysis / gasification properties of biochar may be investigated. Biochar 
adsorption properties may be further tested for removing other metal contaminants. 
iii 
 
Acknowledgments 
 
I would like to express my sincere gratitude to my advisor Dr. Ram B. Gupta, for 
his continued guidance, discussion, motivation, and support during my dissertation work. 
It would have been impossible for me to complete this work without his mentoring and 
moral support. 
I would like to express my sincere thanks to the committee members, Dr. 
Christopher B. Roberts, Dr. Yoon Y. Lee, and Dr. Sushil Adhikari for their valuable 
comments and suggestions. 
Dr. Y.Y. Lee?s research group has been very supportive and accommodating. I 
have frequently used their analytical instruments. I would like to thank all his group 
members. Special thanks to Ms. Urvi Kothari for supporting me in sample analyses and 
also conducting collaborative studies. 
Chemical engineering staff, Karen Cochran, Sue Allen, Georgetta Dennis, and 
Jennifer Harris deserves special thanks for helping me with logistic, academic and 
financial matters. Brian Schwieker (Engineering Technician) needs special thanks for his 
great help in building up the experimental setup and prompt maintenance support. I also 
enjoyed the constant support from Burak Aksoy and David J. Webster in doing the 
industrial projects for the Alabama Centre for Paper and Bioresource Engineering.
iv 
 
My group colleagues, Adam Byrd, Ganesh Sanganwar, Melinda Hemmigway, Dr. 
Lingzhao Kong, Hema Ramsrun, Courtney Ober, and Jin Yang need special appreciation 
for their assistance and creating a joyful environment in the laboratory. 
I would like to extend my sincere thanks to my friend Rajesh Gupta and his 
family for being a great support during my transition period and thereafter. Rajesh has 
provided invaluable suggestions, technical inputs and moral supports. I also thank him for 
doing a collaborative study and co-authoring a research article. I would also like to thank 
my dear friends Chaudhary Devraj Nayak and Vijay A. Loganathan for their constant 
encouragements and emotional supports.  
 My parents deserve credit for all my achievements. They have supported me 
whole heartily in all my endeavors. My mother who passed away recently will remain the 
source of inspiration throughout my life. She inculcated the fighting spirit in me. 
Continued support of my wife, Sweta Kumari is the most important factor which helped 
me in carrying out the family responsibilities along with the research and academic work. 
Love and sweet smile of my son ?Shashank? is always a source of rejuvenation after 
exhaustive work. I wish to thank my brother, sister and entire family members for their 
love and support.  
I want to thank National Science Foundation (grant # DE-FC26-05424.56) and 
Alabama Centre for Paper and Bioresource Engineering for their continued financial 
support for my research study.  
v 
 
vi 
 
Table of Contents 
Abstract 
Acknowledgements 
List of Figures 
List of Tables 
1 Introduction 
1.1 Importance of Biofuels 
1.2 Lignocellulosic Biomass 
1.3 Conversion Routes for Lignocellulosic Biomass to Biofuels 
1.3.1 Thermochemical Pathways 
1.3.2 Biochemical Pathways 
1.4 Pretreatment Methods for Bioethanol 
1.4.1 Physical 
1.4.2 Chemical Pretreatments 
1.4.3 Hydrothermal Pretreatments 
1.4.4 Physiochemical Pretreatments 
1.5 Supercritical Fluid Technology 
1.6 Sub-and Supercritical water 
ii
iv
xii
xviii
1
1
2
4
5
8
8
9
9
13
14
15
16
vii 
 
1.7 Sub- and Supercritical Water Technology for Biofuels 
1.7.1 Biochar 
1.7.2 Bioethanol 
1.7.3 Biocrude 
1.7.4 Hydrothermal Gasification 
1.8 Challenges of Hydrothermal Processing 
1.9 Research Plan 
1.9.1 Pretreatment of Biomass for Bioethanol 
1.9.2 Liquefaction of Biomass for Biocrude 
1.9.3 Carbonization of Biomass for Biochar 
1.10 References 
2 Cellulose pretreatment in subcritical water: Effect of temperature on 
molecular structure and enzymatic reactivity 
2.1 Abstract 
2.2 Introduction 
2.3 Experimental Section 
2.4 Product Analyses 
2.5 Results and Discussion 
2.5.1 Solid Analysis 
2.5.2 Liquid Analysis 
2.6 Conclusion 
2.7 Acknowledgment 
19
21
22
23
26
26
27
28
28
30
31
41
41
42
46
49
52
53
64
66
67
viii 
 
2.8 Nomenclature 
2.9 References 
3 Hydrothermal Pretreatment of Switchgrass and Corn Stover for 
Production of Ethanol and Carbon Microspheres 
3.1 Abstract 
3.2 Introduction 
3.3 Experimental Section 
3.4 Product Characterization 
3.4.1 Pretreatment 
3.4.2 Carbonization 
3.5 Results and Discussions 
3.5.1 Pretreatment 
3.5.2 Hydrothermal Carbonization 
3.6 Conclusion 
3.7 Acknowledgment 
3.8 References 
4 Hydrolysis of Microcrystalline Cellulose in Subcritical and 
Supercritical Water in a Continuous Flow Reactor 
4.1 Abstract 
4.2 Introduction 
4.3 Experimental Section 
4.4 Product Analyses 
67
70
73
73
74
79
84
84
86
87
87
98
106
107
107
115
115
116
119
122
ix 
 
4.5 Results 
4.6 Discussion 
4.7 Conclusion 
4.8 Acknowledgment 
4.9 Product Nomenclature 
4.10 References 
5 Biocrude Production from Switchgrass using Subcritical Water 
5.1 Abstract 
5.2 Introduction 
5.3 Experimental Section 
5.4 Product Separation 
5.5 Product Characterization 
5.6 Results 
5.7 Discussion 
5.8 Conclusion 
5.9 Acknowledgment 
5.10 References 
6 Production of Biochar from Switchgrass by Hydrothermal 
Carbonization 
6.1 Abstract 
6.2 Introduction 
6.3 Experimental Section 
123
133
137
138
138
138
144
144
145
149
152
154
155
165
169
170
170
177
177
178
181
x 
 
6.4 Product Characterization 
6.5 Results and Discussion 
6.6 Conclusion 
6.7 Acknowledgment 
6.8 References 
7 Application of Biochar for the Removal of Heavy Metal Contaminants 
from Ground Water  
7.1 Introduction 
7.2 Experimental Section 
7.3 Results and Discussion 
7.4 Conclusion 
7.5 References 
8 Summary and Future Works 
Appendix A: Structure and Composition of Lignocellulosic Biomass 
A.1 Introduction 
A.2 Composition and Structure 
A.3 References 
Appendix B: Summary of Pretreatment Methods 
Appendix C: Reaction Pathway of Cellulose, Hemicelluloses, and Lignin in 
Hydrothermal Medium 
C.1 Cellulose Reaction Pathways 
C.2 Hemicelluloses Reaction Pathways 
184
185
198
198
199
203
203
205
206
211
212
214
217
217
217
225
227
229
229
230
xi 
 
C.3 Lignin 
C.4 References 
Appendix D: Salt Solubility in Hydrothermal Medium 
D.1 Salt-Water Binary System 
D.2 References 
Appendix E: Journal Publications and Conference Presentations 
231
232
234
234
236
237
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
xii 
 
List of Figures 
Figure 1.1 Annual biomass resource potential from forest and agricultural 
resources. 
2
Figure 1.2 Typical composition of lignocellulosic biomass. 3
Figure 1.3 Major pathways for the conversion of biomass to biofuels. 5
Figure 1.4 Physical properties of water with temperature, at 24 MPa. 18
Figure 1.5 Hydrothermal treatment applications referenced to pressure-
temperature phase diagram of water. 
20
Figure 1.6 Generation of in-situ alkali catalyst nanoparticles in hydrothermal 
reactor. 
29
Figure 2.1 Ionization constant (Kw) of water as a function of temperature 44
Figure 2.2 Apparatus for cellulose pretreatment in subcritical water. 46
Figure 2.3 Degree of polymerization (DPv) versus pretreatment temperature. 54
Figure 2.4 XRD results for MCC and group I samples. 56
Figure 2.5 XRD results for original MCC and group II samples. 57
Figure 2.6a Plot after regression analysis of enzymatic hydrolysis data at 3.5 
FPU and after 1 h. 
59
xiii 
 
Figure 2.6b Plot after regression analysis of enzymatic hydrolysis data at 3.5 
FPU and after 24 h. 
Figure 2.6c Plot after regression analysis of enzymatic hydrolysis data at 60 
FPU and after 1 h. 
Figure 2.6d Plot after regression analysis of enzymatic hydrolysis data at 60 
FPU and after 24 h. 
Figure 2.7. SEM images of samples: (a) MCC, (b) run 5, (c) run 7, and (d) 
run 9. 
Figure 2.8 DSC thermograms of MCC, group I (run 6) and group II (run 7 
and 9) samples. 
Figure 2.9 Simplified reaction scheme of cellulose in subcritical water. 
Figure 3.1 Apparatus for the hydrothermal pretreatment of biomass. 
Figure 3.2 Typical temperature cycle during pretreatment process 
Figure 3.3 Batch reactor for hydrothermal carbonization process 
Figure 3.4 Percentage removal of hemicelluloses and lignin with temperature 
for the switchgrass pretreated at 3.4 MPa and 20 min steady operation time 
Figure 3.5 Glucan digestibility (72 h) of hydrothermally pretreated 
switchgrass 
Figure 3.6 Glucan digestibility (72 h) of switchgrass pretreated at 150?C in 
the presence of K
2
CO
3
 
Figure 3.7 SEM images of untreated and pretreated (run 3) switchgrass taken 
at 600X magnification 
59
60
60
62
63
69
80
82
82
87
89
90
91
xiv 
 
Figure 3.8 Glucan digestibility (72 h) of K
2
CO
3 
(0.9 wt%) assisted 
hydrothermally pretreated switchgrass 
Figure 3.9 Percentage removal of hemicelluloses and lignin with temperature 
for the switchgrass pretreated in the presence of 0.9 wt% of K
2
CO
3
 
Figure 3.10 Material balance for hydrothermally pretreated switchgrass (run 
6) at 190?C 
Figure 3.11 Material balance for K
2
CO
3
 assisted hydrothermally pretreated 
switchgrass (run 7) at 190?C 
Figure 3.12 XRD patter of untreated and pretreated (run 6 and 7) switchgrass 
Figure 3.13 Glucan digestibility (72 h) of untreated and pretreated corn 
stover 
Figure 3.14 SEM images of untreated and pretreated (run 8) corn stover 
taken at 600X magnification 
Figure 3.15 Percentage conversion of liquid hydrolyzate (based on TOC of 
input) of run 6 and glucose solution 
Figure 3.16 SEM images (at 20,000X magnifications) of carbon 
microspheres produced from (a) glucose solution and (b) liquid hydrolyzate 
of run 6 
Figure 3.17 GC-MS spectra of liquid product from (a) run IV, (b) run V, and 
(c) run VI 
Figure 3.18 Simplified reaction pathway of glucose during hydrothermal 
carbonization reactions 
91
92
93
93
96
97
98
100
100
102
104
xv 
 
Figure 3.19 FTIR spectra of solids recovered in run II and V 
Figure 4.1 Apparatus for cellulose hydrolysis in subcritical and supercritical 
water 
Figure 4.2 Yield of hydrolysis products obtained from cellulose treatment in 
subcritical and supercritical water at 27.6 MPa 
Figure 4.3 Yield of glucose obtained from cellulose treatment in subcritical 
and supercritical water at 27.6 MPa 
Figure 4.4 HMF yield in liquids obtained from cellulose treatment in 
subcritical and supercritical water at 27.6 MPa 
Figure 4.5 TOC in liquid products obtained from cellulose treatment in 
subcritical and supercritical water at 27.6 MPa 
Figure 4.6 HPLC chromatograph of run 5 (335?C, and 4.7 s residence time) 
Figure 4.7 HPLC chromatograph of run 7 
Figure 4.8 HPLC chromatograph of run 14 (405 ?C, and 2.5 s residence time) 
Figure 4.9 Products obtained at 302?C, 5.2 s, 27.6 MPa in the presence of 
(a) 0% K
2
CO
3
, (b) 0.22% K
2
CO
3
, and (c) 0.44% K
2
CO
3
. 
Figure 5.1 Ionization constant (K
w
) of water as a function of temperature at 
25 MPa 
Figure 5.2 Major pathways for the conversion of biomass to biofuels 
Figure 5.3 Apparatus for biocrude production 
Figure 5.4 Heating profile in a typical experiment 
Figure 5.5 Products separation scheme 
104
121
125
125
126
127
139
130
131
132
146
147
150
151
153
xvi 
 
Figure 5.6 Total organic content (TOC) of aqueous phase versus steady 
operation time 
Figure 5.7 Conversion of switchgrass (weight loss) for runs 6, 7, and 8 
Figure 5.8 Composition of aqueous phase from run 8 by HPLC analysis 
Figure 5.9 XRD patterns for untreated switchgrass and solid precipitate from 
run 8 
Figure 5.10 XRD pattern of untreated switchgrass and biochars from runs 6 
and 8 
Figure 5.11 FTIR spectra of untreated switchgrass and biochar from run 8 
Figure 5.12 SEM image of (a) switchgrass, and (b) biochar from run 8 
Figure 6.1 Apparatus for HTC process 
Figure 6.2 Products separation scheme in two stage HTC process 
Figure 6.3 Yield (%) of obtained biochar at 200?C and 230?C 
Figure 6.4 HHV of obtained biochar at 200?C and 230?C 
Figure 6.5 Yield (%) of obtained biochar at 200-280?C 
Figure 6.6 HHV of obtained biochar at 200-280?C 
Figure 6.7 Yield (%) of obtained biochar at 280?C 
Figure 6.8 HHV of obtained biochar at 280?C 
Figure 6.9 FTIR spectra of switchgrass and biochar, reaction time 4.0 h 
A-switchgrass, B-200?C, C-230?C, D-260?C, E-280?C 
Figure 6.10 SEM image of switchgrass (A) and biochar (B) obtained at 
260?C, 4.0 h 
Figure 6.11 XRD patterns of untreated switchgrass and biochar obtained at 
156
157
159
 161
162
163
164
182
184
187
187
189
189
190
191
192
194
xvii 
 
280?C, 4.0 h 
Figure 6.12 Conversion efficiency of carbon from biocrude 
Figure 6.13 SEM image of biotar particles (HTC of biocrude obtained at 
200?C) obtained at 200?C, 2.0 h 
Figure 7.1a TG and DTG pattern of switchgrass under inert atmosphere 
Figure 7.1b TG and DTG pattern of biochar under inert atmosphere 
Figure 7.2 Kinetics of U(VI) adsorption on to biochar at different solid 
loadings and initial U(VI) concentrations 
Figure 7.3 Adsorption isotherm of U(VI) on to biochar. SSR: 5 g L
-1
 pH: 3.9 
? 0.2; ionic strength: 0.1 M NaNO
3 
Figure 7.4 Adsorption of U(VI) on to biochar at different SSR. pH: 3.9 ? 0.2; 
ionic strength: 0.1 M NaNO
3 
Figure A.10 Inter-conversion of polymorphs of cellulose  
Figure A.11 Lignin monomers (a) trans-p-coumaryl alcohol, (b) coniferyl 
alcohol, and (c) sinapyl alcohol 
Figure C.12 Key reaction products from cellulose hydrolysis in catalyst free 
hydrothermal medium 
Figure C.13 A segment of hardwood xylan, O-acetyl-4-O-
methylglucuronoxylan
 
Figure C.14 Reaction pathway of lignin in supercritical water 
195
196
197
208
208
209
210
211
219
222
230
231
232
 
xviii 
 
List of Tables 
Table 1.1 Comparison of liquefaction and pyrolysis. 6
Table 1.2 Comparison of ambient and supercritical water. 18
Table 2.1 Summary of experimental conditions (at 27.6 MPa). 48
Table 2.2 Lattice planes of cellulose I and II and the diffraction angles in 
XRD spectra. 
50
Table 2.3 Enzymatic digestibility of MCC, group I, and group II samples. 53
Table 2.4. Crystallinity of samples by XRD analysis. 55
Table 2.5 Yields of different compounds in the liquid product. 64
Table 3.2 Pretreatment details conducted at 3.4 MPa and 20 min steady 
operation time and the weight loss of switchgrass after pretreatment 
81
Table 3.3 Details of hydrothermal carbonization experiments 84
Table 3.4 BET surface area of switchgrass and samples from run 6 and 7 95
Table 3.5 Sugar compounds analyzed by HPLC in liquid hydrolyzate 
collected at 190?C during run 6 99
Table 3.6 Key compounds identified in the product liquid from run IV, V, 
and VI and their trend with reaction time 102
Table 3.7 Functional groups in the FTIR spectra of carbon rich microspheres 
from run II and V 104
xix 
 
Table 4.1 HPLC standards used for analysis of liquid products 
Table 4.2 Liquid products yield from cellulose treatment in subcritical and 
supercritical water at 27.6 MPa 
Table 4.3 Carbon balance in the liquid products obtained from subcritical and 
supercritical treatment of cellulose at 27.6 MPa 
Table 4.4 Experimental conditions of cellulose runs in presence of catalyst 
K
2
CO
3
 in subcritical water 
Table 5.1 Composition (wt%) of switchgrass 
Table 5.2 Experiment details of various runs 
Table 5.3  Biocrude yield from switchgrass conversion (wt%, carbon basis) 
Table 5.4 Identified compounds in biocrude from run 8 by GC-MS analysis  
Table 5.5 Position of infrared absorption bands of switchgrass 
Table 6.1 Comparison of biomass and biochar as solid fuel 
Table 6.2 Main physical and chemical properties of the switchgrass 
Table 6.3 Experiment summary and results 
Table 6.4 FTIR spectrum of the switchgrass and biochar 
Table 6.5 Key compounds in biocrude identified during HPLC analysis 
Table 7.5 Composition of switchgrass and biochar 
Table A.6 Composition and degree of polymerization (DP) of hemicelluloses 
Table B.7 Summary of the important pretreatment methods 
Table D.8 Classification of salt-water binary system 
 
120
124
128
132
149
152
158
160
163
178
182
186
193
196
207
221
227
235
 
1 
 
1. Introduction 
 
1.1 Importance of Biofuels 
Energy security, sustainability, and climate change concerns has led the world to 
look for renewable and alternative energy resources. The large scale substitution of 
petroleum based fuels and products with renewable sources are needed to minimize the 
environmental issues.
1
 Biomass is the world?s fourth largest energy source worldwide, 
following coal, oil and natural gas. Biomass is an attractive feedstock for fuel and 
biomaterials due to the three main reasons. First, it is a carbon-neutral renewable resource 
that could be sustainably developed for the production of bioenergy and biomaterials. 
Second, it is environmentally benign, as it does not add to the green house gas emission, 
and possibly reduces NO
x
 and SO
x
 depending on the fossil-fuels displaced. Though, 
when combusted in traditional stoves, emission of polycyclic aromatic hydrocarbons, 
dioxins, furans, and heavy metals is a concern.
2
 Third, it appears to have a significant 
economic potential given the fluctuating prices of the fossil fuels. Moreover, 
development of bio-based economy brings the opportunity for rural empowerment and 
also the energy security, since biomass resource is distributed all over the world. 
United States is a net importer of petroleum products. As per the department of 
energy sources, in the year 2006, US had 2% of world oil reserve share and used 24% of 
the world supply.
3
 Nearly 60% of the oil is imported to meet the energy demand. 
In recent years, government has sponsored numerous research studies and 
projects, for developing abundantly available biomass resources of the country for 
biofuels applications. The billion ton vision report concludes that land resources of US 
are capable of producing sustainable supply of biomass sufficient to displace 30% or 
more of current petroleum consumption with biofuels by year 2030.
4
  Figure 1.1 show 
that more than a billion ton annual biomass resource potential from forest and agricultural 
resources are available for sustainable biofuels production. 
0
300
600
900
1200
1500
Forest 
Resources
Agricultural 
Resources
Total Resource 
Potential
368
998
1366
M
illio
n
 
d
r
y
 
t
o
n
 p
e
r
 y
e
a
r
 
Figure 1.1 Annual biomass resource potential from forest and agricultural resources.
4
 
Energy Independence and Security Act (EISA) 2007 requires replacing 20% of 
fossil by renewable fuel by 2017. The major focus is on non food resources or so called 
?second? generation biofuels from lignocellulosic biomass.  
1.2 Lignocellulosic Biomass 
Lignocellulose is a generic term for describing the main constituents in most 
plants, namely cellulose, hemicelluloses, and lignin (Figure 1.2). It is the non starch 
2 
 
based fibrous part of plant material. Lignocellulsoic biomass is renewable and abundantly 
available resource, and is considered as potential feedstock for second generation 
biofuels.  
 
Figure 1.2 Typical composition of lignocellulosic biomass.
5
 
The components of lignocelluloses form three-dimensional polymeric composites 
to provide structure and rigidity to the plant. Cellulose is an unbranched chain of 
anhydroglucose (C
6
H
10
O
5
) linked head to tail by ?-glycosidic linkages. The ?-linkages in 
cellulose form linear chains that are highly stable and resistant to chemical attack because 
of the high degree of intra and intermolecular hydrogen bonding.
6
 Hemicelluloses are a 
class of polymers of sugars including the six carbon sugars (mannose, galactose, glucose 
and 4-O-methyl-D-glucuronic acid) and five carbon sugars (xylose and arabinose). 
Unlike cellulose, hemicelluloses possess side chains.
7
 Lignin is a complex network of 
different phenyl propane units (p-cocumaryl alchohol, coniferyl alcohol, sinapyl alcohol). 
In general, cellulose provides rigidity. Hemicelluloses stick and glue to cellulose via 
hydrogen bonds, to create polysaccharide microfibrils. Lignin mainly acts as adhesive or 
3 
 
4 
 
binder that provides strength and structure to the cellular composites of the plant and 
protects against microorganism or chemical attack. It controls the fluid flow, acts as 
antioxidant by absorbing UV lights and stores energy. When lignin binds, it crosslinks 
with the regular structure of the microfibrils made of cellulose and hemicelluloses. The 
composition of lignocellulosic materials varies with several factors such as type of plant, 
growth conditions, on the part of plant, and on the age of harvesting.
8-10
 The structural 
and composition details of the lignocellulosic materials have been discussed further in 
Appendix-A. 
1.3 Conversion Routes for Lignocellulosic Biomass to Biofuels 
Lignocellulosic biomass consists of a variety of materials with distinctive physical 
and chemical characteristics. Typically it is categorized into either woody, herbaceous, or 
crop residues. Most of these biomass are already used, without preliminary conversion, as 
a fuel for heating purpose and also to produce steam for generating electricity. Direct 
combustion is best suited to biomass having low contents of moisture and ash. In fact, 
until the start of 20
th
 century, biomass and coal were the major sources of fuels and 
chemicals. The recent energy crisis, fluctuating oil prices and political factors associated 
with the import of fossil fuels has brought the focus again on the utilization of abundantly 
available biomass resources for producing easy to handle forms of energy such as gases, 
liquids, and charcoal.
11, 12
 Biomass may be converted to energy by many different 
processes, depending on the raw characteristics of the material and the type of energy 
desired. Figure 1.3 shows some of the major competing conversion pathways for 
converting the biomass to liquid fuel, chemicals and hydrogen. 
?
 
Figure 1.3 Major pathways for the conversion of biomass to biofuels. 
1.3.1 Thermochemical Pathways 
 Thermochemical processes depend on the relationship between heat and chemical 
action as a means of extracting and creating products and energy. Pyrolysis, gasification, 
and liquefaction which are conducted at a temperature of several hundred degrees celsius 
are categorized in thermochemical processes. 
Pyrolysis: Pyrolysis is defined as the thermal degradation of biomass in the absence of 
oxygen to produce condensable vapors, gases, and charcoal; in some instances a small 
amount of air may be admitted to promote this endothermic process. The products of 
pyrolysis can be gas, liquid, and/or solid. In flash pyrolysis, biomass is rapidly heated 
(e.g., at rates of 100-10,000?C/s) to 400-600?C, while limiting the vapor residence time to 
less than 2 seconds.
13
  The oil production is maximized at the expense of char and gas. 
Pyrolysis processes typically use dry and finely ground biomass.   Pyrolysis and direct 
liquefaction processes are sometimes confused with each other, and a simplified 
5 
 
6 
 
comparison of the two follows. Both are thermochemical processes in which feedstock 
organic compounds are converted into liquid products. In the case of liquefaction, 
feedstock macro-molecules are decomposed into fragments of light molecules in the 
presence of a suitable catalyst. At the same time, these fragments, which are unstable and 
reactive, repolymerize into oily compounds having appropriate molecular weights.
14
 With 
pyrolysis, on the other hand, a catalyst is usually unnecessary, and the light decomposed 
fragments are converted to oily compounds through homogeneous reactions in the gas 
phase. The differences in operating conditions for liquefaction and pyrolysis are shown in 
Table 1.1. 
Table 1.1 Comparison of liquefaction and pyrolysis. 
Process                       Temperature (?C)            Pressure (MPa)                 Biomass drying 
Liquefaction                       250-350                       5-20                              Not needed 
Pyrolysis                            450-600                     0.1-0.5                             Needed 
 
Gasification: In gasification oxygen-deficient thermal decomposition of organic matter 
primarily produces synthesis gas. Gasification can be thought of as a combination of 
pyrolysis and combustion. Gasification has a good potential for near-term commercial 
application due to the benefits over combustion including more flexibility in terms of 
energy applications, higher economical and thermodynamic efficiency at smaller scales, 
and potentially lower environmental impact when combined with gas cleaning and 
refining technologies. An efficient gasifier decomposes high-molecular-weight organic 
compounds released during pyrolysis into low-molecular-weight, non condensable 
compounds in a process referred to as tar cracking. Undesirable char that is produced 
7 
 
during gasification participates in a series of endothermic reactions at temperatures above 
800?C which converts carbon into a gaseous fuel. Typically gaseous products include: 
CO, H
2
, and CH
4
. Fisher-Tropsch synthesis can be used to convert the gaseous products 
into liquid fuels through the use of catalysts. Gasification requires feedstock that contains 
less than 10% moisture.
15, 16
 
Hydrothermal Liquefaction: Hydrothermal liquefaction of biomass is conducted near 
(subcritical) or above (supercritical) the critical point of water (374?C, 22.1 MPa). 
Density and dielectric constant of the water medium play major role in soubilizing 
organic compounds in subcritical and supercritical water. Dielectric constant of water 
near the critical point decreases considerably, which enhances the solubility of organic 
compounds. Use of sub- and supercritical water (hydrothermal medium) for the 
liquefaction of biomass is one of the most promising technologies to produce aqueous 
carbohydrate solution (biocrude) derived from the direct liquefaction of biomass. Water 
acts both as a reactant and as a solvent. Water as reactant favors the hydrolysis reactions 
over thermal decomposition and rapidly degrades the polymeric structure of biomass 
components to water soluble products. The intermediates formed during the reactions 
show a high solubility in sub- and supercritical water; hence reaction steps are mainly 
homogeneous.
  
Diffusion related problems such as mass transfer through the interface are 
not encountered in homogeneous reaction.
17-19
 The process can utilize mixed biomass 
feedstock without any pretreatment at a comparatively low temperature. Moisture content 
of biomass is not an issue. Moreover, biomass residue generated from the any other 
processes can be utilized to produce biocrude. 
8 
 
1.3.2 Biochemical Pathways 
 Biochemical processes takes place at ambient to slightly higher temperature levels 
using a biological catalyst to bring out the desired chemical transformation. Ethanol from 
lignocellulosic biomass is produced mainly via biochemical routes. The biomass is first 
pretreated by different pretreatment methods (discussed later) for the improving the 
accessibility of enzymes. After the pretreatment, biomass goes through the enzymatic 
hydrolysis for conversion of polysaccharides into monomer sugars such glucose, xylose 
etc. Subsequently sugars are fermented to ethanol by the use of different microorganisms. 
Pretreated biomass can directly be converted to ethanol by using the process called 
simultaneous saccharification and fermentation (SSF).
20
 Therefore, pretreatment is a 
crucial step to overcome lignocellulose recalcitrance in the conversion of biomass to 
ethanol.
21
 Pretreatment of biomass is necessary for enhancing the enzymatic hydrolysis of 
biomass. It alters the physical and chemical properties of biomass and improves the 
enzyme access and effectiveness by: 
? Altering or removing lignin and hemicelluloses, 
? Changing the crystallinity of cellulose, 
? Removing acetyl groups from hemicelluloses, 
? Reducing the degree of polymerization in cellulose, and 
? Increasing internal surface area and pore volume. 
1.4 Pretreatment Methods for Bioethanol 
Lignocellulosics mainly consist of cellulose, hemicelluloses and lignin which are 
bonded together by covalent bonding, various intermolecular bridges, and van der Waals 
forces forming a complex structure, making it resistant to enzymatic hydrolysis and 
9 
 
insoluble in water.
22, 23
  Therefore, an efficient, less energy intensive and cost effective 
pretreatment method is a necessity for producing ethanol at economically viable cost.  
Different pretreatment methods are broadly classified into physical, chemical, 
physiochemical and biological processes. 
1.4.1 Physical 
Physical pretreatments such as chipping, grinding, ball milling and colloid milling 
are mainly done for size reduction, increasing surface area, and mechanical 
decrystallization of biomass.
24-27
 It is not certain that the benefit derived from these 
mechanical treatments is because of reduction in crystallinity or increase in surface 
area.
28
 The non mechanical methods such as high temperature, freeze/thaw cycles, and 
irradiation are attempted to change one or more structural features of the cellulose and 
enhance the hydrolysis. Most of these methods are limited in their effectiveness and often 
expansive.
21, 24, 29, 30
 
1.4.2 Chemical Pretreatments 
Chemical pretreatment employs the use of acid / alkali for fractionating biomass, 
increasing surface area and pore volume, and swelling cellulose. There may be two types 
of swelling of cellulose, intercrystalline and intracrystalline. Inercrystalline swelling can 
be affected by water, whereas intracrystalline swelling requires a chemical agent for 
breaking the hydrogen bonding of the cellulose. Some of the most commonly used 
pretreatment methods are discussed below. 
Dilute Acid Pretreatment: Different acidic reagents such as dilute sulfuric acid, dilute 
nitric acid, dilute hydrochloric acid, dilute phosphoric acid and peracetic acid have been 
used for the pretreatment process. Dilute acid pretreatment mainly removes the 
10 
 
hemicellulose fractions from the lignocellulosics. The removal of hemicelluloses 
increases the porosity in the biomass and, thus, enzymatic accessibility to the cellulosic 
fractions is increased. Dilute acid is an efficient pretreatment method suitable for most of 
the lignocellulosic feedstock. Among the other dilute acids (sulfuric acid, hydrochloric, or 
nitric acids), dilute sulfuric acid (0.5-1.5%)  above 160?C was found to be most suitable 
for the industrial application. The method fractionates majority of hemicelluloses (75-
90%). Two approaches for dilute acid pretreatment are followed. In one approach high 
temperature (more than 160?C) in a continuous flow reactor with low solid loading is 
used, while in another approach, low temperature batch process is used with high solids 
loading. The major disadvantage of the process is the removal of acids or neutralization 
which yields a large amount of gypsum before next step of enzymatic hydrolysis. 
Although close to theoretical yields can be achieved, the process requires high capital 
cost coupled with corrosion problems, acid consumption and recovery costs.
21, 24, 31-33
 
Peracetic Acid (C
2
H
4
O
3
): The use of peracetic acid (2-10%) at 25-75?C causes 
significant delignification of lignocellulosics due to the oxidizing action of peracetic acid. 
Reduction in crystallinity due to structural swelling and dissolution of crystalline 
cellulose are observed. The process retains majority of hemicelluloses in solids.
21, 24
 
Concentrated Sulfuric Acid: Sulfuric acid is strong swelling and a hydrolyzing agent. 
Intracrystalline swelling of cellulose occurs at acid concentration between 62.5-70% and 
above 75%, dissolution and decomposition of cellulose takes place. The dissolved 
cellulose is re-precipitated by the addition of methanol or ethanol. After the treatment 
with 75% sulfuric acid, degree of polymerization (DP) of treated cellulose drops from 
2150 to 300. The re-precipitated cellulose is easily hydrolyzed by acid or enzyme with 
11 
 
high conversions. Sulfuric acid can be reused after the distillation of methanol or ethanol. 
Large scale testing of this process is still needed to determine the permissible recycling of 
sulfuric acid.
24
 
Concentrated Phosphoric Acid: Concentrated phosphoric acid (85%) has been applied 
earlier as a cellulose solvent. Phosphoric acid causes less degradation of the cellulose 
than other acids.
34
 Recently, a novel method for lignocelluloses fractionation was applied 
to hardwoods as well as softwoods. The major advantage of the method lies in the 
moderate process condition (50?C and atmospheric pressure). Fractionation of 
lignocelluloses into highly reactive amorphous cellulose, hemicelluloses sugars, lignin 
and acetic acid was achieved.  After the pretreatment, Enzymatic hydrolysis of Avicel and 
alpha-cellulose were completed within 3 h while corn stover and switchgrass were 
hydrolyzed to the extent of 94%.
24, 35
 
Ionic Liquid: Dissolution of cellulose in ionic liquid and subsequent regeneration as 
amorphous cellulose with the use of anti-solvent such as water, ethanol, or methanol has 
attracted some research interest. Hydrolysis of regenerated cellulose is significantly 
increased and the initial rates of hydrolysis are approximately an order of magnitude 
higher than that of untreated cellulose.
36
 Nearly complete conversion of the carbohydrate 
fraction into water-soluble products is readily observed at 120?C in ionic liquids, which 
is much lower than the temperatures typically applied for aqueous-phase hydrolysis.
37
 
Cost and recycling of ionic liquid are the major disadvantage of the process. 
Alkali Pretreatment: Among the various chemical pretreatments, alkali pretreatment is 
most widely used to enhance enzymatic hydrolysis of lignocellulose. Alkali treatment 
selectively removes majority of lignin and part of hemicelluloses, thus the success of 
12 
 
method depends on the amount of lignin in the biomass. The main reagents used for 
alkali pretreatment are sodium hydroxide, ammonia, calcium hydroxide, and oxidative 
alkali (NaOH + H
2
O
2
 or O
3
). Dilute NaOH treatment causes disruption lignin and 
carbohydrate linkages, swelling of cellulose, removal of lignin and hemicelluloses, 
increase in surface area and decrease in the DP.
30
 The mechanism is considered to be 
saponification of intermolecular ester bonds cross-linking the hemicelluloses and lignin. 
The optimum levels range between 5 and 8 g NaOH/100 g substrate. Process depends on 
the type of substrate. It was observed that digestibility of softwood with high lignin 
content increased slightly as compared to hardwoods.
25
 Considering the cost of chemical, 
Holtzapple et al. used calcium hydroxide as alkali for pretreating corn stover and poplar, 
and applied a longer treatment time (few days) and low temperature. Similar to other 
alkaline pretreatment, calcium hydroxide causes large amount of delignification and 
increases the hydrolysis rate.
24, 31, 38-40
 
Ammonia: Treatment with liquid ammonia was first patented in 1905.
24
 Ammonia causes 
a strong swelling action and cellulose changes its crystal structure from cellulose I to 
cellulose III
1
. Ammonia shows high selectivity for lignin show majority of hemicelluloses 
is retained in the solids. Relatively low cost and possibility of recycling due to its volatile 
nature are the major advantages of this process. Two process based on aqueous ammonia 
known as ammonia recycle percolation (ARP) and soaking in aqueous ammonia (SAA) 
were used by Kim et al. to improve the hydrolysis of corn stover close to the quantitative 
maximum.  ARP process also enhanced the enzymatic digestibility of poplar wood.
21, 24, 41
 
Organic Solvent: Pretreatment using organic solvents with mineral acids as catalysts have 
been reported to an effective method of delignification of biomass. Organic solvents 
13 
 
break attacks the lignin carbohydrate bonds selectively and fractionates the lignin and 
hemicelluloses. Methanol, ethanol, acetone, ethylene glycol, triethylene glycol and 
tetrahydrofurfuryl alcohol have been used to dissolve the lignin from the biomass. 
Organic acids such as oxalic, acetylsalicylic, and salicylic acid can be used as catalysts. 
Organic solvents should be recycled for the cost effective use. Moreover, removal of 
organic solvents from the system after pretreatment is necessary because solvents may be 
inhibitory to the growth of organisms, enzymatic hydrolysis, and fermentation. 
1.4.3 Hydrothermal Pretreatment 
 Hydrothermal pretreatment uses hot water at elevated pressure (more than 
saturation pressure) to ensure that water remain in liquid phase. Water acts as a solvent 
and as a reactant at the same time. Water as reactant is seen as a promising path to green 
chemistry by providing alternatives to corrosive acids and toxic solvents.
42
 Dielectric 
constant of water decreases with temperature, and water behaves like a more non-polar 
solvent. Ionization constant of water increases with temperature below critical point. 
Such changes in properties of water with temperature lead to the cleavage of ether and 
ester bonds and favor the hydrolysis of hemicelluloses.  Hydrothermal treatment of 
lignocellulosic biomass generates acid that arises from the thermally labile acetyl groups 
of hemicelluloses and catalyzes the hydrolysis of hemicelluloses and subsequent 
solubilization. Xylose recovery from biomass can be achieved as high as 88-98%. The 
structural alterations due to the removal of hemicelluloses increase the accessibility and 
enzymatic hydrolysis of cellulose. Process librates organic acids from the biomass, and 
pH of the reaction medium decreases during the treatment.  To avoid the formation of 
inhibitors, the pH should be kept between 4 and 7 during the pretreatment. Maintaining 
14 
 
the pH between 4 and 7 minimizes the formation of monosaccharides, and therefore also 
the formations of degradation products that can further catalyze hydrolysis of the 
cellulosic material during pretreatment.
7, 21, 24, 43-46
 
1.4.4 Physiochemical Pretreatments 
Some of the pretreatment methods combine physical and chemical methods. High 
pressure steaming, with or without rapid decompression (explosion) is considered one of  
the most successful method option for fractionating wood into its three major 
components. 
Steam Treatment (Autohydrolysis) and Steam Explosion: Heating of biomass in the 
presence of steam-saturated water (about 200?C and 1.5 MPa) leads to organic acid 
generation. The evolved acid hydrolyzes some of the hemicelluloses and alters the lignin 
structure. The structural changes in biomass after the treatment enhance its enzymatic 
digestibility.
21
 
If, autohydrolysis is followed by rapid pressure release, the liquid water inside the 
biomass explosively vaporizes which shatters the biomass in popcorn like effect and thus, 
increases surface area. This approach combines both chemical and pretreatment into one 
step. Steam explosion is the widely used method for pretreatment of biomass. The 
process is conducted typically between temperature 190 to 220?C and pressure 1.2 to 4.1 
MPa. Low energy input and negligible environmental effects are the major advantages of 
the process. However, steam explosion does not always break down all the lignin, 
requires small particle size, and also produces some inhibitory compounds for further 
enzymatic hydrolysis and fermentation steps.
21, 24, 47-49
 Sometimes the addition of sulfur 
dioxide or carbon dioxide during steam explosion treatment further improves the 
15 
 
enzymatic hydrolysis of biomass by making the pretreatment environment more acid. 
SO2 forms sulfuric acid and CO2 forms the carbonic acid. The limitation of these 
methods is the lower yield of hemicellulose sugars.
31
 
Ammonia Fiber Explosion (AFEX): In AFEX process, a batch of lignocelluloses is 
contacted with ammonia at high loadings and elevated temperature and pressure. After 
the reaction, the pressure is explosively released (similar to steam explosion) to disrupt 
the biomass structure. The physical disruption due to decompression and alkaline 
hydrolysis improves the enzymatic hydrolysis of biomass. Most of the ammonia (up to 
99%) can be recovered for reuse.
21
 The AFEX treatment does not solubilize much of the 
hemicellulose and effectively results in higher sugar yield than acid catalyzed steam 
explosion. AFEX was found very effective for pretreatment of low lignin substrate 
(Agriculture residue, herbaceous crops and grasses).
31, 50, 51
 
Supercritical Carbon Dioxide (CO
2
) Pretreatment: Supercritical carbon dioxide (critical 
point, 31?C and 7.4 MPa) is nontoxic and relatively cheaper pretreatment agent. The 
method has several advantages such as low pretreatment temperatures, easy recovery of 
CO
2
, and high solids loading. Though the process shoed good results for Avicel, the low 
effectiveness for lignocellulosic biomass is the major drawback.
24
  
Appendix B summarizes the variety of pretreatment methods and process 
conditions. 
1.5 Supercritical Fluid Technology 
The unique physiochemical properties of dense supercritical fluids provide an 
attractive medium for chemical reactions and other processes. Fluids near critical points 
have solvent power comparable to that of liquids and are much more compressible than 
16 
 
dilute gases. The transport properties of such fluids lie intermediate between gas- and 
liquid-like. Supercritical fluids are attracting much attention in various fields of science 
and technology. In the last decades, the number of applications has increased 
continuously in several areas such as: 
? Supercritical critical water and carbon dioxide as an alternative solvents for 
chemical reactions, 
? Supercritical fluid extraction and purification, 
? Fine particle production by supercritical antisolvent (SAS) and rapid expansion of 
supercritical solutions (RESS), 
? Supercritical fluid chromatography for analytical applications, 
? Supercritical steam cycle technology for power plants. 
Supercritical fluids can be advantageously exploited in environmentally benign 
separation and reaction processes, as well as for new kinds of materials processing. 
Although laboratory scale studies show excellent results, there are relatively few 
processes in industrial scale. The high pressure processes are generally expansive to 
design, built, operate, and maintain. Therefore, scaling up from laboratory to industrial 
scale of such processes can only be successful if clear benefits can be achieved in terms 
of high efficiency, conversion ratios, product quality, and cost advantages over the 
conventional processes.
52-54
 
1.6 Sub- and Supercritical Water (hydrothermal treatment) 
Water is an ecologically safe and abundantly available solvent in nature. Water 
has a relatively high critical point (374?C and 22.1 MPa) because of the strong interaction 
between the molecules due to strong hydrogen bond. Liquid water below the critical point 
17 
 
is referred as subcritical water whereas water above the critical point is called 
supercritical water. Density and dielectric constant of the water medium play major role 
in soubilizing different compounds. Water at ambient conditions (25?C and 0.1 MPa) is 
good solvent for electrolytes because of high dielectric constant 78.5), whereas most 
organic matters are poorly soluble at this condition. 
As water is heated, the H-bonding start weakening, allowing dissociation of water 
into acidic hydronium ions (H
3
O
+
) and basic hydroxide ions (OH
?
). Structure of water 
changes significantly near the critical point because of the breakage of infinite network of 
hydrogen bonds and water exists as separate clusters with a chain structure.
55 
In fact, 
dielectric constant of water decreases considerably near the critical point, which causes a 
change in the dynamic viscosity and also increases self-diffusion coefficient of water.
56
 
Supercritical water has liquid-like density and gas-like transport properties, and behaves 
very differently than it does at room temperature. For example, it is highly non-polar, 
permitting complete solubilization of most organic compounds and oxygen.  The 
resulting single-phase mixture does not have many of the conventional transport 
limitations that are encountered in multi-phase reactors.  However, the polar species 
present, such as inorganic salts, are no longer soluble and start precipitating.  The 
physiochemical properties of water, such as viscosity, ion product, density, and heat 
capacity, also change dramatically in the supercritical region with only a small change in 
the temperature or pressure, resulting in a substantial increase in the rates of chemical 
reactions. It is interesting to see (Figure 1.4) that the dielectric behavior of 200?C water is 
similar to that of ambient methanol, 300?C water is similar to ambient acetone, 370?C 
water is similar to methylene chloride, and 50?C water is similar to ambient hexane. 
 
Figure 1.4 Physical properties of water with temperature, at 24 MPa (Kritzer et al.).
57
 
 In addition to the unusual dielectric behavior, transport properties of water are 
significantly different than the ambient water as shown in Table 1.2. 
Table 1.2 Comparison of ambient and supercritical water. 
 Ambient water Supercritical water 
Dielectric constant 78 < 5 
Solubility of organic compounds Very low Fully miscible 
Solubility of oxygen 6 ppm Fully miscible 
Solubility of inorganic compounds Very high ~ 0 
Diffusivity (cm
2
 s
-1
) 10
-5
 10
-3
 
Viscosity (g cm
-1
 s
-1
) 10
-2
10
-4
 
Density (g cm
-3
) 1 0.2 - 0.9 
 
18 
 
19 
 
1.7 Sub- and Supercritical Water Technology for Biofuels 
Hydrothermal media, which can be broadly defined as water-rich phase above 
200?C (sub- and supercritical water) offers several advantages over other biofuel 
production methods.
19
 Some of the major benefits are 
? High energy and separation efficiency (since water remains in liquid phase 
and the phase change is avoided), 
? High throughputs, 
? Versatility of chemistry (solid, liquid and gaseous fuels), 
? Reduced mass transfer resistance in hydrothermal conditions, 
? Improved selectivity for the desired energy products (methane, hydrogen, 
liquid fuel) or biochemicals (sugars, furfural, organic acids etc), 
? Ability to use mixed feedstock as well as wet biomass, 
? No need to maintain specialized microbial cultures, 
? Products are completely sterilized with respect to any pathogens including 
biotoxins, bacteria or viruses, and 
? Processing of post fermentation resides. 
Hydrothermal treatment can be applied to produce solid (biochar), liquid (bioethanol, 
biocrude), and gaseous (methane, hydrogen) fuels depending on the processing 
temperature and pressure as shown in Figure 1.5.  
 
