Investigation of solid-state fungal pretreatment of Miscanthus for biofuels production
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University
By
Juliana Vasco Correa, M.S.
Graduate Program in Food, Agricultural & Biological Engineering
The Ohio State University
2017
Dissertation Committee:
Ajay Shah, Advisor
Thomas Mitchell
Thaddeus Ezeji
Frederick Michel
Copyrighted by
Juliana Vasco Correa
2017
Abstract
Lignocellulosic biomass is an abundant source of renewable energy, but its high recalcitrance to biodegradation needs to be overcome to allow its conversion into biofuels. Thus, pretreatment of the lignocellulosic feedstock is usually required. Fungal pretreatment using white rot fungi is an alternative process to traditional thermo-chemical pretreatments that degrades lignin and enhances the enzymatic digestibility of the lignocellulosic biomass. Fungal pretreatment can be performed in solid-state at low temperature, without added chemicals such as strong acids or bases, and no wastewater is generated. However, in comparison with traditional pretreatments, disadvantages such as long residence times, low yields, and feedstock sterilization requirements make it challenging to implement.
This work investigates the fungal pretreatment of non-sterile biomass with the white rot fungus, Ceriporiopsis subvermispora, for the production of biofuels using the dedicated energy crop Miscanthus × giganteus. For this purpose, solid-state fungal pretreatment of non-sterile Miscanthus was performed in batch using Miscanthus previously colonized with the fungus as inoculum. The process enhanced the enzymatic digestibility of Miscanthus by 3- to 4-fold over that of untreated Miscanthus after 21 days of incubation time. The finished material from this non-sterile pretreatment was used as inoculum for two more generations in a sequential fungal pretreatment process. A
ii propagation of indigenous fungi that outcolonized C. subvermispora was observed through the generations, showing that sterilization is a required step for the stability and reproducibility of fungal pretreatment. The changes in composition and structure of
Miscanthus after fungal pretreatment were compared with those in corn stover, hardwood, and softwood. Fungal pretreatment increased the enzymatic digestibility of hardwood, softwood, and Miscanthus by 2 to 4.5-fold; however, it was not effective for corn stover. Also, fungal pretreatment was effective for Miscanthus harvested in winter or spring, but not for green Miscanthus harvested in fall. Therefore, fungal pretreatment with C. subvermispora showed differential effects that were feedstock-dependent.
Fungal pretreated Miscanthus was used for the production of biogas via solid-state anaerobic digestion, and an increase of 25% in specific methane yield was obtained compared to the untreated Miscanthus. Fungal pretreated Miscanthus was also used for the production of fermentable sugars through enzymatic hydrolysis with commercial hydrolases for butanol production. Results indicated that fungal pretreatment with C. subvermispora does not produce significant amounts of lignocellulosic derived microbial inhibitory compounds that have been shown to inhibit butanol fermentation with
Clostridium beijerinckii; thus, it does not require detoxification and/or washing after pretreatment. A techno-economic analysis of the process to produce fermentable sugars from Miscanthus using fungal pretreatment showed that the process is not feasible at full cellulosic biorefinery scale due to the high capital cost caused by the long residence time, the low bulk density of the material, and the low sugar yields.
iii
Acknowledgments
First and foremost, I would like to express gratitude to my former advisor Dr. Yebo
Li, who was my guide during the first four years of my program. Working with him was the highest privilege and his unconditional support was the foundation for this product. I am forever indebted to him due to the opportunities he offered me, the lesson he taught me, and the high standards he set for me to grow as a researcher. Likewise, I am deeply grateful to Dr. Ajay Shah for his guidance as a committee member throughout these years, and especially for welcoming me into his group for the final part of my work. He believed in me and provided me with autonomy and support, which help me to conclude my dissertation.
I also wish to thank the additional members of my committee, Dr. Tom Mitchell, Dr.
Teddy Ezeji, and Dr. Fred Michel. Dr. Mitchell is one of the most exceptional professors
I have ever come across; his kindness and generosity, along with his passion for science, were a constant source of motivation. Dr. Ezeji and Dr. Michel were available any time I asked for advice, and their knowledgeable comments and suggestions enriched my experience and work immeasurably.
The most valuable lesson of this journey was the opportunity to work with a diverse and bright group of people that were more than colleagues, and to whom I owe my most sincere acknowledgments. I want to thank especially Dr. Johnathon Sheets, Dr. Xumeng
iv
Ge, Dr. Xiaolan Luo, Dr. Fuqing Xu, and Ms. Lo Niee Liew. I will always carry their support, patience, advice, and friendship as the best memory of this time. I also want to thank Dr. Liangcheng Yang, Ms. Long Lin, Ms. Kathryn Lawson, Mr. Lu Zhang, Ms.
Danping Jiang, and other members of the Bioproducts and Bioenergy Research
Laboratory who were vital to that vibrant productive group, of which I was so honored to be part. Finally, I want to express my gratitude to the members of the Biobased Systems
Analysis Laboratory, especially Mr. Ashish Manandhar, Mr. Luis Huezo, and Ms. Asmita
Khanal for their valuable assistance and friendship.
I received unconditional support from the staff of the Department of Food,
Agricultural and Biological Engineering, including Mrs. Peggy Christman, Mrs. Mary
Wicks, Mrs. Candy McBride, Mr. Michael Klingman, Mr. Michael Sword, and Mr. Scott
Wolfe, from whom I received suggestions for my project, administrative and technical assistance, and encouragement during the complex times.