Figure 1.5 Hydrothermal treatment application referenced to pressure-temperature phase 
diagram of water.  
The substantial changes in the physical and chemical properties of water in the 
vicinity of its critical point can be utilized advantageously for converting lignocellulosic 
biomass to desired biofuels.
19, 58
 In fact, reactions in subcritical and supercritical water 
also provide a novel medium to conduct tunable reactions for the synthesis of specialty 
chemicals from biomass.
59
 
 In the subcritical region, the ionization constant (K
w
) of water increases with 
temperature and is about three orders of magnitude higher than that of ambient water 
(Figure 1.4) and the dielectric constant (?) of water drops from 80 to 20.
60
 A low 
dielectric constant allows subcritical water to dissolve organic compounds, while a high 
ionization constant allows subcritical water to provide an acidic medium for the 
hydrolysis reactions. These ionic reactions can be dominant because of the liquid-like 
20 
 
21 
 
properties of subcritical water. Moreover, the physical properties of water, such as 
viscosity, density, dielectric constant and ionic product, can be tuned by small changes in 
pressure and/or temperature in subcritical region.
61-63
 In the supercritical region, density 
of water drops down to lower value.  This means that ionic product of water is much 
lower and ionic reactions are inhibited because of the low relative dielectric constant of 
water. The lower density favors free-radical reactions, which may be favorable for 
gasification.
18
 The following sections discusses about application of hydrothermal 
medium properties for biochar, bioethanol, biocrude, and gaseous fuel production. 
1.7.1 Biochar 
 Biochar is the carbon rich, high energy density solid product resulting from the 
advanced thermal degradation of organic materials such as wood, manure, agricultural 
residues etc. The less fibrous structure and high calorific value of biochar similar to that 
of coal makes it an excellent candidate for solid fuel. Biochar is highly resistant to 
decomposition upon land application and has a number of positive effects relating to soil 
fertility.
64
 Pyrolysis and hydrothermal carbonization (HTC) are the two main processes 
for the production of biochar. Pyrolysis typically utilizes high quality dry biomass for 
biochar production where air pollution is primary concern during traditional pyrolysis 
operation due to the emission of volatile compounds. The HTC is an environment 
friendly and promising process that uses water as solvent. Besides being simple process, 
HTC has a number of other practical advantages. The HTC process does not require dry 
biomass and also the final product can be easily filtered from the reaction solution. This 
way, complicated drying schemes and costly separation procedures are conceptually 
avoided. 
22 
 
 The biomass feedstock typically contains 40-60% oxygen. Therefore, the removal 
of oxygen from biomass is the major objective for upgrading its energy density during 
biochar production. This objective can be accomplished by the removal of oxygen by 
dehydration, which removes oxygen in the form of water, and by decarboxylation, which 
removes oxygen in the form of carbon dioxide. Thermodynamically, water is fully 
oxidized compound and has no residual heating value. Therefore, water makes an ideal 
medium for conducting such reactions. Even in the excess of water, biomass undergoes 
dehydration reaction at elevated temperature and pressure.
19
 As discussed earlier, the 
ionization constant of water increases with temperature and near-critical point, the 
amount of dissociation is three times what it would be at normal temperatures and 
pressures. Due to the increased ionization of water at elevated temperature, the biomass 
components undergo depolymerization mainly by hydrolysis reactions. The temperature 
region 180-250?C in the phase diagram of water (Figure 1.5) can effectively be utilized 
for the production of biochar from biomass. The longer reaction time of the order of 
several hours is typically needed for the substantial removal of oxygen by dehydration 
and for the subsequent breakdown of the fibrous structure of biomass. The dehydration of 
biomass at lower temperature can be accelerated by the addition of small amount of 
acids. 
1.7.2 Bioethanol 
Pretreatment of lignocellulosic biomass is viewed as a critical step to make the 
cellulose accessible to enzymes for achieving an economical yield of fermentable sugars 
for bioethanol production.  As discussed earlier, there are several chemical pretreatment 
methods are available. But these methods are associated with serious the economic and 
23 
 
environmental constraints due to the heavy use of chemicals.
65, 66
 Hydrothermal 
pretreatment of biomass is a non-toxic, environmentally benign and inexpensive medium 
for chemical reactions. Process mainly removes hemicelluloses and a part of lignin.  The 
resulting structural change, increased surface area, and increased pore volume of 
pretreated biomass helps in the acceleration of the rate of enzymatic hydrolysis as well as 
product yield. The increased ionization constant of water in 180-220?C regions (Figure 
1.5) can potentially be used for the fractionation of hemicelluloses. Since, no chemical is 
used during the process; the hydrolyzate can be combined with biomass fraction after pH 
conditioning for fermentation. 
1.7.3 Biocrude 
Biocrude is defined as an aqueous carbohydrate solution (oxygenated 
hydrocarbon) produced from the hydrothermal liquefaction of biomass. Lignocellulosics 
mainly consist of cellulose, hemicelluloses and lignin. Among that approximately 75% of 
the dry weight of herbaceous and woody biomass is composed of holocellulose.
67
 
Hydrolysis of cellulose in hydrothermal medium has been studied extensively. The earlier 
studies show that subcritical and supercritical water can be used under a variety of 
conditions to rapidly (order of seconds) liquefy cellulose to sugar and its degradation 
products.
59, 68, 69
 Hemicellulose, an amorphous structure is much more susceptible than 
cellulose to hydrothermal extraction and hydrolysis. It is easily dissolved in water at 
temperature about 180?C.  Hemicelluloses have been successfully extracted upto 95% of 
its fraction as monomeric sugar and sugar products in hydrothermal medium in the range 
of 200-230?C in a very short reaction time.
7, 70
 Low activation energy of lignin causes 
substantial degradation of lignin in hydrothermal medium at temperature below 200?C. 
24 
 
Reaction proceeds through the cleavage of aryl ether linkages, fragmentation, and 
dissolution.
7
 Lignin depolymerization yields low molecular weight fragments having very 
reactive functional groups such as syringols, guaiacols, catechols and phenols.
71
 The 
density of water within the hydrothermal medium has been found to be a key parameter 
in deciding the product pathways.
19
  
The results of liquefaction studies on model compounds and the actual biomass in 
hydrothermal medium provides an opportunity for converting biomass to biocrude and 
other important chemicals.  The hydrothemal medium (subcritical water) in the range of 
250-350?C regions (Figure 1.5) is a favorable condition for conducting ionic reactions. In 
general, hydrothermal liquefaction conditions range from 250-380?C, 7 to 30 MPa with 
liquid water present, often in presence of alkaline catalyst.
19
 
 Biocrude derived from the direct liquefaction of biomass can be converted to 
liquid fuel, hydrogen gas or chemicals. Aqueous phase reforming processes have been 
successfully utilized for converting the biomass derived water soluble carbohydrates to 
liquid alkanes and hydrogen.
67, 72, 73
 Preliminary studies on the conversion of various 
biomass types into liquid fuels have indicated that hydrothermal liquefaction is more 
attractive than pyrolysis or gasification. In these studies, typically 25% biomass slurry in 
water is treated at temperatures of 300?350?C and 12-18 MPa pressures for 5-20 min to 
yield a mixture of liquid, gas (mainly CO
2
) and water. The liquid is a mixture with a wide 
molecular weight distribution and consists of various kinds of molecules. A large 
proportion of the oxygen is removed as carbon dioxide and the resulting biocrude 
contains only 10-13% oxygen, as compared to 40% in the dried biomass.
74
  
Karagoez  et al. investigated the distribution of hydrothermal liquefaction (280? 
25 
 
for 15 min) products, i.e. liq., gas and solid from wood (sawdust) and non-wood biomass 
(rice husk), and major biomass components, i.e. lignin, cellulose produced by and 
analyzed the liquid hydrocarbons (oils) for the differences in the hydrocarbon 
composition with respect to feed material.  Cellulose showed the highest conversion 
among the four samples investigated.  Sawdust and rice husk had almost similar 
conversions.  Liquid products were recovered with various solvents (ether, acetone, and 
ethyl acetate) and analyzed by GC-MS.  The oil (ether extract) from the hydrothermal 
treatment of cellulose consisted of furan derivatives whereas lignin-derived oil contained 
phenolic compounds.  The composition of oils (ether extract) from sawdust and rice husk 
contained both phenolic compounds and furans, however phenolic compounds were 
dominant.  Rice husk derived oil consists of more benzenediols than sawdust derived oil.  
The volatility distribution of oxygenated hydrocarbons were carried out by C-NP gram 
and it showed that the majority of oxygenated hydrocarbons from sawdust, rice husk and 
lignin were distributed at n-C
11
, whereas they were distributed at n-C
8
 and n-C
10
 in 
cellulose-derived oil.  The gaseous products were carbon dioxide, carbon monoxide, 
methane in sawdust, rice husk, lignin and cellulose.
75, 76
 
Demirbas have reviewed the possible mechanism of liquefaction. Organic 
materials are converted to liquefied products through a series of physical and chemical 
changes such as solvolysis, depolymerization, dehydration, and decarboxylation. 
Solvolysis is a type of nucleophilic substitution where the nucleophile is a solvent 
molecule. This reaction results in micellar-like substructures of the feedstock. 
Depolymerization reactions lead to smaller molecules. Decarboxylation and dehydration 
26 
 
leads to new molecules and the formation of carbon dioxide through splitting off of 
carboxyl groups.
14
 
1.7.4 Hydrothermal Gasification 
 Relatively fast hydrolysis of biomass in hydrothermal medium leads to a rapid 
degradation of polymeric structure of biomass. The subsequent reactions also are rather 
fast, which leads to gas formation at relatively lower temperature compared to dry 
processes.
77
 Above the critical point, the lower density of supercritical fluid favors free 
radical reactions and makes the reaction conditions conducive for the formation of 
methane and hydrogen gas.
18
  Biomass is gasified to mainly methane and carbon dioxide 
in the presence of an added heterogeneous catalyst in near critical or supercritical water 
(350-400?C). At higher temperature in supercritical water (Figure 1.5), biomass is 
converted to hydrogen rich gas without catalyst or with non metal catalysts.
78
 As 
discussed earlier, part of lignin and hemicelluloses fraction of biomass undergo solvolysis 
within few minutes of the exposure to hydrothermal medium. The hydrothemolysis of 
remaining biomass fractions occurs at somewhat higher temperature. The initial products 
subsequently undergo a variety of isomerization, dehydration, fragmentation, and 
condensation reactions, that ultimately form gas and tars.
79
 
1.8 Challenges of Hydrothermal Processing 
 Although in laboratory experiments excellent results have been achieved and the 
technology possesses many potential benefits over the conventional methods of 
processing biomass to biofuels or chemicals, there are certain issues, which need to be 
addressed.  
? Biomass feeding at high pressure: As a ?rule of thumb?, the solid loading in 
27 
 
excess of 15-20 wt% is considered economical on commercial point of view. 
Feeding slurries at high pressure is always challenging especially for the lab scale 
studies since low capacity slurry pumps are rarely available. Pumping slurry at 
large scale is less of a problem, where progressive cavity or similar pumps are 
commercially available.  
? Salt precipitation: Plugging of reactors caused by the precipitation of inorganic 
salts above supercritical temperature and low density conditions. At room 
temperature, water is an excellent solvent for most salts. On the other hand, 
solubility of most salts in very low (typically 1-100 ppm) in supercritical water 
(low density) and precipitating salts may plug the reactors even at high flow 
velocities.
80
 However, the problem may be used as an opportunity to produce a 
valuable fertilizer by-product of the process, if managed properly. 
? Corrosion: The halogens sulfur, or phosphorous present in the organic matters 
are converted to the respective acids, which may cause severe corrosion on the 
reactor wall under harsh reaction conditions. The corrosion problem can be 
reduced or avoided by selecting the right material of construction and or a slightly 
modified rector concept. 
? Coking and deactivation of heterogeneous catalyst: Some catalyst supports 
degrades or oxidize in hydrothermal conditions. Decline in catalyst activity is also 
observed with long period of exposure of catalyst during continuous process.
19
 
1.9 Research Plan 
This research study focuses on the fundamental understanding of the reactions 
involved during hydrothermal treatment (sub- and supercritical water) of lignocellulosic 
28 
 
biomass. A good amount of work has been done on biomass pyrolysis, this knowledge 
base would be utilized in deciphering the mechanisms, while allowing for the alterations 
due to the presence of water as a solvent and high pressure. An experimental facility for 
conducting high temperature and high pressure (sub- and supercritical water) reactions in 
continuous and semi-continuous flow, and batch mode will be designed and tested. 
Following main objectives are being pursued in this study to apply hydrothermal 
treatment technology for biofuels applications: 
1.9.1 Hydrothermal Pretreatment of Biomass for Bioethanol 
Pretreatment of biomass is an important step in bioethanol production. Subcritical 
water as an effective pretreatment medium for biomass is gathering considerable 
attention. However, a clear fundamental understanding is lacking. To compliment, the 
aim of this study is to carry out a fundamental study on how subcritical water condition 
affect the structure and enzymatic digestibility of biomass. Cellulose, switchgrass 
(herbaceous grass), and corn stover (agricultural residue) are chosen for the study. In this 
work, specifically, the effect of temperature on the cellulose structure and its reactivity 
with the cellulase enzyme are studied. The study will help in understanding the fate of 
cellulose molecule during subcritical water pretreatment and will provide key design 
parameters for optimization of the process. 
1.9.2 Hydrothemal Liquefaction for Biocrude 
Liquefaction of biomass in hydrothermal medium provides an environmentally 
benign method for converting biomass to biocrude and other important chemicals. The 
process requires a fundamental understanding of the reactions involved. Most of the work 
is done in the batch reactors. Kinetics obtained from the batch reactor often can give 
misleading information on the design for industrial level continuous reactors. The main 
aim of this study is to examine the possibilities of maximizing hydrolysis products 
(sugar) yields from the liquefaction of cellulose and switchgrass (an energy crop) in a 
flow reactor.  
Another objective of this study is to examine the effect of alkali on the 
liquefaction temperature and also, examine the behavior of solid residues for 
understanding the nature of thermochemical liquefaction. In heterogeneous catalysis, 
high surface is obtained by reducing particle or pore size.  One can think about using 
nanoparticles of alkali as catalysts, but these particles will dissolve in the ambient feed 
water and the later precipitate to form large particles during the heating.  A method for in-
situ generation of catalyst nanoparticles in supercritical water, using antisolvent 
precipitation concept will be applied as shown in Figure 1.6. The method has already 
been successfully tested for sodium carbonate.
81
 The concept will be applied to the 
biomass liquefaction in hydrothermal reactor as shown below: 
 
Figure 1.6 Generation of in-situ alkali catalyst nanoparticles in hydrothermal reactor. 
 The solution of alkali salt in ambient water is injected directly to the reactor. Due 
29 
 
30 
 
to the insolubility in supercritical water, the alkali salts precipitate out rapidly as 
nanoparticles. The key is rapid precipitation due to the high diffusivity of the supercritical 
water and sudden heating upon mixing. The produce nanoparticles will provide a high 
surface area. In addition, the newly generated alkali particle surface is highly catalytic 
due to the high surface roughness. 
1.9.3 Hydrothermal Carbonization for Biochar 
The hydrothermal carbonization process, a traditional but recently revived 
method, presents superior characteristics that make it a promising route of wide potential 
application for converting biomass to biochar. The conversions of biomass in 
hydrothermal conditions are widely examined process with the aim for the recovery of 
liquid or gaseous fuel intermediates (e.g. Glucose, 5-hydroxymethylfurfural, methane, 
hydrogen etc.) from biomass. The solid residues from such process are mostly treated as 
undesirable by-products. However, the acceleration of coalification in HTC process by a 
factor of 10
6
-10
9
 under rather soft conditions, down to a scale of hours, makes it a 
considerable, technically-attractive alternative for the sequestration of carbon from 
biomass on large and ultra-large scales. Essentially, all forms of biomass can be 
converted to biochar. Forest thinning, herbaceous grasses, crop residues, manure, and 
paper sludge are some of the potential feedstock. Presence of inorganic trace elements (K, 
Ca, Si, P etc) and lignin content can affect the biochar yield.
82
 
In general, biochar is produced by heating biomass dispersion in water around 
200?C in a closed reaction vessel in the presence of a weak acid catalyst for 4-24 h. 
Under acidic conditions and below 200?C, most of the original carbon stays bound to the 
final structure. In this work, switchgrass, a major energy crop, is processed by HTC 
31 
 
method for high energy density biochar production. Besides optimizing the process 
parameters for producing high energy density product, the research work will focus on 
enhancing the rate of coalification / dehydration. The process can provide a promising 
route for carbonaceous functional materials and higher energy density fuel for use in a 
variety of applications. 
 
1.10 References 
1. Farrell, A. E.; Plevin, R. J.; Turner, B. T.; Jones, A. D.; O'Hare, M.; Kammen, D. 
M., Ethanol can contribute to enrgy and environmental goals. Science 2006, 311, 506-
508. 
2. Pastircakova, K., Determination of trace metal concentrations in ashes from 
various biomass materials. Energy Edu. Sci. Technol. 2004, 13, 97-104. 
3. http://www-cta.ornl.gov/data/chapter1.shtml. (Feb, 2010) 
4. Perlack, R. D.; Wright, L. L.; Turhollow, A. F.; Graham, R. L.; Stokes, B. J.; 
Erbach, D. C. Biomass as feedstock for a bioenergy and bioproducts induatry: The 
technical feasibility of a billion-ton annual supply; US Department of Energy 
US Department of Agriculture: 2005. 
5. http://www.nrel.gov. (Feb, 2010). 
6. Jarvis, M., Chemistry: Cellulose stacks up. Nature 2003, 426, 611-612  
7. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
Polym. Sci. 1994, 19, 797?841. 
8. Meister, J. J., Chemical modification of lignin. In Chemical modification of 
lignocellulosic materials, Hon, D. N.-S., Ed. Marcel Dekker Inc.: 1996; pp 129-157. 
32 
 
9. Olsson, L.; Jorgensen, H.; Krogh, K. B. R.; Roca, C., Bioethanol production from 
lignocellulosic material. In Polysachharides: Structural diversity and functional 
versatility, second ed.; Dumitriu, S., Ed. Marcel Dekker: 2005; pp 957-993. 
10. Falkehag, S. I., Synthesis of phenolic polymer. Appl. Polym. Symp. 1975, 28, 247-
257. 
11. Zhang, B.; Huang, H.-J.; Ramaswamy, S., Reaction Kinetics of the hydrothermal 
treatment of lignin. Appl Biochem Biotechnol 2008, 147, 119-131. 
12. Knill, C. J.; Kennedy, J. F., Cellulosic biomass-derived products. In 
Polysaccharides: Structural diversity and functional versatility, Dumitriu, S., Ed. Marcel 
and Dekker: 2005; pp 937-956. 
13. Balat, M., Mechanisms of thermochemical biomass conversion processes, Part 1: 
Reactions of pyrolysis. Energy Sources, Part A 2008, 30, 620-635. 
14. Demirbas, A., Mechanisms of liquefaction and pyrolysis reactions of biomass. 
Energy Conversion and Management 2000, 41, 633-646. 
15. Ni, M.; Leung, D. Y. C.; Leung, M. K. H.; Sumathy, K., An overview of hydrogen 
production from biomass. Fuel Processing Technology 2006, 87, 461-472. 
16. Jong, W. d., Sustainable hydrogen production by thermochemical biomass 
processing. In Hydrogen Fuel: Production, transport and storage, Gupta, R. B., Ed. CRC 
Press: 2009; pp 185-225. 
17. Galkin, A. A.; Lunin, V. V., Subcritical and supercritical water: a universal 
medium for chemical reactions. Russian chemical reviews 2005, 74, (1), 21-35. 
18. Kruse, A.; Gawlik, A., Biomass conversion in water at 330-410 ?C and 30-50 
MPa: Identification of key compounds for indicating different chemical reaction 
33 
 
pathways. . Ind. Eng. Chem. Res. 2003, 42, 267-269. 
19. Peterson, A. A.; Vogel, F.; Lachance, R. P.; Froling, M.; Michael Jerry Antal, J.; 
Tester, J. W., Thermochemical biofuel production in hydrothermal media:A review of 
sub- and supercritical water technologies. Energy and Environmental Science 2008, 1, 
32-65. 
20. Ghosh, T. K.; Roychoudhury; Ghosh, P., Simultaneous saccharification and 
fermentation (SSF) of lignocellulosics to ethanol under vacuum cycling and step feeding. 
Biotechnol. Bioeng. 1984, 26, 377-381. 
21. Sierra, R.; Smith, A.; Granda, C.; Holtzapple, M. T., Producing fuels and 
chemicals from lignocellulosic biomass. Chemical Engineering Progress August, 2008, 
pp S10-S18. 
22. O'Sullivan, A. C., Cellulose: the structure slowly unravels. Cellulose 1997, 4, (3), 
173-207. 
23. Schacht, C.; Zetzl, C.; Brunner, G., From plant materials to ethanol by means of 
supercritical fluid technology. Journal of Supercritical Fluids 2008, 46s  (3), 299-321. 
24. Laxman, R. S.; Lachke, A. H., Bioethanol from lignocellulosic biomass, Part 1: 
Pretreatment of the substrates. In Handbook of plant-based biofuels, Pandey, A., Ed. CRC 
Press: 2008; pp 121-139. 
25. Millet, M. A.; Baker, A. J.; Scatter, L. D., Physical and chemical pretreatment for 
enhancing cellulose saccharification. Biotech. Bioeng. Symp. 1976, 6, 125-153. 
26. Caufield, D. F.; Mooer, W. E., Effect of varying crystallinity of cellulose on 
enzymic hydrolysis. Wood Science and Technology 1974, 6, (4), 375?379. 
27. Sintsyn, A. P.; Gusakov, A. V.; Vlasenko, E. Y., Effect of structural and physico-
34 
 
chemical features of cellulosic substrates on the efficiency of enzymatic hydrolysis. Appl. 
Biochem. Biotechnol. 1991, 30, 43-59. 
28. Chang, V. S.; Holtzapple, M. T., Fundamental factors affecting biomass enzymatic 
reactivity. Applied Biochemistry and Biotechnology 2000, 84, 5-37. 
29. Chang, M. M.; Chou, T. Y. C.; Tsao, G. T., Structure, Pretreatment and Hydrolysis 
of Cellulose. Adv. In Biochem. Eng. 1981, 20, 15-42. 
30. Fan, L. T.; Lee, Y. H.; Gharpuray, M. M., The nature of lignocellulosics and their 
pretreatments for enzymatic hydrolysis. Adv. Biochem. Eng. 1982, 23, 157-187. 
31. Gupta, R. Alkaline pretreatment of biomass for ethanol production and 
understanding the factors influencing the cellulose hydrolysis. Auburn University, 
Auburn, 2008. 
32. Grohmann, K. R.; Torget, R.; Himmel, M., Dilute acid pretreatment of biomass at 
high solids concentrations. Biotechnology and Bioengineering Symposium 1985, 15, 59-
80. 
33. Torget, R.; Himmel, M.; Grohmann, K., Dilute-acid pretreatment of two short-
rotation herbaceous crops. Applied Biochemistry and Biotechnology 1992, 34/35, 115-
123. 
34. Welseth, C. S., Influence of fine structure of cellulose on the action of cellulases. 
TAPPI 1952, 35, 233-238. 
35. Zhang, Y.-H. P.; Ding, S.-Y.; Mielenz, J. R.; Cui, J.-B.; Elander, R. T.; Laser, M.; 
Himmel, M. E.; McMillan, J. R.; Lynd, L. R., Fractionating recalcitrant lignocellulose at 
modest reaction conditions. Biotech. Bioeng. 2007, 97, 214-223. 
36. Dadi, A. P.; Varanasi, S.; Schall, C. A. In A novel inic liquid pretreatment strategy 
35 
 
to achieve enhanced cellulose sacchrification kinetics, AIChE Annual meeting, San 
Francisco, CA, 2006; AIChE: San Francisco, CA, 2006. 
37. Sievers, C.; Valenzuela-Olarte, M. B.; Marzialetti, T.; Musin, I.; Agrawal, P. K.; 
Christopher; Jones, W., Ionic-Liquid-Phase Hydrolysis of Pine Wood. Ind. Eng. Chem. 
Res. 2009, 48, (3), 1277-1286. 
38. Playne, M. J., Increased digestibility of bagasse by pretreatment with alkalis and 
steam explosion. Biotechnol. Bioeng. 1984, 26, 426-433. 
39. Chang, V. S.; Nagwani, M.; Kim, C. H.; Holtzapple, M. T., Oxidative lime 
pretreatment of high-lignin biomass. Applied Biochemistry and Biotechnology 2001, 94, 
1-28. 
40. Karr, W. E.; Holtzapple, T., Using lime pretreatment to facilitate the enzymatic 
hydrolysis of corn stover. Biomass & Bioenergy 2000, 18, 189-199. 
41. Yoon, H. H.; Wu, Z. W.; Lee, Y. Y., Ammonia-recycled percolation process for 
pretreatment of biomass feedstock. Appl. Biochem. Biotechnol. 1995, 51-52, 5-19. 
42. Hashaikeh, R.; Fang, Z.; Butler, I. S.; Hawari, J.; Kozinski, J. A., Hydrothermal 
dissolution of willow in hot compressed water as a model for biomass conversion. Fuel 
2007, 86, 1614-1622. 
43. Kohlmann, K. L.; Westgate, P. J.; Sarikaya, A.; Velayudhan, A.; Weil, J.; 
Hendricson, R.; Ladisch, M. R., Enhanced enzyme activities on hydrated lignocellulosic 
substrates. In American Chemical Society National Meeting, ACS Symposium series No. 
618: 1995; Vol. 207, pp 237-255. 
44. Mosier, N.; Hendrickson, R.; Ho, N.; Sedlak, M.; Ladisch, M. R., Optimization of 
pH controlled liquid hot water pretreatment of corn stover. Bioresour. Technol. 2005a, 96, 
36 
 
1986-1993. 
45. Weil, J. R.; Brewer, M.; Hendrickson, R.; Sarikaya, A.; Ladisch, M. R., 
Continuous pH monitoring during pretreatment of yellow poplar wood sawdust by 
pressure cooking in water. Appl. Biochem. Biotechnol. 1997, 68, 21-40. 
46. Hendriks, A. T. W. M.; Zeeman, G., Pretreatments to enhance the digestibility of 
lignocellulosic biomass. Bioresource Technology 2008. 
47. Overend, R. P.; Chornet, E., Fractionation of lignocellulosics by steam-aqueous 
pretreatments. Philosophical Transactions of the Royal Society of London 1987, A321, 
523-536. 
48. Grous, W. R.; Converse, A. O.; Grethlien, H. E., Effect of steam explosion 
treatment on pore size and enzymatic hydrolysis of poplar. Enzyme Microb. Technol. 
1986, 8, 274-280. 
49. Duff, S. J. B.; Murray, W. D., Bioconversion of forest products industry waste 
cellulosics to fuel ethanol: a review. 1996, 55, 1-33. 
50. Dale, B. E.; Leong, C. K.; Pham, T. K.; Esquivel, V. M.; Rios, I.; Latimer, V. M., 
Hydrolysis of lignocellulosics at low enzyme levels: application of the AFEX process. 
Bioresource Technology 1996, 56, (1), 111-116. 
51. Foster, B. L.; Dale, B. E.; Doran-Peterson, J. B., Enzymatic hydrolysis of 
ammonia-treated sugar beet pulp. Appl. Biochem. Biotechnol. 2001, 91, (3), 269-282. 
52. Wellig, B. Transpiring wall reactor for supercritical water oxidation. Swiss 
Federal Institute of Technology, Zurich, 2003. 
53. Savage, P. E.; Gopalan, S.; Mizan, T. I.; Martino, C. J.; Brock, E. E., Reactions at 
supercritical conditions - applications and fundamentals. AIChE J. 1995, 41, (7), 1723-
37 
 
1778. 
54. Eckert, C. A.; Knutson, B. L.; Dobenedetti, P. G., Supercritical fluids as solvents 
for chemical and materials processing. Nature 1996, 383, (6598), 313-318. 
55. Kalinichev, A. G.; Churakov, S. V., Size and topology of molecular clusters in 
supercritical water: a molecular dynamics simulation. Chem. Phys. Letters 1999, 302, 
411-417. 
56. Marcus, Y., On transport properties of hot liquid and supercritical water and their 
relationship to the hydrogen bonding. Fluid Phase Equilibr. 1999, 164, 131-142. 
57. Kritzer, P.; Dinjus, E., An assessment of supercritical water oxidation (SWO). 
Existing problems, possible solutions and new reactor concepts. Chemical Engineering 
Journal (Amsterdam, Netherlands) 2001, 83, (3), 207-214. 
58. Dinjus, E.; Kruse, A., Hot compressed water-a suitable and sustainable solvent 
and reaction medium? Journal of physics: condensed matter 2004, 16, S1161-S1169. 
59. Matsumura, Y.; Sasaki, M.; Okuda, K.; Takami, S.; Ohara, S.; Umetsu, M.; 
Adschiri, T., Supercritical water treatment of biomass for energy and material recovery. 
Combust. Sci. and Tech. 2006, 178, 509-536. 
60. Tester, J. W.; Holgate, H. R.; Armellini, F. J.; Webley, P. A.; Killilea, W. R.; Hong, 
G. T.; Barner, H. E. In Supercritical water oxidation technology ? process development 
and fundamental research, In Emerging technologies in hazardous waste management III, 
Washington D.C., 1993; Amercan Chemical Society: Washington D.C., 1993; pp 35-76. 
61. Franck, E. U., Fluids at high pressures and temperatures. Pure Appl. Chem. 1987, 
59, (1), 25-34. 
62. Miyoshia, H.; Chena, D.; AkaiT., A novel process utilizing subcritical water to 
38 
 
recycle soda?lime?silicate glass. Journal of Non-Crystalline Solids 2004, 337, ( 3), 280-
282. 
63. Savage, P. E., Organic chemical reactions in supercritical water. Chemical 
Reviews 1999, 99, 603-621. 
64. Bruun, S.; Luxhoi, J., Is biochar production really carbon-negative? 
Environmental Science & Technology 2008, 1388. 
65. Hsu, T.-A., Pretreatment of biomass. Handbook on bioethanol: Production and 
Utilization Taylor and Francis, Washington, DC: 1996. 
66. Sun, Y.; Cheng., J. J., Hydrolysis of lignocellulosic material for ethanol 
production: A review. Bioresource Technology 2002, 83, 1-11. 
67. Huber, G. W.; Cheda, J. N.; Barrett, C. J.; Dumesic, J. A., Production of liquid 
alkanes by aqueous processing of biomass derived carbohydrates. Science 2005, 308, 
1446-1450. 
68. Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.; N.Takeda; Adschiri, T.; 
Arai, K., Cellulose hydrolysis in subcritical and supercritical water. J. Supercritical 
Fluids 1998 1998, 13, 261-268. 
69. Deguchi, S.; K. Tsujii; Horikoshi, K., Crystalline-to-amorphous transformation of 
cellulose in hot and compressed water and its implications for hydrothermal conversion. 
Green Chem. 2008, 10, 191-196. 
70. Mok, W. S.; Antal, M. J., Uncatalyzed solvolysis of whole biomass hemicellulose 
by hot compressed liquid water. Ind. Eng. Chem. Res. 1992, 31, 1157-1161. 
71. Fang, Z.; Sato, T.; Jr., R. L. S.; Inomata, H.; Arai, K.; Kozinski, J. A., Reaction 
chemistry and phase behaviour of lignin in high-temperature and super critical water. 
39 
 
Bioresource Technology 2008, 99, 3424-3430. 
72. Huber, G. W.; Iborra, S.; Corma, A., Synthesis of transportation fuels from 
biomass: Chemistry, catalysts, and engineering. Chem. Rev. 2006, 106, 4044. 
73. Valenzuela, M. B.; Jones, C. W.; Agrawal, P. K., Batch aqueous reforming of 
woody biomass. Energy & Fuels 2006, 20, 1744-1752. 
74. Gourdiaan, F.; Peferoen, D., Liquid fuels from biomass via a hydrothermal 
process. Chemical Engineering Science 1990, 45, 2729-2734. 
75. Karagoz, S.; Bhaskar, T.; Muto, A.; Sakata, Y., Hydrothermal upgrading of 
biomass: Effect of K
2
CO
3
 concentration and biomass/water ratio on product distrubution. 
Bioresource Technology 2006, 97, 90-98. 
76. Karagoez, S.; Bhaskar, T.; Muto, A.; Sakata, Y., Comparative studies of oil 
compositions produced from sawdust, rice husk, lignin and cellulose by hydrothermal 
treatment. Fuel 2005, 84, (7-8), 875-884. 
77. Kruse, A., Hydrothermal biomass gasification. J. of Supercritical Fluids 2009, 47, 
(3), 391-399. 
78. Elliott, D. C., Catalytic hydrothemal gasification of biomass. Biofuels, 
Bioproducts and Biorefining 2008, 2, 254-265. 
79. Matsumura, Y.; Minowa, T.; Potic, B.; Kersten, S. R. A.; Prins, W.; Swaaij, W. P. 
M. V.; Beld, B. V. D.; Elliott, D. C.; Neuenschwander, G. G.; Kruse, A.; Jr., M. J. A., 
Biomass gasification in near- and super-critical water: Status and prospects  Biomass and 
Bioenergy 2005, 29, (4), 269-292  
80. Kritzer, P.; Dinjus, E., An assessment of supercritical water oxidation (SCWO): 
Existing problems, possible solutions and new reactor concepts. Chemical Engineering 
40 
 
Journal 2001, 83, 207-214. 
81. Muthukumaraa, P.; Gupta, R. B., Sodium-carbonate assisted supercritical water 
oxidation of chlorinated waste. Ind. Eng. Chem. Res. 2000, 39, 4555-4563. 
82. Amonette, J., An introduction to Biochar: Processes, Properties, and Applications. 
In Harvesting Clean Energy 9, Special Workshop, Billings, MT, 2009. 
 
 
2. Cellulose Pretreatment in Subcritical Water: Effect of Temperature on Molecular 
Structure and Enzymatic Reactivity 
 
2.1 Abstract 
Microcrystalline cellulose (MCC) was pretreated with subcritical water in a 
continuous flow reactor for enhancing its enzymatic reactivity with cellulase enzyme. 
Cellulose/water suspension was mixed with subcritical (i.e., pressurized and heated) 
water and then fed into the reactor maintained at a constant temperature and pressure. 
After the reaction, product was immediately cooled in a double-pipe heat exchanger. The 
solid portion of the product (i.e., treated MCC) was separated and tested for molecular 
structure and enzymatic reactivity. Experiments were conducted at temperatures ranging 
from 200 to 315 ?C, at 27.6 MPa, and for 3.4 to 6.2 seconds reaction times.   
The treated MCC was characterized for degree of polymerization (DP
v
) by 
viscosimetry, and crystallinity by X-ray diffraction (XRD). In addition, differential 
scanning calorimetry and scanning electron microscopy (SEM) analyses were carried out 
to study any transformation in the cellulose structure. As expected, DP
v
 of cellulose 
steadily decreased with increase in the pretreatment temperature, with a rapid drop 
occurring above 300 ?C. On the other hand, XRD analysis did not show any decrease in 
crystallinity upon pretreatment but, partial transformation of cellulose I to cellulose II 
structure was noticed in the MCC treated at ?300 ?C.  Enzymatic reactivity was increased 
after the treatment at ?300 ?C. 
41 
 
2.2 Introduction 
Discovered more than 150 years ago, cellulose is the most abundant organic 
matter on the earth and perhaps is the most studied polymer.
1
  For its utilization as 
transportation fuel, the major research focus is to convert the cellulose into monomeric 
sugar, which can be economically fermented to ethanol. Lignocellulosic biomass, a class 
that includes straw, perennial grasses, plant stalks, and woody biomass is an attractive 
renewable source of holocellulose.
2
 Lignocellulosics mainly consist of cellulose, 
hemicelluloses and lignin which are bonded together by covalent bonding, various 
intermolecular bridges, and van der Waals forces forming a complex structure, making it 
resistant to enzymatic hydrolysis and insoluble in water.
1, 3
 Cellulose is an unbranched 
chain of anhydroglucose (C
6
H
10
O
5
) linked head to tail by ?-glycosidic linkages (Jarvis, 
2003).
4
 The ?-linkages in cellulose form linear chains that are highly stable and resistant 
to chemical attack because of the high degree of intra and intermolecular hydrogen 
bonding. There are six known polymorphs of cellulose (I, II, III
1
, III
II
, IV
I
, and IV
II
) 
which can also interconvert. Cellulose I, also termed as native cellulose, has parallel 
arrangement of chains and is the only polymorph that occurs naturally. Cellulose II is 
converted through mercerization or solubilization-regeneration of native or other 
celluloses. Cellulose II is thermodynamically more stable structure with an antiparallel 
arrangement of the strands and some inter-sheet hydrogen bonding. Cellulose III
I
 and III
II
 
can be obtained from Cellulose I and II, respectively, by treatment with liquid ammonia 
or some amines, whereas polymorphs IV
I
 and IV
II
 can be obtained from heating cellulose 
III
I
 and III
II,
 respectively, to 206?C in glycerol.
1, 5-7
 Two decades ago, it was reported that 
native cellulose exists as a mixture of two crystalline forms I
?
 and
 
I
?, 
having triclinic and 
42 
 
monoclinic unit cells, respectivel.
8
 Cellulose I
?
 is thermodynamically less stable, as 
shown by its conversion to cellulose I
?
 by annealing at 260?C. In both crystalline forms, 
cellobiose is the repeating unit with a strong intra-chain H-bond from 3-OH to the 
preceding ring O5, whereas the inter-chain H-bonding and packing of the crystal are 
slightly different in the two forms.
9
 
Biomass pretreatment with acid or alkali followed by enzymatic saccharification 
is a conventional method to produce fermentable sugars for ethanol production. However, 
this treatment is associated with the serious economic and environmental constraints due 
to the heavy use of chemicals.
10, 11
 Supercritical (>374?C, >22.1 MPa) and subcritical 
water have attracted much attention because of their suitability as a non-toxic, 
environmentally benign and inexpensive media for chemical reactions. Supercritical 
water technology provides a novel method to quickly convert cellulose to sugar and to 
conduct tunable reactions for the synthesis of specialty chemicals from biomass.
12
 The 
solvent property of sub- and super critical water has been discussed in chapter 1 (section 
1.6). In the subcritical region, the ionization constant (K
w
) of water increases with 
temperature and is about three orders of magnitude higher than that of ambient water 
(Figure 2.1). On the other hand, the dielectric constant (?) of water drops from 80 to 20, 
which help solubilize the organic compounds in subcritical water. Therefore subcritical 
water potentially provides an acidic medium for the hydrolysis of cellulose.
13-15
 
43 
 
0 100 200 300 400 500 600 700 800
-26
-24
-22
-20
-18
-16
-14
-12
-10
Ionization con
s
tant
log
 K
w
 (
m
ol
/k
g)
2
Temperature (?C)
 
Figure 2.1 Ionization constant (K
w
) of water as a function of temperature at 25MPa.
16
 
 Experimental studies have been conducted on cellulose hydrolysis in subcritical 
and supercritical water in the temperature ranging 320-400?C and residence time ranging 
0.05 to 10.0 seconds. The results showed that cellulose can be effectively converted to 
hydrolysis products (i.e., oligosaccharides, cellobiose and monosaccharides) and aqueous 
decomposition products of glucose (i.e., levoglucosan, 5-hydroxymethyl furfural, 
erythrose, glycolaldehyde, and dihydroxyacetone) in a fraction of a second within a flow 
type reactor.
17-22
 The hydrolysate (water-soluble portion) and precipitate obtained by 
supercritical water treatment were treated with alkali or wood charcoal for reducing 
inhibitory effects of the various decomposition compounds of cellulose on enzyme 
activity and fermentation. High efficiency of ethanol production was achieved and the 
studies concluded that supercritical water treatment could be a promising pretreatment for 
ethanol production from lignocellulosic biomass.
23-25
 
Hydrolysis of cellulose in supercritical water is very sensitive to residence time. A 
High residence time enhances the formation of organic acids such as acetic acid, formic 
acid and lactic acid. Formation of acids makes the reaction medium more acidic, which is 
44 
 
conducive for further degradation of hydrolysis products.  Indeed after hydrolysis of 
cellulose in water at 320?C and 25 MPa for 9.9 s, more than half of the cellulose was 
converted to organic acids.
21
 The solid cellulose-like residues have been inevitably 
observed in all the studies because of the rapid change in the polarity of water in going 
from reaction condition to room temperature. These residues have been reported to have 
less viscosity-average degree of polymerization (DP
v
) with no significant change in 
crystallinity as compared to untreated cellulose. 
Supercritical water treatment requires additional process of alkaline or wood 
charcoal treatment for reducing the inhibitory effect of degradation products present in 
aqueous solution. To avoid, the degradation products, the use of hot compressed 
(subcritical) water as an effective pretreatment medium for biomass is gathering 
considerable attention.  However, a clear fundamental understanding is lacking. The early 
fundamental work in this area was performed by Bobleter and co-workers in which they 
established that hemicelluloses could be completely solubilized from the lignocellulosic 
biomass, together with lignin.
26-29
 To compliment, the aim of this study is to carry out a 
fundamental study on how subcritical water conditions affect cellulose structure and 
digestibility. 
In this study, pretreatment of microcrystalline cellulose (MCC) was carried out in 
subcritical water in the temperature range of 200-315?C. Treated MCC samples are 
characterized by the use of XRD, SEM, and DSC techniques to investigate the structural 
changes. DP
v 
of the samples was measured to examine the polymer chain length. In this 
work, specifically, the effect of temperature on the cellulose structure and its reactivity 
with the cellulase enzyme are studied. The study will help understand the fate of cellulose 
45 
 
molecule during subcritical water pretreatment and will provide key design parameters 
for optimization of the process. 
2.3 Experimental Section 
Materials: Microcrystalline cellulose powder (size 20 ?m, product number 310697) and 
high purity standard reagents for HPLC analysis were purchased from Sigma Aldrich. 
De-ionized water was used for preparing cellulose slurry.  
Apparatus: Figure 2.2 shows the schematic of the apparatus used for the experiments. It 
consists of high-pressure liquid pumps, a cellulose slurry feeder, an electrical tubular 
furnace, a water-cooled double tube heat exchanger, a backpressure regulator, and a phase 
separator.  
 
Water
High Pressure 
Pump , 1
Water Cooled
Heat Exchanger
TI
TI
PI 
Water
High Pressure 
Pump. 2 
Electrical 
Furnace
Water 
Pre- heating
Reactor
Peak Analysis 
on PC
Gas 
Chomatograph
Data 
Acquisition 
System
6- port injection 
valve for online 
sampling
Gas 
Flowmeter
Slurry
Phase
Separator
Gas Phase
BPR
TI
Insulation
Cellulose slurry
feeder
 
Figure 2.2 Apparatus for cellulose pretreatment in subcritical water. 
Reactor was a high-pressure 14? long Inconel 600 tube having 0.25? OD, 0.11?ID and 
2.2 ml volume.  A custom designed piston-operated high-pressure slurry feeder (0.57? ID, 
20? long) was used to supply cellulose slurry at desired flow rate. Real time temperature 
46 
 
of the reaction zone was measured using a 1/16? thermocouple placed 3? inside the 
reactor from outlet end. This thermocouple measured reaction zone temperature with 
accuracy of ? 0.1?C. There were three additional thermocouples installed at different 
locations in the apparatus. Real time temperature indicated by the thermocouple placed 
inside the reaction zone was used for all the analyses. Reactor pressure was maintained 
using a backpressure regulator. 
Experimental Procedure: MCC was added to continuously stirred hot water (70?C) to 
prepare 24.1 wt% slurry and charged to the slurry feeder.  It was observed that 24.1 wt% 
of cellulose suspension in hot water did not settle appreciably in a similar diameter tube 
for an hour, hence any agitation was not required in the slurry feeder.    
At the start, temperature and pressure of the reaction zone were established under 
water flow. Then, slurry was pumped at a desired flow rate and mixed with 8 fold (mass 
basis) of preheated water prior to the reactor. Products coming out of the reactor were 
rapidly cooled in the heat exchanger, and collected in the phase separator after 
depressurization.  Periodic liquid samples were collected, centrifuged at 3400 rpm to 
remove any solids and analyzed for total organic carbon (TOC) in the Shimadzu TOC-
Vcsn analyzer. Products collected during the steady state phase of experiment were stored 
at 4?C. 
MCC was treated with subcritical water at temperatures ranging from 200 to 
315?C at a constant pressure 27.6 MPa and residence times ranging from 3.4 to 6.2 
seconds. The experiments were divided into two groups (Table 2.1) as: (I) 200-275?C, 
and (II) 300-315?C. One experiment from group I (run 5) and two experiments from 
group II (run 7 and run 9) were conducted in triplicates to confirm the reproducibility of 
47 
 
the cellulose conversion by measuring the total organic contents (TOC) in the 
supernatant.  
Table 2.1 Summary of experimental conditions (at 27.6 MPa). 
 