Finally, I want to acknowledge the financial support provided by Fulbright Colombia,
Colciencias, and The Ohio State University through the Department of Food, Agricultural and Biological Engineering; the Ohio Agricultural Research and Development Center; and the Graduate School.
v
Vita
2009...... B.S. Biological Engineering, National
University of Colombia
2012...... M.S. Food Science and Technology,
National University of Colombia
2012 to present ...... Graduate Fellow, Department of Food,
Agricultural and Biological Engineering,
The Ohio State University
Publications
Vasco-Correa, J., Li, Y. 2015. Solid-state anaerobic digestion of fungal pretreated Miscanthus sinensis harvested in two different seasons. Bioresource Technology 185: 211–217.
Vasco-Correa, J., Ge, X., Li. Y. 2016. Fungal pretreatment of non-sterile miscanthus for enhanced enzymatic hydrolysis. Bioresource Technology 203:118–123.
Vasco-Correa, J., Zapata Zapata, A. 2017. Enzymatic extraction of pectin from passion fruit peel (Passiflora edulis f. flavicarpa) at laboratory and bench scale. LWT - Food Science and Technology 80: 280–285.
Vasco-Correa, J., Ge, X., Li. Y. 2016. Chapter 24 - Biological Pretreatment of Lignocellulosic Biomass, In Biomass Fractionation Technologies for a Lignocellulosic Feedstock Based Biorefinery, edited by S.I. Mussatto, Elsevier, Amsterdam, Pages 561– 585.
vi
Zhao, J., Ge, X., Vasco-Correa, J., Li, Y.2014. Fungal pretreatment of unsterilized yard trimmings for enhanced methane production by solid-state anaerobic digestion. Bioresource Technology 158: 248–252.
Ge, X., Vasco-Correa, J., Li. Y. 2016. Solid-State Fermentation Bioreactor Fundamentals, In Current Developments in Biotechnology and Bioengineering: Bioprocesses, Bioreactor and Controls, edited by M. Sanroman, A. Pandey, G. Du & C. Larroche, Elsevier, Amsterdam, Pages 381–402.
Ge, X., Xu, F., Vasco-Correa, J., Li, Y. 2016. Giant reed: A competitive energy crop in comparison with miscanthus. Renewable & Sustainable Energy Reviews 54:350–362.
Fields of Study
Major Field: Food, Agricultural & Biological Engineering
Study in: Biological Engineering
vii
Table of Contents
Abstract ...... ii
Acknowledgments...... iv
Vita ...... vi
List of Tables ...... xvi
List of Figures ...... xviii
Chapter 1: Introduction ...... 1
1.1 Background ...... 1
1.2 Statement of the problem ...... 5
1.3 Overall goals and objectives...... 7
1.4 Contribution of the dissertation ...... 8
Chapter 2: Literature review ...... 9
2.1 Lignocellulosic biomass ...... 9
2.2 Miscanthus...... 12
2.2.1 Advantages of Miscanthus...... 15
2.2.2 Miscanthus production ...... 17
viii
2.2.3 Feedstock logistics...... 21
2.2.4 Miscanthus potential uses ...... 24
2.3 Pretreatment of lignocellulosic biomass...... 26
2.3.1 Types of pretreatment ...... 27
2.3.2 Pretreatment of Miscanthus ...... 30
2.4 Biological pretreatment ...... 32
2.4.1 Microbial depolymerization of lignin ...... 33
2.4.2 Ligninolytic enzyme system ...... 34
2.4.3 Fungal pretreatment ...... 47
2.4.4 Bacterial pretreatment...... 55
2.4.5 Enzymatic pretreatment ...... 57
2.4.6 Challenges and perspectives of biological pretreatment ...... 59
2.5 Anaerobic digestion...... 61
2.6 Biobutanol ...... 66
2.7 Techno-economic analysis ...... 69
Chapter 3: Solid-state fungal pretreatment for non-sterile Miscanthus ...... 75
Abstract ...... 75
3.1 Introduction ...... 75
3.2 Materials and Methods ...... 77
ix
3.2.1 Feedstock collection and storage ...... 77
3.2.2 Inoculum preparation...... 77
3.2.3 Fungal pretreatment experiments ...... 78
3.2.4 Enzymatic hydrolysis ...... 79
3.2.5 Analytical methods ...... 80
3.2.6 Data analysis ...... 81
3.3 Results and Discussion ...... 81
3.3.1 Cellulose, hemicellulose, and dry matter loss ...... 81
3.3.2 Lignin degradation and enzymatic digestibility ...... 86
3.3.3 Time course of the non-sterile fungal pretreatment...... 91
3.4 Conclusions ...... 93
Chapter 4: Changes in composition, enzymatic digestibility and microbial communities during sequential solid-state fungal pretreatment of non-sterile Miscanthus ...... 94
Abstract ...... 94
4.1 Introduction ...... 95
4.2 Materials and Methods ...... 96
4.2.1 Feedstock collection and storage ...... 96
4.2.2 Inoculum preparation...... 96
4.2.3 Fungal pretreatment experiments ...... 96
x
4.2.4 Enzymatic hydrolysis ...... 99
4.2.5 Microbial community analysis ...... 99
4.2.6 Analytical methods ...... 101
4.2.7 Data analysis ...... 101
4.3 Results and Discussion ...... 102
4.3.1 Enzymatic digestibility and lignin degradation ...... 102
4.3.2 Cellulose, hemicellulose, and dry matter loss ...... 107
4.3.3 Microbial community analysis ...... 112
4.4 Conclusions ...... 123
Chapter 5: Changes in composition and structure during sequential fungal pretreatment of non-sterile feedstocks ...... 124
Abstract ...... 124
5.1 Introduction ...... 124
5.2 Materials and Methods ...... 126
5.2.1 Feedstock collection and storage ...... 126
5.2.2 Inoculum preparation...... 127
5.2.3 Fungal pretreatment experiments ...... 128
5.2.4 Enzymatic hydrolysis ...... 129
5.2.5 Analytical methods ...... 129
xi
5.2.6 Scanning electron microscopy ...... 129
5.2.7 Thermogravimetric analysis ...... 130
5.2.8 Analytic pyrolysis ...... 130
5.2.9 Fourier transform infrared spectroscopy ...... 131
5.2.10 Data analysis ...... 131
5.3 Results and Discussion ...... 131
5.3.1 Enzymatic digestibility ...... 131
5.3.2 Components degradation during fungal pretreatment ...... 135
5.3.3 Structural changes in biomass ...... 138
5.3.4 Thermogravimetric analysis of untreated and pretreated feedstocks ...... 143
5.3.5 Analytic pyrolysis of untreated and pretreated feedstocks ...... 149
5.3.6 FT-IR analysis ...... 160
5.4 Conclusions ...... 164
Chapter 6: Solid-state anaerobic digestion of fungal pretreated Miscanthus harvested in two different seasons...... 165
Abstract ...... 165
6.1 Introduction ...... 165
6.2 Materials and Methods ...... 167
6.2.1 Feedstock collection and preparation ...... 167
xii
6.2.2 Solid-state anaerobic digestion ...... 168
6.2.3 Fungal pretreatment ...... 168
6.2.4 Enzymatic hydrolysis ...... 169
6.2.5 Analytical methods ...... 170
6.2.6 Statistical analysis...... 171
6.3 Results and Discussion ...... 171
6.3.1 Characteristics of Miscanthus and anaerobic digestion inoculum...... 171
6.3.2 Effect of harvest season, TS, and F/I ratio on the SS-AD of Miscanthus ...... 173
6.3.3 Effect of harvest season on fungal pretreatment of Miscanthus ...... 177
6.3.4 Effect of fungal pretreatment on SS-AD of Miscanthus ...... 179
6.3.5 Methane potential per hectare...... 182
6.4 Conclusion ...... 183
Chapter 7: Effect of fungal pretreatment of Miscanthus on microbial inhibitors’ generation and butanol fermentation...... 184
Abstract ...... 184
7.1 Introduction ...... 185
7.2 Materials and Methods ...... 187
7.2.1 Feedstock collection and storage ...... 187
7.2.2 Microorganisms and inoculum preparation ...... 187
xiii
7.2.3 Pretreatment of Miscanthus ...... 188
7.2.4 Enzymatic hydrolysis ...... 189
7.2.5 Fermentation ...... 190
7.2.7 Analytical methods ...... 191
7.2.8 Data analysis ...... 192
7.3 Results and Discussion ...... 192
7.3.1 Pretreatment and enzymatic hydrolysis ...... 192
7.3.2 Fermentation ...... 196
7.4 Conclusions ...... 201
Chapter 8: Techno-economic analysis of solid-state fungal pretreatment of Miscanthus for the production of fermentable sugars ...... 204
Abstract ...... 204
8.1 Introduction ...... 204
8.2 Materials and Methods ...... 207
8.2.1 Modeling overview ...... 207
8.2.2 Miscanthus preparation unit ...... 208
8.2.3 Pretreatment unit...... 209
8.2.4 Enzymatic hydrolysis unit ...... 210
8.2.5 Economic analysis ...... 210
xiv
8.2.6 Sensitivity analysis and evaluation of strategies to reduce sugars production
cost ...... 212
8.3 Results and Discussion ...... 213
8.3.1 Material balance ...... 213
8.3.2 Economic analysis ...... 216
8.3.2 Sensitivity analysis ...... 218
8.3.3 Effect of plant size ...... 220
8.3.4 Potential strategies to reduce sugars production cost ...... 221
8.4 Conclusions ...... 223
Chapter 9: Conclusions and recommendations for future research ...... 224
9.1 Conclusions ...... 224
9.2 Recommendations for future research ...... 226
References ...... 228
Appendix: Supplemental data for Chapter 8 ...... 266
xv
List of Tables
Table 2.1 Properties and composition of harvestable above ground biomass of M. × giganteus ...... 15
Table 2.2 Biomass yield of different Miscanthus species...... 16
Table 2.3 Difference between physical, chemical and biological pretreatments ...... 29
Table 2.4 Pretreatment of Miscanthus ...... 31
Table 2.5 Ligninolytic enzyme system ...... 37
Table 2.6 Recent studies in fungal pretreatment of lignocellulosic biomass ...... 51
Table 3.1 Characteristics of Miscanthus ...... 77
Table 5.1 Composition and total solids of feedstocks ...... 127
Table 5.2 Components degradation during fungal pretreatment...... 137
Table 5.3 Thermogravimetric parameters of untreated and fungal pretreated feedstocks
...... 148
Table 5.4 Relative peak area of components identified in Py-GC-MS of untreated and fungal pretreated feedstocks...... 154
Table 5.5 Lignin-to-carbohydrate and syringil-to-guaiacil ratios based on Py-GC-MS of untreated and fungal pretreated feedstocks ...... 157
Table 5.6 Relative changes (%) in FT-IR spectra after fungal pretreatment ...... 163
xvi
Table 6.1 Characteristics of Miscanthus and inoculum for AD...... 172
Table 6.2 Degradation of Miscanthus components after 28 days of fungal pretreatment
...... 178
Table 7.1 Comparison of fungal, liquid hot water, and alkaline pretreatments on solids recovery and changes in Miscanthus composition...... 193
Table 7.2 Sugar concentrations and yields of fungal, liquid hot water, and alkaline pretreated Miscanthus after enzymatic hydrolysis ...... 196
Table 7.3 ABE fermentation results: final product concentrations, total sugars consumed and yield ...... 199