Run 
Reactor 
temperature, 
(?C) 
Residence 
time, 
(s) 
TOC of 
liquid 
product, 
(ppm) 
pH of 
liquid 
product 
lnR
0
Group I, 200-275?C 
Run 1 200 4.1 667 6.4 4.1 
Run 2 200 6.2 735 6.5 4.5 
Run 3 253 3.9 510 5.5 7.6 
Run 4 251 5.8 596 5.0 7.9 
Run 5 275 3.7 568 4.4 9.1 
Run 6 274 5.6 1151 4.0 9.4 
Group II, 300-315?C 
Run 7 300 3.5 3992 3.8 10.7
Run 8 303 5.2 4309 3.4 11.3
Run 9 315 3.4 4945 3.3 11.7
 
Multiple samples were collected and analyzed for TOC values during each 
experiment to confirm the steady state. The relative standard deviation for runs 5, 7, and 
9 were 4.1, 6.6, and 5.3%, respectively. Regression analysis of enzymatic hydrolysis data 
was carried out with Design of experiment software (DOE PRO 2007. 
The residence time (t) in the reactor was calculated using pure water density at the 
48 
 
reaction temperature and pressure (i.e. fluid phase in the reaction zone is assumed to be 
pure water for the simplicity of calculations) as 
?
?
?
?
?
?
?
?
=
1,1 TP
pump
F
V
t
?
?
    [1] 
where, V (2.2 ml) is reactor volume, F is the combined volumetric flow rate of pumps 1 
and 2, ? 
pump
 is the density of water at pump condition (25?C, 27.6 MPa), and ?
 P1,T1
 is the 
density of water at reactor condition (e.g., T
1
 and P
1
).   
Reynolds number, which varied from 900 to 1700, was calculated assuming pure 
water. Therefore, considering the small diameter tube, relatively high Reynolds number, 
use of very fine cellulose powder (size 20 ?m), short residence time (3.4 to 6.2 s) and 
dilute feed (2.7 wt%) in reactor, ideal plug flow reactor was assumed for the study. 
The severity factor (R
0
) has been used by several researchers to measure the 
combined effect of temperature and residence time in hot water treatment of biomass 
processes.
26, 30, 31
  The severity index is defined as 
?
?
?
?
?
? ?
?=
75.14
100
exp
0
T
tR    [2] 
where t is the residence time in minutes and T is temperature in ?C. The logarithmic 
values of R
0
 are listed in Table 2.1. 
2.4 Product Analyses 
After completion of the experiment, solids from the product slurry were separated 
by centrifugation, which was then freeze dried, weighed and taken for different analyses 
as follows: 
Viscosity Average Degree of Polymerization (DP
v
): DP
v
 of MCC and treated samples 
49 
 
were calculated from the intrinsic viscosity of pulp dissolved in cupriethylenediamine 
(CED) by viscosimetry as per TAPPI method using 
32
 
(DP
v
) = {(1.65 ?
pulp
 ? 116 y)/x}
1.111
  [3] 
where x and y are the weight fractions of cellulose and the hemicelluloses in the 
pulp, respectively, ?
pulp
 is the inherent viscosity of cellulosic pulp. Since, pure cellulose 
was used in the study, no correction was needed for the presence of hemicelluloses or 
other compounds, and hence the value of y was taken as zero.  
X-Ray Diffraction Analysis: Rigaku Miniflex powder X-ray diffractometer equipped with 
a Cu K?
1
 radiation source at 30 kV voltages, 15 mA current and a miniflex goniometer 
was used for the XRD analysis. Diffraction patterns were collected in the 2? range of 10-
28? at a scan speed of 1?/min. and step size of 0.01?. Table 2.2 compares the diffraction 
angle of different planes in Cellulose I and Cellulose II. 
Table 2.2 Lattice planes of cellulose I and II and the diffraction angles in XRD spectra. 
 Plane Angle, 2? (?) 
Cellulose I 101 14.7 
 
 
16.8 
021 20.5 
 002 22.7 
Cellulose II 101 12.1 
 
 
20.0 
002 21.9 
 
The crystallinity index (CI) for cellulose I was determined 
33
 by 
CI = (I
002
 ? I
18.5?
)
 
/ I
002
   [4] 
where I
002 
is the peak intensity corresponding to (002) lattice plane of cellulose molecule, 
50 
 
and I
18.5? 
is the peak intensity observed at 2? equal to 18.5?. 
Enzymatic Hydrolysis: Enzymatic hydrolysis of all the untreated and treated MCC 
samples were carried out simultaneously in 35 ml kmax test tubes with rubber stoppers 
on the top. The protocol used was similar to the NREL Laboratory Analytical Procedure 
(LAP) No. 009,
34
 where a total liquid volume of 10 ml was used instead of 100 ml. 
Cellulase commercial enzyme (brand name: Spezyme CP) used in this study was 
provided by Genencor International Inc. (Palo Alto, CA). Two enzyme loadings were 
used for each substrate: 3.5 FPU/g and 60 FPU/g. Each test tube contained 0.1 g oven 
dried MCC substrate, 1 ml diluted enzyme (with equivalent activity of 3.5 FPU/g or 60 
FPU/g), 0.07 ml of antibiotics (0.04 ml tetracycline solution + 0.03 ml of cycloheximide 
solution) and 0.05 N sodium citrate buffer (rest of the volume) for maintaining the pH 
(4.8) in the reaction. Magnetic stirrer was used for mixing of the reactant in the test tube. 
All the test tubes were placed in the incubator which maintained constant reaction 
temperature of 50?C.  Samples were collected after 1 and 24 h for measurement of 
glucose and cellobiose concentration in the liquid. Sugar concentrations were measured 
using HPLC with a Bio-Rad Aminex HPX-87P column. The reported enzymatic 
digestibility is the ratio of glucan equivalent of glucose + cellobiose found in the liquid 
and initial glucan loading (0.1 g). Glucan equivalent of glucose and cellobiose is 
calculated as (0.9?solublized glucose amount) + (0.95?solublized cellobiose amount). 
Scanning Electron Microscopy (SEM): The samples were prepared onto adhesive carbon 
tape on an aluminum stub followed by sputter coating of gold. Surface morphology of the 
sample was studied using an environmental scanning electron microscopy system (Zeiss 
EVO 50).  
51 
 
Differential Scanning Calorimetry (DSC) Analysis: DSC analysis was performed using a 
TA instruments, model DSC Q2000. Samples (7-10 mg) were placed in an open 
aluminum pan while an empty pan was placed in the reference cell. Heating rate was 
10?C/min from 25 to 350?C. Nitrogen was used as inert gas for removing all oxygen 
from the DSC chamber prior to the heating. 
Liquid Analysis: Carbon content in the supernatant liquid was determined by using TOC 
analyzer. Hydrolysis products (monomers and oligomers) and degradation products in 
liquid samples were determined by following the National Renewable Energy Laboratory, 
biomass laboratory analytical procedure (described in technical report-NREL/TP-510-
42623). The analysis was performed using a HPLC with Bio-Rad Aminex HPX-87P lead-
based column, refractive index detector, water mobile phase, and column temperature of 
85?C.
35
 
2.5 Results and Discussion 
Results were divided in two groups based on the reaction temperature range used 
and the reactivity of the cellulose to cellulase enzyme. Group I experiments represent the 
pretreatment of MCC in the temperature range 200-274?C and group II represents those 
in the temperature range 300-315 ?C. Both groups of experiments were conducted at 27.6 
MPa pressure. Table 2.1 shows the details of the process parameters, TOC, pH value of 
the liquid products and the degree of severity of the treatment. In general, solids 
recovered after subcritical water treatment became fluffier and turned off white in group 
II runs. Liquid products collected during group I experiments were colorless and slightly 
acidic, but the liquid from group II runs were yellowish and more acidic (pH 3.3-3.4). 
There was no improvement in enzymatic digestibility of group I samples even though 
52 
 
severity factor (lnR
0
) was increased from 4.1 to 9.4, on the contrary, group II samples 
showed significant rise in enzymatic reactivity (Table 3.3). 
Table 2.3 Enzymatic digestibility of MCC, group I, and group II samples. 
 Enzyme loading 
3.5 FPU 60 FPU 
Run Digestibility (%) Digestibility (%) 
1 hour 24 hours 1 hour 24 hours 
MCC 7.9 47.2 45.0 75.0 
Group I 
Run 1 (200?C) 6.4 40.0 36.0 68.3 
Run 2 (200?C) 6.7 40.1 46.1 81.1 
Run 3 (253?C) 5.1 37.3 28.5 65.4 
Run 4 (251?C) 5.3 38.5 31.1 69.4 
Run 5 (275?C) 6.3 40.9 31.6 70.9 
Run 6 (274?C) 6.2 42.2 34.7 74.2 
Group II 
Run 7 (300?C) 11.1 48.5 44.4 85.0 
Run 8 (303?C) 13.0 54.6 50.1 90.3 
Run 9 (315?C) 22.0 68.1 60.1 90.6 
  
 As expected, conversion of MCC to water-soluble products increased with 
severity factor. Characterization and enzymatic digestibility of recovered solids and 
liquid products are discussed as follows: 
53 
 
2.5.1 Solids Analysis 
DP
v
 Analysis: DPv analyses for both group I and group II samples, representing the entire 
temperature range of this study, are shown in Figure 2.3.  
50
100
150
200
250
300
350
180 200 220 240 260 280 300 320
DP
v
Temperature (?C)
Run 2
Run 4
Run 6 Run 7
Run 9
 
Figure 2.3 Degree of polymerization (DP
v
) versus pretreatment temperature. 
 The DP
v
 of of untreated MCC was 327. It decreased with temperature as 
expected, but reduced rapidly for treatment above 300?C. Runs 6 and 7 show almost 
similar DP
v
,
 
but there was a distinct difference in the physical appearance of these two 
powders. Run 6 powder looked white like MCC, while run 7 appeared to be off white and 
fluffier than the MCC. The depolymerization of MCC during subcritical water treatment 
at different temperatures can be interpreted to proceed in the following two steps: 
Step 1: At lower pretreatment temperatures (group I), water hydrolyzed mostly the 
amorphous region of MCC, which caused increase in the percentage crystallinity of the 
remaining samples, and slightly decreased DP
v
.  
Step 2: At higher pretreatment temperatures (group II), intermolecular hydrogen bonding 
between the cellulose molecules weakens, which is followed by the cleavage of 
54 
 
glycosidic linkages due to the increased accessibility of water molecules to the reactive 
sites.  
Therefore, reaction at higher temperatures (>300?C) caused the rapid decline of 
DP
v
, increased solubility of cellulose in water, and rapid conversion of cellulose to 
monomers and the degradation products. Liquid analysis of group II experiments showed 
that more than 30% of cellulose converts to water soluble products. Compositions of 
these products in liquid are discussed in the liquid analysis section. 
XRD Analysis: Figures 2.4 and 2.5 show the XRD pattern for group I and II samples. The 
intensity distribution in the region of interest (diffraction angle, 2? = 18? to 24?) usually 
displays two peaks. The peak at 20? is due to (021) line in cellulose I and (101) line in 
cellulose II. The other peak at 22-22.7? is due to the contribution from (002) line of both 
the polymorphs. Using the XRD data, crystallinity of the solids (Table 2.4) was 
determined by Equation 4. Here, each data is derived from a single run; for future work, 
it is recommended that multiple runs should be conducted for each pretreatment condition 
to ensure statistically significant variation in the crystallinity. 
Table 2.4 Crystallinity of samples by XRD analysis. 
Run Crstallinity (%)
MCC 77.8
Group I 
Run 1 79.3 
Run 2 79.3 
Run 3 82.7 
Run 4 82.5 
Run 5 81.9 
Run 6 81.4 
Group II 
Run 7 81.2 
Run 8 82.2 
Run 9 83.6 
 
 
55 
 
MCC crystallinity was 77.8% and increased with the treatment temperature and 
residence time. The maximum TOC value in liquid product of group I sample was 1151 
ppm (run 6), which indicated that there was no significant hydrolysis of MCC at 274?C 
or below by subcritical water in the studied range of residence time. Increase in TOC and 
crystallinity were mainly because of the hydrolysis and removal of the amorphous region 
of MCC. Therefore, it can be concluded that hydrolysis occurred mainly at the crystal 
surface with no minimal swelling or dissolution of MCC. Hence, it is not surprising to 
see that XRD patterns for MCC and group I samples are similar (Figure 2.4). 
TOC values for group II samples were 3992 ppm (run 7) or above, which 
confirmed the sharp rise in the rate of hydrolysis of MCC in this group. As the treatment 
temperature increases, density and dielectric constant of water decreases. Water becomes 
more non-polar and becomes an excellent solvent for organic compounds.
15
 Moreover, 
higher temperature increases the reactivity and kinetic energy of cellulose molecules.  
 
Figure 2.4 XRD results for MCC and group I samples. 
Therefore, a high temperature results in weakening of intermolecular hydrogen bonds. A 
56 
 
change in the reaction condition enhances the partial dissolution of cellulose in water.  
Figure 2.5 shows the gradual shift of diffraction peak of (002) lattice plane from 
2? angle of 22.1? to a slightly lower angle.  
 
Figure 2.5 XRD results for original MCC and group II samples. 
Onset of a new peak at 2? of 19.4? is seen in group II samples. This peak is more 
prominent in run 9, whereas it was very small in runs 7 and 8. Peak at about 20? angle is 
assigned to the (10?) lattice plane of cellulose II.
36
 Based on the XRD pattern of group II 
samples, it can be concluded that cellulose started swelling around the crystal surface of 
MCC at the high temperature. This swollen region should be hydrolyzed to water-soluble 
saccharides and the degradation products. Yields of these products are discussed in the 
liquid analysis section. Cellulose II has antiparallel arrangement of chains and a more 
stable structure with a low energy crystalline arrangement.
37
 After the reaction, products 
are rapidly cooled to the room temperature. Due to the sudden cooling, swollen part of 
cellulose around the crystal surface recrystallizes to cellulose II, which is the 
57 
 
thermodynamically more favorable crystal arrangement.  Therefore, XRD of group II 
samples shows the presence of cellulose II. The ratio between the peak intensities at angle 
about 2? = 20? and 002 (I
20? 
/ I
002
) have been used for estimating the ratio of cellulose I 
cellulose II in a mixture. The ratio (I
20? 
/ I
002
) for MCC (Cellulose I) is 0.37, which upon 
processing increased to 0.44 for run 9. By using the method and the calibration curve 
developed by Gjonnes et al. cellulose II percentage in the run 9 was estimated to be 8.9%, 
where  I
20? 
/ I
002
 of 100% cellulose II was considered 1.16.
36
 
Reaction at the temperatures studied here is mainly heterogeneous and occurs at 
the crystal surface with partial swelling or dissolution of MCC. With increase in the 
temperature, polarity of water decreases considerably, making the reaction medium more 
homogeneous and conducive for swelling and dissolution of cellulose. Consequently, part 
of the undissolved cellulose precipitated as crystal II polymorph. Consistent to our 
observation, Sasaki et al. have also reported the production of cellulose II from native 
cellulose by near- and supercritical water solubilization.
38, 39
 
Enzymatic Digestibility: As shown in Table 2.3, enzymatic digestibility of untreated and 
treated MCC was carried out with two different cellulase loadings (3.5 FPU and 60 FPU). 
Hydrolysis with a low enzyme loading measures the enzymatic reactivity of cellulose, 
while those with the high enzyme loading indirectly measure the total digestible 
cellulose. Regression analysis of enzymatic hydrolysis data was carried out with Design 
of Experiment software (DOE PRO 2007 from Digital Computation). Figures 6a-d show 
the surface plots for different enzymatic digestibility values with respect to the reaction 
temperature and treatment time. 
58 
 
 
Figure 2.6a Plot after regression analysis of enzymatic hydrolysis data at 3.5 FPU and 
after 1 h. 
 
Figure 2.6b Plot after regression analysis of enzymatic hydrolysis data at 3.5 FPU and 
after 24 h. 
 
59 
 
 
Figure 2.6c Plot after regression analysis of enzymatic hydrolysis data at 60 FPU and 
after 1 h. 
 
Figure 2.6d Plot after regression analysis of enzymatic hydrolysis data at 60 FPU and 
after 24 h. 
As evident, enzymatic reactivity (with 3.5 FPU/ g cellulose) of treated solid is not 
much affected by the reaction time but is mainly a function of reaction temperature.  But 
60 
 
overall enzymatic reactivity (with 60 FPU/ g cellulose) of treated solid is a function of 
both reaction temperature as well as time. At, a constant temperature, enzymatic 
digestibility of the treated solid increases with the reaction time.  
Interestingly, it is found that MCC treated in the low temperature range (200-
275?C) shows a lower enzymatic reactivity than the untreated MCC. There are two 
plausible reasoning for this behavior: (a) a slight increase in crystallinity, and (b) a 
reduction in DP
v
 of MCC upon pretreatment. If the crystallinity and enzyme accessibility 
are the main resistances for decreased hydrolysis for these treated samples, that should 
have been overcome in higher enzyme loading of 60 FPU where enzyme is supplied in 
excess. But even with very high enzyme loading, extent of hydrolysis for treated MCC 
samples from runs 1-6 is lower than that for untreated MCC, with the exception of run 2. 
This suggests that it is not the difference in crystallinity that is causing the lower 
hydrolysis but probably some other molecular characteristics such as DP
v
.  As indicated 
elsewhere,
40
 enzymatic hydrolysis of lower DP
v
 amorphous cellulose substrate leads to 
slightly lower hydrolysis rate. Similar trends are shown here with crystalline substrates.  
Another interesting point here is that this trend does not hold true when MCC is 
treated at high temperatures (over 300?C). Even though the DP
v
 of these treated MCC 
substrates is much lower than those treated at lower temperatures, MCC obtained from 
runs 7-9 have higher hydrolysis rates. As discussed earlier, with XRD plot of different 
treated MCC at higher temperature, transformation of cellulose I to cellulose II is 
suspected. Transformation in cellulose structure might be a factor leading to the higher 
enzymatic reactivity of these samples.  
SEM Analysis: MCC and solid samples from runs 5, 7 and 9, which were pretreated at 
61 
 
different temperatures (275, 300, and 315 ?C, respectively), were taken for SEM analysis 
(Figures 2.7a-d). The surface of MCC before subcritical water treatment, shown in Figure 
2.7a, is marked by some boundary edges and does not show the presence of any pores, 
trenches or surface cracks. Figures 2.7b-d show the images of samples produced in runs 
5, 7 and 9. Here, cracks and trenches can be clearly seen on the surface of the samples. 
These cracks, trenches and kinks are very prominent in all images of treated samples seen 
at about 30,000X magnification. Though surface cracks or trenches are seen in the 
images, they are not evenly distributed over the surface of samples. From the several 
images taken at different magnifications of these samples (not shown here), presence of 
steps and terrains was confirmed. These surface modifications might have occurred as a 
result of mild hydrolysis over the crystal surface, whereas cracks are seen, perhaps be due 
to the removal of amorphous region from MCC after subcritical water treatment.  
              
                
1?m
1?m
1?m
1?m
a b
c 
d
 
Figure 2.7 SEM images of samples: (a) MCC, (b) run 5, (c) run 7, and (d) run 9. 
62 
 
DSC Analysis: DSC thermograms MCC and samples from run 6, 7, and 9 are shown in 
Figure 2.8. Despite having different DP
v
, the samples follow similar endothermic heating 
path and show the endotherm minima in the range of 324-331?C, indicating no change in 
thermal stability of the samples after subcritical water treatment. With the removal of the 
amorphous region and subsequently, slight increase in percentage crystallinity as 
observed in XRD, it is expected to increase the thermal stability of the treated cellulose. 
Studies also show that endotherm minimum temperature decreases with DP
v
.
41
 
 
Figure 2.8 DSC thermograms of MCC, group I (run 6) and group II (run 7 and 9) 
samples. 
As discussed earlier, DP
v
 of the samples decreased with temperature and reduced 
to one third of its initial value in case of run 9. In our study, DP
v
 did decreased, but it still 
remained substantially high (>100), and so, lowering of endotherm minimum below 
327?C was not observed. There was no exothermic reaction and also no indication of 
glass transition temperature up to 350?C. The observations of thermograph support the 
fact that there was no substantial change in thermal properties of the samples.  
63 
 
2.5.2 Liquid Analysis 
Table 2.1 shows that TOC values of liquid products from various runs. TOC 
increased rapidly with rise in temperature above 275?C and more than 30% of cellulose 
was converted to hydrolysis or degradation products in group II experiments. Clear liquid 
recovered after centrifugation was analyzed by HPLC for oligosaccharides, monomers 
and degradation products. In this study, we defined (a) hydrolysis products as the 
combination of oligomers (>2 monomer units), cellobiose, glucose, and fructose, and (b) 
degradation products as glycoaldehyde, D-fructose, anhydroglucose, and 5-HMF.  
pH of the liquid products varied from 4.0-6.5 in group I experiments, while 3.3-
3.4 in group II experiments (Table 2.1). Low pH values confirm the presence of organic 
acids (e.g. formic acid, lactic acid).The liquid samples from group I having cellulose 
conversion less than 10% did not show the presence of hydrolysis products during HPLC 
analysis. Liquid samples from group II (runs 8 and 9) which were treated at almost same 
severity factor, but at different temperatures were analyzed.  The yields of different 
products present in the liquid were calculated based on the input of 2.7 wt% of MCC and 
are listed in Table 2.5. These samples primarily showed the presence of hydrolysis 
products, along with a comparatively small amount of degradation products. 
Table 2.5 Yields of different compounds in the liquid product. 
Run Hydrolysis products (%) Degradation products (%) 
 Olg CB Glu Fr GA AG HMF 
Run 8 16.1 2.8 5.0 0.5 0.8 0.9 0.9 
Run 9 20.0 4.1 3.7 0.4 0.7 0.6 0.6 
 
64 
 
Hydrothermal degradation of cellulose is a heterogeneous and pseudo-first-order 
reaction for which detailed chemistry and mechanism have been proposed.
28, 29, 42, 43
 A 
simplified reaction scheme is shown in Figure 2.9.  
 
Figure 2.9 Simplified reaction scheme of cellulose in subcritical water.
44
 
As discussed earlier, the ionization constant of water increases with temperature 
and near-critical point, the amount of dissociation is three times what it would be at 
normal temperatures and pressures. Under such a sever condition, glycosidic linkages of 
cellulose are broken and cellulose starts to convert to water soluble compounds and 
further to glucose. Glucose molecules are unstable at high temperature and so part of the 
glucose further converts to its degradation products such as glycoaldehyde, HMF, organic 
acids etc. which makes the reaction medium more acidic. Once organic acids are formed 
in the reaction mixture, it provides necessary acidity for further degradation of the 
converted hydrolysis products, as an autocatalytic process. Indeed, after hydrolysis of 
cellulose in water at 320 ?C and 25 MPa for 9.9 s, more than half of the cellulose was 
converted to organic acids.
20
 The detailed mechanism of cellulose hydrolysis and product 
65 
 
composition in subcritical and supercritical water above 300?C has been discussed in a 
separate study.
45
 Liquid composition of group II samples clearly indicate that cellulose 
converted to hydrolysis products (oligomers and monomers) and a portion further 
degraded to the aqueous degradation products of glucose, including glycoaldehyde, 
fructose, anhydroglucose, and 5-HMF.  
Based on the results, it is clear that subcritical water can be used as an effective 
pretreatment media for lignocellulosic biomass without degrading or losing the 
significant amount of glucan, and also for increasing the accessibility to enzymes. 
Throughout this study, crystallinity of cellulose was observed to increase with treatment 
temperature while DP
v
 decreased. But enzymatic digestibility improved only after 
considerable lowering of DP
v
. These findings support the theory that to improve the 
enzymatic digestibly, it is not necessary to bring down the crystallinity, rather improving 
the accessibility of enzymes and lowering the DP
v
 are more important factors.   
2.6 Conclusion 
The percentage crystallinity of microcrystalline cellulose (MCC) slightly 
increased after the subcritical water pretreatment in a continuous flow reactor for a short 
residence time (3.4 - 6.2 s) and remained high (> 81%) throughout the treatment range 
(200-315?C). Under the reaction condition, cellulose conversion to water soluble 
products during pretreatment was considerably low (<10 wt.%) at lower treatment 
temperatures (group I).  The enhancement in enzymatic digestibility of cellulose (initial 
and total digestibility, both) was observed for the samples pretreated at higher 
temperatures (?300?C). In fact, MCC pretreated at 315 ?C showed nearly three-fold 
increase in the initial enzymatic reactivity as compared to the untreated MCC at 3.5 
66 
 
FPU/g of glucan enzyme loading. On the other hand, cellulose pretreated at lower 
temperature (? 275?C) had slightly lower digestibility as compared to the untreated 
MCC. The viscosity average degree of polymerization of cellulose though reduced after 
the pretreatment, it decreased rapidly at 315?C.  The onset of cellulose I to cellulose II 
polymorph formation was noticed in XRD spectra for samples pretreated at ?300?C. The 
development of surface cracks and trenches were also observed in the SEM images. 
Improvement in the enzymatic reactivity may be attributed to the partial transformation 
of cellulose I to cellulose II polymorph and some other structural changes in cellulose at 
higher temperature.  
2.7 Acknowledgments 
Author is thankful for the financial supports from the National Science Foundation 
(grant NSF-CBET-0828269) and Alabama Center for Paper and Bioresource Engineering.  
Additionally, the authors are thankful to Rajeev Kumar (CE-CERT, University of 
California, Riverside) for help in DP
v
 analyses. 
2.8 Nomenclature 
Olg: Oligomers 
CB: Cellobiose 
Glu: Glucose 
Fr: D-Fructose 
GA: Glycoaldehyde dimer 
AG: 1,6 Anhydro?beta glucose 
HMF:  5-Hydroxymethyl ?2-furaldehyde 
 
67 
 
2.9 References 
1. O'Sullivan, A. C., Cellulose: the structure slowly unravels. Cellulose 1997, 4, (3), 
173-207. 
2. Chornet, E.; Overend, R. P., Liquid fuels from lignocellulosics. Biomass 1987, 
257-269. 
3. Schacht, C.; Zetzl, C.; Brunner, G., From plant materials to ethanol by means of 
supercritical fluid technology. Journal of Supercritical Fluids 2008, In Press, 23. 
4. Jarvis, M., Chemistry: Cellulose stacks up. Nature 2003, 426, 611-612  
5. Marchessault, R. H.; Sarko, A., X-Ray Structure of Polysaccharides. Advances in 
Carbohydrate Chemistry 1967, 22, 421-482. 
6. Sarko, A., What is the crystalline structure of cellulose ? Tappi 1978, 61, 59-61. 
7. Sarko, A., Cellulose - how much do we know about its structure? Wood and 
Cellulosics: Industrial utilization, biotechnology, structure and properties 1987, 
(J.F.Kennedy ed.), 55-70. 
8. Atalla, R. H.; Vanderhart, D. L., Native Cellulose: A Composite of Two Distinct 
Crystalline Forms. Science  1984, 223, (4633), 283 - 285. 
9. Sinnott, M. L., Chapter 4: Primary structure and conformation of 
oligosaccharides and polysaccharides. RSC publishing: 2007. 
10. Hsu, T.-A., Pretreatment of biomass. Handbook on bioethanol: Production and 
Utilization Taylor and Francis, Washington, DC: 1996. 
11. Sun, Y.; Cheng., J. J., Hydrolysis of lignocellulosic material for ethanol 
production: A review. Bioresource Technology 2002, 83, 1-11. 
12. Matsumura, Y.; Sasaki, M.; Okuda, K.; Takami, S.; Ohara, S.; Umetsu, M.; 
68 
 
Adschiri, T., Supercritical water treatment of biomass for energy and material recovery. 
Combust. Sci. and Tech. 2006, 178, 509-536. 
13. Franck, E. U., Fluids at high pressures and temperatures. Pure Appl. Chem. 1987, 
59, (1), 25-34. 
14. Miyoshia, H.; Chena, D.; AkaiT., A novel process utilizing subcritical water to 
recycle soda?lime?silicate glass. Journal of Non-Crystalline Solids 2004, 337, ( 3), 280-
282. 
15. Savage, P. E., Organic chemical reactions in supercritical water. Chemical 
Reviews 1999, 99, 603-621. 
16. Tester, J. W.; Holgate, H. R.; Armellini, F. J.; Webley, P. A.; Killilea, W. R.; Hong, 
G. T.; Barner, H. E. In Supercritical water oxidation technology ? process development 
and fundamental research, In Emerging technologies in hazardous waste management III, 
Washington D.C., 1993; Amercan Chemical Society: Washington D.C., 1993; pp 35-76. 
17. Ehara, K.; Saka, S., A comparative study on chemical conversion of cellulose 
between the batch-type and flow-type systems in supercritical water. Cellulose 2002, 9, 
301-311. 
18. Ehara, K.; Saka, S., Decomposition behavior of cellulose in supercritical water, 
subcritical water, and their combined treatments. . J. Wood Sci. 2005, 51, 148?153. 
19. Matsumura, Y.; Minowa, T.; Potic, B.; Kersten, S. R. A.; Prins, W.; Swaaij, W. P. 
M. V.; Beld, B. V. D.; Elliott, D. C.; Neuenschwander, G. G.; Kruse, A.; Jr., M. J. A., 
Biomass gasification in near- and super-critical water: Status and prospects  Biomass and 
Bioenergy 2005, 29, (4), 269-292  
20. Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.; N.Takeda; Adschiri, T.; 
69 
 
Arai, K., Cellulose hydrolysis in subcritical and supercritical water. J. Supercritical 
Fluids 1998 1998, 13, 261-268. 
21. Sasaki, M.; Fang, Z.; Fukushima, Y. N.; Adschiri, T.; Arai, K., Dissolution and 
hydrolysis of cellulose in subcritical and supercritical water. Ind. Eng. Chem. Res. 2000, 
39, 2883-2890. 
22. Sakaki, T.; Shibata, M.; Miki, T.; Hirosue, H.; Hayashi, N., Reaction model of 
cellulose decomposition in near-critical water and fermentation of products. Bioresource 
Technology 1996, 197-202. 
23. Hisashi, M.; Toshiki, N.; Katsunobu, E.; Shiro, S., Fermentability of Water-
Soluble Portion to Ethanol Obtained by Supercritical Water Treatment of 
Lignocellulosics Applied Biochemistry and Biotechnology 2005, 124, (1-3), 963-972. 
24. Shiro, S.; Hisashi, M., Bioethanol production from lignocellulosics using 
supercritical water ACS symposium series 2007, 954, 422-433. 
25. Nakata, T.; Miyafuji, H.; Saka, S., Bioethanol from cellulose with supercritical 
water treatment followed by enzymatic hydrolysis Applied Biochemistry and 
Biotechnology 2006, 130, (1-3), 476-485. 
26. Allen, S. G.; Kam, L. C.; Zemann, A. J.; M. J. Antal, J., Fractionation of Sugar 
Cane with Hot, Compressed, Liquid Water. Ind. Eng. Chem. Res. 1996, 35, (8), 2709 -
2715. 
27. Bobleter, O.; Bonn, G.; Concin, R., Hydrothermolysis of biomass-production of 
raw material for alcohol fermentation and other motor fuels. Alt. Energy Sour. 1983, 3, 
323-332. 
28. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
70 
 
Polym. Sci. 1994, 19, 797?841. 
29. Mok, W. S.; Antal, M. J., Uncatalyzed solvolysis of whole biomass hemicellulose 
by hot compressed liquid water. Ind. Eng. Chem. Res. 1992, 31, 1157-1161. 
30. Overend, R. P.; Chornet, E., Fractionation of lignocellulosics by steam-aqueous 
pretreatments. Philosophical Transactions of the Royal Society of London 1987, A321, 
523-536. 
31. Rogalinski, T.; Ingram, T.; Brunner, G., Hydrolysis of lignocellulosic biomass in 
water under elevated temperatures and pressures. Journal of Supercritical Fluids 2007. 
32. Perez, D. d. S.; van, A. R. P., Determination of cellulose degree of polymerization 
in chemical pulps by viscosimetry. 7th EWLP conference 2002, 393-396. 
33. Segal, L.; Creely, J. J.; Martin, A. E.; Conrad, C. M., An empirical method for 
estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. 
Textile Res. J. 1959, 29, 786?794. 
34. Larry, B.; Robert, T., Enzymatic saccharification of lignocellulosic biomass,LAP-
09. http://www.eere.energy.gov/biomass/analytical_procedures.html 1996. 
35. Sluiter, A.; Hammes, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D., 
Determination of sugars, byproducts, and degradation products in liquid fraction process 
samples. NREL Biomass Program, Laboratory Analytical Procedure (LAP) 2006. 
36. Norman, J. G. a. N., X-ray investigation on cellulose II and mixtures of cellulose I 
and II: A method for characterizing and determining the relative contents of the new 
modifications. Acta Chem. Scand. 1960, 14, (3), 683-688. 
37. Simon, I. G., L.; Scheraga, H.A.; Manley, R.S.J., Structure of Cellulose. Part 2. 
Low-Energy Crystalline Arrangements. Macromolecules 1988, 21, 990-998. 
71 
 
72 
 
38. Sasaki, M.; Adschiri, T.; Arai, K., Production of cellulose II from native cellulose 
by near- and supercritical water solubilization. J Agric Food Chem. 2003 51, (18), 5376-
81. 
39. Sasaki, M.; Adschiri, T.; Arai, K., Kinetics of cellulose conversion at 25 MPa in 
sub- and Supercritical water. AIChE J. 2004, 50, (1), 192?202. 
40. Gupta, R.; Lee, Y. Y., Cellulase reaction mechanism with pure cellulosic 
substrates. Biotechnology and Bioengineering 2008. 
41. Bouchard, J.; Abatzoglou, N.; Chornet, E.; Overend, R. P., Characterization of 
depolymerized cellulosic residues. Wood Science and Technology 1989, 23, (4), 343-355. 
42. Mok, W. S. L.; M.J., A. J.; Varhegyi, G., Productive and parasitic pathways in 
dilute-acid-catalyzed hydrolysis of cellulose. Ind. Eng. Chem. Res. 1992, 31, 94-100. 
43. Schwald, W.; Bobleter, O., Hydrothermolysis of cellulose under static and 
dynamic conditions at high temperatures. . J. Carbohydr. Chem. 1989, 8, (4), 565-578. 
44. Kruse, A.; Gawlik, A., Biomass conversion in water at 330-410 ?C and 30-50 
MPa: Identification of key compounds for indicating different chemical reaction 
pathways. . Ind. Eng. Chem. Res. 2003, 42, 267-269. 
45. Kumar, S.; Gupta, R. B., Hydrolysis of Microcrystalline Cellulose in Subcritical 
and Supercritical Water in a Continuous Flow Reactor. Industrial and Engineering 
Chemistry Research 2008, 47, (23), 9321-9329.  
73 
?
3. Hydrothermal Pretreatment of Switchgrass and Corn Stover for Production of Ethanol 
and Carbon Microspheres 
 
3.1 Abstract 
 Pretreatment of biomass is viewed as a critical step to make the cellulose 
accessible to enzymes and for an adequate yield of fermentable sugars in ethanol 
production. Recently, hydrothermal pretreatment has attracted much attention as a non-
toxic, environmentally benign and inexpensive media for chemical reactions. The 
hydrothermal pretreatment of switchgrass and corn stover was conducted in a flow 
through reactor to enhance and optimize the enzymatic digestibility. More than 80% of 
glucan digestibility was achieved by pretreatment at 190?C. Addition of small amount of 
K
2
CO
3
 (0.45-0.9 wt%) can enhance the pretreatment and allow use of a lower 
temperature. Switchgrass pretreated at 190?C only with water had higher internal surface 
area than that pretreated in the presence of K
2
CO
3
, but both the biomasses showed similar 
glucan digestibility. In comparison to switchgrass, corn stover required milder 
pretreatment conditions. The liquid hydrolyzate generated during pretreatment was 
converted into carbon microspheres by hydrothermal carbonization, providing a value-
added byproduct. The carbonization process is further examined by GC-MS analysis to 
understand the mechanism for microsphere formation. 
74 
?
3.2 Introduction 
 Lignocellulosic biomass has emerged as a potential renewable biomass resource 
for the bioethanol production
1, 2
. Biomass, which mainly contains cellulose, 
hemicelluloses, and lignin biopolymers, can be used as fuel by direct combustion or by 
first gasifying and then burning the gas. However, being only renewable resource which 
can be converted to liquid fuel for transportation, there is a great deal of interest in 
converting these resources to bioethanol and other chemicals. The concept is to hydrolyze 
the cellulose and hemicelluloses to recover C
5
 and C
6
 sugars and then ferment the sugars 
to bioethanol. 
3
 The recovered lignin in the process which has relatively higher heating 
value in the range of 24-26 MJ/kg is typically used for generating steam or providing the 
process heat. Switchgrass, a summer perennial grass native to North America that grows 
on marginal land, is considered to be a major lignocellulosic biomass crop of interest for 
the second generation biofuels. The advantages of switchgrass as energy crop include low 
requirement of chemical inputs and the use of marginal land which does not compete 
with the food production. The switchgrass plants are immense biomass producer that can 
reach heights of ?10 feet and provide 6-8 dry-tons/acre/year (as compared to 2.5 
tons/acre/year for hay) with a high cellulosic content. 
4
 
 The biochemical pathways which can be realized at a very moderate process 
conditions using cellulase enzyme to convert holocelluloses to fermentable sugars are the 
most promising ones for large scale bioethanol production. But, the efficiency of this 
technology is limited due to the complex chemical structure of lignocellulose biomass 
and the inaccessibility of ?-glycosidic linkages to cellulase enzymes because of the low 
surface area and small size of pores in multicomponent structure. Hence, pretreatment is 
75 
?
nowadays viewed as a critical step in lignocellulose processing. Pretreatment alters both 
the structural barrier (removal of lignin and hemicelluloses) and physical barrier (surface 
area, crystallinity, pore size distribution, degree of polymerization) which help in 
improving the accessibility of enzyme for hydrolysis.
5-7
 The pretreatment enhances the 
rate of production and the yield of monomeric sugars from biomass. But the pretreatment 
is among the most costly step in the bioethanol conversion process as it may account for 
up to 40% of the processing cost. Moreover, it also affects the cost of upstream and 
downstream processes.
8-11
 Hence, an efficient, less energy intensive and cost effective 
pretreatment method is a necessity for producing ethanol at an economically viable cost. 
Different pretreatment methods are broadly classified into physical, chemical, 
physiochemical and biological processes. The conventional pretreatment, by using acids 
or alkalis, is associated with the serious economic and environmental constraints due to 
the heavy use of chemicals and chemical resistant materials.
12-14
  
 Recently, hydrothermal pretreatment employing subcritical water has attracted 
much attention because of its suitability as a non-toxic, environmentally benign and 
inexpensive media for chemical reactions.
15, 16
 One of the most important benefit of using 
water instead of acid as pretreatment media is that there is no need of acid recovery 
process and related solid disposal and handling cost.
14, 17
 Liquid water, below the critical 
point is conducive for conducting ionic reactions. Ambient water is polar, has extensive 
network of H-bonding, and does not solubilize most organics. As water is heated, the H-
bonds start weakening, allowing dissociation of water into acidic hydronium ions (H
3
O
+
) 
and basic hydroxide ions (OH
?
). Below the critical point, the ionization constant of water 
increases with temperature and is about three orders of magnitude higher than that of 
76 
?
ambient water. Also, the dielectric constant of water decreases with temperature. A low 
dielectric constant allows liquid water to dissolve organic compounds, while a high 
ionization constant provides an acidic medium for the hydrolysis of biomass components 
via the cleavage of ether and ester bonds and favor the hydrolysis of hemicelluloses.
18-21
 
The structural alterations due to the removal of hemicelluloses increase the accessibility 
and enzymatic hydrolysis of cellulose. Enzyme accessibility is increased as a result of the 
increase in mean pore size of the substrate which enhances the probability of the 
hydrolysis of glycosidic linkage.
22
  
 Hydrothermal pretreatment is typically conducted in the range of 150-220?C. The 
temperature range, aiming for the fractionation of hemicelluloses, are decided based on 
the fact that at temperature below 100?C less/small extent of hydrolytic reaction is 
observed whereas cellulose hydrolysis and degradation become significant above 
210?C.
23, 24
 Ether bonds of the hemicelluloses are most susceptible to breakage by the 
hydronium ions. Depending on the operational conditions, hemicelluloses are 
depolymerized to oligosaccharides and monomers, and the xylose recovery from biomass 
can be as high as 88-98%. For example, Suryawati et al. has reported 90% removal of 
hemicelluloses from Kanlow switchgrass at 200?C.
25
 Acetic acid is also generated from 
the splitting of thermally labile acetyl groups of hemicelluloses. In further reactions, the 
hydronium ions generated from the autoionization of acetic acid also acts as catalyst and 
promotes the degradation of solubilized sugars. In fact, the formations of hydronium ions 
from acetic acid is much more than from water.
23, 24
 
 The low pH (< 3) of the medium causes the precipitation of solubilized lignin and 
also catalyzes the degradation of hemicelluloses. To avoid the formation of inhibitors, the 
77 
?
pH should be kept between 4 and 7 during the pretreatment. This pH range minimizes the 
formation of monosaccharides, and therefore the formations of degradation products that 
can further catalyze hydrolysis of the cellulosic material during pretreatment.
6, 22, 26-30
 
Maintaining the pH near neutral (5-7) helps in avoiding the formation of fermentation 
inhibitors during the pretreatment. 
 In general, the concentrations of solubilized products are lower in hydrothermal 
pretreatment compared to the steam pretreatment.
26
 Since the hot compressed water is 
used instead of steam, the latent heat of evaporation is saved which makes it easier to 
apply for a continuous process.
31
 Earlier, Yang and Wyman have reported that flow 
through process fractionated more hemicelluloses and lignin from corn stover as 
compared  to batch system under the conditions of similar severity.
32
 In a flow through 
system, the product is continuously removed from the reactor which reduces the risk on 
condensation and precipitation of lignin components, making the biomass less digestible. 
The soluble lignin are very reactive at the pretreatment temperature and if not removed 
rapidly part of these compounds re-condense and precipitate on the biomass.
33, 34
 
 The main objective of this work is to carry out the hydrothermal pretreatments of 
Auburn switchgrass and corn stover in a flow through reactor to enhance and optimize 
the enzymatic digestibility. Another objective is to maintain the pH of liquid hydrolyzate 
near to neutral (5-6) by conducting alkali salt (K
2
CO
3
) assisted hydrothermal 
pretreatment. The purpose was to retain hemicelluloses, fractionate lignin to minimize the 
loss of carbohydrates in liquid. Hence, the effect of pretreatment temperature (150-
190?C) and addition of K
2
CO
3 
(0-0.9 wt%) is studied. The enzymatic digestibility of the 
resulting pretreated biomass is investigated with cellulase and ?-glucosidase enzymes. In 
78 
?
addition, the changes in crystallinity, surface area, and surface morphology are examined 
by X-ray diffraction (XRD), BET surface area measurement, and electron microcopy 
(SEM), respectively.  
 As discussed earlier, the liquid hydrolyzate byproduct from pretreatment contains 
several compounds resulting from the partial solubilization of lignin besides hydrolysis of 
hemicelluloses. Depending upon the pretreatment temperature, 25-45% of the initial 
biomass carbon is solubilized in hydrolzate, mostly in the form of sugars from 
hemicelluloses, degradation products such as furfural, lignin derived phenolic 
compounds, and carboxylic acids. These degradation compounds are the potential 
inhibitors during fermentation.
22
 Fermentation of cellulose (C
6
 sugar) for bioethanol 
production is a well-established process which is carried out by using Baker?s yeast 
Saccharomyces cerevisiae. However, yeast or the high ethanol yielding bacterium 
Zymomonas mobilis cannot ferment multiple sugar substrates e.g. xylose and arabinose to 
ethanol.
35
 To maximize the yield of ethanol from lignocellulosic biomass, it is proposed 
that liquid hydrolyzate containing sugars from hemicelluloses should be co-fermented 
with cellulosic sugars.
36
 Though many pentose-utilizing strains such as Escherichia coli, 
Pichia stipititis, and Zymomonas mobilis have been developed, the detoxification of 
hydrolyzate are needed.
37
 Detoxification of hydrolyzate improves their fermentability, 
however, the cost associated in the process is often higher than the benefit.
38
 
Furthermore, genetically engineered microorganisms which can utilize all sugars in 
hydrolyzate are still in developmental stage.
37
 Depending upon the type of aqueous 
pretreatment, pH of liquid hydrlozate may vary. The dilute concentration of sugars in 
liquid hydrolyzate is another challenge. 
79 
?
 In this work an alternate pathway is developed to recover soluble organic carbon 
in liquid hydrolyzate.  Hydrolyzate is subjected to hydrothermal carbonization to produce 
carbon microsphere byproduct. The properties of microspheres are analyzed by 
calorimetry, SEM and FTIR.  The reaction mechanism involved in hydrothermal 
carbonization is studied using gas chromatography and mass spectrometry of the 
hydrolyzate. 
3.3 Experimental Section 
Materials:  Switchgrass was locally grown at Auburn University in a test plot (Prof. 
David I. Bransby, Department of Agronomy and Soil) and corn stover was obtained from 
National Renewable Energy Laboratory (NREL, Colorado). Both feedstocks were 
reduced to (+) 60 mesh to (?) 20 mesh size before utilization.  Moisture content of both 
the feedstocks was 8-9 wt%. The composition analysis of switchgrass and corn stover are 
shown in Table 3.1.  
Table 3.1 Feedstock composition (oven dry basis). 
Components  Switchgrass  
wt% 
Corn stover  
wt%  
Cellulose  38.9  34.5  
Hemicelluloses  28.9  27.7  
Klason lignin  30.4  16.5  
Acid soluble lignin  1.3  1.3  
Ash  3.0  5.6  
  
The selected lot of Auburn switchgrass contained unexpectedly high lignin content.  
Potassium carbonate (K
2
CO
3
) and high purity standard reagents for HPLC analysis were 
purchased from Sigma Aldrich. De-ionized water was used for all the experiments.  
Apparatus: Figure 3.1 shows the schematic of the hydrothermal pretreatment apparatus. 
It consists of two high-pressure liquid pumps, an electrical tubular furnace, a water-
cooled double tube heat exchanger, a backpressure regulator, and a phase separator. 
Reactor is a 0.57? ID, 65 ml volume high-pressure stainless steel tube.  
?
Liquid 
Hydrolyzate
?
Figure 3.1 Apparatus for the hydrothermal pretreatment of biomass. 
 