Table 7.4 Concentration of selected LDMICs in the fermentation media at the beginning of the fermentation (T = 0)...... 200
Table 8.1 Economic evaluation parameters used in this study ...... 211
Table 8.2 Capital investment for fungal pretreatment facility producing fermentable sugars from Miscanthus for a 30 million gallon/year cellulosic biorefinery ...... 217
Table 8.3 Annual operating cost of fungal pretreatment facility producing fermentable sugars from Miscanthus for a 30 million gallon/year cellulosic biorefinery ...... 218
xvii
List of Figures
Figure 2.1 Structure of lignocellulosic biomass in plant cell walls. Reproduced from
Yang et al. (2015) (Yang et al., 2015), with permission of Elsevier...... 11
Figure 2.2 Miscanthus × giganteus. Zanesville, OH, December 2015...... 13
Figure 2.3 Miscanthus annual cycle. Source: adlib.everysite.co.uk ...... 21
Figure 2.4 Miscanthus swathing. Source: www.energycrops.com ...... 22
Figure 2.5 Miscanthus chopping. Source: www.quinns.ie ...... 23
Figure 2.6 Monolignols: precursors of lignin ...... 35
Figure 2.7 Common bonds in lignin structure ...... 36
Figure 2.8 Formation of phenoxyl radicals by laccases in phenolic lignin...... 39
Figure 2.9 Mechanism of lignin oxidation by laccase using mediators ...... 40
Figure 2.10 Formation of cation radicals by lignin peroxidases in non-phenolic lignin .. 42
Figure 2.11 Catalytic cycle of peroxidases ...... 43
Figure 2.12 Formation of phenoxyl radicals by manganese peroxidase in phenolic lignin
...... 44
Figure 3.1 Cellulose loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture)...... 83
xviii
Figure 3.2 Hemicellulose loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture)...... 85
Figure 3.3 Dry matter loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture)...... 86
Figure 3.4 Lignin loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture)...... 87
Figure 3.5 Enzymatic digestibility of fungal pretreated Miscanthus after 28 days. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture)...... 88
Figure 3.6 Correlation between lignin loss and enzymatic digestibility of fungal pretreated Miscanthus after 28 days. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture)...... 90
Figure 3.7 Time course study of fungal pretreatment of non-sterile Miscanthus during 28 days at 50% inoculum ratio and 60% moisture content. Components loss (cellulose, hemicellulose, lignin, and dry matter) is a percentage (in dry matter) of amount present in the original mixture (unsterilized Miscanthus plus inoculum). Glucose yield is calculated as the amount of sugar released after enzymatic hydrolysis of the pretreated material, over the glucose present as cellulose in the original Miscanthus...... 92
xix
Figure 4.1 Experimental flowchart for the sequential fungal pretreatment of non-sterile
Miscanthus...... 98
Figure 4.2 Enzymatic digestibility of fungal pretreated Miscanthus after 28 days of incubation at 45%, 60%, and 75% initial moisture content. Negative control: incubated alown treatment without inoculation. Postive control: sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture...... 104
Figure 4.3 Lignin degradation after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment without inoculation. Postive control: sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture...... 106
Figure 4.4 Correlation between lignin degradation and enzymatic digestibility of fungal pretreated Miscanthus...... 107
Figure 4.5 Cellulose loss after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment without inoculation. Postive control: sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture...... 109
Figure 4.6 Hemicellulose loss after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment without inoculation. Postive control: sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture...... 110
Figure 4.7 Dry matter loss after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment
xx without inoculation. Postive control: sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid culture...... 112
Figure 4.8 Alpha diversity of the fungal (ITS) data. a) Observed species (OTUs), b)
Chao1 richeness index, c) Shannon evenness index, and d) Simpson diversity index. .. 114
Figure 4.9 Fungal community composition of untreated and fungal pretreated Miscanthus expressed as relative abundance at the phylum level. UN: unidentified...... 116
Figure 4.10 Fungal community composition of untreated and fungal pretreated
Miscanthus expressed as relative abundance at the genus level. UN: unidentified. OTU assignments with less than 1% abundance are represented as ‘other”...... 119
Figure 4.11 Principal component analysis (PCA) for the fungal community in the untreated and fungal pretreated Miscanthus...... 121
Figure 4.12 Bacterial communities in untreated and pretreated Miscanthus. a) Alpha diversity of bacterial (16S) data. B) Bacterial community composition as relative abundance at the order level...... 122
Figure 5.1 Enzymatic digestibility of untreated and fungal pretreated feedstocks.
Negative control: unsterile uninoculated feedstocks incubated along treatments...... 134
Figure 5.2 Correlation between lignin degradation and increase in enzymatic digestibility in fungal pretreated feedstocks...... 138
Figure 5.3 SEM images of untreated and fungal pretreated Miscanthus and corn stover
...... 141
Figure 5.4 SEM images of untreated and fungal pretreated softwood and hardwood .... 142
xxi
Figure 5.5 TG and DTG curves for untreated feedstocks (TG: dashed lines, DTG: solids lines)...... 144
Figure 5.6 TG and DTG curves for first generation non-sterile fungal pretreated feedstock
(TG: dashed lines, DTG: solid lines)...... 146
Figure 5.7 TG and DTG curves for positive control fungal pretreated feedstocks (TG: dashed lines, DTG: solid lines)...... 147
Figure 5.8 TG and DTG curves for untreated and fungal pretreated Miscanthus...... 148
Figure 5.9 Py-GC-MS of untreated Miscanthus (a), fungal pretreated Miscanthus (b), untreated corn stover (c), and fungal pretreated corn stover (d)...... 152
Figure 5.10 Py-GC-MS of untreated softwood (a), fungal pretreated softwood (b), untreated hardwood (c), and fungal pretreated hardwood (d)...... 153
Figure 5.11 PCA score plot of untreated and fungal pretreated feedstocks characterized by Py-GC-MS...... 160
Figure 5.12 FT-IR spectra of untreated and fungal pretreated hardwood...... 161
Figure 6.1 Cumulative methane yield obtained during 60 days of SS-AD of fall and spring harvested Miscanthus at F/I ratio of 2...... 174
Figure 6.2 Cumulative methane yield obtained during 60 days of SS-AD of fall and spring harvested Miscanthus at F/I ratio of 4...... 174
Figure 6.3 Daily methane yield during 60 days of SS-AD of fall and spring harvested
Miscanthus at F/I ratio of 2...... 176
Figure 6.4 Daily methane yield during 60 days of SS-AD of fall and spring harvested
Miscanthus at F/I ratio of 4...... 177
xxii
Figure 6.5 Enzymatic digestibility of untreated and fungal pretreated Miscanthus...... 178
Figure 6.6 Cumulative methane yield during 60 days of SS-AD of untreated and pretreated Miscanthus...... 180
Figure 6.7 Degradation of components of untreated and fungal pretreated Miscanthus after 60 days of SS-AD...... 182
Figure 7.1 Fermentation profile of Miscanthus hydrolysates. a) Cell growth of C. beijerinckii estimated by optical density at 600 nm, b) pH of fermentation media, c) butanol concentration, d) total ABE concentration, e) acetate concentration, f) butyric acid concentration...... 198
Figure 8.1 Simplified overview of the fungal pretreatment process including major equipment. Fungal pretreatment is performed in pack bed bioreactor. Fungal inoculum is prepared in air-lift seed fermentor. Feedstock is sterilized by autoclaving. Miscanthus is used as feedstock and sugars solution is the main product...... 208
Figure 8.2 Material balance requirement to produce 1 ton of fermentable sugars in a fungal pretreatment facility (values in kg)...... 215
Figure 8.3 Fermentable sugars production cost from Miscanthus with fungal pretreatment process at biorefinery scale...... 217
Figure 8.4 Sensitivity analysis of sugars production cost in fungal pretreatment facility from Miscanthus for a 30 million gallon/year cellulosic biorefinery ...... 219
Figure 8.5 Effect of plant size on fermentable sugars production cost from Miscanthus in a fungal pretreatment facility...... 221
xxiii
Figure 8.6 Potential strategies and impacts for the reduction of fermentable sugars production cost in a fungal pretreatment facility...... 223
xxiv
Chapter 1: Introduction
1.1 Background
The search for alternative sources of renewable energy, such as energy from biomass, has been driven by the need to reduce fossil fuel imports and the effect of these fossil fuels on greenhouse gases emissions (U.S. DOE, 2005). Lignocellulosic biomass has gained attention because it can provide an abundant renewable feedstock supply for the bioenergy industry (Kumar et al., 2008; McKendry, 2002).
To estimate the feasibility of replacing 30% of the current petroleum consumption with biofuels, the United States Department of Energy and the United States Department of Agriculture determined in 2005 the likelihood to produce one billion tons of biomass per year (U.S. DOE, 2005). The latest update of this study shows that to reach this goal by 2040, several new sources of biomass should be considered, including forestry, agricultural and waste resources, in addition to dedicated energy crops. Energy crops, in particular, are renewable, consistent, high volume supply for the bioenergy industry and they could constitute about 35-50% of the total biomass supply by 2040 according to U.S.
Department of Energy’s updated billion ton report (U.S. DOE, 2016). For the last three decades, the most studied energy crops have been perennial grasses (Arundale et al.,
2014; Feng et al., 2017; Ge et al., 2016; Khanna et al., 2008; Lewandowski et al., 2003)
1 such as switchgrass and Miscanthus, because they usually require less agronomic inputs than the annual crops (U.S. DOE, 2011).
For example, Miscanthus is a tall perennial grass that generates high biomass yields compared to crops such as switchgrass and maize (Heaton et al., 2008a), has a remarkable adaptability to different environments, and requires low fertilizer inputs
(Jones and Walsh, 2001). Miscanthus has high water and nitrogen use efficiency
(Lewandowski and Clifton-Brown, 2000), it sequesters carbon into the soil (Dondini et al., 2009), and it can be grown in marginal land (Sanderson and Adler, 2008). Also,
Miscanthus can be harvested from senesce (in fall) to early spring, allowing flexibility in the biomass logistics. Miscanthus biomass has several uses, but it is typically combusted to generate electricity and heat (Clifton-Brown et al., 2008). However, many recent studies have investigated the transformation of Miscanthus into liquid and gaseous fuels, via thermochemical and biological processes, highlighting the growing interest in using
Miscanthus as a bioenergy feedstock (Agu et al., 2016a; Baldini et al., 2017; Boakye-
Boaten et al., 2016; Bomberg et al., 2014; Frydendal-Nielsen et al., 2016; Jurado et al.,
2013; Kang et al., 2013a; Mayer et al., 2014; Menardo et al., 2013; Michalsk and
Ledakowicz, 2014; Scordia et al., 2013; Theuretzbacher et al., 2014; Wafiq et al., 2016;
Whittaker et al., 2016; Yeh et al., 2016; Zhang and Ezeji, 2014).
Biological processes can be used to produce liquid fuels such as ethanol and butanol
(via fermentation) or gaseous biofuels such as methane (via anaerobic digestion).