 Temperature of the reaction zone was measured using a 1/16? thermocouple 
placed inside the biomass bed from the inlet. There were three additional thermocouples 
installed at different locations in the apparatus. Real time temperature indicated by the 
thermocouple placed inside the reaction zone was used for all the analyses. Reactor 
pressure was maintained using a backpressure regulator.  
Experimental Procedure: Study was conducted in a flow through scheme in the apparatus 
shown in Figure 3.1. In a typical experiment, reactor was placed inside the furnace after 
80 
?
81 
?
packing with 11.0 g of biomass.  Moisture of the biomass (switchgrass or corn stover) 
was measured before charging into the reactor. A stainless steel frit (2 ?m pore size) was 
placed in the reactor outlet to prevent the entrainment of fine solids with the liquid 
stream. Water and K
2
CO
3 
solution was pumped using high pressure pumps 1 and 2 
(Figure 3.1) and total liquid inlet flow rate was maintained 2.5 ml/min for the duration of 
study. Pretreated biomass retained in the reactor was washed with water to neutral pH in 
the reactor during the cooling cycle and collected for the solid composition analysis and 
subsequent enzymatic hydrolysis. Experiments were conducted using pure water and 
K
2
CO
3 
(0-0.9 wt%) dissolved in water for temperature range of 150-190?C (Table 3.2).  
Table 3.2 Pretreatment details conducted at 3.4 MPa and 20 min steady operation time 
and the weight loss of switchgrass after pretreatment (oven dry basis). 
Switchgrass Temperature 
(?C) 
K
2
CO
3
 
(wt%) 
Weight loss 
(wt%) 
Run 1 150 0 17.0 
Run 2 150 0.6 17.8 
Run 3 150 0.9 17.1 
Run 4 175 0 34.0 
Run 5 175 0.9 35.0 
Run 6 190 0 40.0 
Run 7 190 0.9 50.0 
Corn stover    
Run 8 150 0 25.6 
Run 9 150 0.45 34.4 
Run 10 190 0.45 50.0 
 After completing the desired (20 minute) steady operation, heating was stopped 
and reactor was rapidly cooled to below 50?C. In a flow through process, biomass is 
subjected to a typical temperature cycle as shown in Figure 3.2.  
0
50
100
150
200
0 10203040506
T
e
m
p
er
at
u
r
e
 (
?
C)
Time (min)
0
Steady operation
time
Cooling 
time
Preheating
time
?
Figure 3.2 Typical temperature cycle during pretreatment process. 
 
 The liquid products collected during all three stages of operation - preheating, 
steady operation, and cooling/washing - were analyzed for total organic carbon (TOC) 
content. Liquid samples were centrifuged at 3600 rpm before analyzing TOC to remove 
suspended solids. The gaseous products which were appreciably low in the temperature 
range of study were vented without analysis. 
 Hythothermal carbonization experiments were conducted in a 45 ml batch reactor 
(Figure 3.3).  
82 
?
Heater
Reactor
Thermocouple
?
Figure 3.3 Batch reactor for hydrothermal carbonization process. 
 
 The liquid hydrolyzate was charged to reactor and heated to 200?C for the desired 
reaction time under autogeneous pressure condition. After the reaction, reactor was 
rapidly cooled to ambient condition and the precipitated solids were recovered using 
centrifuge. Obtained powder was washed with water, freeze dried for 48 h and preserved 
for further characterization. The liquid fraction recovered after centrifugation was 
analyzed for the TOC content to determine the conversion of organic carbon into powder 
and gaseous fraction. Again, at the studied temperature condition (200?C), gaseous 
products were appreciably low and so were not collected. 
 Since, liquid hydrolyzate comprised of mainly sugar compounds, glucose solution 
of equivalent TOC value was also used as model compound for the similar experiments 
and the results were compared. Table 3.3 shows the details of experiment conducted for 
hydrothermal carbonization. 
?
?
?
83 
?
84 
?
Table 3.3 Details of hydrothermal carbonization experiments. 
Run no Input Solution Time (h) 
Run I Liquid hydrolysate 1 
Run II Liquid hydrolysate 2 
Run III Liquid hydrolysate 4 
Run IV Glucose 1 
Run V Glucose 2 
Run VI Glucose 4 
 
 Liquid hydrolyzate collected during the hydrothermal pretreatment experiments 
(run 6) typically contained 1.1-1.2 wt% of organic carbon and this solution was used for 
the carbonization study. 
Experimental and analytical accuracy: All the thermocouples (? 0.1?C) were calibrated 
before the start of experiment. High precision HPLC pumps (Chrome Tech, USA) were 
used for the liquid flow. Pump flow rates were calibrated before every experiment and 
showed less than 1% variation. Multiple liquid samples were collected and analyzed for 
TOC and pH values during every experiment. Pretreatment experiments were duplicated 
or done in triplicates to confirm the reproducibility of the data. Composition analysis of 
solids after the pretreatment experiment was done in duplicates as per the NREL standard 
procedures. The standard deviation of the reported values of solid composition in 
biomass is less than 3%. Enzymatic digestibility is also performed as per NREL standard 
procedures using duplicate samples within the error of ? 5%.  
 
85 
?
3.4 Product Characterization 
3.4.1 Pretreatment 
 Pretreated switchgrass and corn stover were analyzed for carbohydrate and lignin 
composition, and enzymatic hydrolysis published in NREL technical report NREL/TP-
510-42618 (www.eere.energy.gov./biomass/analytical_procedures.html).   
Chemical composition of pretreated solids/liquid: Composition analysis of solid biomass 
was done as per NREL laboratory analytical procedures (LAP) ??Preparation of samples 
for compositional analysis,?? ??Determination of structural carbohydrates and lignin in 
biomass?, and ?Determination of ash in biomass?.   Pretreatment liquor was analyzed for 
different monomer and oligomer sugars as per NREL LAP: ?Determination of sugars, 
byproducts and degradation products in liquid fraction process samples.??  HPLC with 
Bio-Rad Aminex HPX-87P column was employed for determination of sugar 
concentrations in liquid samples while Bio-Rad Aminex HPX-87H column was used for 
acids and degradation products.  
Enzymatic hydrolysis of pretreated solids: Enzymatic hydrolyses of the original and 
pretreated biomasses were carried out in 125 ml flasks with rubber stoppers on the top. 
The protocol used was similar to the NREL LAP ?Enzymatic saccharification of 
lignocellulosic biomass.? Cellulase commercial enzyme (Spezyme CP, Genencor) with 
initial loading of 15 FPU/g glucan and ?-glucosidase (Novo 188
?
, Novozyme) 30 CBU/g 
glucan were used for the study. Other test conditions were maintained as per the protocol: 
0.05 N sodium citrate buffers were used and the initial pH was maintained as 4.8. 
Antibiotics tetracycline and cyclohexamide were used to avoid contamination. Flasks 
were placed in an incubator which maintained constant reaction temperature of 50?C at 
86 
?
150 rpm. Liquid samples were collected every 1, 12, 24, and 72 h of hydrolysis and then 
were placed in boiling water for 5 minutes to kill the reaction. Then the samples were 
centrifuged for 10 minutes at 14000 rpm to remove solids and analyzed for released 
glucose in the liquid phase by HPLC with Bio-Rad Aminex HPX-87P column and 
refractive index detector. The enzymatic digestibility is calculated as the ratio of glucan 
equivalent of glucose + cellobiose in the liquid phase to the initial glucan loading (1 g). 
Glucan equivalent of glucose and cellobiose is calculated as 0.9?solublized glucose 
amount + 0.95?solublized cellobiose amount. Though, periodic hydrolysis samples were 
collected as discussed earlier only 72 h glucan digestibility data are discussed in the study 
for all the pretreatment conditions.  
XRD analysis: Rigaku Miniflex powder X-ray diffractometer equipped with a Cu K?
1
 
radiation source at 30 kV voltages, 15 mA current and a miniflex goniometer was used 
for the XRD analysis. Diffraction patterns were collected in the 2? range of 10-35? at a 
scan speed of 1?/min. and step size of 0.01?. The cellulose crystallinity index (CI) in the 
switchgrass was determined
39
 by 
 CI = (I
002
 ? I
AM
)
 
/ I
002
     [1]  
where I
002 
is the peak intensity corresponding to (002) lattice plane of cellulose molecule, 
and I
AM  
is the peak intensity observed at 2? equal to 18?. I
002
 represents both crystalline 
and amorphous material while I
AM
 represents amorphous material only. 
BET surface area measurement: Surface area is a measure of the exposed surface of a 
solid sample on the molecular scale. The method of Brunauer, Emmett, and Teller (BET) 
is used to determine the total surface area of materials. Raw and pretreated switchgrass 
87 
?
samples were analyzed for the BET surface area using Krypton as adsorptive gas by 
multi-point analysis method at Micromeritics Analytical Services (Norcross, GA). 
Electron microscopy: The untreated and pretreated samples were prepared on adhesive 
carbon tape on an aluminum stub followed by sputter coating of gold. Surface 
morphology of the sample was studied using an environmental scanning electron 
microscopy system (Zeiss EVO 50). 
3.4.2 Hydrothermal Carbonization: 
 After hydrothermal carbonization of hydrolyzate (and glucose solution for 
comparison) product liquid and solid were separated.   Solids were characterized using 
SEM (Zeiss EVO 50), FTIR (Nicolet IR100), elemental carbon analysis, and calorimetry 
(IKA-C200 Calorimeter).  Liquids were analyzed by TOC and GC-MS as following.  
Agilent model HP-5 column with 5% phenyl methyl siloxane (30 m long, 0.25 mm 
diameter, 0.25 ?m film thickness) was utilized for identifying the compounds in liquid 
samples. Oven was programmed as: 45?C (hold 5 min) and ramped up (rate 5?C/min) to 
250?C (hold 8 min). Injector was kept at 250?C in split mode (split ratio 50:1) with 
helium as carrier gas. Sample volume was 0.2 ?l and compounds were identified using 
NIST library of mass spectra. 
3.5 Results and Discussion 
3.5.1 Pretreatment 
Hydrothermal pretreatment: Figure 3.4 shows the percentage removal (oven dry basis) of 
hemicelluloses and lignin from switchgrass upon pretreatment 3.4 MPa and 20 min of 
steady operation time (run 1,4 and 6) and varying temperature. 
 
0
20
40
60
80
100
150?C 175?C 190?C
W
e
ig
h
t
 lo
s
s
 
(%
)
Pretreatment temperature
Hemicellulose
Lignin
?
Figure 3.4 Percentage removal of hemicelluloses and lignin with temperature for the 
switchgrass pretreated at 3.4 MPa and 20 min steady operation time. 
?
  It was observed that hemicelluloses hydrolysis to water soluble products 
increased from 30% to 77% as the pretreatment temperature increased from 150 to 
190?C. Part of the lignin (~40%) was also solubilized, but it remained nearly constant in 
the temperature range of 150-190?C. And pH of the liquid hydrolyzate varied between 
3.5 to 4.5 depending on the pretreatment temperature. Typically, pH liquid hydrolyzate 
decreases with increase in temperature due to the increased formation of organic acids.
40
 
Low activation energy of lignin causes partial degradation of lignin in hydrothermal 
medium at temperature below 200?C.
26, 41
 The ether bonds in lignin start breaking above 
150?C in a weakly acidic medium.
42
 Cellulose remained stable at below 190 
o
C and about 
10% of weight loss was observed at 190?C. Cellulose composition in washed pretreated 
solid biomass was increased to 51-58 wt% compared to 38.9 wt% in native switchgrass. 
These results are in agreement with those for Almo and Kanlow switchgrass.
25, 43
 In 
general, cellulose hydrolysis is mainly observed above 210?C in hydrothermal medium. 
88 
?
Hemicelluloses are the most susceptible in hydrothermal medium because of hydronium 
ions generated from water autoionization. Hemicelluloses possess side groups, including 
acetic acid, pentoses, hexuronic acids, and deoxyhexoses. The lower stability of 
hemicelluloses is mainly due to the fact that their side chains reduces the aggregation of 
their molecules and inhibits the formation of H-bonds.
26
 
 Figure 3.5 shows the 72 h glucan digestibility for pretreated switchgrass. The 
untreated switchgrass has almost negligible enzymatic reactivity.  
0
20
40
60
80
100
Untreated 150?C 175?C 190?C
G
lu
c
a
n
 d
ig
e
s
t
ib
ilit
y
 
(
w
t
%
)
Pretreatment temperature
?
Figure 3.5 Glucan digestibility (72 h) of hydrothermally pretreated switchgrass. 
 
 However, as a result of partial removal of hemicelluloses and lignin during the 
pretreatment, physical and chemical properties of switchgrass are altered. The 
pretreatment helped in improving the enzyme accessiblity and effectiveness for cellulose 
hydrolysis by increasing the internal surface area and pore volume. As a result, cellulose 
depolymerization to sugars substantially enhanced. Glucan digestibility of switchgrass 
89 
?
pretreated at 150?C was still low (25%) but with the increase of pretreatment 
temperature, more than 82% of digestibility could be achieved.  
Alkali salt (K
2
CO
3
) assisted hydrothermal pretreatment: Figure 3.6 shows the effect of 
K
2
CO
3
 addition during the hydrothermal pretreatment at 150?C.  
0
15
30
45
60
0% 0.6 wt% 0.9 wt% 
G
lu
c
a
n
 d
ig
e
s
t
ib
ilit
y
 
(
w
t
%
)
Potassium carbonate
 
Figure 3.6 Glucan digestibility (72 h) of switchgrass pretreated at 150?C in the presence 
of K
2
CO
3
. 
 
 With the addition of 0.9 wt% K
2
CO
3
 in water during pretreatment, enzymatic 
digestibility of the substrate doubled compared to that without K
2
CO
3
. On the component 
basis, more than 40% of lignin was removed during run 3, and nearly 90% of 
hemicelluloses were retained in solid fraction. In runs 2 and 3, which are conducted in the 
presence of K
2
CO
3
,
 
pH of liquid hydrolyzate increased to 7 as compared to 5 in run 1 
(i.e., without K
2
CO
3
). As expected, the hydrolysis of hemicelluloses could be minimized 
mainly due to the relatively low temperature and near neutral pH. Alkali has selectivity 
towards lignin and cause lignin degradation by breaking aryl ether linkages which 
constitutes approximately 50-70% of total linkages. On the other hand, diaryl ethers and 
carbon-carbon bonds are relatively stable. Hydroxyl ion catalyzes the cleavage of ether 
linkages in the lignin and thus liberating the soluble phenolics in the liquid. The breakage 
90 
?
of these bonds increases the hydrophilicity of lignin. Alkaline medium also causes the 
solvation of hydroxyl groups in carbohydrates which in turn swells the sugar residue, thus 
increasing the diffusivity of reagent through the capillaries.
44
 These all combined factors 
would help in increasing the effective adsorption of enzyme molecule on pretreated 
biomass and thus increasing the accessibility and digestibility of carbohydrate fraction in 
the biomass.
45, 46
 Figure 3.7 compares the SEM images of raw and the pretreated 
switchgrass (run 3), where additional pores created after the pretreatment can be seen. 
?
Figure 3.7 SEM images of untreated and pretreated (run 3) switchgrass taken at 600X 
magnification. 
 
 Figure 3.8 shows the effect of pretreatment temperature on the enzymatic 
digestibility of pretreated switchgrass. These samples were pretreated in the presence of 
0.9 wt% of K
2
CO
3
 at 3.4 MPa and 20 min steady operation time. 
91 
?
0
20
40
60
80
100
150?C175?C190?C
G
l
u
c
a
n
 D
i
g
e
s
t
ib
ilit
y
 
(w
t
%
)
Pretreatment temperature
?
Figure 3.8 Glucan digestibility (72 h) of K
2
CO
3 
(0.9 wt%) assisted hydrothermally 
pretreated switchgrass. 
 
 Glucan digestibility increased with pretreatment temperature and 85.5% of glucan 
digestibility could be achieved in run 7. This was marginally higher compared to run 6. 
To understand the phenomena, it is important to observe the Figure 3.9 which shows the 
component wise weight loss during runs 3, 5 and 7.  
0
20
40
60
80
100
150?C175?C190?C
W
e
i
ght
 l
o
s
s
 
(
%
)
Pretreatment temperature
Hemicellulose
Lignin
 
Figure 3.9 Percentage removal of hemicelluloses and lignin with temperature for the 
switchgrass pretreated in the presence of 0.9 wt% of K
2
CO
3
. 
 
 Though delignification increased from 40.8 to 67.3% in these experiments, 
hemicelluloses were also hydrolyzed substantially during these runs. Percentage removals 
92 
?
of hemicelluloses were 10.5, 50, and 75% for the runs 3, 5, and 7, respectively. Liquid 
pH for run 7 dropped down to 5.  Increase in pretreatment temperature increases the 
delignification as well as hemicelluloses loss from the solids. Hemicelluloses are known 
to undergo numerous physical and chemical changes such as swelling, dissolution, 
saponification, reprecipitation, peeling and glycosidic cleavage reactions at higher 
temperature in the presence of hydroxyl ions.
44
 Presence of several biomass derived 
components in the reaction medium makes the chemistry very complex. 
 If we compare the glucan digestibility of the run 6 to run 7, it can be concluded 
that at the higher pretreatment temperature, hydrothermal treatment alone can enhance 
the enzymatic digestibility of pretreated switchgrass higher than 80%. Figure 3.10 and 
3.11 compares the overall material balance based on the 100 kg oven dry switchgrass for 
run 6 and 7.  
100 kg dry 
Switchgrass
37.95 kg Glucan
28.20 kg Xylan
30.93 kg Lignin
3.01 kg Glucan
19.28 kg Xylan
12.80 kg Lignin
Liquid Hydrolyzate
Hydrothermal 
Pretreatment
59.88 kg 
Pretreated Solids
34.58 kg Glucan
6.55 kg Xylan
18.13 kg Lignin
Enzymatic
Hydrolysis
Enzymes 0.721 kg Cellulase
0.216 kg beta-glucosidase
Cellulase: 20.84 mg protein / g Glucan
?-G: 6.25 mg protein / g Glucan
32.45 kg 
Solid Residues
6.24 kg Glucan
5.86 kg Xylan
18.13 kg Lignin
28.37 kg Glucan
0.65 kg Xylan
Fermentable Sugars
 
Figure 3.10 Material balance for hydrothermally pretreated switchgrass (run 6) at 190?C. 
 
 
93 
?
100 kg dry 
Switchgrass
37.95 kg Glucan
28.20 kg Xylan
30.93 kg Lignin
6.24 kg Glucan
18.67 kg Xylan
20.53 kg Lignin
8.2 kg K2CO3
K
2
CO
3
Assisted 
Hydrothermal 
Pretreatment
Liquid Hydrolyzate
8.2 kg K
2
CO
3
50.05 kg 
Pretreated Solids
Enzymes 0.646 kg Cellula se
0.193 kg beta-glucosidase
Cellulase: 20.84 mg protein / g Glucan
?-G: 6.25 mg protein / g Glucan
Enzymatic
Hydrolysis
26.50 kg Glucan
2.02 kg Xylan
4.50 kg Glucan
5.24 kg Xylan
10.40 kg Lignin
22.87 kg 
Solid Residues
Fermentable Sugars
 
Figure 3.11 Material balance for K
2
CO
3
 assisted hydrothermally pretreated switchgrass 
(run 7) at 190?C. 
 
 Presence of K
2
CO
3
 supported the delignification even at lower temperature (150-
175?C), and so the enzymatic digestibility of switchgrass was comparatively higher than 
that of hydrothermal pretreatment. Pressure is another important factor, which may be 
considered. Constant pressure 3.4 MPa, which is higher than the vapor pressure of water 
in the range of study (150-190?C) was chosen in this study to maintain water in liquid 
phase. Previous studies on the effect of pressure have reported that the influence of 
pressure on hydrolysis of biomass is normally limited to keeping water in liquid phase at 
the operating temperature.
47
 In fact, ionization constant of water shows small dependence 
on pressure (ionization slightly increases with increasing pressure).
48
 But, pressure and 
temperature can be used to influence the density and transport properties (viscosity, 
thermal conductivity, and diffusivity) of hydrothermal medium. Change in pressure 
mainly influences the density-dependent physical properties such as partition coefficient 
and dielectric constant.
49
 To understand the effect of pressure during pretreatment, an 
experiment (not shown in Table 3.2) at higher pretreatment pressure (13.8 MPa) keeping 
94 
?
95 
?
other process parameters same as run 3 was conducted using switchgrass.  The glucan 
digestibility after pretreatment increased from 53% (run 3) to 71.2%. It was observed that 
the overall weight loss after pretreatment increased with pretreatment pressure from 17.1 
to 39.4%. On the component basis, removal of hemicelluloses and lignin both increased 
substantially, whereas almost no loss (< 5%) of cellulose was observed.  
BET surface area: The purpose of the pretreatment is to remove resistance in biomass 
which limits the enzyme accessibility to cellulosic sites. Enzymatic hydrolysis of 
lignocellulosic biomass is limited by several factors such as crystallinity of cellulose, 
degree of polymerization, available surface area and the lignin content.
22, 50
 Among these 
factors, the accessible surface area is regarded as one of the most important factors 
affecting the effectiveness of enzymatic digestibility.
35
 After the pretreatment, internal 
surface area may increase because of following changes: (i) due to the fragmentation and 
development of cracks (Figure 3.7), (ii) removal of hemicelluloses and lignin which 
diminishes shielding effects and opens up additional pores.
35
 To understand the changes 
in switchgrass internal surface area after the pretreatment, samples from run 6 and run 7 
were analyzed for the BET surface area. It is observed that internal surface area of 
pretreated switchgrass increased by 3 to 5 folds (Table 3.4).  
Table 3.4 BET surface area of untreated switchgrass and pretreated samples from run 6 
and run 7. 
Biomass BET surface area 
(m
2
/g) 
Untreated switchgrass 0.8251 
Pretreated switchgrass, Run 6 4.4184 
Pretreated switchgrass, Run 7 2.8300 
 
 An interesting and contradicting observation is that though pretreated switchgrass 
from run 6 has 55% more internal surface area than that of run 7, their glucan 
digestibility are comparable. Notably, the lignin content in run 6 and run 7 solids are 30.3 
and 20.8%, respectively. Nonproductive and irreversible adsorption of enzymes to lignin 
is often claimed to be responsible for the need for high enzyme loadings. 
51
 Lignin limits 
the rate and extent of enzymatic hydrolysis due to its shielding effect. If only the internal 
surface area is the reason, run 6 pretreated sample should have higher digestibility than 
run 7, which not the case here. This suggests that it is not just the internal surface area or 
lignin content in pretreated biomass affecting enzymatic digestibility of the substrate but 
probably some other molecular characteristics such as degree of polymerization, 
crystallinity, and other internal rearrangement of the structure are also responsible. 
XRD pattern: Figure 3.12 shows the XRD pattern of untreated switchgrass and pretreated 
switchgrass from runs 6 and 7.  
?
?????????????????????????
12 16 20 24 28 32 36
Angle (2?)
Untreated
Only water
(Run 6)
0.9 wt% K
2
CO
3
(Run 7)
I
002
I
18.5?
 
Figure 3.12 XRD patter of untreated and pretreated (run 6 and 7) switchgrass. 
 
96 
?
 The untreated switchgrass has a broader peak (I
002
) with cellulose crystallinity of 
53.5%. Peak (I
002
) became sharper after the pretreatment and also percentage crystallinity 
increased. The crystallinity of cellulose in runs 6 and 7 samples were almost same and 
was 67.1 and 67.6%, respectively. Increase in percentage crystallinity may be attributed 
to the hydrolysis and removal of the amorphous part of cellulose during pretreatment.
52
  
Comparison with corn stover:  The composition of lignocellulosic biomass is a key factor 
affecting efficiency of biofuel production during conversion processes. The structural and 
chemical compositions of lignocelluloses are highly variable because of genetic and 
environmental influences and their interactions. Nevertheless the chemical composition 
of switchgrass and corn stover is relatively similar (Table 3.1) except the fact that the 
switchgrass used in this study had much higher lignin content than corn stover. Figure 
3.13 compares the glucan digestibility of corn stover pretreated in different process 
condition.  
?
0
20
40
60
80
100
Untreated 150?C 150?C 190?C 
G
l
u
c
a
n
 D
i
g
e
s
t
ib
il
it
y
 
(
w
t
%
)
0% 
K
2
CO
3
0.45 wt% 
K
2
CO
3
0.45 wt% 
K
2
CO
3
?
Figure 3.13 Glucan digestibility (72 h) of untreated and pretreated corn stover. 
 
97 
?
 Interestingly, corn stover required less severe pretreatment than switchgrass. Corn 
stover pretreated at 190?C had 92.6% of glucan digestibility even though K
2
CO
3
 
concentration in water was only 0.45 wt%. Under the similar pretreatment condition, 
switchgrass (run 7) pretreated in the presence of 0.9 wt% K
2
CO
3
 had only 85.5% of 
digestibility. The SEM images of corn stover (Figure 3.14) reveal the fact that even 
untreated corn stover has relatively more porous structure and also morphologically it 
shows more structural changes than switchgrass after the pretreatment.  
?
Figure 3.14 SEM images of untreated and pretreated (run 8) corn stover taken at 600X 
magnification. 
3.5.2 Hydrothermal Carbonization: 
98 
?
 At present, co-fermentation of liquid hydrolyzate for ethanol production is the 
major focus of research. The challenges associated with the detoxification and 
development of pentose-utilizing strains for co-fermentation has already been discussed 
in introduction section. On the other hand, there have been very few studies conducted on 
the development of alternate pathway for utilizing the liquid hydrolyzate excluding the 
studies on hydrogen production through dark fermentation and on biogas production 
through anaerobic digestion.
36
 However, the hydrogen yield obtained by dark 
fermentation is relatively low (approximately 20-30% of the energy content in the 
organic matter) while the remaining organic matter is converted to volatile fatty acids.
36
  
99 
?
The biogas production by anaerobic digestion has the possibility to utilize all types of 
organic molecules however; setting up another bioreactor for the purpose may bring some 
additional cost to the process. Hydrothermal carbonization of liquid hydrolyzate is 
another potential alternative, where majority of organic content of the liquid hydrolyzate 
can be recovered as carbon rich microspheres.
53
 The process is typically conducted at 
180-200?C, so there is no additional heat requirement since hydrothermal pretreatment is 
also conducted in the similar range for achieving the optimal enzymatic digestibility of 
pretreated biomass. Table 3.5 shows the typical sugar composition of liquid hydrolyzate 
collected at 190?C during experiment run 6.  
Table 3.5 Sugar compounds analyzed by HPLC in liquid hydrolyzate collected at 190?C 
during run 6. 
Compounds  Before hydrolysis 
 g/L  
After hydrolysis 
g/L  
Cellobiose  0.5  -  
Glucose  0.7  2.7  
Xylose  2.6  18.1  
Galactose  0.4  1.8  
Arabinose  1.1  2.4  
Furfural  1.1  -  
Acetic acid  1.5  -  
 
 Xylose is the major component besides the host of other inhibitory compounds 
present in the liquid. GC-MS analysis of the sample confirmed the presence of several 
other phenolic compounds such as furfural, 2,3-dihydrobenzofuran, 2-
furancarboxaldehyde, benzaldehyde 3-hydroxy- , 1,4-benzenediol 2-methyl-, vanillin etc. 
It can be seen that by carbonizing this liquid at 200?C for just 2 h, nearly 60% of the 
organic compounds (based on TOC) could be precipitated as solids (Figure 3.15).  
0
20
40
60
80
1 h 2 h 4 h
C
onver
s
i
on
 (
%
)
Time
Liquid Hydrolysate
Glucose
?
Figure 3.15 Percentage conversion of liquid hydrolyzate (based on TOC of input) of run 
6 and glucose solution. 
 Though, maintaining the temperature for 4 h (Table 3.3, run III), the conversion 
did not increase. Similar results were observed when glucose solution was used for 
comparison. The precipitated solids were globular with their diameter ranging from 0.2-2 
?m (Figure 3.16).  
a b
?
Figure 3.16 SEM images (at 20,000X magnifications) of carbon microspheres produced 
from (a) glucose solution and (b) liquid hydrolyzate of run 6. 
 
100 
?
 The microspheres produced from glucose solution (Figure 3.16a) were mostly of 
uniform size but they were much finer (submicron size, 0.2-0.4 ?m) compared to the 
101 
?
microspheres obtained from the liquid hydrolyzate. This may be mainly because of the 
many phenolic compounds present in liquid which also contributed in polymerization 
process. These microspheres appeared brown and were fluffy. Elemental carbon in 
microspheres produced in run II and run V were 64.88 and 66.06%, respectively. Titirici 
et al. had also studied the hydrothermal carbonization of carbohydrate model compounds 
such as glucose and xylose. There study reported the elemental carbon in the microsphere 
as 64.47 and 68.58% from glucose and xylose, respectively.
54
 The HHV of these carbon 
rich microspheres from run II was 24.8 MJ/kg which is comparable to lignin?s HHV (24-
26 MJ/kg).  
 Glucose conversion to its degradation products and the carbon microspheres was 
almost complete (99%) just in an hour of reaction time. Only traces (~1%) of glucose was 
observed in the output liquid from run IV. The concentration of acetic acid in liquid 
product was found to increase with reaction times, i.e run VI had higher acetic acid 
concentration than run IV. GC-MS spectra of the liquid samples from runs IV, V, and VI 
are shown in Figure 3.17.  
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
500000
1000000
1500000
2000000
2500000
3000000
3500000
Time-->
Abundance
TIC: 091809-S-3.D\ data.ms
Levoglucosan
5 HMF
LA
Furfural
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
50000
100000
150000
200000
250000
300000
350000
400000
450000
Time-->
Abundance
TIC: 091809-S-4.D\ data.ms
Levoglucosan
5 HMF
levulinic acid
Furfural
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
50000
100000
150000
200000
250000
300000
350000
400000
450000
Time-->
Abundance
TIC: 092009-S-5.D\ data.ms
levulinic acid
(a) Run IV
(b) Run V
(c) Run VI
Levulinic acid
?
Figure 3.17 GC-MS spectra of liquid product from (a) run IV, (b) run V, and (c) run VI. 
 
 Some of the key compounds indentified in the output liquid from glucose 
experiments and their trend for run IV, V, and VI (based on GC-MS spectra) are shown 
in Table 3.6.   
Table 3.6 Key compounds identified in the product liquid from run IV, V, and VI and 
their trend with reaction time. 
No. Compounds Retention time (min) Trend 
1 Furfural 12.86 Initially increased (run IV) and then started 
decreasing. Not observed in (run VI) 
2 pentanoic acid, 4-oxo-,  
(levulinic acid) 
22.67 Increased with reaction time i.e. from run IV to run 
VI 
3 2-Furancarboxaldehyde, 5-
(hydroxymethyl)-  
(5HMF) 
27.39 Initially increased (run IV) and then started 
decreasing. Not observed in (run VI) 
4 1,6-Anhydro-.beta.-D-
glucopyranose (levoglucosan) 
34.99 Initially increased (run IV) and then started 
decreasing. Not observed in (run VI) 
102 
?
 It is interesting to note from Figure 3.17 that the ring compounds (e.g. furfural, 
5HMF, and levoglucosan) started to disappear with increase in reaction time from 1 h to 
4 h (run IV, V, and VI), whereas concentration of organic acids (e.g. levulinic acid) 
increased. In the run VI (Figure 3.17c), none of the ring compounds discussed in Table 6 
are present. Glucose in hydrothermal medium at relatively low temperature (180-200?C) 
range undergoes mainly dehydration and partial fragmentation (C-C bond breaking) 
reactions. The intermediate compounds are mainly furan like compounds, organic acids 
and aldehydes.
55
 Furan like ring compounds may undergo polymerization via aldol 
condensation to form soluble polymers. Aromatization of soluble polymers takes place 
under the reaction condition and when the aromatic clusters in aqueous solution reach the 
critical super-saturation point, they precipitate as carbon rich microspheres.
56
 Figure 3.18 
shows the simplified reaction scheme of glucose to carbon rich microspheres in 
hydrothermal medium.  
O
CHO
HOH
2
C
O
CHO
OH
O
OH
OH
HO
HO
O
OH
OHHO
HO
HOH
2
C CH2OH
Glucose
Fructose
5-HMF
Furfural
OH
OH
O
O
O
O
OO
H
Carbon 
microspheres
Intermediates
Phenols
Acids
Acetic acid
Levolunic acid
Lactic acid
Formic acid
?
Figure 3.18 Simplified reaction pathway of glucose during hydrothermal carbonization 
reactions.
54, 56, 58
 
 
103 
?
 The carbon rich microspheres produced in run II and V were further analyzed by 
FTIR to understand the presence of functional groups. Figure 3.19 compares the FTIR 
spectra from run II and V. Also, Table 3.7 shows the position of IR absorption bands and 
their assignments for the solid samples from run II and V.  
?
4000 3500 3000 2500 2000 1500 1000 500 0
Wave number (cm
-1
)
Run V
Run II  
?
Figure 3.19 FTIR spectra of solids recovered in run II and V. 
 
 
Table 3.7 Functional groups in the FTIR spectra of carbon rich microspheres from run II and V.
59
 
 
Position of the bands, cm
-1
 Functional group Belongs 
3405  O?H stretching Run II and V 
2930-2850  CH, CH
2
 stretching Run II and V 
1680-1580  C=C stretching  Run II and V 
1510 Benzene ring stretching  Run II and V 
1456 C-O-H stretching Run II and V 
1428 C-C-H stretching Run II and V 
1375-1317 C-H benching  Run II and V 
1281 C?C-H stretching Run II and V 
1235 C-O-H stretching Run II and V 
1112 C-O-H stretching Run II and V 
1056 C?O stretching  Run II and V 
897 Alienation Zone ?-bond absorption Run V 
803 C=H Benzene ring stretching Run II 
700-400 C-H benching - 
 
104 
?
105 
?
 The broad absorbance peak of O?H stretching vibration between 3,200 and 3,600 
cm
-1
 indicate the presence of water impurities and other polymeric O?H in the solids. The 
presence of alkanes was indicated by the strong absorbance peak of C?H vibration at 
2,930 cm
-1
 and the C?H deformation vibration at 1,460 cm
-1
. The absorbance peak at 
1,700 cm
-1
 represented the C=O stretching vibration indicating the presence of ketones 
and aldehydes. The presence of both O?H and C=O stretching vibrations also indicated 
the presence of carboxylic acids and their derivative esters. The possible presence of 
alkenes was indicated by the absorbance peaks between 1,680 and 1,580 cm
-1
. The peaks 
between 1,300 and 950 cm
-1
 were due to the presence of primary, secondary, and tertiary 
alcohols, phenols, ethers, and esters showing the C?O stretching and O?H deformation 
vibration. Absorbance peaks between 900 and 650 cm
-1 
indicated the possible presence of 
single, polycyclic, and substituted aromatic groups. The presence of aromatic rings (900 
and 650 cm
-1
) and C=O stretching as well as C?O stretch indicated the presence of 
aromatic esters. The results were confirmed by the peaks 1281, 1428 and 1513cm
-1
 in the 
FTIR. At the same time, 1600 and 1700 cm
-1
, the peaks of polymeric product were also 
shown in the spectra.  The results proved that the carbon rich solids were composed of 
the aromatics and polymeric product which are produced in the hydrothermal 
carbonization process. 
57
 
 Microspheres which are fine and easy to compact seem to have good solid fuel 
value based on their elemental carbon and HHV.  These powders can be mixed and co-
fired with lignin. The study have shown that just by maintaining the liquid hydrolyzate 
temperature for another two hours, nearly 60% of the organic matters can be converted as 
high energy density powders. The product liquid after this process mainly contains the 
106 
?
organic acids. Though the results only from run 6 liquid hydrolyzate are discussed here, 
carbonization of liquid hydrolyzate from run 7 yielded the similar results. This confirms 
that liquid hydrolyzates recovered after aqueous pretreatment processes can be used 
potentially for recovering soluble carbons as high energy density products.  
3.6 Conclusion 
 Switchgrass can be pretreated effectively in a flow through reactor in 
hydrothermal medium. Enzymatic digestibility of pretreated biomass depends on the 
pretreatment temperature. Addition of small amount of K
2
CO
3
 increased the glucan 
digestibility even at low temperatures (150-175?C). At a higher pretreatment temperature 
(190?C), the study showed that even though switchgrass pretreated just with water had 
more internal surface area than that pretreated in the presence of K
2
CO
3
, both the 
substrate had almost similar glucan digestibility. Therefore, it is not only the internal 
surface area, but many other factors such as degree of polymerization, structural 
rearrangements, lignin content after pretreatment are responsible in improving the sugar 
yields. Corn stover required milder pretreatment conditions than switchgrass. 
 An alternative pathway for the utilization of liquid hydrolyzate is developed, 
where dissolved sugars can be recovered as carbon rich microspheres by hydrothermal 
carbonization. Sugars (C
5
 and C
6
) degrade to furan based intermediates (ring compounds) 
and organic acids during the process. Ring compounds further undergo condensation 
polymerization and precipitate as carbon rich microspheres. The energy densities of these 
solids are comparable to that of lignin, which make them suitable as solid fuels. The 
hydrothermal carbonization process can potentially be used to recover valuable byproduct 
with a good energy value. 
107 
?
3.7 Acknowledgments 
Author is thankful for the financial supports from the National Science Foundation (grant 
NSF-CBET-0828269) and Alabama Center for Paper and Bioresource Engineering.  In 
addition, the authors are thankful to Prof. David I. Bransby for supplying switchgrass, 
and to Mr. Ramesh V. Pallapolu, Prof. Sushil Ashikari, and Mrs. Suchithra T. 
Gopakumar for help with some analyses. 
 
3.8 References 
1. Taherzadeh, M. J.; Karimi, K., Enzyme based hydrolysis processes for ethanol 
from lignocellulosic materials: A review. BioResources 2007, 2, (4), 707-738. 
2. Perlack, R. D.; Wright, L. L.; Turhollow, A. F.; Graham, R. L.; Stokes, B. J.; 
Erbach, D. C. Biomass as feedstock for a bioenergy and bioproducts induatry: The 
technical feasibility of a billion-ton annual supply; US Department of Energy 
US Department of Agriculture: 2005. 
3. Petchpradab, P.; Yoshida, T.; Charinpanitkul, T.; Matsumura, Y., Hydrothermal 
Pretreatment of Rubber Wood for the Saccharification Process. Ind. Eng. Chem. Res. 
2009, 48, (9), 4587-4591. 
4. Rinehart, L., Switchgrass as a bioenergy crop. www.attra.ncat.org 2006. 
5. Mosier, N.; Wyman, C.; Dale, B.; Elander, R.; Lee, Y. Y.; Holtzapple, M.; 
Ladisch, M., Features of promising technologies for pretreatment of lignocellulosic 
biomass. Bioresource Technology 2005, 96, (6), 673-686. 
108 
?
6. Laxman, R. S.; Lachke, A. H., Bioethanol from lignocellulosic biomass, Part 1: 
Pretreatment of the substrates. In Handbook of plant-based biofuels, Pandey, A., Ed. 
CRC Press: 2008; pp 121-139. 
7. Gupta, R.; Lee, Y. Y., Mechanism of cellulase reaction on pure cellulosic 
substrates. Biotechnology and Bioengineering 2008, 102, (6), 1570-1581. 
8. Zhang, Y.-H. P.; Berson, E.; Sarkanen, S.; Dale, B. E., Sessions 3 and 8: 
Pretreatment and Biomass Recalcitrance: Fundamentals and Progress. Appl Biochem 
Biotechnol 2009, 153, 80-83. 
9. Wyman, C. E.; Dale, B. E.; Elander, R. T.; Holtzapple, M.; Ladisch, M. R.; Lee, 
Y. Y., Coordinated development of leading biomass pretreatment technologies. 
Bioresour. Technol. 2005, 96, 1959-1966. 
10. Lynd, L. R., Overview and evaluation of fuel ethanol from cellulosic biomass: 
technology, economics, the environment, and policy. Ann Rev Energy Environ 1996, 21, 
403-465. 
11. P?rez, J. A.; Ballesteros, I.; Ballesteros, M.; S?ez, F.; Negro, M. J.; Manzanares, 
P., Optimizing Liquid Hot Water pretreatment conditions to enhance sugar recovery from 
wheat straw for fuel-ethanol production. Fuel 2008, 87, 3640-3647. 
12. Hsu, T.-A., Pretreatment of biomass. Handbook on bioethanol: Production and 
Utilization Taylor and Francis, Washington, DC: 1996. 
13. Sun, Y.; Cheng., J. J., Hydrolysis of lignocellulosic material for ethanol 
production: A review. Bioresource Technology 2002, 83, 1-11. 
14. Diaz, M. J.; Cara, C.; Ruiz, E.; Romero, I.; Moya, M.; Castro, E., Hydrothermal 
pre-treatment of rapeseed straw. Bioresour. Technol. 2010, 101, (2010), 2428-2435. 
109 
?
15. Kumar, S.; Gupta, R.; Lee, Y. Y.; Gupta, R. B., Cellulose pretreatment in 
subcritical water: Effect of temperature on molecular structure and enzymatic reactivity. 
Bioresour. Technol. 2009, 101, (2010), 1337-1347. 
16. Matsumura, Y.; Sasaki, M.; Okuda, K.; Takami, S.; Ohara, S.; Umetsu, M.; 
Adschiri, T., Supercritical water treatment of biomass for energy and material recovery. 
Combust. Sci. and Tech. 2006, 178, 509-536. 
17. Kadam, K. L.; Chin, C. Y.; Brown, L. W., Continuous biomass fractionation 
process for producing ethanol and low-molecular-weight lignin. Environmental Progress 
& Sustainable Energy 2009, 28, (1), 89-99. 
18. Franck, E. U., Fluids at high pressures and temperatures. Pure Appl. Chem. 1987, 
59, (1), 25-34. 
19. Miyoshia, H.; Chena, D.; AkaiT., A novel process utilizing subcritical water to 
recycle soda?lime?silicate glass. Journal of Non-Crystalline Solids 2004, 337, ( 3), 280-
282. 
20. Savage, P. E., Organic chemical reactions in supercritical water. Chemical 
Reviews 1999, 99, 603-621. 
21. Kumar, S.; Gupta, R. B., Biocrude Production from Switchgrass using Subcritical 
Water. Energy & Fuels 2009, 23, (10), 5151-5159. 
22. Hendriks, A. T. W. M.; Zeeman, G., Pretreatments to enhance the digestibility of 
lignocellulosic biomass. Bioresour. Technol. 2009, 100, (1), 10-18. 
23. Heitz, M.; Carrasco, F.; Rubio, M.; Chauvette, G.; Chornet, E.; Jaulin, L.; 
Overend, R. P., Generalized correlations for aqueous liquefaction of lignocellulosics. 
Can. J. Chem. Eng. 1986, 64, 647-650. 
110 
?
24. Garrote, G.; Dominguez, H.; Parajo, J. C., Hydrothermal processing of 
lignocellulosic materials. Holz als Roh- und Werkstoff 1999, 57, (1999), 191-202. 
25. Suryawati, L.; Wilkins, M. R.; Bellmer, D. D.; Huhnke, R. A.; Maness, N. O.; 
Banat, I. M., Simulteneous sacchrification and fermentation of Kanlow Switchgrass 
pretreated by hydrothermolysis using Kluyveromyces marxianus IMB4. Biotechnology 
and Bioengineering 2008, 101, (5), 894-902. 
26. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
Polym. Sci. 1994, 19, 797?841. 
27. Sierra, R.; Smith, A.; Granda, C.; Holtzapple, M. T., Producing fuels and 
chemicals from lignocellulosic biomass. Chemical Engineering Progress August, 2008, 
pp S10-S18. 
28. Kohlmann, K. L.; Westgate, P. J.; Sarikaya, A.; Velayudhan, A.; Weil, J.; 
Hendricson, R.; Ladisch, M. R., Enhanced enzyme activities on hydrated lignocellulosic 
substrates. In American Chemical Society National Meeting, ACS Symposium series No. 
618: 1995; Vol. 207, pp 237-255. 
29. Mosier, N.; Hendrickson, R.; Ho, N.; Sedlak, M.; Ladisch, M. R., Optimization of 
pH controlled liquid hot water pretreatment of corn stover. Bioresour. Technol. 2005a, 
96, 1986-1993. 
30. Weil, J. R.; Brewer, M.; Hendrickson, R.; Sarikaya, A.; Ladisch, M. R., 
Continuous pH monitoring during pretreatment of yellow poplar wood sawdust by 
pressure cooking in water. Appl. Biochem. Biotechnol. 1997, 68, 21-40. 
31. kobayashi, N.; Okada, N.; Hirakawa, A.; Sato, T.; J.Kobayashi; Hatano, S.; Itaya, 
Y.; Mori, S., Characteristics of solid residues obtained from hot-compressed-water 
111 
?
treatment of woody biomass. Industrial and Engineering Chemistry Research 2009, 48, 
373-379. 
32. Yang, B.; Wyman, C. E., Effect of xylan and lignin removal by batch and flow 
through pretreatment on the enzymatic digestibility of Corn Stiver Cellulose. Biotech. 
Bioeng. 2004, 86, (1), 88-95. 
33. Liu, C.; Wyman, C. E., The effect of flow rate of compressed hot water on xylan, 
lignin and total mass removal from corn stover. Ind. Eng. Chem. Res. 2003, 42, 5409-
5416. 
34. Rogalinski, T.; Ingram, T.; Brunner, G., Hydrolysis of lignocellulosic biomass in 
water under elevated temperatures and pressures. Journal of Supercritical Fluids 2008, 
47, (1 ), 54-63  
35. Olofsson, K.; Bertilsson, M.; Lid?n, G., A short review on SSF ? an interesting 
process option for ethanol production from lignocellulosic feedstocks. 
 1: 7 (doi: 10.1186/1754-6834-1-7). Biotechnology for Biofuels 2008, 1, (7), 1-14. 
36. Kaparaju, P.; Serrano, M.; Angelidaki, I., Effect of reactor configuration on 
biogas production from wheat straw hydrolysate. Bioresour. Technol. 2009, 100, 6317-
6323. 
37. Klinke, H. B.; Thompsen, A. B.; Ahring, B. K., Inhibition of ethanol-producing 
yeast and bacteria by degradation products produced during pre-treatment of biomass. 
Appl. Microbiol Biotechnol 2004, 66, 10-26. 
38. Palmqvist, E.; Hahn-Hagerdal, B., Fermentaion of lignocellulosic hydrolyzates. 
II. Inhibitors and mechanism of inhibition. Bioresour. Technol. 2000b, 74, 25-33. 
112 
?
39. Segal, L.; Creely, J. J.; Martin, A. E.; Conrad, C. M., An empirical method for 
estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. 
Textile Res. J. 1959, 29, 786-794. 
40. Negro, M. J.; Manzanares, P.; Ballesteros, I.; Oliva, J. M.; Cabanas, A.; 
Ballesteros, M., Hydrothermal pretreatment conditions to enhance ethanol production 
from poplar biomass. Applied Biochemistry and Biotechnology 2003, 105-108, 87-100. 
41. Lai, Y.-Z., Chemical Degradation. In Wood and Cellulosic Chemistry, David N.-
S. Hon, N. S., Ed. Marcel Dekker, Inc.: New York . Basel, 2001; pp 443-512. 
42. Rodr?guez, A.; Moral, A.; S?nchez, R.; Requejo, A.; Jim?nez, L., Influence of 
variables in the hydrothermal treatment of rice straw on the composition of the resulting 
fractions. Bioresour. Technol. 2009, 100, 4863-4866. 
43. Suryawati, L.; Wilkins, M. R.; Bellmer, D. D.; Huhnke, R. A.; Maness, N. O.; 
Banat, I. M., Effect of hydrothermolysis on ethanol yield from Alamo switchgrass using a 
thermotolerant yeast. In 2007 ASABE Annual International Meeting, American Society of 
Agricultural and Biological Engineers: Minneapolis, Minnesota, 2007; Vol. Paper 
Number: 077071, St. Joseph, Mich.:ABASE, pp 1-13. 
44. Gupta, R. Alkaline pretreatment of biomass for ethanol production and 
understanding the factors influencing the cellulose hydrolysis. Auburn University, 
Auburn, 2008. 
45. Carr, M. E.; Doane, W. M., Modification of wheat straw in a high- shear mixer. 
Biotech. and Bioeng. 1984, 26, 1252-1257. 
113 
?
46. Detroy, R. W.; Lindenfelser, L. A.; Orton, W. L., Bioconversion of wheat straw to 
ethanol: Chemical modification, enzymatic hydrolysis, and fermentation. Biotech. and 
Bioeng. 1981, 23, 1527-1535. 
47. Schacht, C.; Zetzl, C.; Brunner, G., From plant materials to ethanol by means of 
supercritical fluid technology. Journal of Supercritical Fluids 2008, 46s  (3), 299-321. 
48. http://en.wikipedia.org/wiki/Self-ionization_of_water Self-ionization of water. 
(Jan 19, 2010),  
49. Rezaei, K.; Temelli, F.; Jenab, E., Effects of pressure and temperature on 
enzymatic reactions in supercritical fluids. Biotechnology Advances 2007, 25, 272-280. 
50. Chang, V. S.; M. T, H., Fundamental factors affecting biomass enzymatic 
reactivity. Applied Biochemistry and Biotechnology 2000, 84-86, 5-37. 
51. Kumar, R.; Wyman, C. E., Access of cellulase to cellulose and lignin for poplar 
solids produced by leading pretreatment technologies. Biotechnol. Prog. 2009, 25, (3), 
807-819. 
52. Kumar, S.; Gupta, R.; Lee, Y. Y.; Gupta, R. B., Cellulose pretreatment in 
subcritical water: Effect of temperature on molecular structure and enzymatic reactivity. 
Bioresour. Technol. 2010, 101, 1337-1347. 
53. Titirici, M.-M.; Thomas, A.; Antonietti, M., Back in the black: hydrothermal 
carbonization of plant material as an efficient chemical process to treat the CO
2
 problem. 
New J. Chem. 2007, 31, 787-789. 
54. Titirici, M.-M.; Antonietti, M.; Baccile, N., Hydrothermal carbon from biomass: a 
comparision of the local structure from poly- to monosaccharides and pentoses/hexoses. 
Green Chem. 2008, 10, 1204-1212. 
114 
?
55. Knez?evic?, D.; Swaaij, W. P. M. v.; Kersten, S. R. A., Hydrothermal Conversion 
of Biomass: I, Glucose Conversion in Hot Compressed Water. Ind. Eng. Chem. Res. 
2009, 48, 4731-4743. 
56. Sevilla, M.; Fuertes, A. B., The production of carbon materials by hydrothermal 
carbonization of cellulose. Carbon 2009, 47, 2281-2289. 
57. Almendros, G.; Sanj, J.; Sobrados, L., Characterization of synthetic carbohydrate-
derived humic-like polymers. The Science of the Total Environment 1989, 81-82, 91-98. 
58. Chuntanapum, A.; Matsumura, Y., Formation of Tarry Material from 5-HMF in 
Subcritical and Supercritical Water. Ind. Eng. Chem. Res. 2009, 48, (22), 9837-9846. 
59. Baeza, J.; Freer, J., Chemical characterization of wood and its components. In 
Wood and cellulosic chemistry, second ed.; Hon, D. N.-S.; Shiraishi, N., Eds. Marcel 
Dekker, Inc.: New York. Basel, 2001; pp 275-384. 
 