Pretreatment of lignocellulosic biomass is usually required to reduce recalcitrance and allow efficient hydrolysis mediated by enzymes, which releases sugars that can be
2 converted by microorganisms into biofuels (Kim, 2013). Miscanthus has been found to be particularly recalcitrant to enzymatic hydrolysis and several recent studies have been focused on developing efficient pretreatments for this feedstock (Cha et al., 2016;
Kärcher et al., 2015; Li et al., 2016; Lou et al., 2016; Nkemka et al., 2016; Padmanabhan et al., 2016; Soares Rodrigues et al., 2016; Yu et al., 2014; Zhou et al., 2017). The most common pretreatments for lignocellulosic biomass employ strong chemicals (acids, alkali) and/or high temperature and pressure, which can be energy intensive and harmful to the environment in addition to generating a variety of molecules such as furans, acids, and aromatic compounds, which can inhibit the subsequent fermentation process
(Hendriks and Zeeman, 2009). As an alternative, microbial pretreatment uses microorganisms that grow on the biomass to reduce its recalcitrance to biodegradability
(Kumar et al., 2009). Microbial pretreatment can significantly improve enzymatic digestibility of lignocellulosic biomass and it is performed at near room temperature and with minimal use of water and chemicals compared to other pretreatment methods (Keller et al., 2003; Wan and Li, 2010a). The most promising microorganisms for microbial pretreatment are white rot fungi, which produce a pool of oxidative enzymes that degrade lignin (Chen et al., 2010; Wan and Li, 2012). Indeed, Ceriporiopisis subvermispora is a white rot fungi that selectively degrades lignin over cellulose, and has been extensively studied for its potential for biopulping in the paper industry and for fungal pretreatment of lignocellulosic biomass (Cui et al., 2012a; Ferraz et al., 2008; Ge et al., 2015; Giles et al., 2012; Nazarpour et al., 2013; Tanaka et al., 2009; Wan and Li, 2011a, 2011b, 2010b;
Yaghoubi et al., 2008; Zhao et al., 2014b). However, these studies are usually performed
3 on sterile feedstocks, since C. subvermispora as well as other white rot fungi are very sensitive to microbial competition (Akin et al., 1995). This sterilization requirement is one of the main bottlenecks that restrict fungal pretreatment implementation at larger scale.
Anaerobic digestion is a simple and robust technique for the production of biogas from organic feedstocks, typically animal manure and food waste. These feedstocks are usually processed by liquid anaerobic digestion – the standard process used in the United
States. To digest lignocellulosic biomass such as Miscanthus, a solid-state process that operates at more than 15% of total solid content is more suitable (Li et al., 2011).
Compared to the liquid anaerobic digestion, solid-state anaerobic digestion (SS-AD) requires a smaller reactor volume for the same solids loading, produces higher volumetric yields, needs lower energy input for heating and mixing, and the generates easier-to- handle byproduct (digestate with a lower moisture content) (Li et al., 2011). However,
SS-AD usually has lower productivities and it is highly affected by the nitrogen content of the feedstock, which tends to be low in lignocellulosic biomass (Xu et al., 2013), and, in the case of energy crops, varies significantly with the harvesting time (Lehtomäki et al., 2008; Seppälä et al., 2009). SS-AD of various lignocellulosic biomasses, such as switchgrass, corn stover, and wheat straw, have been studied previously (Brown et al.,
2012; Liew et al., 2012).
Biobutanol is one of the most promising liquid biofuels. It has higher energy content than ethanol – closer to that of gasoline. It also has a low vapor pressure and is less hygroscopic and corrosive, which makes it safer to transport and allows its use in existing
4 gasoline distribution infrastructures and engines, either in pure form or in any mixed proportion with gasoline (Qureshi et al., 2013a). The use of lignocellulosic biomass for butanol production could potentially provide a low cost substrate. However, most of the microorganisms used for butanol production, namely solventogenic clostridia, are very sensitive to inhibitors produced during the pretreatment processes; hence, usually an additional detoxification step is required to obtain efficient butanol production (Ezeji and
Blaschek, 2010), and concerns about the cost of this additional process arise (Qureshi et al., 2013a).
In conclusion, Miscanthus has a high potential as a biomass source for bioenergy, but it is still underutilized. Fungal pretreatment has the potential to reduce the recalcitrance of
Miscanthus for its use as substrate for biofuels production, but some disadvantages, such as low yields, long residence time and sterilization requirements, still need to be overcome, and a better understanding of the process is needed to facilitate its implementation. After fungal pretreatment, Miscanthus could be used for the production of liquid or gaseous biofuels, such as biogas through SS-AD and biobutanol via enzymatic hydrolysis and fermentation. Fungal pretreatment has the potential to overcome one of the main drawbacks of the butanol process from lignocellulosic biomass: the microbial inhibition produced by lignocellulosic derived molecules.
1.2 Statement of the problem
Miscanthus is considered a promising crop for biomass production with several advantages that address the concerns of other biomass sources. It generates high biomass
5 yields with low inputs and its storage needs are more flexible because the harvest window is longer. But its utilization is still limited because, processes to convert Miscanthus efficiently, sustainably, and cost-effectively into biofuels, are lacking. There is a necessity to develop processes that can utilize the potential of Miscanthus sustainably and cost-effectively, and the optimization of those processes can be achieved by deep understanding of the scientific principles behind them. Fungal pretreatment has the potential to reduce the recalcitrance of Miscanthus and facilitate its conversion into biofuels with low energy input, no added chemicals, and low waste generation. However, this technology has some disadvantages, including low retention time, low yields, and feedstock sterilization requirements. In this dissertation, fungal pretreatment of
Miscanthus is evaluated and compared with that of other feedstocks. A process that reduces the needs for feedstocks sterilization is developed, and a deep understanding of the structure, chemistry, and microbiology changes caused by fungal pretreatment of the non-sterile Miscanthus was sought. The effect of Miscanthus harvesting date on fungal pretreatment and SS-AD was evaluated, and fungal pretreated Miscanthus was used for the production of biogas via SS-AD, and butanol via enzymatic hydrolysis and anaerobic fermentation. This study shows, for the first time, that fungal pretreatment of unsterilized
Miscanthus with C. subvermispora can be performed successfully using fungal colonized
Miscanthus as inoculum, yielding similar results as the fungal pretreatment of sterile
Miscanthus. However, this process cannot be performed sequentially, using the pretreated material as inoculum for subsequent generations, because C. subvermispora is rapidly outcolonized by other fungi. This study also shows that the effects of fungal pretreatment
6 with C. subvermispora on the composition and structure of the feedstocks are feedstock- dependent, and that, unlike some traditional biomass pretreatments, fungal pretreatment of Miscanthus with C. subvermispora does not generate significant concentrations of microbial inhibitory compounds that could inhibit butanol fermentation with C. beijerinckii. Finally, this dissertation demonstrates that the main bottlenecks that limit the feasibility of fungal pretreatment at commercial scale are sugar yields, feedstock bulk density, and pretreatment time.
1.3 Overall goals and objectives
The hypothesis of this dissertation is that fungal pretreatment of unsterilized
Miscanthus can be performed using fungal colonized Miscanthus as inoculum for the production of biofuels. Therefore, the main goal of this dissertation was to investigate fungal pretreatment of unsterilized lignocellulosic biomass for biofuel production, using
Miscanthus as a model feedstock. To accomplish this goal, the objectives were to:
1. Optimize a solid-state fungal pretreatment batch process for non-sterile Miscanthus
that enhances its enzymatic digestibility.
2. Study the behavior of a sequential fungal pretreatment process for non-sterile
Miscanthus using fungal-colonized Miscanthus as inoculum for multiple generations.
3. Assess the changes in composition and structure of Miscanthus during sequential
fungal pretreatment in comparison to corn stover, hardwood and softwood.
4. Evaluate the influence of Miscanthus harvesting date on fungal pretreatment and
biogas production by solid-state anaerobic digestion.
7
5. Study the effects of fungal pretreatment of Miscanthus in the microbial inhibitors’
generation and the butanol production by anaerobic fermentation.
6. Evaluate the techno-economics of the developed fungal pretreatment process for
fermentable sugar production.
1.4 Contribution of the dissertation
The outcome of this research is expected to further enhance the understanding of the fungal pretreatment process, including (1) the effects of fungal pretreatment on
Miscanthus, (2) the feasibility of reducing feedstock sterilization requirements for fungal pretreatment, and (3) the potential of using fungal pretreatment for the production of biofuel from Miscanthus via anaerobic digestion and butanol fermentation. This research provides insights into the complexity of the fungal pretreatment process and generates baseline data for the potential design and optimization of sustainable processes for the production of biofuels at larger scale.
Three articles (from Chapters 2, 3, and 6) have been published in peer-reviewed journals. Two book chapters were also published with materials from Chapter 2. Four additional articles are being prepared based on Chapters 4, 5, 7, and 8.
8
Chapter 2: Literature review
2.1 Lignocellulosic biomass
Concerns about energy dependence on fossil fuels and greenhouse gases emissions have prompted the development of the biofuel industry worldwide. In 2015, 35 billion gallons of liquid biofuels were produced, and they met 2.8% of the world’s transportation fuel demand (REN21, 2016). The vast majority of the liquid biofuels produced to date are
“first-generation” or conventional biofuels, which are produced from biomass that could be otherwise used for food or feed, such as corn, sugarcane, wheat, sugar beet, or vegetable oils. An alternative to those are the “advance” or “second-generation” biofuels that used renewable non-food biomass as feedstock. Most of that biomass is estimated to come from lignocellulosic feedstocks (U.S. DOE, 2016).
The use of lignocellulosic biomass as a renewable feedstock for biorefineries has gained attention because of its potential to replace a significant fraction of petroleum for the generation of bioenergy and bioproducts (Cherubini, 2010). Ideally, biorefineries will sustainably transform a variety of lignocellulosic feedstocks through different multi-step processes into fuels, power, chemicals, and materials (FitzPatrick et al., 2010).
Biorefineries’ success depends on the availability of renewable, consistent, high volume feedstock supply, which can be provided by dedicated energy crops, such as switchgrass or Miscanthus (U.S. DOE, 2011).
9
United States is the main producer of biofuels, with a share of 46% of the global production in 2015 (REN21, 2016). In 2005 the United States Department of Energy and the United States Department of Agriculture attempted to determine the availability of one billion ton of dry biomass, necessary to displace 30% of the U.S. fossil fuels consumption (U.S. DOE, 2005). The last update of this study, the 2016 Billion-Ton
Report, concluded that there is a potential to produce 0.9-1.1 billion tons of biomass by
2030, for the considered base-case and high-yield scenarios, respectively. This includes forest biomass, agricultural resources, waste resources, and energy crops, with the latter encompassed 24-33% of the total biomass resources (U.S. DOE, 2016).
Lignocellulosic biomass is composed mainly of cellulose, hemicellulose, and lignin
(Figure 2.1). It usually contains low amounts of proteins, monosaccharides, starch, and/or oils (Potumarhi et al., 2013; Wu et al., 2010). For thermochemical conversion of biomass into bioenergy, most of the components of the lignocellulosic biomass are utilized; however, the biological conversion is based mostly in the transformation of the sugars contained in the holocellulose (cellulose and hemicellulose). The cell wall components in the lignocellulosic biomass are attached in an intricate arrangement that protects the plant structure from pathogens, but it also obstructs the access of hydrolytic enzymes to holocellulose and hinders the liberation of sugars (Mosier et al., 2005).
10
Figure 2.1 Structure of lignocellulosic biomass in plant cell walls. Reproduced from
Yang et al. (2015) (Yang et al., 2015), with permission of Elsevier.
Cellulose is a homopolymer of cellobiose (glucose disaccharide) attached by β(1-4) bonds. Up to 10,000 units of D-glucose can be attached to form one cellulose chain.