 
4. Hydrolysis of Microcrystalline Cellulose in Subcritical and Supercritical Water in a 
Continuous Flow Reactor 
 
4.1 Abstract 
For cellulosic ethanol production, efficient hydrolysis of crystalline cellulose to 
easily fermentable sugars is important. Focus of this study is to maximize the yield of 
cellulose hydrolysis in subcritical and supercritical water at practically achievable 
reaction times. Microcrystalline cellulose is treated with subcritical and supercritical 
water in temperature range 302 to 405?C, at pressure 27.6 MPa, and residence times of 
2.5 to 8.1 s. Cellulose-water slurry of 2.7 wt% after mixing with preheated water is 
rapidly heated to the reaction temperature and then the reaction product is rapidly cooled 
in a continuous reactor. Cellulose partially dissolves in subcritical water at 302?C and 
completely dissolves at 330?C. About 65% of cellulose converts to the oligomers and 
monomers at 335?C in 4.8 s, and also at 354?C in 3.5 s.  Upon increase in the reaction 
time or temperature to supercritical region, oligomers and monomers partially degrade to   
glycoaldehyde dimer, D-fructose, 1, 3 dihydroxyacetone dimer, anhydroglucose, 5-HMF 
and furfural. The effect of temperature, pressure, and reaction time on formation of 
various products is studied.  In addition, the effect of a base catalyst, K
2
CO
3
, is examined. 
The catalyst increases cellulose gasification in the temperature range studied (302-
333?C).  
115 
 
4.2 Introduction 
Ethanol from lignocellulosic biomass is an attractive and renewable fuel for 
transportation. Globally, the means to utilize this abundantly available resource in a cost-
effective way is a major research focus. Cellulose, a linear polymer of anhydroglucose 
units is one of the three major components of lignocellulosic biomass. Acid/alkali pre-
treatment of biomass followed by enzymatic hydrolysis is a conventional method to 
produce glucose from cellulose. Glucose can be used to synthesize various specialty 
chemicals or fermented to produce ethanol. Supercritical water, an environmentally 
benign solvent for organic matters, can be used under a variety of conditions to rapidly 
convert cellulose to sugar or to convert biomass into a bio-crude. Bio-crude, which 
contains only 10-13% oxygen provides excellent opportunities for conversion of biomass 
to a transportable form of energy.
1 
Also, reactions in subcritical and supercritical water 
provide a novel medium to conduct tunable reactions for the synthesis of specialty 
chemicals from biomass.
2
 
Density and dielectric constant of the water medium play major role in soubilizing 
organic compounds in subcritical and supercritical water. Dielectric constant of water 
near the critical point decreases considerably, which enhances the solubility of organic 
compounds.
3 
Direct observation of phase behavior of cellulose-water system in a 
diamond anvil cell showed that at higher temperatures (300-320?C) cellulose particles 
disappeared with increasing transparency, and shrinking rate was significantly increased 
with temperature above 300?C.
4 
Recently, it was shown that crystalline cellulose 
undergoes transformation to an amorphous state in subcritical water, which is followed 
by complete dissolution at temperature between 330-340?C at 25 MPa.
5, 6  
116 
 
Most of the work on cellulose decomposition in sub- or supercritical water has 
been done in the batch reactors. Due to the formation of gaseous products during the 
reactions, it becomes difficult to maintain the constant density of the reaction medium 
and so, kinetics obtained from the batch reactor often can give misleading information on 
the design for industrial level continuous reactors.  Also, difficulty arises when one want 
to limit the reaction time to few seconds. In addition, catalytic effects of the wall are not 
translated accurately to the continuous reactor. A comparative study on conversion of 
cellulose between batch-type and flow-type systems in supercritical water concluded that 
flow type systems can hydrolyze cellulose with minimum pyrolyzed products, whereas 
batch-type systems give higher yields of the pyrolyzed products due to the longer 
treatment time.
7 
It has been reported that combined process of short supercritical water 
treatment followed by subcritical water treatment increases the yield of hydrolyzed 
products.
8 
 
In context to studies in batch reactors, Minowa et al.
9, 10 
conducted hydrolysis of 
cellulose with and without catalyst. Using a conventional autoclave, slurry solutions of 
cellulose were heated in a nitrogen atmosphere to various temperatures between 200-
350?C. Results indicate that decomposition of cellulose increased with reaction 
temperature. Temperatures of 240?C and below lead to the formation of water-soluble 
products. Temperatures above 260?C leads to the formation of oil, char, and gas. At 
temperatures above 260?C non-recoverable char begins to form decreasing the carbon in 
the liquid phase. The maximum oil yield was obtained at 280?C; increasing temperature 
leads to increase in the char and gas formation but decrease in the oil yields, indicating a 
secondary decomposition of oil to char and gas. Based on the same product distribution 
117 
 
obtained from glucose and cellulose decomposition, it was concluded that cellulose 
decomposition starts with hydrolysis step.  Sakaki et al.
11
 reported nearly 80 wt% 
conversion of cellulose to water soluble products with glucose yields 40% within 15 s by 
heating the cellulose and water mixture in a sealed reactor above 355?C.  
Sasaki et. al.
4, 12
 developed a high-pressure slurry feeder to feed the cellulose 
water slurries in a flow reactor and studied the decomposition of microcrystalline 
cellulose in subcritical and supercritical water (320-400?C, and 0.05-10.0 s). It was 
reported that 75% yield of hydrolysis products were obtained in supercritical water, while 
aqueous decomposition products of glucose were the main products in subcritical water.
 
Cellobiose degradation in sub- and super-critical water in a flow apparatus proceeds via 
the hydrolysis of the glycosyidic bond and via pyrolysis of the reducing end.
13
 
The results of earlier studies in flow reactors show that a higher hydrolysis 
product yield can be achieved by treating cellulose with supercritical water in a fraction 
of a second residence time. To develop the process for large scale applications, it is 
necessary to optimize both from process point of view (e.g., temperature, pressure, 
residence time etc.) and engineering point of view (e.g., feeding, heating, etc.)
2
 Cellulose 
being a poor conductor of heat will require some innovative heating techniques for 
heating to the reaction temperature in such a short time. Therefore, the main aim of this 
study is to examine the possibilities of maximizing hydrolysis products yield from the 
hydrolysis of crystalline cellulose at a relatively lower temperature and at a practically 
achievable residence time in a continuous flow apparatus. 
Lignocellulosic biomass naturally contains alkali metals such as K, Ca and Na, 
which can increase the rate of gas formation at the supercritical water temperatures. At 
118 
 
less than 350?C, it has been reported that cellulose or wood can be liquefied in subcritical 
water in presence of Na
2
CO
3
.
14 
Also, direct liquefaction of wood in the presence of 
K
2
CO
3 
exhibits a marked catalytic effect.
14, 15 
Sinag et al. reported that hydropyrolysis of 
glucose in the presence of K
2
CO
3
 at residence time in the range of 1.6-6.3 minutes leads 
to increased gas formation.
 
The gaseous products mainly consist of hydrogen, carbon 
dioxide, methane and smaller amount of carbon monoxide.
16, 17 
Carbohydrate structures 
of the solid residues start to disappear at liquefaction temperatures of 290-300?C in the 
presence of Na
2
CO
3. 
Above 300?C the amount of methane shows a highly upward trend 
indicating an increasing carbonization.
18
 
The second objective of this study is to examine the effect of heterogeneous 
catalyst (K
2
CO
3
) in flow reactor on the liquefaction temperature and also, examine the 
behavior of solid residues for understanding the nature of thermochemical liquefaction 
when cellulose is used. Dielectric constant of water decreases from 78.5 at 25?C to 5 in 
the near critical region which causes reduced solubility of inorganic compounds, as a 
result alkali salts precipitates as fine powder. These fine particles provide a high surface 
area during the heterogeneous reactions.
19
 The effect of catalysis on the decomposition of 
cellulose is examined in the temperature range of 302-330?C in subcritical water. 
4.3 Experimental Section 
Materials: Microcrystalline cellulose powder (size 20 ?m, degree of polymerization 229, 
Product Number 310697), K
2
CO
3
, high purity standard reagents used for HPLC analysis 
(see Table 4.1) were purchased from Sigma Aldrich. De-ionized water was used for 
preparing cellulose slurry.  
119 
 
Table 4.1 HPLC standards used for analysis of liquid products. 
No. Chemicals Formula Conc.,  (g/l) MW 
1 Cellobiose C
12
H
22
O
11
 2 342 
2 Glucose C
6
H
12
O
6
 2 180 
3 Glycoaldehyde,dimer C
4
H
8
O
4 
1 120 
4 D- Fructose C
6
H
12
O
6
 2 180 
5 1,3 Dihydroxyacetone,dimer C
6
H
12
O
6
 2 180 
6 1,6 Anhydro - beta glucose C
6
H
10
O
5
 1 162 
7 5-(hyhroxymethyl)-furaldehyde C
6
H
6
O
3
 1 126 
8 Furfural C
5
H
4
O
2
 2 96 
 
Apparatus: Figure 4.1 shows the schematic of apparatus used for the experiments. It 
consists of high-pressure liquid pumps, a cellulose slurry feeder, an electrical tubular 
furnace, a water-cooled double tube heat exchanger, a backpressure regulator, and a phase 
separator. Reactor is a high-pressure 14? long Inconel 600 tube having 0.25? OD, 0.11? 
ID and 2.18 mL volume.  Specially designed piston operated high-pressure slurry feeder 
(0.57? ID, 20? long) was used to supply cellulose slurry at desired flow rate. Real time 
temperature of the reaction zone was measured using a 1/16? thermocouple placed 3? 
inside the reactor from outlet end. This thermocouple measures the reaction zone 
temperature with accuracy of ?0.1?C. There are three additional thermocouples installed 
at different locations in the apparatus. Real time temperature indicated by the 
thermocouple placed inside the reaction zone was used for all the analyses. Reactor 
pressure was maintained using a Stra-val backpressure regulator. 
120 
 
 
Figure 4.1 Apparatus for cellulose hydrolysis in subcritical and supercritical water. 
Experimental Procedure: Microcrystalline cellulose (MCC) was added to continuously 
stirred hot water (70?C) to prepare 24.1 wt% slurry, which was charged in the slurry 
feeder.  It was observed that 24.1 wt% of cellulose suspension in hot water did not settle 
appreciably in a similar diameter tube for an hour, which helped us in avoiding the 
requirement of any agitation in the slurry feeder.  
At the start, temperature and pressure of the reaction zone are established under 
water flow. Then, slurry is pumped at desired flow rate and mixed with 8 fold (mass 
basis) of preheated water prior to entering the reactor. Products coming out of the reactor 
are cooled in the heat exchanger, depressurized in the back pressure regulator and 
collected in the phase separator.  Periodic liquid samples are collected, centrifuged at 
3400 rpm to remove any solids and analyzed for total organic carbon (TOC) in Shimadzu 
TOC-Vcsn analyzer.  
 MCC was treated with subcritical and supercritical water at temperatures ranging 
121 
 
from 302 to 405?C, at a constant pressure 27.6 MPa and residence times of 2.5 to 8.1 s. 
Experiments using K
2
CO
3 
catalyst were conducted at 302?C (5.2 s) and 330?C (4.8 s) 
temperature. The catalyst was introduced as 8 wt% solution in ambient water, pumped to 
the mixing point as shown in the Figure 4.1. 
The experiments were divided into four groups as: (1) 302-335?C, (2) 350-366?C, 
(3) 376-405?C, and (4) 302, 330?C in the presence of K
2
CO
3
. One experiment from each 
group was repeated to confirm the reproducibility of the cellulose conversion by 
measuring the total organic contents (TOC). Multiple samples were collected and 
analyzed for TOC values during each experiment to confirm the steady state. TOC value 
varied within ? 2% during the reaction. Liquid products collected during the steady state 
phase of experiment are stored in the refrigerator for chromatographic (HPLC) analyses. 
Yields were calculated based on the 2.7 wt % of pure cellulose input to the reactor. 
The residence time (t) in the reactor was calculated using pure water density at the 
reaction temperature and pressure (i.e. fluid phase in the reaction zone is assumed to be 
pure water for the simplicity of calculations) as 
?
?
?
?
?
?
?
?
=
1,1 TP
pump
F
V
t
?
?
 
where, V (2.18 ml) is reactor volume, F is the combined volumetric flow rate of pumps 1, 
2 and 3, ? 
pump
 is the density of water at pump condition (25?C, 27.6MPa), and ? 
P1,T1
 is 
the density of water at reactor condition (e.g., T
1
 and P
1
).   
4.4 Product Analyses 
 Residual solids from the product were separated by centrifugation, oven dried at 
70?C, weighed and taken for analysis. Carbon content in the supernatant liquid was 
122 
 
determined by TOC analyzer. Hydrolysis products (monomers and oligomers) and 
degradation products in liquid samples were determined by following the National 
Renewable Energy Laboratory, biomass laboratory analytical procedure (described in 
technical report-NREL/TP-510-42623). The analysis was done using HPLC having Bio-
Rad Aminex HPX-87P (lead based) column, refractive index detector, water mobile 
phase, and column temperature of 85?C. Additionally analysis was done using Bio-Rad 
Aminex HPX-87H column, refractive index detector, 0.01 N sulfuric acid mobile phase 
and column temperature of 65?C.
20 
 Solid residues collected during the catalytic experiments were oven dried and 
analyzed for elemental carbon using Elementar Vario Macro CNS analyzer. 
21
 
4.5 Results  
Four groups of experiments were performed in the continuous reactor at constant 
pressure 27.6 MPa and residence time ranging from 2.5 to 8.1 s, at (1) 302-335?C, (2) 
350-366?C, (3) 376-405?C, and (4) 302, 330?C in the presence of K
2
CO
3
. At temperature 
275?C or below cellulose conversion to water soluble products were not significant (< 
10%) for the 3-6 s residence time. Longer residence times would be needed to hydrolyse 
cellulose in lower temperature range which promotes rapid degradation of converted 
hydrolysis products. Cellulose conversion increased sharply above 275?C. Therefore 
hydrolysis study was mainly focused at higher temperatures in the range 302-405?C. The 
products from cellulose hydrolysis in subcritical and supercritical water were identified 
and quantified in different runs as shown in Table 4.2. 
123 
 
Table 4.2 Liquid products yield from cellulose treatment in subcritical and supercritical 
water at 27.6 MPa. 
 
Figure 4.2, 4.3, and 4.4 show the yield of hydrolysis products, glucose and 5-
HMF for runs 1-14. Here, product yields were calculated based on 2.7 wt% of pure 
cellulose input in the reactor at 27.6 MPa.  
124 
 
 
Figure 4.2 Yield of hydrolysis products obtained from cellulose treatment in subcritical 
and supercritical water at 27.6 MPa. 
 
Figure 4.3 Yield of glucose obtained from cellulose treatment in subcritical and 
supercritical water at 27.6 MPa. 
125 
 
 
Figure 4.4 HMF yield in liquids obtained from cellulose treatment in subcritical and 
supercritical water at 27.6 MPa. 
In this study, we defined (a) hydrolysis products as the combination of oligomers 
(>2 monomer units), cellobiose, glucose, and fructose, and (b) degradation products as 
glycoaldehyde, D-fructose, 1,3 dihydroxyacetone, anhydroglucose, 5-HMF, and furfural. 
These products are quite similar to those obtained by Sasaki et al.
4
 HPLC 
chromatographs show the presence of pyruvaldehyde, glyceraldehydes and erythrose in 
most of the liquid products, these intermediate products were not quantified due to lack 
of accurate calibration. The organic acid (e.g., formic acid, lactic acid and acetic acid) 
presence were confirmed by HPLC analysis in Bio-Rad Aminex HPX-87H column, but 
were not quantified. 
Figure 4.5 shows that the TOC of liquid product increased rapidly with rise in 
temperature from 302 to 330?C, and more than 90% of cellulose was converted to 
hydrolysis or degradation products above 330?C, except for run 4 (333?C, 3.48 s).  
126 
 
 
Figure 4.5 TOC in liquid products obtained from cellulose treatment in subcritical and 
supercritical water at 27.6 MPa. 
The nine quantified compounds in 14 runs are complied in Table 4.3. It is 
observed that as percentage of other compounds (OC), which is defined in Table 4.3 as 
the difference between TOC and total identified carbon (mmol/l), increases with increase 
in temperature or residence time. The OC are mostly organic acids, which is confirmed 
by the low pH (2.3 to 3) for all the runs. 
127 
 
Table 4.3 Carbon balance in the liquid products obtained from subcritical and 
supercritical treatment of cellulose at 27.6 MPa. 
 
Solids from all the runs were oven dried at 70?C, before analysis.  Solids from 
runs 1, 3, 4 and 5 were brown fluffy, whereas from others were black. The elemental 
carbon in the solids from run 1 was 41.9%. 
Experiment Group 1 (302-335?C): Runs 3, 5 (Table 4.2), conducted at almost the same 
process conditions, show reproducible results in terms of hydrolysis product yield, which 
are 66.7% and 65.8%, respectively. HPLC chromatograph from run 5 is shown in Figure 
4.6.  
128 
 
 
Figure 4.6 HPLC chromatograph of run 5 (335?C, and 4.7 s residence time). 
Glucose yield in run 3 is 37.6%. OC (Table 4.3) for runs 3 and 5 were 17.4 and 
10.9%, respectively. 5-HMF and furfural yields in product liquid were almost same for 
both the runs.  Liquid products of run 3 and 5 were hazy, and turned turbid even after the 
centrifuging. Residual solids were brown, fluffy and sticky. Degradation products yield in 
run 1 is minimum among all the runs, but it has higher yield of OC because of the 
comparatively longer residence time (5.2 s), as organic acids increase with the increase in 
residence time. Run 4 conducted at relatively low residence time (3.5 s) had only 3.3% of 
OC as shown in Table 4.3. 
Experiment Group 2 (350-366?C): For the experiments conducted in this temperature 
range, run 7 (354?C, 3.5 s) had the highest hydrolysis products yield (64.9%) and glucose 
yield (32.9%). HPLC chromatograph from run 7 is shown in Figure 4.7.  
129 
 
 
Figure 4.7 HPLC chromatograph of run 7 with (354?C, and 3.5 s residence time). 
This experiment was duplicated to verify the reproducibility. Liquid products 
were clear yellowish liquid having very little solids. Increase in residence time from 3.5 
to 4.4 s for run 8 resulted in a rapid decline in yield of hydrolysis products, which again 
decreased for run 9, where reactor temperature was increased from 354 to 366?C. OC 
yields for runs 7 and 9, shown in Table 3.5 were 22.0% and 25.6%, respectively. 
Experiment Group 3 (376-405?C): Degradation products were the main products for the 
experiments conducted in supercritical water. The highest hydrolysis products and 
glucose yields were 15.2% and 11.6% respectively, in run 10 (376 ?C, 3.7 s). HPLC 
chromatograph of run 14 is shown in Figure 4.8.  
130 
 
 
Figure 4.8 HPLC chromatograph of run 14 (405 ?C, and 2.5 s residence time). 
Glycoaldehyde dimer was the major product among the entire identified products. 
OC yield (Table 4.4) increased to more than 50%, due to increase in the organic acids 
formation. Liquid products in supercritical range were almost clear, yellowish liquid with 
very little solids. 
Experiment Group 4 (302 and 330?C with K
2
CO
3
 catalyst): For these experiments, Table 
4.4 shows the process parameters and the results in terms of carbon balance.  Notably, the 
gas formation is high in the presence of K
2
CO
3
. Gas flow almost doubled, when K
2
CO
3
 
concentration was increased from 0.22 wt% (run C1) to 0.44 wt% (run C2) at 302?C.  
131 
 
 Table 4.4 Experimental conditions of cellulose runs in presence of catalyst K
2
CO
3
 in 
subcritical water. 
 
In run C3, 43.4 wt% of input carbon goes to gaseous products at 330?C, 4.8 s in 
presence of 0.13 wt% of K
2
CO
3
. Figure 4.9 shows the photographs of liquid products 
obtained from run 1 (in absence of catalyst at 303?C) and those from runs C1 and C2, 
which were conducted in presence of 0.22 wt% and 0.44 wt% of K
2
CO
3
, respectively.  
 
Figure 4.9 Products obtained at 302?C, 5.2 s, 27.6 MPa in the presence of 
(a) 0% K
2
CO
3
, (b) 0.22% K
2
CO
3
, and (c) 0.44% K
2
CO
3
. 
 Color of liquid turned light yellowish to brownish with the increase of K
2
CO
3
. 
Solids from runs C1 and C2 were dense, but the carbon analysis did not show rise in 
%carbon.  Liquid products, from runs C1 and C2, show small peaks for glycoaldehyde 
132 
 
dimer in HPLC analysis. However, no such peaks were observed for run C3.  Small peaks 
due to formic acid and acetic acid were observed in all the three runs, but were not 
quantified. 
4.6 Discussion 
  Structure of water changes significantly near the critical point because of the 
breakage of infinite network of hydrogen bonds and water exists as separate clusters with 
a chain structure.
22 
In fact, dielectric constant of water decreases considerably near the 
critical point, which causes a change in the dynamic viscosity and also increases self-
diffusion coefficient of water.
23
 These properties of supercritical water provide a 
completely different homogeneous reaction atmosphere. Hot compressed water acts as 
both reactant and reaction medium. Water as reactant leads to hydrolysis reactions; 
rapidly degrading the polymeric structure of cellulose to hydrolysis products, instead of 
pyrolysis. The intermediates formed during the decomposition reactions show a high 
solubility in sub- and supercritical water; hence reaction steps are mainly homogeneous.
24 
Diffusion related problems such as mass transfer through the interface are not 
encountered in homogeneous reaction.
3
 
 Hydrothermal degradation of cellulose in 200-300?C range is a heterogeneous and 
pseudo-first-order reaction for which detailed chemistry and mechanism were proposed 
by Antal and coworkers.
25-30 
The models are based on the sequence cellulose ?  glucose 
? decomposition products, in which hydrolysis to glucose is an important step. It is 
assumed that that the unconverted cellulose retains its chemical integrity with the extent 
of the hydrolysis.
31 
Results of the studies clearly indicate that cellulose converts to 
hydrolysis products (oligomers and monomers) and then further degrades to aqueous 
133 
 
degradation products of glucose, producing glycoaldehyde, fructose, 1,3 
dihydroxyacetone, anhydroglucose, 5-HMF and furfural. 
In a very recent study, it has been observed that crystalline to amorphous 
transition of cellulose is observed between 320-330?C at 25 MPa and cellulose is 
completely dissolved above 340?C.
5, 6 
Hydrolysis products are intermediate products in 
cellulose decomposition in subcritical and supercritical water and they convert further to 
various degradation products, if allowed a higher residence time.
 
To minimize 
degradation, most of the studies have been conducted for a very short residence times, 
typically less than a second.
4, 7
 However, this extremely short residence time may not be 
enough for crystalline to amorphous transformation of cellulose.
6
 
All of our studies in subcritical water were conducted for the residence time 3.3-
8.1 s. For the runs 3 and 5, residence times were 4.8 and 4.7 s, respectively, which appear 
to be sufficient for the crystalline to amorphous transformation of cellulose. Therefore, a 
yield of hydrolysis products as high as 66.8% was achieved in this region. Glucose 
molecules are unstable at high temperature and immediately degrade to conversion 
products. Earlier study of glucose reactions in hot compressed water (200-500?C, 4-40 
MPa) in a batch-type reactor has shown that more than 80% of the glucose was consumed 
above 300?C
 
 and 60 s heating period. Gasification of glucose was promoted with 
increasing temperature and the heating rate.
32
  
Optimizing the residence time in this temperature zone may further increase the 
hydrolysis products yields. By increasing the reactor temperature from 335 to 354?C and 
reducing the residence time from 4.7 to 3.5 s achieved similar yields of hydrolysis 
products. In our observation, yields of hydrolysis products depend on the temperature and 
134 
 
residence times at which crystalline to amorphous transformation of cellulose is 
completed in subcritical water, and the yield of degradation products may be minimized 
by optimizing the residence time around this temperature. A longer residence time 
enhances the degradation rate of converted hydrolysis products. For example, in a recent 
study, majority of water soluble compounds from cellulose decomposition in subcritical 
water at 300-310?C for the residence time 0.5-1.5 min were converted to degradation 
products, and the yield of hydrolysis products such as oligomers, glucose was less than 
20%.
33 
In our study, pH of all the liquid products varied between 2.3 to 3. Acidic medium 
is conducive for the degradation of converted hydrolysis products at the reaction 
temperatures studied here. Once organic acids are formed in the reaction mixture, it 
provides necessary acidity for further degradation of the converted hydrolysis products, 
as an autocatalytic process. Indeed after hydrolysis of cellulose in water at 320?C and 25 
MPa for 9.9 s, more than half of the cellulose was converted to organic acids.
 
Sasaki et al.
4 
reported that cellohexaose, cellopentaose, cellotetraose, cellotriose, 
cellobiose was derived from cellulose by superciritical water treatment. However, in runs 
10-14, other oligosaccharides, which differ from the results of Sasaki et al., were 
apparently obtained from cellulose. These oligosaccharides decreased and glucose 
degradation products (glycoaldehyde, fructose, 1,3-dihydroxyacetone, anhydroglucose, 5-
HMF and furfural) increased upon the prolonged treatment (2.5-6.6 s). The liquid 
products obtained in this region are transparent yellow with almost 100% cellulose 
conversion.  
Among the identified products (runs 10-14) in supercritical region, glycoaldehyde 
dimer is the major product yield. It is known that the reducing ends of oligosaccharides 
135 
 
are fragmented to erythrose and glycoaldehyde in supercritical water. Low supercritical 
water density favors retro-aldol condensation of glucose.
34, 35
 Main reaction pathway of 
glucose decomposition in subcritical and supercritical water explains the formation routes 
for degradation products, such as glycoaldehyde, erythrose, glyceraldehydes, 
dihydroxyacetone, pyruvaldehyde, 1,6 anhydrous glucose, 5-HMF, furfural and lactic 
acid.
13, 36 
In fact, 5-HMF was observed to increase with temperature and residence time. 
Effect of K
2
CO
3 
catalyst: Addition of alkali metal salts (e.g. KHCO
3
, KOH, Na
2
CO
3
, 
K
2
CO
3
) reduces coke formation and catalyzes the water-gas shift reaction.
37
 For example, 
the addition of KHCO
3 
leads to an increase in gas formation and a decrease in the amount 
of carbon monoxide due to water-gas shift reaction.
16 
Also, methane formation is 
dominant at relatively low temperatures in catalytic gasification of cellulose in 
supercritical water. 
38
 We also observed the increased formation of gas in presence of 
K
2
CO
3
, based on carbon balance. Biomass contains about 0.5 wt% potassium and other 
alkalis such as Ca and Na. Therefore, gasification of biomass in supercritical water will 
be enhanced by the presence of these natural elements. It has been reported that 
carbohydrate structure of solid residues start to disappear at liquefaction temperature of 
290-300?C and according to elemental analysis, the carbon content of the solid residues 
increases significantly.
18
 Since the residence time in our study was short (4.8-5.2 s), 
carbonization of solids might not have started. With the increase in the concentration of 
K
2
CO
3
, color of the liquid turned dark. Conversion of cellulose to liquid products without 
K
2
CO
3
 in similar conditions (run 1) was higher than the conversion in presence of K
2
CO
3 
(runs C1- C3), use of much lower concentration of K
2
CO
3
 may improve liquefaction of 
cellulose at low temperatures. 
136 
 
4.7 Conclusion 
Crystalline cellulose can be successfully converted to monomers (mainly glucose) 
and oligomers by hydrolysis in subcritical water. Earlier studies show that high yields of 
hydrolysis products (65-77%) are mainly obtained in supercritical water (at 400?C, 25 
MPa) between 0.1-0.5 s residence times in a continuous flow reactor. Some of the studies 
suggest a combination of supercritical and subcritical water treatment of cellulose to 
obtain a high yield of hydrolysis products. 
More than 90% of the cellulose converts to water soluble products above 330 ?C. 
Reaction medium becomes acidic, which is conducive for the degradation of converted 
oligomers and monomers (i.e., hydrolysis products) at the reaction temperature. Thus 
optimizing the residence time and rapid cooling after the reaction are key to minimizing 
the degradation.  Our findings show that a high yield of hydrolysis products can be 
achieved at comparatively lower temperature (335?C) in subcritical water. For example, 
upto 66.8% of
 
crystalline
 
cellulose can be converted to
 
hydrolysis products at 335?C and 
27.6 MPa in merely 4.7 second reaction time. With increase in the reaction time, the 
hydrolysis products degrade to glycoaldehyde, fructose, 1,3 dihydroxyacetone, 
anhydroglucose, 5-HMF and furfural. Yield of glycoaldehyde, a retro-aldol condensation 
product of glucose, increases with a decrease in the density of supercritical water, and the 
yield of degradation products, 5-HMF and organic acids, increases with temperature and 
residence time.  In supercritical water conditions, studied here, more than 80% of the 
cellulose converts into the degradation products and organic acids. 
      Use of K
2
CO
3 
catalyst in the reaction mixture during depolymerization of 
cellulose in subcritical water substantially enhances the gas formation. However, during 
137 
 
the short residence time (4.8-5.2 s), carbonization of residue solid does not occur. 
4.8 Acknowledgment 
Author is thankful for the financial supports from the U.S. Department of Energy 
(grant DE-FC26-05424.56) and Alabama Center for Paper and Bioresource Engineering.  
Additionally, the authors are thankful for technical discussions and assistance from Prof. 
Y. Y. Lee and Prof. Ayhan Demirbas. 
4.9 Products Nomenclature 
Olg: Oligomers 
CB: Cellobiose 
Glu: Glucose 
Fr: D-Fructose 
GA: Glycoaldehyde dimer 
DA: 1,3 Dihydroxyacetone  dimer 
AG: 1,6 Anhydro - beta glucose 
HMF:  5-Hydroxymethyl ?2-furaldehyde 
FUR: Furfural 
OC: Other compounds 
 
 
 
 
138 
 
4.10 References 
1. Gourdiaan, F.; Peferoen, D., Liquid fuels from biomass via a hydrothermal 
process. Chemical Engineering Science 1990, 45, 2729-2734. 
2. Matsumura, Y.; Sasaki, M.; Okuda, K.; Takami, S.; Ohara, S.; Umetsu, M.; 
Adschiri, T., Supercritical water treatment of biomass for energy and material recovery. 
Combust. Sci. and Tech. 2006, 178, 509-536. 
3. Galkin, A. A.; Lunin, V. V., Subcritical and supercritical water: a universal 
medium for chemical reactions. Russian chemical reviews 2005, 74, (1), 21-35. 
4. Sasaki, M.; Fang, Z.; Fukushima, Y. N.; Adschiri, T.; Arai, K., Dissolution and 
hydrolysis of cellulose in subcritical and supercritical water. Ind. Eng. Chem. Res. 2000, 
39, 2883-2890. 
5. Deguchi, S.; Tsujii, K.; Horikoshi, K., Cooking cellulose in hot and compressed 
water. Chem. Commun. 2006, 3293-3295. 
6. Deguchi, S.; K. Tsujii; Horikoshi, K., Crystalline-to-amorphous transformation of 
cellulose in hot and compressed water and its implications for hydrothermal conversion. 
Green Chem. 2008, 10, 191-196. 
7. Ehara, K.; Saka, S., A comparative study on chemical conversion of cellulose 
between the batch-type and flow-type systems in supercritical water. Cellulose 2002, 9, 
301-311. 
8. Ehara, K.; Saka, S., Decomposition behavior of cellulose in supercritical water, 
subcritical water, and their combined treatments. . J. Wood Sci. 2005, 51, 148?153. 
9. Minowa, T.; Zhen, F.; Ogi, T., Cellulose decomposition in hot-compressed water 
with alkali or nickel catalyst. J. Supercritical Fluids 1997, 13, 253-259. 
139 
 
10. Minowa, T.; Zhen, F.; Ogi, T.; Varhegyi, G., Decomposition of cellulose and 
glucose in hot-compressed water under catalyst-free conditions. J Chem. Eng. Japan 
1998, 31, 131-134. 
11. Sakaki, T.; Shibata, M.; Miki, T.; Hirosue, H.; Hayashi, N., Reaction model of 
cellulose decomposition in near-critical water and fermentation of products. Bioresource 
Technology 1996, 197-202. 
12. Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.; N.Takeda; Adschiri, T.; 
Arai, K., Cellulose hydrolysis in subcritical and supercritical water. J. Supercritical 
Fluids 1998 1998, 13, 261-268. 
13. Kabyemela, B. M.; Takigawa, M.; Adschiri, T.; Malaluan, R. M.; Arai, K., 
Mechanism and Kinetics of Cellobiose Decomposition in Sub- and Supercritical Water. 
Ind. Eng. Chem. Res. 1998, 37, 357-361. 
14. Ogi, T.; Yokayama, S.; Koguchi, T., Direct liquefaction of wood by catalyst (Part 
1). Sekiyu Gakkaishi 1985, 28, 239-245. 
15. Yokayama, S.; Ogi, T.; Koguchi, K.; Murakami, M.; Suzuki, A., Direct 
liquefaction of wood by catalyst (Part 2). Sekiyu Gakkaishi 1986, 29, 262-266. 
16. Sinag, A.; Kruse, A.; Schwarzkopf, V., Key compounds of the hydropyrolysis of 
glucose in supercritical water in the presence of K2CO3. Ind. Eng. Chem. Res. 2003, 42, 
3516-3521. 
17. Sinag, A.; Kruse, A.; Schwarzkopf, V., Formation and degradation pathways of 
intermediate products formed during the hydropyrolysis of glucose as a model substance 
for wet biomass in a tubular reactor. Eng. Life Sci. 2003, 3, 469-473. 
140 
 
18. Szabo, P.; Minowa, T.; Ogi, T., Liquefaction of cellulose in hot compressed water 
using sodium carbonate. Part 2: thermogravimetric-mass spectrometric study of the solid 
residues. J. Chem. Eng. of Japan 1998, 31, 571-576. 
19. Muthukumaraa, P.; Gupta, R. B., Sodium-carbonate assisted supercritical water 
oxidation of chlorinated waste. Ind. Eng. Chem. Res. 2000, 39, 4555-4563. 
20. Sluiter, A.; Hammes, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D., 
Determination of sugars, byproducts, and degradation products in liquid fraction process 
samples. NREL Biomass Program, Laboratory Analytical Procedure (LAP) 2006. 
21. Wright, A. F.; Bailey, J. S., Organic carbon, total carbon, and total nitrogen 
determinations in soils of variable calcium carbonate contents using a Leco CN-2000 dry 
combustion analyzer. Communications in soil science and plant analysis 2001, 32, 3243-
3258. 
22. Kalinichev, A. G.; Churakov, S. V., Size and topology of molecular clusters in 
supercritical water: a molecular dynamics simulation. Chem. Phys. Letters 1999, 302, 
411-417. 
23. Marcus, Y., On transport properties of hot liquid and supercritical water and their 
relationship to the hydrogen bonding. Fluid Phase Equilibr. 1999, 164, 131-142. 
24. Kruse, A.; Gawlik, A., Biomass conversion in water at 330-410 ?C and 30-50 
MPa: Identification of key compounds for indicating different chemical reaction 
pathways. . Ind. Eng. Chem. Res. 2003, 42, 267-269. 
25. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
Polym. Sci. 1994, 19, 797?841. 
141 
 
26. Schwald, W.; Bobleter, O., Hydrothermolysis of cellulose under static and 
dynamic conditions at high temperatures. . J. Carbohydr. Chem. 1989, 8, (4), 565-578. 
27. Mok, W. S. L.; M.J., A. J.; Varhegyi, G., Productive and parasitic pathways in 
dilute-acid-catalyzed hydrolysis of cellulose. Ind. Eng. Chem. Res. 1992, 31, 94-100. 
28. Mok, W. S.; Antal, M. J., Uncatalyzed solvolysis of whole biomass hemicellulose 
by hot compressed liquid water. Ind. Eng. Chem. Res. 1992, 31, 1157-1161. 
29. Antal, J. M. J.; Mok, W. S. L.; Richards, G. N., Mechanism of formation of 5-
(hydroxymethyl)-2-furfuraldehyde from D-fructose and sucrose. Carbohydr. Res. 1990, 
199, 91-109. 
30. Antal, J. M. J.; Leesomboon, T.; Mok, W. S. L.; Richards, G. N., Mechanism of 
formation of 2-furaldehyde from D-xylose. Carbohydr. Res. 1991, 217, 71-85. 
31. Bouchard, J.; Abatzoglou, N.; Chornet, E.; Overend, R. P., Characterization of 
depolymerized cellulosic residues. Wood Science and Technology 1989, 23, (4), 343-355. 
32. Watanabe, M.; Aizawa, Y.; Iida, T.; Levy, C.; Aida, T. M.; Inomata, H., Glucose 
reactions within the heating period and the effect of heating rate on the reactions in hot 
compressed water. Carbohydrate Research 2005, 340, 1931-1939. 
33. Xiuyang, L.; Sakoda, A.; Motoyuki, S., Kinetics and product distributions of 
cellulose decomposition in near critical water. J. Chem. Ind. and Eng. China 2001, 52, 
(6), 556-559. 
34. Sasaki, M.; Furukawa, M.; Adschiri, T.; Arai, K., Kinetics and mechanism of 
cellobiose hydrolysis and retro-aldol condensation in subcritical and supercritical water. 
Ind. Eng. Chem. Res. 2002, 42, 6642-6649. 
142 
 
143 
 
35. Sasaki, M.; Goto, K.; Tajima, K.; Adschiri, T.; Arai, K., Rapid and selective retro-
aldol condensation of glucose to glycolaldehyde in supercritical water. Green Chem. 
2002, 4, 285-287. 
36. Sasaki, M.; Hattori, H.; Arai, K., Rapid and selective productions of valuable 
chemical intermediates from cellulose using supercritical water. Cellulose Commun. 
2003, 10, 63-68. 
37. Hao, X. H.; Guo, L. J.; Mao, X.; Zhang, X. M.; Int., X. J. C., Hydrogen 
production from glucose used as a model compound of biomass gasified in supercritical 
water. J. Hydrogen Energy 2003, 28, 55-64. 
38. Guan, Y.; Pei, A.; Lie-jin, P.; Guo, L., Hydrogen production by catalytic 
gasification of cellulose in supercritical water. J. Chem. Eng. of Chinese Univ. 2007, 21, 
436-441. 
 
 
5. Biocrude Production from Switchgrass using Subcritical Water 
 
5.1 Abstract 
Subcritical water is an environmentally attractive solvent for organic matters and 
can be used to liquefy biomass to biocrude, which is a mixture of oxygenated 
hydrocarbons of varying molecular weights.  Liquefaction of switchgrass in subcritical 
water is studied using a semi-continuous reactor in the temperature range of 235-260?C. 
Subcritical water is pumped through a tubular reactor packed with switchgrass particles 
of 40-60 mesh size. Effects of reaction temperature and catalysis by K
2
CO
3
 are 
examined. Potassium carbonate significantly enhances the hydrolysis of macromoleculer 
components of switchgrass into water soluble products. More than 50 wt% of organic 
carbon available in switchgrass was converted to biocrude after 20 min of steady 
operation at 235?C in the presence of 0.15 wt% of K
2
CO
3
. At the high temperature 
(260?C), dehydration of biomass was favored over hydrolysis reactions. Based on 
chromatography and mass spectrometry analyses, biocrude contains lignin derived 
products, sugars and its decomposition products. Based on the infrared spectroscopy and 
electron miscopy of residue solid, the subcritical water treatment causes complete 
breakdown of lignocellulosic structure of switchgrass. In fact, the residue solid mainly 
contained lignin fractions. 
144 
 
5.2 Introduction 
The increased demand for energy, and concerns about climate change and energy 
security have led to an explosion in interest in alternative and renewable energies. Among 
the several renewable energy options available, biomass is the only renewable source, 
which is capable of producing petroleum compatible products.
1
 In recent years, U.S. 
government has sponsored numerous research studies and demonstration projects, for 
developing abundantly available biomass resources of the country for biofuels 
applications. The major focus is on non-food resources or so called second generation 
biofuels from lignocellulosic biomass, which is mainly composed of cellulose, 
hemicelluloses, and lignin. Switchgrass, a summer perennial grass native to North 
America that grows on marginal land, is considered to be a major energy crop of interest 
for the second generation biofuels. The advantages of switchgrass as energy crop include 
low requirement of chemical inputs and the use of marginal land which does not compete 
with the food production. The switchgrass plants are immense biomass producer that can 
reach heights of ?10 feet and provide 6-8 dry-tons/acre/year yield (as compared to 2.5 
tons/acre/year for hay) with a high cellulosic content.
2, 3
  
In the subcritical region, the ionization constant (K
w
) of water increases with 
temperature and is about three orders of magnitude higher than that of ambient water 
(Figure 5.1) and the dielectric constant (?) of water drops from 80 to 20.
4
 A low dielectric 
constant allows subcritical water to dissolve organic compounds, while a high ionization 
constant allows subcritical water to provide an acidic medium for the hydrolysis of 
biomass components.
5-7
  
145 
 
0 100 200 300 400 500 600 700 800
-26
-24
-22
-20
-18
-16
-14
-12
-10
Io
nization co
nstan
t
lo
g Kw (
m
o
l/kg)
2
Temperature (?C)
 
Figure 5.1 Ionization constant (K
w
) of water as a function of temperature at 25 MPa.
4
 
 
Although the dielectric constant of subcritical is low enough for organics to be 
soluble, it remains high enough to allow salt dissolution.
8
  
There are several competing pathways for converting the biomass to liquid fuel, 
chemicals, and/or hydrogen (Figure 5.2). Among them, hydrothermal (subcritical) 
liquefaction process has attracted much attention due its versatility to utilize mixed 
biomass feedstock without any pretreatment or drying, at a comparatively low 
temperature. Since water is used both as a reactant and as the reaction medium, the 
moisture content of biomass is not an issue. Moreover, biomass residue generated from 
the any other process can be utilized to produce biocrude.
9
 Liquefaction of biomass in 
subcritical water proceeds through a series of structural and chemical transformations 
involving
10
  
? Solvolysis of biomass resulting in micellar-like structure 
? Depolymerization of cellulose, hemicelluloses, and lignin 
146 
 
? Chemical and thermal decomposition of monomers to smaller molecules  
 
Figure 5.2 Major pathways for the conversion of biomass to biofuels. 
 
In this study, biocrude is defined as the solubilized fraction of biomass after 
subcritical water treatment. Biocrude, which is an aqueous carbohydrate solution derived 
from the direct liquefaction of biomass can be converted to liquid fuel, hydrogen, or 
chemicals. For example, Huber et. al. produced liquid alkanes by aqueous phase 
processing of carbohydrates derived from biomass.
11
 In another study, hydrogen was 
produced by aqueous-phase reforming of woody biomass in a batch reactor using two 
step processes: hydrolysis of biomass using sulfuric acid followed by reforming in the 
presence of Pt/Al
2
O
3
 catalyst.
12
  Glucose, a biocrude component, was catalytically 
reformed in supercritical water to produce hydrogen with yields as high as 12 mol of 
H
2
/mol of glucose reaching the stoichiometric limit.
13
  
Bobleter et al., in their pioneering work, have developed hydrothermal process for 
pretreatment of biomass and subsequent enzymatic hydrolysis.
14-16
 Several studies have 
been conducted using model compounds such as cellulose, hemicelluloses and lignin in 
147 
 
sub-and supercritical water. Minowa et al.
17, 18 
conducted hydrolysis of cellulose with and 
without catalyst. Using a conventional autoclave, water/cellulose slurry was heated in a 
nitrogen atmosphere to various temperatures between 200-350?C. Results indicate that 
decomposition of cellulose increases with the reaction temperature. For example, 
reactions below 240?C lead to the formation of water-soluble products, and those above 
260?C lead to the formation of oil, char, and gas products. Hydrolysis of microcrystalline 
cellulose in subcritical and supercritical water (320-400?C, and 0.05-10.0 seconds) in a 
continuous reactor were conducted to produce upto 75% yield of hydrolysis products 
(glucose and oligomers) along with aqueous decomposition products of glucose.
19-21
 
Most of the hemicelluloses together with lignin was solubilized in hot compressed water 
using a tubular percolating reactor in the pretreatment of biomass.
22
 Conversion of lignin 
model compound under hydrothermal conditions have also been studied earlier.
23, 24
 
Majority of previous studies have focused mainly on model compounds and have 
established the reaction kinetics and reaction pathways of such compounds in 
hydrothermal medium. However, lignocellulosic biomass is mainly a mixture of cellulose 
(38-50 wt%), hemicelluloses (23-32 wt%), and lignin (15-25 wt%) which are held 
together by covalent bonding, various intermolecular bridges, and van der Waals forces 
forming a complex structure.
25, 26
 These components interact with each other in reaction 
medium and lead to a very complex chemistry. The presence of different inorganic salts 
in biomass can also influence the reaction pathways.
27
 For example, Karagoz et al. and 
Midgett observed enhanced hydrothermal upgrading of biomass in batch reactor for the 
producing oils using alkali catalyst.
28-31
 However, there are not many studies conducted 
with the real biomass particularly with the energy crops except a recent study by Cheng 
148 
 
et al. on the rapid conversion of switchgrass in subcritical water in a batch reactor in the 
temperature range of 250-350?C and pressure of 20 MPa.
32
 
The main aim of this study is to explore the possibility of depolymerizing the 
entire holocellulose and lignin fraction of switchgrass in subcritical water to produce 
biocrude, and converting them to sugar and lignin derived components, in a semi-
continuous reactor. Effects of temperature and K
2
CO
3
 on the decomposition of 
switchgrass are examined. The major objective is to understand the decomposition 
reaction of swichgrass in subcritical water and characterization of biocrude for 
identifying the formation of value-added chemicals. In addition, the physical and 
chemical characteristics of the solid residue are investigated. 
5.3 Experimental Section 
Materials: Alamo switchgrass (panicum virgatum species), moisture content 7.78%, and 
size + 60 to - 40 mesh (0.25 to 0.42 mm) was used in this study.  The composition 
analysis of switchgrass is shown in Table 5.1. The higher heating value (HHV) of the 
switchgrass was measured in IKA C200 calorimeter as 17.10 MJ/kg. Potassium carbonate 
(K
2
CO
3
) and high purity standard reagents for HPLC analysis were purchased from 
Sigma Aldrich. De-ionized water was used for conducting the experiments.  
 
Table 5.1 Composition (wt%) of switchgrass. 
Glucan Xylan Galactan Arabinan Klason 
Lignin 
Acid soluble 
lignin 
Ash Acetic 
acid 
39.5 22.4 1.4 2.9 22.2 1.3 2.2 4.1 
C N Ca K 
44.61 0.74 0.22 0.57 
 
149 
 
Apparatus: Figure 5.3 shows the schematic of the biocrude production apparatus. It 
consists of two high-pressure liquid pumps, an electrical tubular furnace, a water-cooled 
double tube heat exchanger, a backpressure regulator, and a phase separator. Reactor was 
a 0.57? ID, 20? long high-pressure stainless steel tube.  
 
Figure 5.3 Apparatus for biocrude production. 
 
 Temperature of the reaction zone was measured using a 1/16? thermocouple 
placed inside the biomass bed from the inlet. This thermocouple measured reaction zone 
temperature with accuracy of ?0.1?C. There were three additional thermocouples 
installed at different locations in the apparatus. Real time temperature indicated by the 
thermocouple placed inside the reaction zone was used for all the analyses. Reactor 
pressure was maintained using a backpressure regulator. All the thermocouples were 
calibrated before the start of experiment. High precision HPLC pumps (Chrome Tech, 
USA) were used for the liquid flow. Pump flow rates were calibrated before every 
experiment. Multiple samples were collected and analyzed for TOC values during each 
experiment. 
150 
 
Experimental Procedure: In a typical experiment, reactor was placed inside the furnace 
after packing with 13 g of switchgrass. A stainless steel frit (pore size 2 ?m) was placed 
in the reactor outlet to prevent the entrainment of fine solids with the liquid stream. Water 
and K
2
CO
3 
solution was pumped using high pressure pumps 1 and 2 (Figure 5.3) and 
total liquid inlet flow rate was maintained 1.5 ml/min for the entire study. Experiments 
were conducted using only water and also adding K
2
CO
3 
solution for some experiments 
in the temperature range of 235-260?C. The typical heating cycle of the process is shown 
in Figure 4. After completing the desired steady operation time (t) of steady operation, 
heating was stopped and reactor was rapidly cooled to below 50?C. Biocrude produced at 
the reaction temperature was collected from the phase separator and stored at 4?C. The 
liquid products collected during all three stages of operation (Figure 5.4) i.e. preheating, 
steady operation, and cooling/washing period were analyzed for total organic carbon 
(TOC) content.  
0
40
80
120
160
200
240
280
0 102030405060
T
e
m
p
er
a
t
u
r
e (
?
C)
Time (min)
Steady operation
Cooling 
zone
Preheating
zone
t
 
Figure 5.4 Heating profile in a typical experiment. 
 
 Gaseous products were noticeably low (< 10 wt.%) at 235?C. The gaseous 
products were vented without analysis. As per the Cheng et al. study the gas yield was 
17.2 wt% at 250?C.
32
 Biomass residue remained in the reactor was washed with de-
ionized water and dried for the further analysis.  
151 
 
Two groups of experiments were conducted as listed in Table 5.2. Group I 
experiments were conducted at 260?C, 23.5 MPa in subcritical water and also in the 
presence of 0.5 wt% of K
2
CO
3
. Group II experiments were conducted at 235?C and 13.8 
MPa.  
Table 5.2 Experiment details of various runs. 
Run 
No. 
Temperature 
(?C) 
Steady operation 
time (min) 
K
2
CO
3
 
(wt%) 
Switchgrass weight loss 
dry basis (wt%) 
Group I ( at 23.5 MPa) 
1 260 50 0 82.0 
 2 260 25 0.50 81.9 
Group II (at 13.8 MPa) 
3 235 90 0 65.0 
4 235 80 0.15 69.7 
5 235 20 0 61.9 
6 235 5 0.15 42.3 
7 235 10 0.15 59.6 
8 235 20 0.15 70.8
* 
* Relative standard deviation (RSD) is 5.3% 
 
5.4 Product Separation 
Reaction products were separated into three components named as aqueous phase, 
solid precipitates, and biochar as per the scheme shown in Figure 5.5. In this study, the 
aqueous phase (water soluble fraction) and the solid precipitates are collectively called as 
Biocrude. 
152 
 
Product 1
Aqueous 
phase
Switchgrass
Biochar
(solid residue)
Solid 
precipitates
In reactorAt phsae separator
Biocrude
Product 2
Product 3
SolidLiquid
Gas (vent)
 
Figure 5.5 Products separation scheme. 
Aqueous Phase (Product 1): After removing the suspended solids by centrifugal 
separation, aqueous phase samples collected at the reaction temperature were analyzed 
for organic carbon. Sugar and degradation products were determined as per the NREL 
biomass laboratory analytical procedure (described in technical report-NREL/TP-510-
42623). Ultraviolet (UV) spectra were collected in the range of 200-400 nm. Sample was 
also taken for gas chromatography-mass spectroscopy (GC-MS) for identifying the 
phenolic compounds.  
Solid precipitates (Product 2): Because of the change in property of water, some of the 
compounds soluble in subcritical condition were precipitated as fine solids upon cooling 
to ambient temperature. These fine solids, named solid precipitates were separated by 
centrifugation and dried in oven at 105?C to the constant weight. Solid precipitate was 
analyzed for the carbohydrate composition as per the earlier discussed procedure for 
biochar.  
153 
 
Biochar (Product 3): The residue solids collected from reactor after water washing at the 
end of experiment were named biochar. Biochar was analyzed for carbohydrates 
compositions by chemical analysis, and further characterized by X-ray diffraction, FT-IR, 
and SEM techniques. Composition analysis was performed as per the National 
Renewable Energy Laboratory (NREL) biomass laboratory analytical procedure 
(described in technical report-NREL/TP-510-42618).  
5.5 Product Characterization 
TOC analysis: After separating the suspended solids from liquid by centrifugation, liquid 
samples were analyzed for TOC in Shimadzu TOC-Vcsn analyzer, and an average of 
analyses is reported in this work. 
High pressure liquid chromatography (HPLC): Monomers and oligomers of sugars and 
its degradation products in aqueous phase samples were determined as per the NREL 
biomass laboratory analytical procedure (described in technical report-NREL/TP-510-
42623). The analysis was done using HPLC having Bio-Rad Aminex HPX-87P (lead 
based) column, refractive index detector, water mobile phase, and column temperature of 
85?C. Additionally, analysis was done using Bio-Rad Aminex HPX-87H column, 
refractive index detector, 0.01 N sulfuric acid mobile phase and column temperature of 
65?C.
33
 
GC-MS analysis: Agilent model HP-5 column with 5% phenyl methyl siloxane (30 m 
long, 0.25 mm diameter, 0.25 ?m film thickness) was utilized. Oven was programmed as: 
45?C (hold 5 min) and ramped up (rate 5?C/min) to 250?C (hold 8 min). Injector was 
kept at 250?C in split mode (split ratio 50:1) with helium as carrier gas. Sample volume 
was 0.2 ?l and compounds were identified using NIST library of mass spectra. 
154 
 
XRD pattern: Rigaku Miniflex powder X-ray diffractometer equipped with a Cu K?
1
 
radiation source at 30 kV voltages, 15 mA current and a miniflex goniometer was used 
for the XRD analysis. Diffraction patterns were collected in the 2? range of 10-35? at a 
scan speed of 1?/min. and step size of 0.01?. The cellulose crystallinity index (CI) in the 
switchgrass was determined
34
 by 
 CI = (I
002
 ? I
AM
)
 
/ I
002
     [1]  
where I
002 
is the peak intensity corresponding to (002) lattice plane of cellulose molecule, 
and I
AM  
is the peak intensity observed at 2? equal to 18?. I
002
 represents both crystalline 
and amorphous material while I
AM
 represents amorphous material only. 
FT-IR spectra: Infrared spectra of the samples were obtained using the FT-IR model 
Avatar 360, Nicolet.  The sample discs were prepared by mixing 3 mg of oven dried (at 
105?C) sample with 200 mg of spectroscopy grade KBr in an agate mortar. The final 
spectra were composed of 64 scans performed in the range of 4000 and 400 cm
?1 
with 2 
cm
?1
 resolution. 
SEM: The samples were prepared onto adhesive carbon tape on an aluminum stub 
followed by sputter coating of gold. Surface morphology of the sample was studied using 
an environmental scanning electron microscopy system (Zeiss EVO 50). 
5.6 Results 
The initial experiments (Group I) were conducted at 260?C, 23.5 MPa in 
subcritical water and also in the presence of K
2
CO
3
. Eighty two percent (oven dry basis) 
of the switchgrass was solubilized in run 1, and the maximum TOC value in the biocrude 
was observed as 31,520 ppm within 10 minutes of steady operation. With the use of 
merely 0.5 wt% K
2
CO
3
, TOC in the biocrude increased to 44,500 ppm and 81.9% of the 
155 
 
switchgrass was solubilized in 25 minutes of steady operation (run 2). Therefore, the use 
of K
2
CO
3
 showed the marked catalytic effect on switchgrass decomposition. But the use 
of catalyst at this temperature frequently plugged the reactor because of the rapid 
formation of char-like solids at the frit (reactor outlet), due to the dehydration and 
pyrolysis reactions favored at the high temperature along with the catalytic effect of 
K
2
CO
3
. Therefore, subsequent studies were conducted at less severe conditions, i.e. at 
235?C and 13.8 MPa.  
Figure 5.6 shows the TOC profile in the aqueous phase with steady operation time 
for the studies conducted at 235?C with liquid flow rate 1.5 ml/min (runs 3 and 4).  
0
10000
20000
30000
40000
0 102030405060
T
O
C
 (
ppm
)
Steady operation time (min)
0.15 wt %  K
2
CO
3
(Run 4)
No  K
2
CO
3
(Run 3)
 
Figure 5.6 Total organic content (TOC) of aqueous phase versus steady operation time. 
The rate of decomposition of switchgrass increased significantly with the use of 
0.15 wt% of K
2
CO
3
 (run 4) where the maximum TOC value in aqueous phase, observed 
within 10 minutes of reaction time, was increased by more than 55%. It was observed 
that the TOC profile in the followed a bell shaped curve. TOC started increasing rapidly 
in the initial 10 min of the steady operation and then decreased with the depletion of 
156 
 
switchgrass in the reactor. Run 1-4 experiments were either conducted at higher 
temperature 260?C or for a very long steady operation time. These experiments provided 
the basis for optimizing the parameters based on the switchgrass conversion (%wt loss) 
and operational difficulties at higher temperature in flow reactor. To investigate further 
for the rate of solubilization of switchgrass at 235?C in presence of 0.15 wt% K
2
CO
3
, 
separate experiments were conducted for the 5, 10 and 20 min of steady operation (runs 
5-8). Figure 5.7 shows the percentage weight loss of swithchgrass with steady operation 
time in subcritical water in the presence of K
2
CO
3
 for runs 6, 7, and 8.  
0
20
40
60
80
100
42.3
59.6
71.6
We
i
g
h
t
 l
o
s
s
 
(
%
)
Run 7 Run 8Run 6
 
Figure 5.7 Conversion of switchgrass (weight loss) for runs 6, 7, and 8. 
 
The weight loss increased steadily and 71.6% of the switchgrass was solubilized 
after 20 minutes of steady operation in run 8. Triplicates of run 8 were conducted to 
confirm the reproducibility of the results and the average value of weight loss (run 8) is 
shown in Table 5.2; the relative standard deviation of in weight loss was 5.3%. Aqueous 
phase and solid precipitates from runs 5-8 were separated to determine the biocrude yield. 
Table 5.3 shows the yields of biocrude from switchgrass conversion for the runs 5-8 
based on the organic carbon in liquid product and solid precipitate. More than 50 wt% 
157 
 
(run 8) of carbon present in switchgrass could be converted to biocrude in the presence of 
K
2
CO
3
 at 235?C.  
Table 5.3 Biocrude yield from switchgrass conversion (wt%, carbon basis). 
 
Run no. 
Aqueous 
Phase  
(1) 
Solid  
Precipitate
* 
(2) 
Biocrude 
yield 
(1) + (2) 
Run 5 31.8 7.2 39.0 
Run 6 31.4 16.4 47.8 
Run 7 32.4 14.6 47.0 
Run 8 36.9 14.2 51.1 
* 
Carbon in solid precipitate was considered as 66.67 wt%. 
 
Cheng et al.?s study on switchgrass conversion in subcritical water using a batch 
apparatus showed that percentage weight loss of switchgrass based on initial mass varied 
between 41.3 - 54.8 wt% at 250?C depending upon the reaction time (1-60 s). Conversion 
increased with temperature to 64.6 - 78.9 wt% at 300?C.
32
  The results of the different 
products as per the Figure 5.4 products separation scheme follow. 
Aqueous Phase (Product 1): The pH of the aqueous phase ranged from 3 to 5 throughout 
the study depending on the use of K
2
CO
3
, and the liquid product appeared dark in color. 
Table 5.3 shows the percentage of carbon in switchgrass converted to water soluble 
compounds (aqueous phase). On the basis of carbon balance, 31.8 wt% of switchgrass 
(run 5) was converted to aqueous phase in the absence of K
2
CO
3
. With the use of K
2
CO
3
, 
switchgrass decomposition to water-soluble products increased to 36.9 wt% (run 8). 
158 
 
Similar effect was observed, when steady operation at 235?C was increased from 5 to 20 
min in the presence of K
2
CO
3
. The TOC value in aqueous phase during steady operation 
reaches to nearly 4 wt% (equivalent to 10 wt% of glucose solution) in the presence of 
0.15 wt% of K
2
CO
3
. Continuous liquefaction of biomass in subcritical water at large 
scale, where progressive cavity pumps can be used for feeding the biomass slurry 
continuously may provide even higher carbon concentration in aqueous phase. The 
aqueous phase from run 8 collected at steady operation was analyzed for the sugar 
compounds and organic acids by HPLC. Figure 5.8 shows the composition analysis of 
aqueous phase.  
 
Figure 5.8 Composition of aqueous phase from run 8 by HPLC analysis. 
 
The total oligomers and monomers of five carbon and six carbon sugars were 15 
wt%. More than 25 wt% of the compounds was composed of sugar degradation products 
(5-hydroxymethyl-2-furaldehyde (5-HMF), and furfural) and organic acids (lactic, 
formic, and acetic acids). GC-MS analysis was conducted for identifying the products 
159 
 
which could not be identified by HPLC analysis. Table 5.4 lists some of the important 
compounds identified with more than 85% of confidence level.  
Table 5.4 Identified compounds in biocrude from run 8 by GC-MS analysis (> 85% of 
confidence level). 
No. GC Retention time 
(minutes) 
Compound Area 
(%) 
Quality 
(%) 
1 9.0 Furfural 10.5 86 
2 21.0 1,2-benzenediol 3.8 94 
3 21.6 2,3-Dihydrobenzofuran 2.6 86 
4 21.9 2-Furancarboxaldehyde 30.2 94 
5 24.3 2-Methoxy-4-vinylphenol 2.3 93 
6 24.9 1,4-Benzenediol, 2-methyl- 2.1 96 
7 25.3 Phenol, 2,6-dimethoxy- 1.5 96 
8 25.4 Benzaldehyde, 4-hydroxy- 1.4 95 
9 26.5 Vanillin 3.0 97 
10 29.7 Homovanillyl alcohol 1.0 87 
11 32.6 Benzaldehyde, 4-hydroxy-3,5-
dimethoxy- 
1.0 91 
 
Furfural and 2-furancarboxaldehyde showed a large peak and they contributed 
10.5% and 30.2% of the peak area, respectively. Other compounds identified in GC-MS 
are mainly phenolics which might have been derived from the lignin depolymerization. 
The UV spectra of aqueous phase from runs 5 and 8 showed the strong absorption at 202 
and 280 nm, which is the characteristic of lignin or lignin derivatives.
35
  
160 
 
Solid precipitates (Product 2): The composition analysis of the solid precipitate (run 8) 
indicated only 23.3 wt% of sugar components in the precipitate and majority were the 
lignin. XRD spectrum of solid precipitate from run 8 is compared with that of untreated 
switchgrass in Figure 5.9.  
 
Figure 5.9 XRD patterns for untreated switchgrass and solid precipitate from run 8. 
 
The distinct sharp crystalline cellulosic peak (I
002
) was absent in the solid 
precipitate, which confirms that it contained mainly less crystalline or amporphous 
components. Table 5.3 shows the amount of solid precipitates collected in runs 5-8. 
Since, majority of solid precipitate contained lignin compounds, carbon percentage in 
solid precipitate was considered to be 66.67 wt% for calculating the yield.
36
 Solid 
precipitate yield was 7.2 wt% in run 5, which almost doubled (14.2 wt%) with the use of 
K
2
CO
3
 in case of run 8. Heating value of the oven dried solid precipitate (runs 5 and 8) 
was measured by IKA C 200 calorimeter using Isoperibol measuring procedure. The 
HHV of solid precipitates from runs 5 and 8 were 25.1 and 26.2 MJ/kg, respectively. 
161 
 
Solid precipitates seem to be formed via re-condensation process and precipitated as 
water insoluble fine particles. 
Biochar (Product 3): The solid residue had two distinct colors. The solids remained in 
the initial 15% of the reactor length appeared to be less reacted with light color similar to 
the original switchgrass, whereas solid collected near the outlet of the reactor was 
completely black and char like. The char-like solid (biochar) from runs 6 and 8 were 
analyzed for carbohydrates and lignin contents. As expected, biochar contained only 
cellulose and lignin fractions, and no hemicelluloses were detected. Biochar from runs 6 
and 8 contained 33 and 32 wt% cellulose, respectively. These samples were further 
investigated for structural transformations. 
Figure 5.10 compares x-ray diffraction patterns of biochar from runs 6 and 8 with 
untreated switchgrass.  The untreated switchgrass has a broader peak (I
002
) with cellulose 
crystallinity of 51.6%. The peaks became sharper after the subcritical water treatment due 
to the increase in the crystallinity to 56.9% and 64.3% for biochar from runs 6 and 8, 
respectively. 
 
Figure 5.10 XRD pattern of untreated switchgrass and biochars from runs 6 and 8. 
162 
 
 The FT-IR spectrum of biochar from run 8 is compared with that of untreated 
switchgrass in Figure 5.11. Table 5.5 shows the position of IR absorption bands and their 
assignments.  
Table 5.5 Position of infrared absorption bands of switchgrass.
32, 37, 38
  
Position of the band, cm
-1
 Functional group 
1740 C=O stretch, uncongugated (xylan) 
1635 Adsorbed  water 
1600 Aromatic skeletal vibration plus CO stretch 
1510 Benzene ring stretching (lignin) 
1425 CH
2
 shearing (cellulose), CH
2
 benching (lignin) 
1370 CH benching (cellulose, Hemicelluloses) 
1330 Phenolic OH (syringical nuclei) 
1270 C?O?C stretching in alkyl aromatic (lignin) 
1166 C?O?C asymmetry stretching (cellulose, Hemicelluloses)
1053 C?O stretching (cellulose and Hemicelluloses) 
893 ?-anomers or ?-linked (cellulose) 
 
 
Figure 5.11 FT-IR spectra of untreated switchgrass and biochar from run 8. 
 
163 
 
The remarkable changes of absorbance peaks mainly appeared in the range of 
1800 - 650 cm
-1
. Absorbance peak at 1740 cm
-1
 which represents the C=O stretching 
vibration indicating the presence of uncongugated xylan (hemicelluloses component) is 
absent in biochar spectra. The peaks at 1425, 1510 and 1600 cm
-1
 typically used for the 
lignin evaluation did not change. Peak at 1330 cm
-1 
phenolic OH syringical nuclei is also 
absent in the biochar. In biochar, peak at 1166 cm
-1
, typical of C?O?C asymmetry 
stretching in ester group, becomes less sharp, and a sharper peak at 1270 cm
-1
 due to 
guaiacyl nuclei of lignin plus CO stretch is observed. The peak at 893 cm
-1 
is 
characteristic of ?-anomers or ?-linked glucose polymers and the peak at 1053 cm
-1
 is C?
O stretching. These peaks are present in the untreated switchgrass but are absent in 
biochar. Further, the absence of peak around 1635 cm
-1
, which indicates the adsorbed 
moisture, suggested that the lignocellulosic structure of switchgrass was totally broken 
after the treatment.
32, 37, 38
  
SEM images of untreated switchgrass and biochar from run 8 are shown in Figure 
5.12a-b. These images further confirm the disruption of lignocellulosic matrix after the 
treatment. Small globular particles can be seen precipitated over thin fibers. 
           
a b
Figure 5.12 SEM image of (a) untreated switchgrass, and (b) biochar from run 8. 
 
164 
 
5.7 Discussion 
The ionization constant of water increases with temperature and reaches a 
maximum near 250?C where it is about 1000-fold that at ambient temperature. Therefore, 
subcritical water in the 220-300?C region offers opportunities both as a benign solvent 
and as a self-neutralizing catalyst. Here, water acts as both a reactant and the reaction 
medium. Water as reactant leads to hydrolysis reactions and rapidly degrades the 
polymeric structure of biomass to water-soluble products. As explained earlier, 
switchgrass is composed mainly of hemicelluloses, cellulose and lignin, which are 
arranged in a complex manner within the natural feedstock. The chemical structure of 
these polymers is different and so is their individual reactivity as described below.   
Hemicelluloses possess side groups, including acetic acid, pentoses, hexuronic 
acids, and deoxyhexoses, which are responsible for the solubility in water and/or alkali. 
These side groups reduce aggregation of hemicelluloses molecules making them more 
susceptible to chemical degradation than cellulose. Hemicelluloses in alkali medium 
undergo both physical changes and chemical reactions including swelling, dissolution, 
saponification, reprecipitation, peeling, and glycosidic cleavage. Most of the 
hemicelluloses along with considerable part of lignin are dissolved in water at 180?C. In 
the hydrothermal pretreatment of Kanlow switchgrass, more than 90% of hemicelluloses 
were solubilized at 200?C.
39
 Low activation energy of lignin causes substantial 
degradation of lignin in hydrothermal medium at temperature below 200?C.  Reaction 
proceeds through the cleavage of aryl ether linkages, fragmentation, and dissolution.
14, 40
 
Lignin depolymerization yields low molecular weight fragments having very reactive 
functional groups such as syringols, guaiacols, catechols and phenols.
41
 
165 
 
Cellulose, which exists as a mixture of two crystalline forms I
?
 and
 
I
?
 is an 
unbranched chain of anhydroglucose linked head to tail by ?-glycosidic linkages.
42, 43
 The 
?-linkages in cellulose form linear chains that are highly stable and resistant to chemical 
attack because of the high degree of intra- and inter-molecular hydrogen bonding. 
Hydrothermal degradation of cellulose is a heterogeneous and pseudo-first-order reaction 
for which detailed chemistry and mechanism have been proposed.
14, 22, 44, 45
 Cellulose 
converts to hydrolysis products (oligomers and monomers) and the aqueous degradation 
products of glucose, including glycoaldehyde, fructose, anhydroglucose, 5-HMF, and 
furfural. The presence of alkali catalyzes degradation reactions via numerous side 
reactions, and comparatively lower monomer sugar (< 30%) was obtained after the 
treatment of cellobiose with 0.1N NaOH.
46
  
As observed in this work, K
2
CO
3
 acted as a catalyst in the decomposition of 
switchgrass during the subcritical water treatment. With the hydrolytic cleavage of 
glycosidic linkages, catalyst also promoted the endwise depolymerization reactions 
(peeling) of hemicelluloses and cellulose molecules, leading to the formation of organic 
acids. The autohydrolysis reactions with the generation of organic acids enhance the 
degradation of cellulose and lignin. Alkali in the reaction medium acts as hydrolytic 
reagent for the disintegration of macromolecular lignin in biomass.
14
 In hydrothermal 
reaction medium, most of the hemicelluloses and part of the lignin are removed below 
200?C, thus increasing the accessibility of water to cellulose molecules and remaining 
free lignin, which were initially associated with hemicelluloses. At the higher 
temperature, dehydration reactions starts competing with the hydrolysis reactions; though 
hydrothermal media has excess water in the reactor.
8
 For example, experiments at 260?C 
166 
 
(Group I) mainly promoted the dehydration reaction and so, reactor was frequently 
plugged due to the presence of char-like solids at the reactor outlet.  
In the case of Group II experiments, hydrolysis reactions were favored over 
dehydration. The presence of small amount of catalyst (0.15 wt% K
2
CO
3
), in fact, acted 
like a molecular scissor in breaking down the complex polymeric structure of biomass to 
water soluble products by hydrolysis and peeling reaction.  Cellulose and hemicelluloses 
depolymerize to their respective monomeric sugar units and forms a homogeneous 
medium in hydrothermal conditions. Under the reaction condition these sugars start 
degrading, via several reaction pathways, to organic acids, aldehydes, furfural, phenols 
and also the heavy molecular weight compounds.
47
 Lignin decomposition products have 
very reactive functional groups. Therefore, part of these compounds might again be 
cross-linked with the available organic acids and re-condense together to form high-
molecular-weight, water-insoluble fragments. In fact, SEM image (Figure 12b) of biochar 
obtained in run 8 confirms the re-condensation process and globular particles can be seen 
deposited over the thin fibrous structure. Re-condensed particles over the fibers could 
have limited the accessibility of water molecules to the remaining cellulose. 
Temperatures above 200-220?C induce faster lignin solubilization, and the solubilized 
lignin is subjected to re-condensation because of longer reaction time.
48
 Reaction kinetics 
of the hydrothermal treatment of lignin obtained from different sources was studied in 
detail by Zhang et al. Switchgrass residue obtained after two-stage dilute acid hydrolysis 
process was also used in the study. The liquid yield from the switchgrass residue treated 
at 374?C was 29.7 wt%.
49
 
167 
 
Due to the numerous side reactions in the presence of alkali salts, it is difficult to 
obtain a high sugar yield
46
 and nearly impossible to quantify or even identify all 
substances formed during conversion of biomass;
27, 50
 therefore, only key compounds are 
analyzed for representing the different reaction pathways.  Liquid composition from run 8 
sample clearly indicates that cellulose and hemicellulose fractions of switchgrass were 
initially converted to hydrolysis products (oligomers and monomers). Since these 
products are not stable at high temperature for a long residence time, the major part of 
these products were further degraded to the aqueous degradation products of sugars, 
including glycoaldehyde, fructose, anhydroglucose, 5-HMF and furfural.  
Lignin is an aromatic polymer mainly built of phenylpropanoid precursors, which is a 
rich source of phenolic compounds. Several phenolic compounds such as 1,2 
benzenediol, vanillin, and others (Table 4) were identified in the aqueous phase. The 
yield of phenols increases with temperature and degradation rate of phenols is slower 
than the other aqueous products such as 5-HMF, furfurals.
51
 Typically, aromatic 
compounds are assumed to originate from lignin precursors during hydrothermal 
treatment of biomass. But, earlier study with model compounds such as fructose, 5-HMF, 
and cellulose in hydrothermal medium has shown that aromatics can also be formed from 
sugar compounds.
52, 53
 The GC-MS result (run 8) showed the presence of low molecular 
weight compounds (C
5
-C
9
). Furfural and 2-furancarboxaldehyde formed the major 
components among the identified compounds. The compounds identified in aqueous 
products are in agreement with Cheng et al.?s study.
32
 Most of the compounds are 
oxygenated hydrocarbons of five to nine carbon atoms. Carbon percentages in these 
compounds vary from 40% (for sugars) to 66.7% (for furfural). 
168 
 
 Biochar from runs 6 and 8 still composed nearly 32 wt% of cellulose, which 
suggest that the complete hydrolysis of cellulose was not be achieved in these 
experiments. The cellulose crystallinity in the biochar increased mainly due the removal 
of amorphous portion of cellulose and some structural rearrangements after the 
treatments. Increased crystallinity may also reduce the accessibility of water molecules. 
As discussed earlier, FT-IR spectra of biochar (run 8) showed the presence of peaks used 
for the lignin evaluation, whereas peaks typical of polysaccharides were either absent or 
became weak. 
5.8 Conclusion 
Switchgrass can be effectively liquefied to produce biocrude in subcritical water 
at a comparatively low temperature (235?C) in a flow reactor.  Biocrude is composed of 
aqueous phase (water-soluble compound) and solid precipitates. The aqueous phase 
contains oligomers and monomers of five and six carbon sugars, degradation products (5-
HMF and furfural), organic acids (lactic, formic and acetic acid), 2-furancarboxaldehyde 
and other phenolic products containing 5 to 9 carbon atoms. Solid precipitates seem to be 
formed via re-condensation process and precipitated as water insoluble fine particles due 
to the decrease in the solubility upon cooling. A small amount of potassium carbonate 
can catalyze the liquefaction, which can be used to lower the severity of the reaction 
conditions, and to enhance the decomposition of biomass to water-soluble products. The 
residual solid (i.e., biochar) contained mainly lignin fractions and a small amount of 
cellulose. Based on the infrared spectroscopy and electron microscopy, it was confirmed 
that subcritical water treatment leads to a breakdown of lignocellulosic structure. Char-
169 
 
like color of residue biomass is caused by partial dehydration of switchgrass even in 
hydrothermal medium. 
5.9 Acknowledgment 
Author is thankful for the financial supports from the National Science 
Foundation (grant NSF-CBET-0828269) and Alabama Center for Paper and Bioresource 
Engineering.  In addition, the authors thank Prof. Sushil Adhikari for GC-MS analysis, 
Ms. Urvi Kothari for HPLC analysis, Mr. Brian Schwieker for the assembly of the 
apparatus, and Prof. Y.Y. Lee for fruitful discussions. 
 
5.10 References 
1. Perlack, R. D.; Wright, L. L.; Turhollow, A. F.; Graham, R. L.; Strokes, B. J.; 
Erbach, D. C., Biomass as a feedstock for a bioenergy and bioproducts industry:The 
technical feasibility of a billion-ton annual supply. A Joint report sponsored by US 
department of energy and US department of agriculture 2005, 78. 
2. Bransby, D. Switchgrass  profile.  
 http://bioenergy.ornl.gov/papers/misc/switchgrass-profile.html. (May 10, 2009), 
3. Rinehart, L., Switchgrass as a bioenergy crop. www.attra.ncat.org. (May 10, 
2009) 
4. Tester, J. W.; Holgate, H. R.; Armellini, F. J.; Webley, P. A.; Killilea, W. R.; 
Hong, G. T.; Barner, H. E. In Supercritical water oxidation technology ? process 
development and fundamental research, In Emerging technologies in hazardous waste 
management III, Washington D.C., 1993; Amercan Chemical Society: Washington D.C., 
1993; pp 35-76. 
170 
 
5. Franck, E. U., Fluids at high pressures and temperatures. Pure Appl. Chem. 1987, 
59, (1), 25-34. 
6. Miyoshia, H.; Chena, D.; AkaiT., A novel process utilizing subcritical water to 
recycle soda?lime?silicate glass. Journal of Non-Crystalline Solids 2004, 337, ( 3), 280-
282. 
7. Savage, P. E., Organic chemical reactions in supercritical water. Chemical 
Reviews 1999, 99, 603-621. 
8. Patrick, H. R.; Griffith, K.; Liotta, C. L.; Eckert, C. A., Near critical water: A 
benign medium for catalytic reactions. Ind. Eng. Chem. Res. 2001, 40, 6063-6067. 
9. Peterson, A. A.; Vogel, F.; Lachance, R. P.; Froling, M.; Michael Jerry Antal, J.; 
Tester, J. W., Thermochemical biofuel production in hydrothermal media:A review of 
sub- and supercritical water technologies. Energy and Environmental Science 2008, 1, 
32-65. 
10. Chornet, E.; Overend, R. P., Biomass liquefaction: An overview. In 
Fundamentals of Thermochemical Biomass Conversion, Overend, R. P.; Milne, T. A.; 
Mudge, L. K., Eds. Elsevier Applied Science: New York, 1985; pp 967-1002. 
11. Huber, G. W.; Cheda, J. N.; Barrett, C. J.; Dumesic, J. A., Production of liquid 
alkanes by aqueous processing of biomass derived carbohydrates. Science 2005, 308, 
1446-1450. 
12. Valenzuela, M. B.; Jones, C. W.; Agrawal, P. K., Batch aqueous reforming of 
woody biomass. Energy & Fuels 2006, 20, 1744-1752. 
13. Byrd, A. J.; Pant, K. K.; Gupta, R. B., Hydrogen production from glucose using 
Ru/Al2O3 catalyst in supercritical water. Ind. Eng. Chem. Res. 2007, 46, 3574-3579. 
171 
 
14. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
Polym. Sci. 1994, 19, 797?841. 
15. Bobleter, O.; Niesner, R.; Rohr, M., The hydrothermal degradation of cellulosic 
matter to sugars aand their fermentative conversion to protein. J. Appli. Polym Sci. 1976, 
20, 2083. 
16. Bobleter, O.; Bonn, G.; Concin, R., Hydrothermolysis of biomass-production of 
raw material for alcohol fermentation and other motor fuels. Alt. Energy Sour. 1983, 3, 
323-332. 
17. Minowa, T.; Zhen, F.; Ogi, T., Cellulose decomposition in hot-compressed water 
with alkali or nickel catalyst. J. Supercritical Fluids 1997, 13, 253-259. 
18. Minowa, T.; Zhen, F.; Ogi, T.; Varhegyi, G., Decomposition of cellulose and 
glucose in hot-compressed water under catalyst-free conditions. J Chem. Eng. Japan 
1998, 31, 131-134. 
19. Sasaki, M.; Fang, Z.; Fukushima, Y. N.; Adschiri, T.; Arai, K., Dissolution and 
hydrolysis of cellulose in subcritical and supercritical water. Ind. Eng. Chem. Res. 2000, 
39, 2883-2890. 
20. Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.; N.Takeda; Adschiri, T.; 
Arai, K., Cellulose hydrolysis in subcritical and supercritical water. J. Supercritical 
Fluids 1998 1998, 13, 261-268. 
21. Kumar, S.; Gupta, R. B., Hydrolysis of Microcrystalline Cellulose in Subcritical 
and Supercritical Water in a Continuous Flow Reactor. Industrial and Engineering 
Chemistry Research 2008, 47, (23), 9321-9329. 
172 
 
22. Mok, W. S.; Antal, M. J., Uncatalyzed solvolysis of whole biomass hemicellulose 
by hot compressed liquid water. Ind. Eng. Chem. Res. 1992, 31, 1157-1161. 
23. Saisu, M.; Sato, T.; Watanabe, M.; Adschiri, T.; Arai, K., Conversion of lignin 
with supercritical water-phenol mixtures. Energy &Fules 2003, 17, 922-928. 
24. Wahyudiono; Kanetake, T.; Sasaki, M.; Goto, M., Decomposition of a lignin 
model compound under hydrothermal conditions. Chem. Eng. Technol 2007, 30, (8), 
1113-1122. 
25. O'Sullivan, A. C., Cellulose: the structure slowly unravels. Cellulose 1997, 4, (3), 
173-207. 
26. Schacht, C.; Zetzl, C.; Brunner, G., From plant materials to ethanol by means of 
supercritical fluid technology. Journal of Supercritical Fluids 2008, 46s  (3), 299-321. 
27. Behrendt, F.; Neubauer, Y.; Oevermann, M.; Wilmes, B.; Zobel, N., Direct 
liquefaction of biomass. Chem. Eng. Technol 2008, 31, (5), 667-677. 
28. Karagoz, S.; Bhaskar, T.; Muto, A.; Sakata, Y.; Uddin, M. A., Low temperature 
hydrothermal treatment of biomass: Effect of reaction parameters on products and boiling 
point distributions. Energy & Fuels 2004, 18, 234-241. 
29. Karagoz, S.; Bhaskar, T.; Muto, A.; Sakata, Y., Effect of Rb and Cs carbonates 
for production of phenols from liquefaction of wood biomass. Fuel 2004, 83, 2293-2299. 
30. Karagoz, S.; Bhaskar, T.; Muto, A.; Sakata, Y., Hydrothermal upgrading of 
biomass: Effect of K
2
CO
3
 concentration and biomass/water ratio on product distrubution. 
Bioresource Technology 2006, 97, 90-98. 
173 
 
31. Midgett, J. S. Assessing a hydrothermal liquefaction processing using biomass 
feedstocks. Louisiana State University And Agricultural and Mechanical College, Baton 
Rouge, 2005. 
32. Cheng, L.; Ye, X. P.; He, R.; Liu, S., Investigation of rapid conversion of 
switchgrass in subcritical water. Fuel Processing Technology 2009, 90, 301-311. 
33. Sluiter, A.; Hammes, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D., 
Determination of sugars, byproducts, and degradation products in liquid fraction process 
samples. NREL Biomass Program, Laboratory Analytical Procedure (LAP) 2006. 
34. Segal, L.; Creely, J. J.; Martin, A. E.; Conrad, C. M., An empirical method for 
estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. 
Textile Res. J. 1959, 29, 786-794. 
35. Ando, H.; Sakaki, T.; Kokusho, T.; Shibata, M.; Uemura, Y.; Hatate, Y., 
Decomposition behaviour of plat biomass in hot-compressed water. Ind. Eng. Chem. Res. 
2000, 39, 3688-3693. 
36. http://www.nrel.gov/biomass/glossary.html#L. 
37. Baeza, J.; Freer, J., Chemical characterization of wood and its components. In 
Wood and cellulosic chemistry, second ed.; Hon, D. N.-S.; Shiraishi, N., Eds. Marcel 
Dekker, Inc.: New York. Basel, 2001; pp 275-384. 
38. kobayashi, N.; Okada, N.; Hirakawa, A.; Sato, T.; J.Kobayashi; Hatano, S.; Itaya, 
Y.; Mori, S., Characteristics of solid residues obtained from hot-compressed-water 
treatment of woody biomass. Industrial and Engineering Chemistry Research 2009, 48, 
373-379. 
174 
 
39. Suryawati, L.; Wilkins, M. R.; Bellmer, D. D.; Huhnke, R. A.; Maness, N. O.; 
Banat, I. M., Simulteneous sacchrification and fermentation of Kanlow Switchgrass 
pretreated by hydrothermolysis using Kluyveromyces marxianus IMB4. Biotechnology 
and Bioengineering 2008, 101, (5), 894-902. 
40. Lai, Y.-Z., Chemical Degradation. In Wood and Cellulosic Chemistry, David N.-
S. Hon, N. S., Ed. Marcel Dekker, Inc.: New York . Basel, 2001; pp 443-512. 
41. Fang, Z.; Sato, T.; Jr., R. L. S.; Inomata, H.; Arai, K.; Kozinski, J. A., Reaction 
chemistry and phase behaviour of lignin in high-temperature and super critical water. 
Bioresource Technology 2008, 99, 3424-3430. 
42. Jarvis, M., Chemistry: Cellulose stacks up. Nature 2003, 426, 611-612  
43. Atalla, R. H.; Vanderhart, D. L., Native Cellulose: A Composite of Two Distinct 
Crystalline Forms. Science  1984, 223, (4633), 283 - 285. 
44. Schwald, W.; Bobleter, O., Hydrothermolysis of cellulose under static and 
dynamic conditions at high temperatures. . J. Carbohydr. Chem. 1989, 8, (4), 565-578. 
45. Mok, W. S. L.; M.J., A. J.; Varhegyi, G., Productive and parasitic pathways in 
dilute-acid-catalyzed hydrolysis of cellulose. Ind. Eng. Chem. Res. 1992, 31, 94-100. 
46. Bobleter, O., Hydrothermal degradation and fractionation of saccharides and 
polysaccharides In Polysaccharides: Structural diversity and functional versatility, 
Dumitriu, S., Ed. Marcel Dekker: 2005; pp 893-936. 
47. Kruse, A.; Gawlik, A., Biomass conversion in water at 330-410 ?C and 30-50 
MPa: Identification of key compounds for indicating different chemical reaction 
pathways. . Ind. Eng. Chem. Res. 2003, 42, 267-269. 
175 
 
176 
 
48. Kim, T. H.; Lee, Y. Y., Pretreatment and fractionation of corn stover by ammonia 
recycle percolation process Bioresource Technology 2005, 96, (18), 2007-2013. 
49. Zhang, B.; Huang, H.-J.; Ramaswamy, S., Reaction Kinetics of the hydrothermal 
treatment of lignin. Appl Biochem Biotechnol 2008, 147, 119-131. 
50. Loppinet-Serani, A.; Aymonier, C.; Cansell, F., Current and foreseeable 
applications of supercritical water for energy and the environment. ChemSusChem 
2008, 1, 486-503. 
51. B?hler, W.; Dinjus, E.; Ederer, H. J.; Kruse, A.; Mas, C., Ionic reactions and 
pyrolysis of glycerol as competing reaction pathways in near- and supercritical 
water. Journal of Supercritical Fluids 2002, 22, 37?53. 
52. Luijkx, G. C. A.; Rantwijk, F. V.; Bekkum, H. V., Hydrothermal formation of 
1,2,4-benzenetriol from 5-hydroxymethyl-2-furaldehyde and D-fructose. 
Carbohydrate Research 1993, 242, 131-139. 
53. Nelson, D. A.; Molton, P. M.; Russell, J. A.; Hallen, R. T., Application of direct 
thermal liquefaction for the conversion of cellulosic biomass. Ind. Eng. Chem. 
Prod. Res. Dev. 1984, 23, (3), 471-475. 
 
 
177 
?
6. Production of Biochar from Switchgrass by Hydrothermal Carbonization 
 
6.1 Abstract 
 Hydrothermal carbonization (HTC), a traditional but recently revived process 
using hot compressed water, is a promising route for converting biomass to biochar. HTC 
of switchgrass, a major energy crop, is studied to produce the high energy density biochar in 
a batch reactor at 200-280 ?C. The effects of temperature, residence time, and pressure 
are examined. The obtained biochar is characterized for molecular structure by FTIR, 
XRD, and SEM analyses. The results indicate that temperature plays an important role in 
the conversion of switchgrass to biochar. At the higher temperatures, the lignocellulosic 
crosslink structure of switchgrass is broken and biochar having a high energy density is 
obtained. Also, the micro-crystalline structure of cellulosic components present in 
switchgrass is destroyed and biochar contained mainly amorphous fractions as confirmed 
by the XRD spectra. The liquid fraction recovered after the HTC of switchgrass, termed 
biocrude, was further subjected to HTC in the temperature range of 200-280?C to recover 
soluble carbon fractions.  The sugar and its degradation products present in biocrude 
precipitated as uniform carbon rich microspheres (biotar particles) as a result of 
polycondensation and dehydration reactions. The results demonstrate that the HTC 
process can provide a promising route for high energy density (28 MJ/kg) solid fuel and 
carbonaceous functional materials for use in a variety of applications. 
178 
?
6.2 Introduction 
 Depletion of fossil fuels and environmental concerns are prompting search for 
alternative and renewable energy resources. Various options include wind, hydro/oceanic, 
geothermal, solar, and biomass energies, among which biomass is the only renewable 
source capable of directly producing solid, liquid and gaseous fuels. Additional advantage 
of biomass utilization is in its wider distribution in the world which may eliminate the 
regional dependence for the energy sources. Biomass is the CO
2
 neutral, as during the 
growth and utilization process of biomass, same amount of carbon dioxide and other 
forms of carbon is absorbed and released, respectively. Hence, the concentration of 
carbon dioxide in the atmosphere theoretically remains constant in this cycle.
1
 The 
conversion of biomass to biofuels can be realized in several pathways such as pyrolysis, 
gasification, fermentation and hydrothermal reaction.
2
 Hydrogen,
3
 ethanol,
4
 organic 
acid,
5-7
 biofuel,
8
 and biochar 
9
 are produced in different processes. Recently, the research 
attention was focused on the conversion of biomass to biochar which has higher heating 
valve (HHV) typically over 28 kJ/g and is richer in carbon content as compared to raw 
biomass. Biochar is defined here as the non-liquefied carbonaceous solid product having 
a high energy density resulting from the thermal treatment of biomass. Table 6.1 
compares some of the advantages of biochar as solid fuels to biomass. 
Table 6.1 Comparison of biomass and biochar as solid fuel. 
Biomass Biochar 
High moisture retention Low moisture retention and easily dried 
Low heating value, so high transportation cost  High heating value, so less ($)/MJ cost during 
transportation 
Perishable on storage Not perishable, and can be stored longer 
Fibrous and so, difficult material handling Friable, easier to compact and handle 
Poor compatibility with coal for co-firing Better compatibility with coal for co-firing 
179 
?
 In nature, coal is formed from plant material undergoing heat and pressure 
treatment over millions of years. The acceleration of coalification of biomass by a factor 
of 10
6
-10
9
 in hydrothermal medium under milder process conditions can be a 
considerable and technically-attractive alternative for biochar production.
10
 Essentially, 
all forms of biomass can be converted to biochar. Forest thinning, herbaceous grasses, 
crop residues, manure, and paper sludge are some of the potentially attractive feedstocks. 
The enhanced transportation and solubilization properties of sub- and super-
critical water (hydrothermal medium) play an important role in the transformation of 
biomass to high energy density fuels and functional materials.
11
 Here, water acts as 
reactant as well as reaction medium which helps in performing hydrolysis, 
depolymerization, dehydration and decarboxylation reactions. The proton-catalyzed 
mechanism, direct nucleophilic attack mechanism, hydroxide ion catalyzed mechanism, 
and the radical mechanism play important roles in the conversion of biomass in 
hydrothermal medium. 
12, 13
 Under the umbrella of hydrothermal process, the conversion 
of a variety of biomass to chemicals, including organic monomer, biofuel, hydrogen, and 
biochar have been studied widely. Through hydrothermal carbonization, a carbon-rich 
black solid as insoluble product is obtained from biomass in 180?350?C range. 
10, 14
  
 The earliest research focused on the analyzing the changes in O/C and H/C atomic 
ratio to understand the chemical transformations taking place during HTC.
15
 For 
example, Marta et al. studied the hydrothermal carbonization of three different 
saccharides (glucose, sucrose, and starch) at temperatures ranging from 170 to 240 ?C. 
The result showed that a carbon-rich solid product made up of uniform spherical 
micrometer-sized particles of diameter 0.4?6 mm range could be synthesized by 
180 
?
modifying the reaction conditions. The formation of the carbon-rich solid through the 
HTC of saccharides was the consequence of dehydration, condensation, polymerization 
and aromatization reactions. In a recent study, the same group of researchers used 
cellulose as starting material and successfully established that highly functionalized 
carbonaceous materials can be produced by HTC from cellulose in the range of 220 to 
250?C.
16
 The formation of this material follows essentially the path of a dehydration 
process, similar to that previously observed for the hydrothermal transformation of 
saccharides such as glucose, sucrose or starch. In another recent study, Titirici et al. 
compared hydrothermal carbons synthesized from diverse monomeric sugars and their 
derivatives (5-hydroxymethyl-furfural-1-aldehyde (HMF) and furfural) under 
hydrothermal conditions at 180?C with respect to their chemical structures. The results 
showed that type of sugars has an effect on the structure of carbon rich solids.
17
 Majority 
of the research studies have utilized monomeric sugars instead of real biomass as model 
compounds for understanding the reaction mechanism and the structural differences in 
solids formed from the diverse sugar sources in HTC process. 
 The traditional method for biochar production from biomass sources is slow 
pyrolysis, where dry biomass is used for the purpose in the range of 500 to 800?C.  Antal 
Jr. and coworkers has developed another method for charcoal production which is named 
as flash carbonization. The process is conducted at elevated pressure by the ignition and 
control of a flash fire within a packed bed of biomass.
18
 Considering the relatively high 
energy consumption needed in the pyrolysis process,?hydrothermal carbonization process 
which is typically conducted in subcritical water below 300?C might be an economical 
181 
?
and efficient option for biochar production. The process is particularly important since it 
can utilize the wet biomass. 
 In this work our primary interest is the production of carbonaceous functional 
materials and higher energy density solid fuels. Our work relates to a process for the 
hydrothermal carbonization of biomass in shorter residence time (order of minutes to few 
hours) and moderate temperature condition. The effect of reaction temperature and 
pressure in the range of 200-280?C and 2-8 MPa, respectively are studied for 
understanding the carbonization of an energy crop switchgrass. The liquid product 
(biocrude) from HTC process is a complex mixture organic compounds soluble in 
water,
2,19
 mainly resulting from the decomposition sugars and acid soluble lignin. In this 
work, biocrude is also investigated as a source for producing biotar particles (a carbon 
rich solid) in hydrothermal medium, by polymerization of the organic compounds in 
biocrude. The FTIR, XRD and SEM analytical techniques were used to characterize the 
products and to understand the transformation process during HTC. The higher heating 
value (HHV) of the products are also evaluated and discussed. 
6.3 Experimental Section 
Materials: Auburn grown switchgrass (chopped to 5-10 mm long size) having moisture 
content of 7-8 wt% was used for the study. The main physical and chemical properties of 
the switchgrass are listed in the Table 6.2. De-ionized water was used for conducting the 
experiments.  
 
 
 
Table 6.2 Main physical and chemical properties of switchgrass. 
Properties Mass  
Moisture (%) 2.6 
Lignin (%) 23.5 
Cellulose (%) 39.5 
Hemicellulose (%) 26.7 
Carbon (%) 44.6 
HHV(kJ/g) 17.1 
 
Apparatus and procedure: Hydrothermal carbonization of switchgrass was carried out in 
a batch apparatus (Figure 6.1). The reactor was a high pressure stainless steel tube having 
an internal volume of 65.0 cm
3 
(internal diameter 14.5 mm).  
 
Figure 6.1 Apparatus for HTC process. 
  
 The study was conducted for the temperature range of 200-280?C and reaction 
times of 0.5-4.0 h as described in Table 3. To maintain the water in liquid phase during 
the reaction, water fill was fixed at 75% of reactor volume. In a typical run, the desired 
amount of switchgrass and water was charged into the reactor and sealed. The reactor 
was heated up to a reaction temperature in an electric furnace. After the desired residence 
time, the reactor was rapidly cooled to quench the reaction. The residence time was 
defined as the time at reaction temperature excluding preheating and cooling time. The 
182 
?
183 
?
preheating and cooling time typically varied between 10-15 minutes. To study the effect 
of pressure on biochar formation at constant temperature, reactor pressure was increased 
to desired pressure by pumping water into the reactor. Similarly, the reactor was filled 
with biocrude obtained after the HTC process of switchgrass for experiments related to 
biotar particles production.  
The solid products (biochar and biotar) recovered after the HTC processes were 
washed with deionized water and oven dried. The biochar samples were characterized for 
molecular structure by infra-red spectroscopy (FTIR),? X-ray diffraction (XRD) and 
scanning electron microscopy (SEM) techniques. The higher heating value of biochar 
was analyzed using a calorimeter. Liquid products were analyzed for total organic carbon 
(TOC) content and for the organic compounds by? high performance liquid 
chromatography (HPLC) before subjecting to the second stage HTC process for 
producing biotar particles. 
Product Separation Scheme: Figure 6.2 shows the two stage process and the product 
separation scheme. In the first stage, biochar is produced after HTC process using 
switchgrass as starting material. The biotar particles are recovered during the second 
stage HTC process where biocrude from the first stage HTC is used as starting material. 
Gaseous products which were appreciably low in both the stages were vented out. 
Liquid
Water
Switchgrass
HTC Reactor HTC Reactor 
Biochar
(product 1)
Gas (vent)
(biocrude)
Biotar Particles 
(product 2)
Gas (vent)
Liquid
Stage 1 Stage 2
 
Figure 6.2 Products separation scheme in two stage HTC process. 
6.4 Product Characterization 
SEM analysis: The biochar samples were prepared onto adhesive carbon tape on an 
aluminum stub followed by sputter coating of gold. Surface morphology of the sample 
was studied using an environmental scanning electron microscopy system (Zeiss EVO 
50). 
 XRD analysis: Rigaku Miniflex powder X-ray diffractometer equipped with a Cu K?
1
 
radiation source at 30 kV voltages, 15 mA current and a miniflex goniometer was used 
for the XRD analysis. Diffraction patterns were collected in the 2? range of 10-35? at a 
scan speed of 1?/min. and step size of 0.01?.  
FTIR analysis: Infrared spectra (4000?400 cm
-1
) were recorded by a Nicolet IR100 FTIR 
spectrometer which was equipped with a TGS/PE detector and a silicon beam splitter 
with 1 cm
-1
 resolution. The sample discs were prepared by mixing oven dried (at 105?C) 
sample with spectroscopy grade KBr in an agate morter. 
HHV analysis: The higher heating valve of sample was determined by IKA-C200 
Calorimeter. The samples were dried (at 105?C) and compressed as small pellet for HHV 
analysis. 
184 
?
185 
?
HPLC analysis: Liquid analysis was performed with an HPLC system equipped with a 
refractive index detector. Two columns were used in the HPLC analysis. Monomers and 
oligomers of sugars were determined by Bio-Rad Aminex HPX-87P (lead based) column, 
water mobile phase, and column temperature of 85?C. The other column was Bio-Rad 
Aminex HPX-87H column with IR detector, to identify polar compounds with a polar 
functional group (-OH, -C=O, -COOH) and separate more intermediate products expect 
sugars. The column was held at 65?C. The element containing 0.01mol/L H
2
SO
4
 was 
applied at a flow rate of 1.0 ml/min. The peak identification was accomplished by 
comparison of sample peak retention times with those of standard solutions of pure 
compounds. 
TOC analysis: The total organic carbon concentration (TOC) in clear liquid products, 
obtained after centrifugation at 3600 rpm, was measured using a TOC analyzer 
(Shimadzu TOC 5000A). 
6.5 Results and Discussions 
 Table 6.3 shows the details of the experiment conditions, biochar yield, and HHV 
of biochar.  
 
 
 
 
 
 
 
 
 
186 
?
Table 6.3 Experiment summary and results. 
Run 
 
Temperature 
(?C) 
Reaction Time 
 (h) 
Pressure 
(MPa) 
Biochar Yield 
(wt%) 
Biochar HHV 
(kJ/g) 
1 200 2.0 17.2 59.8 21.5 
2 200 2.0 27.6 66.7 20.9 
3 200 4.0 17.2 63.2 21.5 
4 200 4.0 27.6 60.6 21.8 
5 230 2.0 17.2 48.9 26.9 
6 230 2.0 27.6 45.6 24.3 
7 230 4.0 17.2 52.8 25.7 
8 230 4.0 27.6 49.1 25.4 
9 200 4.0 2.4 66.4 20.3 
10 230 4.0 4.1 37.8 23.5 
11 260 4.0 6.5 28.4 25.9 
12 280 4.0 8.3 19.0 28.1 
13 280 0.5 7.2 20.7 24.7 
14 280 1.0 7.2 19.2 28.3 
 
 In general, biochar yield (wt%) decreased with increase in temperature, whereas 
HHV of biochar showed opposite trend. Qualitatively, the biochar produced at higher 
temperatures retains less moisture and is more fragile and could easily be powdered just 
by hand. The results are discussed in following sections: 
Effect of residence time and pressure: Experiments at the temperature 200?C and 230?C 
were conducted to study the effect of residence time and pressure on the carbonization of 
switchgrass. The residence time and pressure varied were changed from 2 to 4 h and 17.6 
and 27.2 MPa at these temperatures. Figures 6.3 and 6.4 show the yield (wt%) and HHV 
of biochar.   
2h,17.2MPa 2h,27.6MPa 4h,17.2MPa 4h,27.6MPa
0
10
20
30
40
50
60
70
Yield
(
%)
Conditions
 
230?C
200?C 
 
Figure 6.3 Yield (wt%) of obtained biochar at 200?C and 230?C. 
2h,17.2MPa 2h,27.6MPa 4h,17.2MPa 4h,27.6MPa
0
5
10
15
20
25
30
230?C
HHV(k
J/g)
Conditions
 
 
200?C 
 
Figure 6.4 HHV of obtained biochar at 200?C and 230?C. 
 The results of runs 2 and 4 (Table 6.3) showed that at constant pressure and 
temperature, the yield of biochar decreased from 66.7 to 60.6 wt% with residence time 
whereas heating value of biochar increased from 20.9 to 21.8 kJ/g. Increase in the 
temperature to 230?C produced higher HHV of biochar, the range of 24.3-26.9 kJ/kg (run 
5-8). As expected, the yield of biochar showed the descending trend with temperature.  
But even with 4 h of residence time at 230?C, the HHV of biochar was just comparable to 
lignin?s higher heating value (24-26 kJ/g). These results confirme that lignocellulosic 
187 
?
188 
?
structure of switchgrass was not decomposed at 200-230?C range and the increase in 
heating value may be mainly attributed to the removal of amorphous components 
(hemicelluloses and part of lignin) from switchgrass. It can be concluded that higher 
residence time though favored the carbonization process; the rate of carbonization was 
still slower due to the moderate temperature condition.  
 The set of experiments represented by run 9, 3, and 4 were conducted to 
investigate the effect of pressure in the range of 2.4-27.6 MPa at 200?C and at 4 h 
residence time. There is gradual decrease in biochar yield, from 66.4% (run 9) to 60.6 
wt% (run 4). The heating value of biochar increased marginally with pressure from 20.3 
kJ/g (run 9) to 21.8 kJ/g (run 4). It is important to note that even after more than ten fold 
rise in pressure, the increase in heating value was less than 10% when comparing run 9 to 
run 4.  Even at 230?C (run 10), the HHV of biochar obtained under autogeneous pressure 
was 23.5 kJ/g which is slightly lower than the data under controlled pressure.  
 The increased pressure of the reaction medium keeps water in liquid phase and 
influences mainly the density dependent physical properties of water.
20
 At 200?C, the 
density of water which is 0.86 g/ml at 2.4 MPa increases slightly with pressure to 0.88 
g/ml at 27.6 MPa (source: National Institute of Standards and Technology). Therefore, 
effect of pressure in this temperature range is not pronounced except it helps in 
improving the contact between biomass components and water molecules.  
Switchgrass contains more than 40 wt% oxygen. The major objective of HTC is 
the deoxygenation of biomass. The removal of oxygen can be achieved most readily by 
dehydration reaction, which removes oxygen in the form of water, and by 
decarboxylation, which removes oxygen in the form of carbon dioxide.
21
 Hence, higher 
reaction temperature should favor HTC more than the increased residence time or 
pressure.  
Effect of temperature: To examine the effect of temperature on biochar production from 
switchgrass further experiments at the residence times 4.0 h (runs 9-12) were performed. 
The effects of temperature from 200 to 280?C were investigated and the results are shown 
in the Figures 6.5 and 6.6.  
?
200 220 240 260 280
0
10
20
30
40
50
60
Temperature(
O
C)
Yi
el
d(
%
)
Run 9
Run 10
Run 11
Run 12
 
Figure 6.5 Yield (wt%) of obtained biochar at 200-280?C.   
 
 
200 220 240 260 280
0
5
10
15
20
25
30
HHV
(
kJ/g
)
Temperature(
O
C)
Run 9
Run 10
Run 11
Run 12
 
Figure 6.6 HHV of obtained biochar at 200-280?C. 
189 
?
 The reactor pressure was autogenous and was higher than the vapor pressure of 
water due to the formation of gaseous products during the reaction. As expected, the 
influence of temperature was noteworthy. With increase in temperature, biochar became 
richer in carbon content indicted by its higher heating value. For example, the HHV value 
reached to 28.1 kJ/g at 280?C (run 12), which is comparable to bituminous grade coal.  
Though biochar yield (wt%) decreased with temperature.   
To quantify the amount of energy being retained in solid biochar with respect to 
the initial energy input from switchgrass, energy conversion ratio (ECR) is defined as: 
sswitchgras of HHV  sswitchgras of massdry 
biochar of HHV biochar  of massdry 
?
?
=ECR  
The ECR for the runs 10-12 was 52.4%, 43.0%, and 31.2% at 230, 260, and 280?C, 
respectively. The results indicate that energy conversion efficiency decreases with the 
temperature. Hence, temperature can play an important role in optimizing the ECR and 
HHV of biochar. 
Figures 6.7 and 6.8 show the biochar yield and HHV with residence time at 
280?C under autogeneous pressure (runs 12-14).  
0.5h 1h 4h
0
5
10
15
20
Yield
(
%)
Time(h)
Run 12Run 14
Run 13
 
190 
?
Figure 6.7 Yield (wt%) of obtained biochar at 280?C. 
0.5h 1h 4h
0
5
10
15
20
25
30
HH
V
(
kJ
/
g
)
Time(h)
Run 12Run 14
Run 13
   
Figure 6.8 HHV of obtained biochar at 280?C. 
 
Interestingly, the yield of biochar and HHV for run 12 and 14 did not changed 
appreciably after increasing the residence time from 1 to 4 h. The biochar yield for run 13 
was higher but the HHV was unexpectedly lower when compared to run 12 or 14. 
Switchgrass yielded less biochar during run 14 (1 h and 280?C) than during run 11 (4 h 
and 260?C), but the HHVs for obtained biochar were 28.4 and 25.9 kJ/g, respectively. 
These results show that the temperature plays a key role for the conversion of switchgrass 
to biochar. The higher temperature favors the dehydration of polymeric components of 
biomass including cellulose. In fact, dehydration reactions start competing with the 
hydrolysis reactions; even though hydrothermal media has excess water in the reactor. 
22
 
Biochar yield can be further optimized by increasing the selectivity towards 
dehydration and decarboxylation reactions. The hydrolysis needs to be supressed which 
causes the loss of organic carbon as water soluble compounds in liquid product 
(biocrude). The second part of this study discusses about the process to recover these 
191 
?
water soluble compounds as biotar particle again by conducting reactions in 
hydrothermal medium.  
Biochar characterization: The switchgrass and biochar are characterized by FTIR in the 
near IR region (wave number: 4000-400 cm
-1
). Typical FTIR spectra of switchgrass and 
biochar are shown in Figure 6.9 and the various band assignments are listed in Table 6.4. 
The FTIR spectrum of biochar was somewhat similar to that of switchgrass and the peaks 
of lignin were remained in the biochar. These results indicate that the biochar has lingnin-
like molecular structure. The appearance of absorption peaks at 1,060 and 1,160 cm
-1 
in 
the samples produced at 200?C (run 9) and 230?C (run 8) confirmed that switchgrass was 
not decomposed completely under these reaction conditions.  
4000 3000 2000 1000 0
E
W (cm
-1
)
A
B
C
D
Run 12
Run 11
Run 08
Run 09
Switchgrass
 
Figure 6.9 FTIR spectra of switchgrass and biochar, reaction time 4 h: 
Switchgrass (A), biochar from HTC at 200?C (B), biochar from HTC at 230?C (C), biochar 
from HTC at 260?C (D), biochar from HTC at 280?C (E). 
 
 
 
192 
?
193 
?
Table 6.4 FTIR spectra of the switchgrass and biochar. 
Position of the bands (cm
-1
) Functional group
3401 O?H stretching 
2920 CH, CH
2
 stretching 
1700 Carbonyl C=O stretching (polyxylose) 
1600 C=C stretching
1513 Benzene ring stretching (lignin) 
1424 CH
2
 shearing (cellulose), CH2 benching (lignin) 
1383 CH benching (cellulose, semi-cellulose) 
1267 C?O?C stretching in alkyl aromatic (lignin) 
1160 C?O?C asymmetry stretching (cellulose, semi-cellulose) 
1060 C?O stretching (cellulose and semi-cellulose) 
 
 The spectra of switchgrass and the samples from runs 8 and 9 were similar as 
confirmed the presence of absorption peaks attributed to the hemicelluloses and cellulose 
components of switchgrass. As for 260?C (run 11) and 280?C (run 12), the FTIR spectra 
have peaks for the biochar which are different than switchgrass or lower temperature 
biochar. This difference suggests that the aromatic ring started to recombine at the higher 
temperatures. The possible presence of alkenes was indicated by the absorbance peaks 
between 1,680 and 1,580 cm
-1
. The ether linkages around 1,200 cm
-1
 and 1,000 cm
-1
 
between the cellulose skeleton units were hydrolyzed for the biochar samples produced at 
260?C to 280?C.
23-26
 It is noted that cellulose, which has four OH groups in its unit 
structure, exhibited a higher hydrolysis conversion to the aqueous sample due to its facile 
solvation and decoagulation with sub-critical water. Macropores located within the 
cellulose structure are readily altered during cross-linking, whereas the associated 
mesopores are much more stable. The particle size and porosity of the substrates should 
be taken into account for the comparison of their hydrolysis reactivity, although they may 
be relatively minor factors compared to their different chemical structures and intra- and 
intermolecular interactions through the functional groups.
24
 More detailed information 
and experimental data on such physical factors are necessary for the confirmation. As the 
skeleton structure of switchgrass, the lignin was remained in the biochar when the 
treatment temperature increased from 200?C to 280?C. The results were confirmed by the 
lignin peaks 1265, 1424 and 1513cm
-1
 in the spectra. At the same time, 1600 and 1700 
cm
-1
, the peaks of polymeric product are also present.  The FTIR results prove that the 
biochar was composed of lignin and polymeric products produced during HTC.  
 The SEM images (Figure 6.10) suggest that biochar has smaller particle size and 
different microstructure than switchgrass.  
A B
 
Figure 6.10 SEM images of switchgrass (A), and biochar (B) from run 11. 
 
 The images show that the lignocellulosic structure of switchgrass was destroyed 
in HTC process.
2, 27
 Compared to lignin, the microcrystal structure of cellulose is 
relatively easier to change at higher temperatures as also demonstrated by FTIR analysis.  
 The XRD patterns for biochar from run 12 (280?C, 4.0 h) and switchgrass are 
compared in Figure 6.11.  
194 
?
?
10 15 20 25 30 35
1000
2000
3000
4000
5000
6000
7000
8000
9000
Switchgrass
Biochar
Inte
nsity(CPS)
(Run 12)
Angle (2?)
 
Figure 6.11 XRD images of untreated switchgrass and biochar obtained at 280?C, 4 h. 
 The distinctly sharp crystalline cellulosic peak (at 2? = 22.7?) due to the crystal 
structure of cellulose is present in switchgrass, but is absent in biochar.
28
 The XRD 
pattern of biochar from run 12 confirmed that it contained mainly amorphous components. 
Hence, cellulose microcrystalline feature converted to non-crystal structure during HTC 
process. The element analysis of the biochar (run 12) indicate that its carbon content was 
over 70 wt%. The results of HHV for different samples listed in Table 3.3 conclude that 
O/C ratio starts decreasing with the treatment temperature from about 0.90 for the 
untreated switchgrass to 0.32 for hydrothermally produced biochar at 280?C, due to 
deoxygenation. With the breakdown of lignocellulosic structure and removal of some 
oxygen from the structure, the moisture retention capacity of the product biochar is 
significantly reduced. Qualitatively, it was observed that biochar was a fluffy, less 
hydrophilic, and high energy density powder, which can be easily compacted.  
HTC of biocrude to produce biotar particles: Typically, the TOC value of biocrude 
ranged between 1-1.5 wt%. The main sugar and organic acid in biocrude obtained at 
195 
?
different temperature are listed in Table 6.5, with the main hydrolysis product being 
glucose, acetic acid, lactic acid, HMF and furan, etc.  
Table 6.5 The main sugar and acid in biocrude. 
Composition 200?C 230?C 260?C 280?C 
Glucose (g/L) 2.71 1.18 0.36 0.28
Lactic acid (g/L) 3.21 3.52 3.56 2.67 
Acetic acid (g/L) - 0.64 1.10 0.88 
HMF (g/L) - - - 0.66 
Furan (g/L) - - - 1.88 
 
 With the increase in temperature, the amount of glucose decreases whereas 
amounts of organic acids increase. It is well known that a low temperature is favorable 
for the production of sugars from the hydrolysis of hemicelluloses and cellulose, and 
these sugars further transform to organic acids at higher temperatures.  
  Biocrude obtained from the HTC of switchgrass at 200-280?C were again 
subjected to the hydrothermal conditions at 200?C for 2 h in a batch reactor. The 
dissolved organic carbon started precipitating as biotar particles. The conversion 
efficiency of carbon from biocrude to biotar particles based on TOC value was shown in 
the Figure 6.12.  
180 200 220 240 260 280 300
0
5
10
15
20
25
30
35
40
Temperature(
O
C)
Con
v
e
r
si
on f
o
r carbon (
%
)
 
196 
?
Figure 6.12 Conversion efficiency of carbon from biocrude. 
 
 As for the biocrude obtained at 200?C, the conversion efficiency of carbon was 
35.5%. The value decreases to mere 4.4% when the biocrude obtained at 280?C from run 
12 was used. The conversion efficiency of carbon for biocrude obtained from 230?C and 
260?C were 26.2% and 6.5%, respectively. The experimental data indicated that biocrude 
produced at lower temperature is favorable for the transformation of carbon to biotar 
particles due to the presence sugars. In the reaction condition, the sugars are expected to 
partially convert to ring compound (furans) which further precipitate through 
polycondensation process as biotar particles. The SEM images of biotar particles 
obtained at 200?C are shown in Figure 6.13. The biotar particles are globules of about 2 
?m in size.  
A B
?
Figure 6.13 SEM images of biotar particles obtained at 200?C, 2 h. 
 
 Comparing biochar from switchgrass to biotar from biocrude processes, it is 
obvious different formation mode can provide various product properties. The biochar 
can potentially be used as high energy density solid fuel. On the other hand, due to the 
homogeneous characteristics, biotar particles can be utilized for producing carbonaceous 
functional materials. 
197 
?
198 
?
6.6 Conclusion 
 Switchgrass was converted to biochar by hydrothermal carbonization at 200-
280?C. Compared with the reaction time and pressure, the temperature plays an important 
role in the conversion.  With increase in temperature, biochar yield decreases but the 
heating value increases.  Bases on infrared spectroscopy and electron microscopy of 
biochar, complete breakdown of lignocellulosic crosslink structure of switchgrass was 
realized at higher temperatures and biochar contained mainly lignin fractions. Based on 
x-ray diffraction, microcrystalline component is converted to non-crystalline components.  
The HHV of biochar produced at 280?C in just 1 h of HTC was comparable to the 
bituminous grade coal. 
 Biotar microparticles can be produced from HTC of biocrude. Biocrude produced 
at a lower temperature is favorable for the biotar production where sugar compounds 
present in the biocrude undergoes polycondensation and dehydration processes resulting 
in carbon rich microspheres. The biotar microparticles are globular and uniform in 
particle size.? 
6.7 Acknowledgments 
 Author is thankful for the financial supports from the National Science 
Foundation (grant NSF-CBET-0828269) and Alabama Center for Paper and Bioresource 
Engineering.   
 
 
 
 
199 
?
6.8 References 
1. Matsumura, Y.; Sasaki, M.; Okuda, K.; Takami, S.; Ohara, S.; Umetsu, M.; 
Adschiri, T., Supercritical water treatment of biomass for energy and material recovery. 
Combust. Sci. and Tech. 2006, 178, 509-536. 
2. Kumar, S.; Gupta, R. B., Biocrude Production from Switchgrass using Subcritical 
Water. Energy & Fuels 2009, 23, (10), 5151-5159. 
3. Lee, I.; Kim, M.; Ihm, S., Gasification of glucose in supercritical water. Ind. Eng. 
Chem. Res. 2002, 41, 1182-1188. 
4. Motonobu, G.; Ryusaku, O.; Tsutomu, H., Hydrothermal conversion of municipal 
organic waste into resources. Bioresource Technology 2004, 93, 279-284. 
5. Kishida, H.; Jin, F.; Zhou, Z., Conversion of glycerin into lactic acid by alkaline 
hydrothermal reaction. Chemistry Letters 2005, 34, (11), 1560-1561. 
6. Kong, L. Z.; Li, G. M.; Wang, H., Hydrothermal catalytic conversion of biomass 
for lactic acid production. Journal of Chemical Technology & Biotechnology 2008, 83, 
383-388. 
7. Lourdes, C.; David, V., Formation of organic acids during the hydrolysis and 
oxidation of several wastes in sub- and supercritical water. Ind. Eng. Chem. Res. 2002, 41, 
6503-6509. 
8. Selhan, K.; Thallada, B.; M. Akinori  Low-temperature catalytic hydrothermal 
treatment of wood biomass: analysis of liquid products. Chemical Engineering Journal 
2005, 108, 127-13. 
200 
?
9. Abdullah, H.; Wu, H. W., Biochar as a fuel: 1. Properties and grindability of 
biochars produced from the pyrolysis of mallee wood under slow-heating conditions. 
Energy & Fuels 2009, 23, (8), 4174-4181. 
10. Titirici, M.-M.; Thomas, A.; Antonietti, M., Back in the black: hydrothermal 
carbonization of plant material as an efficient chemical process to treat the CO
2
 problem. 
New J. Chem. 2007, 31, 787-789. 
11. Hu, B.; Yu, S.-H.; Wang, K.; Liu, L.; Xu, X.-W., Functional carboneceous 
materials from hydrothermal carbonization of biomass: an effective chemical process. 
Dalton Transactions 2008. 
12. Phillip, E., Organic chemical reactions in supercritical water. Chemical Reviews 
1999, 99, 603-621. 
13. Masaru, W.; Takafumi, S.; Hiroshi, I., Chemical reactions of C1 compounds in 
near-critical and supercritical water. Chemical Reviews 2004, 104, 5803-5821. 
14. Savovaa, D.; Apakb, E.; Ekinci, E.; Yardimb, F.; Petrov, N.; Budinovaa, T.; 
Razvigorovaa, M.; Minkovaa, V., Biomass conversion to carbon adsorbents and gas. 
Biomass and Bioenergy 2001, 21, (2), 133-142. 
15. Bergius, F.; Specht, H., Die Anwendung hoher Drucke bei chemischen 
Vorg?ngen und eine Nachbildung des Entstehungsprozesses der Steinkohle.Verlag 
Wilhelm Knapp. Halle an der Saale 1913, 58. 
16. Marta, S.; Antonio, B. F., The production of carbon materials by hydrothermal 
carbonization of cellulose. Carbon 2009, 47, 2281-2289. 
201 
?
17. Titirici, M.-M.; Antonietti, M.; Baccile, N., Hydrothermal carbon from biomass: a 
comparision of the local structure from poly- to monosaccharides and pentoses/hexoses. 
Green Chem. 2008, 10, 1204-1212. 
18. Varhegyi, G.; Szabo, P.; Till, F.; Zelei, B.; Antal, M. J. J.; Dai, X., TG, TG-MS, 
and FTIR Characterization of high-yield biomass charcoals. Energy & Fuels 1998, 12, 
969-974. 
19. Cheng, L.; Ye, X. P.; He, R.; Liu, S., Investigation of rapid conversion of 
switchgrass in subcritical water. Fuel Processing Technology 2009, 90, 301-311. 
20. Rezaei, K.; Temelli, F.; Jenab, E., Effects of pressure and temperature on 
enzymatic reactions in supercritical fluids. Biotechnology Advances 2007, 25, 272-280. 
21. Peterson, A. A.; Vogel, F.; Lachance, R. P.; Froling, M.; Michael Jerry Antal, J.; 
Tester, J. W., Thermochemical biofuel production in hydrothermal media:A review of 
sub- and supercritical water technologies. Energy and Environmental Science 2008, 1, 
32-65. 
22. Patrick, H. R.; Griffith, K.; Liotta, C. L.; Eckert, C. A., Near critical water: A 
benign medium for catalytic reactions. Ind. Eng. Chem. Res. 2001, 40, 6063-6067. 
23. Sakanishi, K.; Ikeyama, N.; Sakaki, T., Comparison of the hydrothermal 
decomposition reactivities of chitin and cellulose. Ind. Eng. Chem. Res. 1999, 38, 2177-
2181. 
24. Kong, L. Z.; Li, G. M.; He, W., Reutilization disposal of sawdust and maize straw 
by hydrothermal reaction. Energy Sources, Part A 2009, 31, 876-887. 
202 
?
25. Biagini, E.; Barontini, F.; Tognotti, L., Devolatilization of biomass fuels and 
biomass components studied by TG/FTIR technique. Ind. Eng. Chem. Res. 2006, 45, 
4486-4493. 
26. Qian, Y. J.; Zuo, C. J.; Tan, J., Structural analysis of bio-oils from sub-and 
supercritical water liquefaction of woody biomass. Energy & Fuels 2007, 32, (3), 196-
202. 
27. Kobayashi, N.; Okada, N.; Hirakawa, A.; Sato, T.; J.Kobayashi; Hatano, S.; Itaya, 
Y.; Mori, S., Characteristics of solid residues obtained from hot-compressed-water 
treatment of woody biomass. Industrial and Engineering Chemistry Research 2009, 48, 
373-379. 
28. Segal, L.; Creely, J. J.; Martin, A. E.; Conrad, C. M., An empirical method for 
estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. 
Textile Res. J. 1959, 29, 786-794. 
 
 
203 
?
7. Application of Biochar for the Removal of Heavy Metal Contaminants from 
Ground Water  
 
7.1 Introduction 
 Besides the use of biochar as solid fuels, this carbonaceous material can be 
potentially used for the removal of heavy metals contamination from ground water.
1-3
 
Nano-porous structure of biochar with available oxygen functional groups on the surface 
provides an excellent opportunity to adsorb heavy metal ions from the aqueous solution.
4-
6
 The cost of the adsorbent is the most important criteria in its large scale application. 
Abundantly available lignocellulosic biomass is a promising raw material for the 
production of activated carbons considering their low cost. But the main disadvantage of 
activated carbon is its relatively high production cost due to the high expense of energy 
used during its production.
7
  As a traditional method for biochar production, pyrolysis 
processes typically use dry biomass sources and temperature range from 500 to 800?C.  
So the higher energy consumption is needed in the pyrolysis process for biochar 
production. Considering the cost factor, biochar produced by subcritical water 
carbonization technology, which is relatively less energy intensive process is viewed as 
the potentially low-cost process for obtaining organic adsorbents. The pyrogenic organic 
matters (eg. biochar, charcoal, soot and activated carbon) have been studied extensively 
for their high affinity and capacity of absorbing organic compounds.
8
  
204 
?
 Raw biomass such as peat, poplar sawdust, coconut shells have been reported to 
show affinity towards heavy metal ions even at low ion concentration.
9, 10
 However use of 
raw biomass as adsorbent is associated with the risk of leaching of the organic 
pollutants.
3
 Raw biomass is also biodegradable. On the other hand, biochar as sorbent is 
useful in avoiding such risks. The earlier studies have been reported on the use of biochar 
for removing the metal contaminants such as lead, mercury, and arsenics from aqueous 
solutions.
11-13
 Among the several metal contaminants of concern, uranium is considered 
as top priority metal contaminant by the U.S. Department of Energy. Ground water 
contamination caused by the radionuclides, in particular uranium is a serious 
environmental problem worldwide.
14
  Uranium, a radiotoxic also possesses the risk 
related to chemical toxicity. Among the various oxidation states of uranium, +4 and +6 
oxidation states are the most important states in geological environments.  The chemical 
toxicity of uranium is predominantly caused by the highly reactive hexavalent uranyl 
ions.  
In the early 1940?s the naturally occurring radioactive element, uranium (U), was 
heavily used in the nuclear weapons program and energy production. Ever since, the wide 
use of uranium has left a legacy of contaminants in the subsurface environment especially 
at most of the U.S. Department of Energy sites. Considering some of the U contaminated 
sites adjoin river boundaries, the subsurface mobility of uranium poses a huge concern to 
both humans and the environment. The common remediation strategies used to address 
U(VI) contamination are precipitation of uranium to its least soluble minerals, 
abiotic/biotic reduction of U(VI) to its less soluble form U(IV), sorption of U(VI) to 
strongly adsorbing agents (zero valent iron, fish bone etc.).  
205 
?
In this study, we focus on the adsorption of uranium on to biochar prepared from 
switch grass in subcritical water. Although the scientific literature is replete with studies 
focusing on the adsorption of contaminants on to biochar obtained from agricultural 
products, the application of biochar to U(VI) adsorption is not studied. Moreover, the 
effective use of wastes from biorefinery processes for contaminant remediation would 
enhance the goal of the green process. The batch adsorption experiments were conducted 
where uranium [U(VI)] was loaded as a solute at both the natural pH of biochar (~4.0) 
and at a higher pH (~5.7). 
7.2 Experimental Section 
Biochar Preparation: Auburn grown switchgrtass (chopped to 5-10 mm long size) was 
used for producing biochar. De-ionized (D.I.) water was used for conducting the 
experiments. Conversion of switchgrass to biochar was conducted in a high pressure 
batch type reactor. The set up consists of a 500 ml high temperature, high pressure metal 
reactor equipped with a state of art control system comprising Proportional-Integral-
Differential (PID) controllers. Switchgrass (50 g) and 350 ml of water were charged into 
the reaction vessel and the reactor temperature was raised to 300?C with a heating rate of 
7?C/min. After maintaining the temperature at 300?C for 30 min reactor under 
autogeneous pressure condition, reactor was rapidly cooled to ambient condition using 
water through cooling coil. The product biochar is separated from the liquid after reaction 
and further washed with D.I. water. The biochar was air dried and used for the further 
characterization.  
Batch adsorption kinetic experiments: Batch kinetic experiments were performed at two 
different  solid to solution ratio (SSR: 4 and 5 g L
-1
) and initial U(VI) concentration (ca. 
206 
?
30 and  10 mg L
-1
) to estimate the equilibrium time for the adsorption reaction. The 
experiments were performed at the natural pH of biochar (ca. pH 3.9 ? 0.2) with 0.1 M 
NaNO
3
 as the background solution. The experiments were carried out for 72 hours and 
samples were taken at regular intervals by sacrificing the tubes. U(VI) was measured 
using kinetic phosphorescence analyzer (KPA, Chemchek Instruments, Inc.) after 
filtration of the samples through 0.45 ?m PTFE filer units and acidifying the samples to 
pH 1.0. The difference between the initial U(VI) concentration and aqueous U(VI) 
concentration was attributed to the adsorption of U(VI) onto biochar. Both the 
experiments had duplicate test tubes for each sampling interval, blank [no U(VI)], and 
controls (no biochar).  
Batch adsorption equilibrium experiment: Batch isotherm experiment was performed at a 
solid loading of 5g L
-1 
and at a pH of about 3.9 ? 0.2. The initial concentration of U(VI) 
was varied between 5 to 30 mg L
-1
 and the experiment was conducted in a 0.1 M NaNO
3
 
background solution. Moreover, to study the effect of solid to solution ratio on 
adsorption, duplicate tubes were set up at 1 g of biochar L
-1
 containing an initial U(VI) 
concentration 10 mg L
-1
.  The test tubes were prepared in duplicates and tumbled for 
about 34 hours at 25 ?C.  At the end of the equilibrium time, aqueous samples were 
filtered and acidified and analyzed. Similar to the kinetic experiments, the difference 
between the initial U(VI) concentration and aqueous U(VI) concentration was attributed 
to the adsorption of U(VI) onto biochar. 
7.3 Results and Discussion 
 Table 7.1 compares the elemental composition and ash content in raw switchgrass 
and the biochar. It was observed that elemental carbon in biochar increased to 70.46% as 
a result of subcritical water carbonization. The oxygen content in raw switchgrass is 
typically 40-43%.
15
 As result of carbonization process in subcritical water more than 70% 
of deoxygenation could be achieved. 
Table 7.1 Composition of switchgrass and biochar. 
Table 1: Composition of switchgrass and biochar  
Elements C Ash Ca K Mg P Fe Mn Na Pb 
 (wt%) (ppm) 
Switchgrass 44.61 4.5 2105 4082 4514 941 115 48 701 1 
Biochar 70.46 3.7 2029 665 2215 481 258 53 395 <0.1 
?
 
  The biochar yield and higher heating value (HHV) of biochar produced from 
switchgrass was 48.2 wt% and 27.5 MJ/kg on oven dry basis, respectively. The effects of 
temperature, pressure and residence time on the yield and HHV of biochar have been 
studied separately (Chapter 6). The biochar had specific surface area of 18 m
2
/g, as 
determined by nitrogen physisorption using five point Brunauer-Emmett-Teller (BET) 
methods at 77 K using Autosorb-1 instrument (Quantachrome).   
 Figures 7.1(a,b) show the results of thermogravimetric (TG) and derivative 
thermogravimetric (DTG) analyses conducted under helium gas flow to understand the 
thermal stability of materials upto 700?C. TG was conducted using a TGAQ 5000 
instrument under high purity (99.99%) helium gas flow. Samples were placed in a sample 
pan and heated from room temperature to 700?C with a heating rate of 10?C/min. The 
TGA curves were obtained directly from the apparatus while the derivative 
thermogravimetric (DTA) curves were obtained by the software Universal V4.3A.  
207 
?
 Weight loss in switchgrass was observed over a wide range of temperature upto 
700?C, but majority of weight loss (70%) happened in the range of 200 to 400?C (Figure 
7.1a). The rapid weight loss below 400?C is mainly due to the decomposition of 
holocellulose fractions of the switchgrass.
16
 On contrary, the weight loss in biochar is 
relatively gradual upto 700?C. In fact, nearly 45% of biochar remains even at 700?C 
which confirmed the thermal stability of biochar relative to the switchgrass. 
Switchgra ss
 
Figure 7.1a TG and DTG pattern of switchgrass under inert (helium) atmosphere. 
Biochar
 
Figure 7.1b TG and DTG pattern of biochar under inert (helium) atmosphere. 
208 
?
 The kinetic experiments indicate about 90% of initial U(VI) adsorption resulting 
within 8 hours, at both the SSR (Figure 7.2). Based on the reaction kinetics, about 34 
hours of equilibrium time was chosen for the batch isotherm experiments.  
0
5
10
15
20
25
30
35
0 1020304050607080
A
que
ou
s
 U
 
(m
g L
-1
)
Time (h)
SSR: 4 g/L
SSR: 5 g/L
 
Figure 7.2 Kinetics of U(VI) adsorption on to biochar at different solid loadings and 
initial U(VI) concentrations. Circles represent a SSR of 5 g L
-1
 and an initial U(VI). 
 
 The batch adsorption isotherm is shown in Figure 7.3. The adsorption of U(VI) on 
to biochar followed a H-type isotherm.   
209 
?
0.0
0.5
1.0
1.5
2.0
2.5
0 5 10 15 20 25
A
ds
orbe
d
 U
 
(m
g
 g
-1
)
Aqueous U (mg L
-1
)
 
Figure 7.3 Adsorption isotherm of U(VI) on to biochar. SSR: 5 g L
-1
 pH: 3.9 ? 0.2; ionic 
strength: 0.1 M NaNO
3.
 
 This could have been due to the filling of biochar pores by the uranyl cations. 
Once the pore volume gets completely filled, the biochar ceases to trap more uranium 
resulting in a plateau at about 2 mg g
-1
. Furthermore, the isotherm indicates an adsorption 
capacity of ca. 0.2% (w/w). Figure 7.4 indicates the variation in adsorption with respect 
to solid loading. The distribution coefficient is about three times higher at a SSR of 1 g L
-
1
 when compared with 5 g L
-1
.   
210 
?
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
012345
K
d
(L
 g
-1
)
SSR (g biochar L
-1
)
6
 
Figure 7.4 Adsorption of U(VI) on to biochar at different SSR. pH: 3.9 ? 0.2; ionic 
strength: 0.1 M NaNO
3. 
 
7.4 Conclusion 
 The study clearly shows that the adsorption of uranium on to biochar could be 
considered as one of the possible alternatives to treat U(VI) contaminated water. 
Kinetically, the sorption process is fast and the extent of adsorption is fixed. Compared to 
other remediation strategies, feasibility of biochar as U adsorbent is supported by its 
environmental benign nature. Thus the study illustrates the proof of concept that biochar 
could serve as an effective adsorbent of uranium containing groundwater without causing 
environmental damages. But the practical application of biochar in the field needs further 
studies including desorption and flow through column experiments. 
 
 
211 
?
212 
?
7.5 References 
1. Bailey, S. E.; Olin, T. J.; Bricka, R. M.; Adrian, D. D., A review of potentially 
low-cost sorbents for heavy metals. Water Res. 1999, 33, 2469?2479. 
2. Dias, J. M.; Alvim-Ferraz, M. C. M.; Almeida, M. F.; Rivera-Utrilla, J.; Sanchez-
Polo, M., Wastematerials for activated carbon preparation and its use in aqueous-phase 
treatment: a review. J. Environ. Manage. 2007, 85, 833-846. 
3. Liua, Z.; Zhang, F.-S., Removal of lead from water using biochars prepared from 
hydrothermal liquefaction of biomass Journal of Hazardous Materials 2009, 167, (1-3), 
933-939. 
4. Tan, W. T.; Ooi, S. T.; Lee, C. K., Removal of Cr(VI) from solution by coconut 
husk, palm pressed fibers. Environ. Technol 1993, 14, 277?282. 
5. Sevilla, M.; Fuertes, A. B., The production of carbon materials by hydrothermal 
carbonization of cellulose. Carbon 2009, 47, 2281-2289. 
6. Hu, B.; Yu, S.-H.; Wang, K.; Liu, L.; Xu, X.-W., Functional carboneceous 
materials from hydrothermal carbonization of biomass: an effective chemical process. 
Dalton Transactions 2008. 
7. Savovaa, D.; Apakb, E.; Ekinci, E.; Yardimb, F.; Petrov, N.; Budinovaa, T.; 
Razvigorovaa, M.; Minkovaa, V., Biomass conversion to carbon adsorbents and gas. 
Biomass and Bioenergy 2001, 21, (2), 133-142. 
8. Smernik, R. J., Biochar and sorption of organic compounds. In Biochar for 
Environmental Management: Science and Technology, Lehmann, J.; Joseph, A., Eds. 
Earthscan: London, 2009; pp 289-300. 
213 
?
9. McKay, J. F. P. G., Equilibrium parameters for the sorption of copper, cadmium 
and zinc ions onto peat. J. Chem. Technol. Biotechnol. 1997, 69, 309-320. 
10. Sciban, M.; Radetic, B.; Kevresan, Z.; Klasnja, M., Adsorption of heavy metals 
from electroplating wastewater by wood sawdust. Bioresour. Technol. 2007, 98, 402-409. 
11. Budinova, T.; Ekinci, E.; Yardim, F.; Grimm, A.; Bjornbom, E.; Minkova, V.; M. 
Goranova, Characterization and application of activated carbon produced by H3PO4 and 
water vapor activation. Fuel Process. Technol. 2006, 87, 899?905. 
12. Kalderis, D.; Koutoulakis, D.; Paraskeva, P.; Diamadopoulos, E.; Otal, E.; J.O. 
Valle d.; Fernandez-Pereira, C., Adsorption of polluting substances on activated carbons 
prepared from rice husk and sugarcane bagasse. Chem. Eng. J. 2008, 144, (42-50). 
13. Amuda, O. S.; Giwa, A. A.; I.A. Bello, Removal of heavy metal from industrial 
wastewater using modified activated coconut shell carbon. Biochem. Eng. J. 2007, 36, 
174-181. 
14. Srivastava, S.; Bhainsa, K. C.; D?Souza, S. F., Investigation of uranium 
accumulation potential and biochemical responses of an aquatic weed Hydrilla verticillata 
(L.f.) Royle. Bioresour. Technol. 2010, 101, 2573-2579. 
15. Berg, D. V. D.; Visser, P. D. Final Report FAIR 5-CT97-3701 ?Switchgrass?. 
BTG (Biomass Technology Group), Enschede, The Netherlands: pp 50-52. 
16. Chen, B.; Chen, Z., Sorption of naphthalene and 1-naphthol by biochars of orange 
peels with different pyrolytic temperatures. Chemosphere 2009, 76, 127-133. 
?
214 
 
8. Summary and Future Works 
 
 The dissertation has successfully demonstrated that hydrothermal technology can 
play an important role in developing future biorefineries. The inherently high moisture 
content (considered as a major drawback) of biomass can be used advantageously to 
produce commercially viable biofuels without prior drying or using toxic/expensive 
chemicals. The dissertation encompasses and concludes the important finding in the area 
of pretreatment for bioethanol, liquefaction of biomass for biocrude and hydrothermal 
carbonization process for biochar production.  
 Lignocellulosic biomass can be efficiently pretreated in subcritical water in a 
short residence time. The process is not dependent on the type of biomass. Further, the 
research work has also established an alternative pathway for the utilization of liquid 
hydrolyzate generated during the pretreatment process.  In other studies, cellulose and 
switchgrass were liquefied in subcritical water to water-soluble products. In fact, nearly 
70 wt% yield of sugar compounds from cellulose liquefaction in subcritical water was 
achieved. Similarly, switchgrass was converted to biocrude and high energy density 
biochar in subcritical water. Fundamental study on the conversion of cellulose in sub- and 
super-critical water was completed. The use of potassium carbonate helped in catalyzing 
the depolymerization/liquefaction process. Further, switchgrass was converted a high 
heating value (coal-like) biochar for solid fuels applications.   
215 
 
 The nano-porous structure of biochar available with oxygen functional groups on 
the surface provided an excellent opportunity to develop this material as an adsorbent for 
heavy metal contaminants in ground water. The present research has opened up several 
avenues for continuing the fundamental and development research works on biofuels. 
Following research area may be further explored to support the future biorefineries. 
i. Pretreatment of Biomass by Transient Heating in Hydrothermal Medium: 
Pretreatment is an important and first step in the bioconversion process of biomass to 
ethanol. Typically it accounts for up to 40% of the total processing cost. Conventional 
pretreatment methods employ variety of chemicals for fractionating lignin and/or 
hemicelluloses to improve the enzymatic reactivity of biomass. Majority of these 
processes limit the exposure of biomass to below 220?C, which may not be sufficient 
to cause substantial structural changes in cellulose. In the present study on pure 
cellulose, it was observed that cellulose depolymerizes rapidly above 300?C and also 
undergoes changes in hydrogen bonding arrangements. Therefore, rapid heating and 
quenching (i.e. transient heating) of switchgrass in subcritical water above 300?C is 
expected to cause some structural changes because of re-orientation / changes in 
hydrogen bonding patterns. The very short time (order of few seconds) exposure of 
switchgrass in hydrothermal medium will also help in minimizing the generation of 
inhibitory compound. The physical alteration in cellulose structure caused by 
exposure to high temperature is expected to increase its enzymatic reactivity by 
manifolds. 
ii. Upgradation of biocrude: Rapid depolymerization of biomass components in 
subcritical water results in the formation of oxygenated hydrocarbons (biocrude). The 
216 
 
liquid mixture, which consists of cyclic molecules (e.g., glucose, furans, and 
oligosaccharides) and aromatic cyclic molecules (e.g, phenols) can be upgraded to 
liquid fuel by aqueous phase or supercritical water reforming. The presently available 
aqueous phase reforming technology is mostly applicable to model compound such as 
glucose, HMF, furfural, etc. and employs costly noble metals catalysts. The 
development of an economical catalyst which can be used to upgrade the mixture of 
biomass derived liquid products will immensely help in supporting the commercial 
development of petroleum like fuels from biomass. 
iii. Biochar related research: Hydrothermally produced biochar is a high energy density 
(coal-like) powder which has a potential to be compatible with coal infrastructure. 
The oxygen content in biochar was typically 22-23 wt% which is much lower 
compared to original biomass (> 40 wt%). Therefore, pyrolysis products from such 
biochar are expected to have a higher heating value compared to the conventional 
pyrolysis oil. Due to the high moisture content and the low surface area of untreated 
biomass, conventional pyrolysis or gasification processes require drying and size 
reduction of biomass prior to processing. On the other hand, hydrothermally produced 
biochar can be potentially developed as a feedstock for pyrolysis/gasification process 
due to almost uniform C, H, and O composition, fine particle size and reduced 
moisture retention capacity (i.e. hydrophobicity). The process may substantially 
eliminate the current costs associated with biomass drying and size reduction.  
 In view of biochar?s sorption properties, it is recommended to continue research 
for employing it as an economical alternative for removing subsurface contaminants. 
217 
 
Appendix A: Structure and Composition of Lignocellulosic Biomass 
 
A.1 Introduction 
Lignocelluloses are derived from wood, grass, agricultural residues, forestry 
waste and municipal solid wastes. The major components of lignocellulosic material are 
cellulose, lignin, hemicelluloses and others (extractives and ash). Small percentage of 
extractives and ash are also present in lignicellulosic biomass. Cellulose and 
hemicelluloses are polymers based on different sugars, whereas lignin is an aromatic 
polymer mainly built of phenylpropanoid precursors. Composition of these polymers 
within single plant varies with age, and stage of growth.
1
 
A.2 Composition and Structure 
Cellulose: Discovered 150 years ago, cellulose is most abundant organic matter on the 
earth. It is the main structural constituent of plants and algal cell walls. Cellulose in wood 
is mixed with many polymers such as hemicelluloses and lignin. Fluffy fiber of cotton 
ball is the purest naturally occurring form of cellulose. It is an unbranched chain and 
homopolymer of ?-D-glucopyranose units liked together by (1?4)-glycosidic bonds with 
a repeating unit of C
6
H
10
O
5
 strung together by ?-glycosidic linkages. The ?-linkages in 
cellulose form long linear chains (called elemental fibril) that are highly stable and 
resistant to chemical attack because of the high degree of hydrogen bonding that can 
218 
 
occur between chains of cellulose. 
Hemicellulose and lignin cover microfibrils (which are formed by elemental 
fibrils). Hydrogen bonding between cellulose chains makes the polymers more rigid, 
inhibiting the flexing of the molecules that must occur in the hydrolytic breaking of 
glycosidic linkages. Hydrolysis can reduce cellulose to a cellobiose repeating unit, 
C
12
H
22
O
11
, and ultimately to glucose.
2-4
  
There are six known polymorphs of cellulose (I, II, III
1
, III
II
, IV
I
, and IV
II
) which 
can also interconvert. The interconversion of cellulose is shown in Figure A.1. Cellulose 
I, also termed as native cellulose, has parallel arrangement of chains and is the only 
polymorph that occurs naturally. Cellulose II is converted through mercerization or 
solubilization-regeneration of native or other celluloses. Cellulose II is 
thermodynamically more stable structure with a low energy crystalline arrangement 
having an antiparallel arrangement of the strands (two cellulose chains lie antiparallel to 
one another) and some inter-sheet hydrogen bonding. Cellulose III
I
 and III
II
 can be 
obtained from Cellulose I and II, respectively, by treatment with liquid ammonia or some 
amines, whereas polymorphs IV
I
 and IV
II
 can be obtained from heating cellulose III
I
 and 
III
II,
 respectively, to 206?C in glycerol.
2-5
 Two decades ago, it was reported that native 
cellulose (cellulose I) exists as a mixture of two crystalline forms I
?
 and
 
I
?, 
having triclinic 
and monoclinic unit cells, respectively.
6
 Cellulose I
?
 is thermodynamically less stable, as 
shown by its conversion to cellulose I
?
 by annealing at 260?C. In both crystalline forms, 
cellobiose is the repeating unit with a strong intra-chain H-bond from 3-OH to the 
preceding ring O5, whereas the inter-chain H-bonding and packing of the crystal are 
slightly different in the two forms.
7
 
Cellulose I Cellulose II
Regeneration
Mercerization
NH
3(l) NH
3(l)
-NH
3(g)
-NH
3(g)
Cellulose III
I
Cellulose III
II
Cellulose IV
I
Cellulose IV
II
 
Figure A.1 Inter-conversion of polymorphs of cellulose (redrawn from O?Sullivan).
3
 
 
The comparatively lower stability of I
?
 may provide the the site of initial reaction 
in the microfibril. Cellulose molecules have reducing end group, as chemical linkage 
between C
1
 carbon and the ring oxygen. It is a hemiacetal and allows ring to open to form 
an aldehyde. The other end-group is a secondary alcohol and commonly referred as non 
reducing end group. The structural unit of cellulose has three hydroxyl groups, one 
primary and two secondary. These groups undergo chemical modifications (e.g. 
easterifications and etherifications) during the reactions.  But the accessibility of the 
reactants is limited because of the high degree of crystallinity of the native cellulose.
8
 
Cellulases enzyme hydrolyze the ?-1,4-glycosidic linkages of cellulose. They are 
divided in two classes and referred as endoglucanases (endo-1,4-?-glucanases, EGs) and 
cellobiohydrolases (exo-1,4-?-glucanases, CBHs). EGs hydrolyze preferably the 
amorphous region of cellulose and releases new terminal ends. Whereas CBHs act on 
219 
 
220 
 
existing or EGs generated chain ends. Both enzymes can degrade amorphous cellulose, 
but only CBHs enzyme act efficiently on crystalline cellulose. Thus CBHs and EGs work 
synergistically in cellulose hydrolysis and releases cellobiose molecules. ?-glucosidases 
are required to further break down the cellobiose to two glucose molecule.
1
 
Hemicelluloses: Hemicelluloses are important carbohydrate fraction in plants made of 
mixed polysaccharides and stick to cellulose via hydrogen bonds, to create 
polysaccharide microfibrils.  Together these give a strong rigidity to plant cell walls but 
the added lignin improves this strength greatly.  The molecules are much smaller than the 
cellulose (degree of polymerization ? 10
2
) and are dominated by hydrogen bond from the 
3-OH of one sugar to the ring oxygen of preceding sugar. Hemicelluloses are a class of 
polymers of sugars including the six carbon sugars mannose, galactose, glucose and 4-O-
methyl-D-glucuronic acid and five carbon sugars xylose and arabinose.  Unlike cellulose, 
hemicelluloses possess side chains. The side groups include acetic acid, pentoses (?-D-
glucose, ?-D-mannose, ?-D-galactose), hexuronic acids (?-D-glucoronic acid, ?-D-4-O-
methylglucoronic acid, ?-D-galacturonic acid) and deoxyhexoses (?-L-rhamnose ?-L-
fructose) and are responsible for the solubility of hemicelluloses in water and / or alkali. 
These side groups stop hemicelluloses molecules to aggregate and make them more 
susceptible to chemical degradation than cellulose. Within the plant, hemicelluloses are 
bound to cellulose and lignin component by covalent & non-covalent bonds in the cell 
wall and are thus fixed in the fiber structure.
9, 10
 
Xylans, mannans (glucomannans) and galactans are the major categories of 
hemicelluloses. The xylans have backbone of ?-(1?4)-glycosidic linked xylose units and 
some of the xylose molecules have ?-(1?2)-bonded 4- O-methylglucoronic acids. The 
xyloses also contain acetyl groups. In softwoods, xylose molecules are connected to 
arabinose by ?-(1?3)-glycosidic bonds. Corncobs are highly concentrated in xylans. 
Hardwoods have mannose and glucose units liked by ?-(1?4)-glycosidic bonds. Acetyl 
and galactose groups are connected to glucose-mannose backbone structure in softwoods. 
A summary of hemicelluloses composition in hardwoods (deciduous trees) and in 
softwoods (coniferous trees) are given in Table A.1.  
Table A.1 Composition and degree of polymerization (DP) of hemicelluloses. 
9
 
 
Acetyl groups present in hardwood xylans and softwood galatoglucomannans are 
hydrolysable by acid at elevated temperatures. The acetic acid released provides the 
acidity to the reaction media. The presesnce of uronic acid groups reduces the hydrolysis 
rate of glycosisdic linkages appreciably.
9
 
221 
 
Hemicelluloses are biodegraded to monomeric sugars and acetic acid. Xylan is the 
main carbohydrate found in hemicelluloses and degradation requires the cooperative 
action of a variety of hydrolytic enzymes. Endo-1,4-?-xylanse generates olisaccharides 
from the cleavage of xylan and 1,4-?-xylosidase produces xylose from the xylan 
oligosachharides. Hemicelluloses biodegradation requires accessory enzymes such as 
xylane esterases, ferulic and p-coumeric esterases, ?-l-arabinofuranosidases, and ?-4-O-
methyl glucuronosidases to work synergistically on wood xylans and mannans.
1, 11
 
Lignin: Lignin is a complex aromatic polymer. It is synthesized by the generation of free 
radicals released during peroxidise-mediated dehydrogenation of three phenylpropionic 
alcohols: trans-p-coumaryl (p-hydroxyphenyl propanol), coniferyl alcohol (guaiacyl 
propanol), and sinapyl alcohol (syringyl propanol). The structure of ligin monomers are 
shown in Figure A.2. The polymerization results in a heterogeneous structure whose basic 
units are linked by C-C and aryl-ether linkages, with aryl-glycerol ?-aryl ether being the 
predominant structure. Lignin is an amorphous heteropolymer, non-water soluble and 
optically inactive.
1
 
OH
CH
2
OH
(a)
OH
CH
2
OH
(b)
COH
3
H
OH
CH
2
OH
(c)
COH
3COH
3
 
Figure A.2 Lignin monomers (a) trans-p-coumaryl alcohol, (b) coniferyl alcohol, and (c) 
sinapyl alcohol (redrawn from Bobleter).
9
 
 
The structural complexity of the lignin and its high molecular weight is the main 
reason why it is so hard to degrade by enzymes.  Lignin content in wood or 
lignocellulosic varies based on the species type, growing conditions, part of the plant 
tested and numerous other factors. Lignin mainly acts as adhesive or binder in wood that 
222 
 
223 
 
provides strength and structure to the cellular composites of the plant and protects against 
microorganism or chemical attack. It controls the fluid flow, acts as antioxidant by 
absorbing UV lights and stores energy. When lignin binds, it crosslinks with the regular 
structure of the microfibrils made of cellulose and hemicelluloses.
12, 13
 Isolated lignin 
shows maximum solubility in the solvents including dioxane, acetone, methyl cellosolve 
(ethylene glycol monomethyl ether), tetrahydrofuran, dimethyl formaldehyde (DMF) and 
dimethyl sulfoxide (DMSO). 
Carbon-carbon linkages are very resistant to chemical attack, and degradation of 
lignin is largely limited to the cleavages of ether units at ?- and ?-positions.  Functional 
groups of lignin follows 
14
: 
? Hydrolyzable ether linkages: ?-aryl, ?-aryl, and ?-alkyl ether linkages are the 
main hydrolysable ether units in lignin. Lignin may also contain some ?-ether 
linked l to carbohydrate, which are hydrolyzed at relatively lower rates. 
? Phenolic hydroxyl groups: It plays an important role in promoting alkali-catalyzed 
cleavages of interunitary ether linkages, oxidative degradation of lignin, and in 
lignin modification reactions.  
? Aliphatic hydroxyl groups: Two major aliphatic hydroxyl groups in lignin are 
located at the ?- and ?- positions of the side chains. Aliphatic hydroxyl group at ?-
position is a benzyl alcohol, which is very reactive and plays a dominant role in 
lignin reactions. 
? Uncondensed Units: Units at position C2, C3, C5, and C6 are free or substituted 
by methyl groups are defined as uncondensed units. Hardwood lignin which 
contains high syringical units, have high content of uncondensed units. 
224 
 
? Unsaturated groups: Lignin contains some unsaturated groups, mainly as 
coniferyl alcohol and coniferaldehyde end groups.  
? Easter group: Grass lignins contain significant amount of p-coumeric acid and 
ferulic acid moieties, which are mainly esterified. These functional groups are 
liable to mild alkali treatment and mainly present at the ?-position. 
? Methoxyl groups: These groups are relatively resistant to both acidic and alkaline 
hydrolysis. 
? Accessibility: Lignin have very high tendency to form hydrogen bonding like 
hemicelluloses and cellulose. 
Lignin empirical formulae are based on ratios of methoxy groups to 
phenylpropanoid groups (MeO:C
9
). The general empirical formula for lignin monomers 
is C
9
H
10
O
2
 (OCH
3
)n, where n is the ratio of MeO to C9 groups. Where no experimental 
ratios have been found, they are estimated as follows: 0.94 for softwoods; 1.18 for 
grasses; 1.4 for hardwoods. These are averages of the lignin ratios found in the literature. 
Paper products, which are produced primarily from softwoods, are estimated to have an 
MeO:C9 ratio of 0.94 (source: NREL).  
Extractives: Extractives are low to moderately high molecular weight compounds, which 
are soluble in water or organic solvents. They impart color, odor, and taste to the biomass. 
The composition of extractives varies widely based on the species and class of wood. 
Some of the important classes of extractives are Terpenes, triglycerides, fatty acids, and 
phenolic compounds.
15
 
Ash: Ash contains metallic ions of sodium, potassium, calcium, and the corresponding 
ions of carbonate, phosphate, silicate, sulfate, chloride etc. 
225 
 
A.3 References 
1. P?rez, J.; Mu?oz-Dorado, J.; Rubia, J. T. D. l.; Mart?nez, J., Biodegradation and 
biological treatments of cellulose, hemicellulose and lignin: an overview Int Microbiol 
2002, 5, 53-63. 
2. Sarko, A., Cellulose - how much do we know about its structure? Wood and 
Cellulosics: Industrial utilization, biotechnology, structure and properties 1987, 
(J.F.Kennedy ed.), 55-70. 
3. O'Sullivan, A. C., Cellulose: the structure slowly unravels. Cellulose 1997, 4, (3), 
173-207. 
4. Marchessault, R. H.; Sarko, A., X-Ray Structure of Polysaccharides. Advances in 
Carbohydrate Chemistry 1967, 22, 421-482. 
5. Sarko, A., What is the crystalline structure of cellulose ? Tappi 1978, 61, 59-61. 
6. Atalla, R. H.; Vanderhart, D. L., Native Cellulose: A Composite of Two Distinct 
Crystalline Forms. Science  1984, 223, (4633), 283 - 285. 
7. Sinnott, M. L., Chapter 4: Primary structure and conformation of 
oligosaccharides and polysaccharides. RSC publishing: 2007. 
8. Dumitriu, S., Preparation and properties of cellulose bicomponent fibers. CRC 
Press: 2004. 
9. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
Polym. Sci. 1994, 19, 797?841. 
10. Saha, B. C., Hemicellulose Bioconversion. J Ind Microbiol Biotechnol 2003, 30, 
279-291. 
226 
 
11. Sukumaran, R. K., Bioethanol from lignocellulosic biomass: part II production of 
cellulases and hemicellulases. In Hand book of plant based biofuels, 2009; pp 141-157. 
12. Meister, J. J., Chemical modification of lignin. In Chemical modification of 
lignocellulosic materials, Hon, D. N.-S., Ed. Marcel Dekker Inc.: 1996; pp 129-157. 
13. Sjostrom, E., Wood chemistry: Fundamentals and applications. Academic press, 
New York: 1981. 
14. Hon, D. N.-S.; Shiraishi, N., Wood and cellulosic chemistry. second ed.; 2000; p 
275-384. 
15. Biermann, C. J., Handbook of pulping and paper making. Academic press: 1996. 
 
 
227 
?
Appendix B: Summary of Pretreatment Methods 
?
Table B.1 Summary of the important pretreatment methods. 
Methods Treatment  time Temperature Pressure Remarks 
Acid Catalyzed 
Dilute Acid (H
2
SO
4
) 
Concentrated Acid (H
3
PO
4
) 
Peracetic Acid (C
2
H
4
O
3
) 
Base Catalized 
Sodium Hydroxide (NaOH)  
Lime (Ca(OH)
2
) 
Low Lignin (12-18%) 
Medium Lignin (18-24%) 
High Lignin (> 24%) 
Ammonia Recycle Percolation (ARP) 
Ammonia Fiber Explosion (AFEX) 
 
5-30 min 
30-60 min 
1-180 h 
 
24-96 h 
 
24-96 h 
 ?30 days 
2 h 
?15 min 
? 5 min 
 
140-190?C 
50?C 
25-75?C 
 
25?C 
 
100-120?C 
?55?C 
? 150?C 
180?C 
60-100?C 
 
0.4-1.3 MPa 
0.1 MPa 
0.1 MPa 
 
0.1 MPa 
 
0.1-0.2 MPa 
0.1 MPa 
1.5 MPa 
?2.0MPa 
?2.0 MPa 
 
0.5 - 1.5% 
85% 
2-10% C
2
H
4
O
3 
 
1% NaOH, 0.1 g NaOH/g biomass
 
 
0.10 g Ca(OH)
2
/g biomass 
0.10-0.15 g Ca(OH)
2
/g biomass 
0.15-0.20 g Ca(OH)
2
/g biomass 
15% NH
3 
1 g NH
3
/g biomass 
228 
?
Oxidative Alkali (NaOH + H
2
O
2
 or O
3
) 
 
Ionic Liquid 
Organosolv 
Steam Explosion 
Hydrothermal 
Supercritical CO
2
 
6-25 h 
 
5-12 h 
60 min 
0.3-50 min 
1-15 min 
1 h 
20-60?C 
 
50-150?C 
180?C 
190-250?C 
160-220?C 
35-80?C 
0.1 MPa 
 
- 
3.5-7.0 MPa 
1.2-4.0 MPa 
0.6-3.5 MPa 
7-27 MPa 
1-15% H
2
O
2
 or O
3
, pH =11.5 
 
Ionic liquids 
1.25% H
2
SO
4
, 60% ethanol 
Steam 
Subcritical water 
CO
2
 
 
229 
??
Appendix C: Reaction Pathways of Cellulose, Hemicelluloses, and Lignin in 
Hydrothermal Medium 
?
 Lignocellulosic biomass is a mixture of cellulose (38-50 wt%), hemicelluloses 
(23-32 wt%), and lignin (15-25 wt%) which are held together by covalent bonding, 
various intermolecular bridges, and van der Waals forces forming a complex structure. 
Several studies have been conducted using model compounds such as cellulose, 
hemicelluloses, and lignin in sub-and supercritical water to establish the reaction 
pathways of these compounds in hydrothermal medium. The chemistry behind the 
reactions of the individual components of biomass under hydrothermal conditions is well 
understood. The generalized individual reaction pathways of the major components of 
biomass (cellulose, hemicelluloses, and lignin) are discussed in following section: 
C.1 Cellulose Reaction Pathways 
 Hydrothermal degradation of cellulose is a heterogeneous and pseudo-first-order 
reaction for which detailed chemistry and mechanism have been proposed earlier.
1-4
 
Cellulose reaction in hydrothermal and catalyst free medium mainly proceeds via 
hydrolysis of glycosidic linkages. The long chain of cellulose starts breaking down in 
such condition to smaller molecular weight water soluble compounds (oligomers) and 
further to glucose (monomer). Glucose is water soluble and undergoes rapid degradation 
in hydrothermal medium at elevated temperature.  
 The key products from the glucose decomposition are shown in Figure C.1. These 
products are formed mainly via dehydration, isomerization, reverse aldol condensation, 
and fragmentation reactions. 
OH
2
3
4
5
6
O
1
O
OH
HO
HO
OH
O
OH
OH
O
HO
O
OH
OH
OH
n-2
H
y
d
r
o
l
y
s
i
s
OH
O
OH
OH
HO
HO
Glucose
O
OH
OHHO
HO
HOH2C CH2OH
Fructose
Isomerization
C
1
2
3
4
5
6
OHH
OHH
CH2OH
OH
OH
HO
H
H
Ring openning
CH2
O
OH
HO
HO
O
1,6-Anhydroglucose
O
CHOHOH
2C
O
CHO
5-HMF
Furfural
D
e
h
y
d
r
a
t
i
o
n
Dehydration
R
e
v
e
r
s
e
 
a
l
d
o
l
 
c
o
n
d
e
n
-
s
a
t
i
o
n
 
a
t
 
C
4
-
C
5
R
e
v
e
r
s
e
 
a
l
d
o
l
 
c
o
n
d
e
n
-
s
a
t
i
o
n
 
a
t
 
C
3
-
C
4
5
OHH
3
4
OH
6
H
CH2OH
CHO
CH2OH
2
CHO
1
+
CH2OH
CH2OH
OOHH
CHO
CH2OH
+
CH
3
CHO
O
CH
2OH
CHO
CH2OH
CHO
+
O
OH
OHHO
R
e
v
e
r
s
e
 
a
l
d
o
l
c
o
n
d
e
n
s
a
t
i
o
n
D
e
h
y
d
r
a
t
i
o
n
D
e
h
y
d
r
a
t
i
o
n
Dehydration
Phenols
Glycolaldehyde
Erythrose
Pyruvaldehyde
Dihydroxyacetone
Glyceraldehyde
Organic acids
Organic acids
Gaseous
products
Gaseous
products
Organic acids
Acids/Aldehydes
Gaseous
products
Char
Cellulose
D
e
h
y
d
r
a
t
i
o
n
 
Figure C.1 Key reaction products from cellulose hydrolysis in catalyst free hydrothermal 
medium.
5-8
 
?
C.2 Hemicelluloses Reaction Pathways 
 Hemicelluloses are polysaccharides of five carbon sugar such as xylose or six 
carbon sugars other than glucose. They are usually branched and have much lower degree 
of polymerization. Figure C.2 shows a hardwood xylan with branch chain. 
230 
??
O
O
O
O-Ac
OH
O
OH
O
OH
OH
O
O
O
O-Ac
OH
O
OH
O-Ac
O
O
OH
OH
O
COOH
H
3
CO
 
Figure C.2 A segment of hardwood xylan, O-acetyl-4-O-methylglucuronoxylan (redrawn 
from bobleter et. al.).
1
 
 Branches in the chains do not allow the formation of tightly packed fibrils. 
Hemicelluloses are not crystalline and easily hydrolysable to their respective monomers. 
In fact, about 95% of hemicelluloses were extracted as monomeric sugars using water at 
34.5 MPa and 200-230?C in a span of just few minutes.
2
 In hydrothermal medium, 
Hemicelluloses are hydrolyzed to sugars, which subsequently degrade into furfural and 
other degradation compounds. Furfural (2-furaldehyde) is commercially produced from 
hemicelluloses-derived xylose.
5
 
 
C.3 Lignin Reaction Pathways 
 Lignin is a complex and high molecular weight polymer of phenylpropane 
derivatives (p-coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol). The density of 
hydrothermal medium is found to be a key parameter in lignin decomposition. In 
hydrothermal reaction medium, most of the hemicelluloses and part of the lignin are 
solubilized below 200?C. Lignin fragments have high chemical reactivity. Part of these 
fragments again cross links and re-condenses to form high molecular weight water 
insoluble products.
1
 Recently, Fang et al. have proposed the reaction pathway for lignin 
in supercritical water. The reaction steps consist of four phases: oil phase, aqueous phase, 
231 
??
gas phase, and a solid residue phase.
9, 10
 Their study concluded that lignin can be 
completely dissolved and undergoes homogeneous hydrolysis and pyrolysis preventing 
further re-polymerization. The proposed (Fang et al.) reaction pathway of lignin in 
supercritical water is shown in Figure C.3. 
?
Figure C.3 Reaction pathway of lignin in supercritical water (taken from Fang et al.).
10
 
?
C.4 References 
1. Bobleter, O., Hydrothermal degradation of polymers derived from plants. Prog. 
Polym. Sci. 1994, 19, 797?841. 
2. Mok, W. S.; Antal, M. J., Uncatalyzed solvolysis of whole biomass hemicellulose 
by hot compressed liquid water. Ind. Eng. Chem. Res. 1992, 31, 1157-1161. 
3. Mok, W. S. L.; M.J., A. J.; Varhegyi, G., Productive and parasitic pathways in 
dilute-acid-catalyzed hydrolysis of cellulose. Ind. Eng. Chem. Res. 1992, 31, 94-100. 
232 
??
233 
??
4. Schwald, W.; Bobleter, O., Hydrothermolysis of cellulose under static and 
dynamic conditions at high temperatures. . J. Carbohydr. Chem. 1989, 8, (4), 565-578. 
5. Peterson, A. A.; Vogel, F.; Lachance, R. P.; Froling, M.; Michael Jerry Antal, J.; 
Tester, J. W., Thermochemical biofuel production in hydrothermal media:A review of 
sub- and supercritical water technologies. Energy and Environmental Science 2008, 1, 
32-65. 
6. Sasaki, M.; Goto, K.; Tajima, K.; Adschiri, T.; Arai, K., Rapid and selective retro-
aldol condensation of glucose to glycolaldehyde in supercritical water. Green Chem. 
2002, 4, 285-287. 
7. Kabyemela, B. M.; Adschiri, R.; Malaluan, R. M.; Arai, K.; Ohzeki, H., Rapid 
and selctive conversion of glucose to erythrose in supercritical water. Ind. Eng. Chem. 
Res. 1997, 36, 5063-5067. 
8. Watanabe, M.; Aizawa, Y.; Iida, T.; Levy, C.; Aida, T. M.; Inomata, H., Glucose 
reactions within the heating period and the effect of heating rate on the reactions in hot 
compressed water. Carbohydrate Research 2005, 340, 1931-1939. 
9. Behrendt, F.; Neubauer, Y.; Oevermann, M.; Wilmes, B.; Zobel, N., Direct 
liquefaction of biomass. Chem. Eng. Technol 2008, 31, (5), 667-677. 
10. Fang, Z.; Sato, T.; Jr., R. L. S.; Inomata, H.; Arai, K.; Kozinski, J. A., Reaction 
chemistry and phase behaviour of lignin in high-temperature and super critical water. 
Bioresource Technology 2008, 99, 3424-3430. 
 
 
Appendix D: Salt Solubility in Hydrothermal Medium 
?
D.1 Salt-water Binary System 
 Ambient water is an excellent solvent (typically several 100 g/l) for most of the 
salts. On the other hand, solubility of these salts goes down drastically (1-100 ppm) in the 
low density supercritical water. The reduced solubility in supercritical region causes salts 
to precipitate as fine-crystalline solids, which may lead to plugging of the reactor even at 
high flow velocities.
1
 Salt's solubility in supercritical water is reduced to such an extent 
that electrolytes can be effectively recovered through the precipitation and solid 
separation.
2
 
 The solubility behavior of water-salt systems differ considerably in sub- and 
supercritical water conditions. In the temperature range of 100-200?C and at varying 
pressure, the diversity of solubility criteria is mainly due to the abundance of crystalline 
hydrates. The crystalline hydrate ceases to exist at higher temperature and precipitate as 
solid phase with immiscible fluid phase.
3
 The phase behavior of salt-water systems at 
high temperature depends on the nature of the salt as well as its initial concentration. 
Salts are classified as Type 1 or Type 2 based on their liquid phase solubility changes 
with temperature in binary salt-water system saturated with the salt. Type 1 binary 
systems are characterized by increasing salt solubility in the liquid phase with 
temperature (positive temperature coefficient of solubility) at vapor and higher pressure 
234 
?
upto the melting point of salt. Type 1 systems exhibit high salt solubility in the vicinity of 
critical point of water.
2-4
 
 In case of Type 2 binary systems, the salt solubility in liquid solution decreases 
with temperature (negative temperature coefficient of solubility) under subcritical 
conditions. Type 2 system exhibits very low salt solubility of the order of parts per 
million near the critical point. Generally, Type 1 salts have melting points below 800-
1000?C (can be lower or higher than the critical temperature of water), while Type 2 salts 
have melting points above 700-800?C (always higher than the critical temperature). Table 
D.1 shows the classification of binary mixtures of salt-water systems. 
Table D.1 Classification of salt-water binary system (taken from Valyashko).
5
 
Type 1 salts Type 2 salt 
KF, RbF, CsF 
LiCl, LiBr, LiI 
NaCl, NaBr, NaI 
K
2
CO
3
, Rb
2
CO
3 
Rb
2
SO
4
, 
Na
2
SeO
4 
K
2
SiO
3
 
K
3
PO
4
 
CaCl
2, 
CaBr
2
, CaI
2
 
SrCl
2
, SrBr
2
 
BaCl
2, 
BaBr
2 
LiF, NaF 
Li
2
CO
3
, Na
2
CO
3
 
Li
2
SO
4
, Na
2
SO
4
, K
2
SO
4
 
Li
2
SiO
3
, Na
2
SiO
3
 
Li
3
PO
4
, Na
3
PO
4
 
CaF
2
 
SaF
2
 
BaF
2
 
?
 
 In brief, Type 1 salts have increasing solubility in water with temperature, where 
as Type 2 salts shows opposite trend. For example, Type 2 salt, sodium carbonate 
(Na
2
CO
3
) has about 30 wt% solubility in ambient water and its solubility near the critical 
temperature of water approaches zero.
6
 On the other hand, sodium chloride (NaCl), an 
235 
?
236 
?
example of Type 1 salt has 37 wt% solubility in subcritical water at 300?C and about 120 
ppm at 550?C.
7
 
 
D.2 References 
1. Kritzer, P.; Dinjus, E., An assessment of supercritical water oxidation (SCWO): 
Existing problems, possible solutions and new reactor concepts. Chemical Engineering 
Journal 2001, 83, 207-214. 
2. Wofford, W. T.; Dell'Orco, P. C.; Gloyna, E. F., Solubility of potassium 
hydroxide and potassium phosphate in supercritical water. J. Chem. Eng. Data 1995, 40, 
968-973. 
3. Valyashko, V.; Urusova, M., Solubility behaviour in ternary water-salt systems 
under sub= and supercritical conditions. Monatshefte fur Chemie 2003, 134, 679-692. 
4. Hodes, M.; Marrone, P. A.; Hong, G. T.; Smith, K. A.; Tester, J. W., Salt 
precipitation and scale control in supercritical water oxidation-Part A: fundamentals and 
research. J. of Supercritical Fluids 2004, 29, 265-288. 
5. Valyashko, V. M., Phase equilibria in water-salt systems: some problems of 
solubility at elevated temperature and pressure. High temperature and high pressure 
electrochemistry in aqueous solutions, Natl. Assoc. Corrosion Eng. 1976, 4, 153. 
6. Keevil, N. B., Vapor pressures of aqueous solutions at high temperature. J. Am. 
Chem. Soc. 1942, 64, 841-850. 
7. Sourirajan, S.; Kennedy, G. C., The System of H
2
O-NaCl at elevated 
temperatures and pressures. Am. J. Sci. 1962, 260, 115-141. 
 
237 
?
Appendix E: Journal Publications and Conference Presentations 
 
PATENT APPLICATION 
1. Biomass to biochar conversion in subcritical water (provisional patent # 6124591, 
USPTO, September 2009). 
JOURNAL PUBLICATIONS  
1. Kumar, S.; Gupta, R. B.; Lee, Y. Y.; Gupta, R. B., Cellulose Pretreatment in 
Subcritical Water: Effect of temperature on molecular structure and enzymatic 
reactivity. Bioresour. Technol., 2010, 101 (4) 1337-1347. 
2. Kumar, S.; Gupta, R. B., Biocrude Production from Switchgrass using 
Subcritical Water. Energy & Fuels, 2009, 23 (10) 5151-5159. 
3. Kumar, S.; Gupta, R. B., Hydrolysis of Cellulose in Sub- and Supercritical Water 
in a Continuous Flow Reactor. Ind. Eng. Chem. Res.; 2008, 47, 9321-29. 
4. Kumar, S.; Kothari, U.; Lee, Y. Y.; Gupta, R. B., Hydrothermal Pretreatment of 
Switchgrass and Corn Stover for Production of Ethanol and Carbon Microspheres 
Biomass & Bioenergy, Feb. 2010 (submitted). 
5. Kumar, S.; Kong, L.; Gupta, R. B., Production of Biochar from Switchgrass by 
Hydrothermal Carbonization. Energy & Fuels, Feb. 2010 (submitted). 
6. Kumar, S.; Loganathan, V. A.; Barnett, M. O.; Gupta R. B.; Heavy Metal 
Adsorption on Biochar prepared from Switchgrass (to be submitted). 
238 
?
CONFERENCE PRESENTATIONS 
1. Kumar, S.; Kothari, U.; Lee, Y.Y.; Gupta, R.B., Hydrothermal Pretreatment of 
Switchgrass for Ethanol Production. AIChE annual conference, Nov. 2009. 
2. Kumar, S.; Kong, L.; Gupta, R.B., Hydrothermal Carbonization of Switchgrass 
for Biochar Production. AIChE annual conference, Nov. 2009. 
3. Kumar, S.; Byrd, A.; Gupta R.B., Sub- and Super-critical Water Technology for 
Biofuels: Swtichgrass to Ethanol, Biocrude, and Hydrogen, at 31st Symposium on 
Biotechnology for Fuels and Chemicals, San Francisco, May 2009. 
4. Kumar, S.; Kothari, U; Lee, Y.Y.; Gupta, R.B., Hydrothermal pretreatment of 
switchgrass: Effect of temperature and potassium carbonate on enzymatic 
reactivity, (poster) 31st Symposium on Biotechnology for Fuels and Chemicals, 
San Francisco, May 2009. 
5. Kumar, S.; Byrd, A.; Kong, L.; Cullinan, H.; Gupta, R.B.? Hydrothermal (sub- 
and super-critical water) technology for biofuel production from switchgrass, 
(poster) tcbiomass 2009, The International Conference on Thermochemical 
Conversion Science, Chicago, Sept. 16-18, 2009. 
6. Kumar, S.; Gupta, R; Lee, Y. Y.; Gupta R. B., Enhancement of Enzymatic 
Digestibility of Microcrystalline Cellulose by Treatment in Subcritical Water, 
AIChE conference, Nov. 2008. 
7. Kumar, S.; Demirbas A. Gupta, R. B., Hydrolysis of Microcrystalline Cellulose 
in Subcritical and Supercritical Water in a Continuous Flow Reactor, AIChE 
annual conference, Nov. 2008. 
239 
?
8. Kumar, S.; Gupta, R. B., Biocrude Production from Switchgrass using 
Subcritical Water, AIChE annual conference, Nov. 2008. 
9. Byrd, A.; Kumar, S.; Gupta, R. B., Cellulose Liquefaction and Hydrogen 
Production in Supercritical Water, AIChE annual conference, Nov. 2008. 
10. Kothari, U; Kumar, S.; Gupta, R.B.; Lee, Y.Y., Improvement of Enzymatic 
Hydrolysis for Hydrothermally-Treated Switchgrass Using Mixed-Enzymes, 
submitted for 32
nd
  Symposium on Biotechnology for Fuels and Chemicals, May 
2010.