About 36 of those chains are arranged uniformly and linked by hydrogen bonds to form a microfibril. This creates a crystalline structure that is both flexible and strong (Faik,
2004). Some parts of the cellulose are also non-crystalline, and are usually easier to hydrolyze. Cellulose corresponds to around 40-50% of the lignocellulosic biomass and it is one of the most abundant polymers on earth (Potumarhi et al., 2013).
Hemicellulose is a heteropolymer with an amorphous structure of short and branched chains (Potumarhi et al., 2013), which surrounds the cellulose and connects it with the lignin, giving rigidity to the whole arrangement. Hemicellulose is composed of different monosaccharides, such as xylose, mannose, galactose, and arabinose (Faik, 2004). These characteristics make it easier to hydrolyze than cellulose (Kim, 2013).
11
Lignin is an amorphous three-dimensional polymer made of phenylpropanoid units derived from p-coumaryl, guaiacyl, and syringyl alcohol (Faik, 2004) attached by ether and C-C bonds (Potumarhi et al., 2013). Lignin structure is highly heterogeneous and it is resistant to hydrolysis, thus it is extremely resilient to chemical and microbial degradation (Kim, 2013). Lignin is covalently and non-covalently cross-linked to the holocellulose, acting mainly as a protective insoluble layer (Axelsson et al., 2012). This gives it the ability to act as shield for the holocellulose, while it holds together the cellulose microfibrils and attaches the hemicellulose, in a complex structure that forms the plant cell wall (Figure 2.1) (Zheng et al., 2014). While lignin contains about one third of the carbon fixed by plants, this carbon usually cannot be assimilated as an energy source by microorganisms, making the presence of lignin undesirable for the production of biofuels by biochemical conversion (Zeng et al., 2014). Also, lignin has the potential of binding cellulolytic enzymes, thus reducing their availability for hydrolysis (Zeng et al., 2014).
2.2 Miscanthus
Miscanthus is a tall perennial grass that has high annual biomass yields, low nutrients and water requirements, and the ability to adapt to a wide variety of climate and soil conditions, (Jones and Walsh, 2001), compared to similar energy crops such as switchgrass and maize (Heaton et al., 2008a). Therefore, Miscanthus has been identified as one of the most promising energy crops (Heaton et al., 2010).
12
Miscanthus is a perennial rhizomatous grass that has its origin in East-Asia
(Lewandowski et al., 2003). The term Miscanthus (or miscanthus) is frequently used as the common name for any grass of the genus Miscanthus, which includes around 14 species, and belongs to the family Poaceae (Jones and Walsh, 2001). It produces a tall grass (Figure 2.2) that generates one crop per year (Clifton-Brown et al., 2008) and it is likely to originate natural hybrids (Pyter et al., 2009).
Figure 2.2 Miscanthus × giganteus. Zanesville, OH, December 2015.
Three species have been identified as the most promising for biomass production: M. sinensis, M. sacchariflorus, and M. × giganteus (Jones and Walsh, 2001). M. × giganteus, the most studied species for bioenergy, is a triploid sterile hybrid (Linde
Laursen, 1993) which makes it a non-invasive plant (Atkinson, 2009), but makes it more difficult to propagate. It was introduced to Europe in the 1930s as an ornamental grass, surviving the cold temperatures in Denmark (Scurlock, 1999). It can be adapted to a wide 13 variety of climates, because its natural zones are very broad and include from Siberia to
Polynesia, and even some varieties can be found in Africa (Clifton-Brown et al., 2008;
Jones and Walsh, 2001). Other species, as M. sinensis, can produce fertile seeds, and even though they can be invasive, their establishment can be less costly (Christian et al.,
2005).
M. × giganteus has a significant content of cellulose, of around 50% (dry matter basis). It also contains hemicellulose of around 25% and a relatively low concentration of lignin of approximately 12% (Table 2.1) (Hodgson et al., 2011). These values fluctuate significantly with harvesting time and genotype (Brosse et al., 2012). The mineral concentration also varies considerably with the genotype and harvesting period, additionally to the type of soil and fertilization (Brosse et al., 2012). Mineral concentration decreases during the growth period, probably because of a dilution effect and the nutrient translocation to the rhizomes (Jørgensen, 1997).
14
Table 2.1 Properties and composition of harvestable above ground biomass of M. ×
giganteus
Properties Range Heating value (MJ kg-1)* 16-19 Ash (% DW) 2.0 Lignin (% DW) 9-12 Cellulose (% DW) 43-58 Hemicellulose (% DW) 16-34 Extractives (% DW) 9-17 Crude protein (% DW) 3-7 C (% DW) 44.7-47.4 N (% DW) 0.3-2.1 H (% DW) 5.9-6.0 O (% DW) 39.8-43.6 *based on dry biomass. DW: Dry weight. Adapted from (Ge et al., 2016).
One of the updates of the 2016 Billion-Ton Report with respect to the previous studies was the inclusion of Miscanthus as an independent modeled feedstock, which showcases the recently increased interest in this energy crop as biomass source in the
U.S. (U.S. DOE, 2016).
2.2.1 Advantages of Miscanthus
M. × giganteus has higher yields than other similar crops such as switchgrass and maize (Heaton et al., 2008a). It has been recently confirmed that Miscanthus’ yields consistently double those of switchgrass over a wide range of conditions and environments (LeBauer et al., 2017). Miscanthus yields vary with species, environmental
15 conditions, nutrients and water availability, age of the crop, and harvest date, among others (LeBauer et al., 2017) (Table 2.2).
Table 2.2 Biomass yield of different Miscanthus species
Fall Winter Species harvest harvest Location Ref. (ton ha-1) (ton ha-1) M. × giganteus 11-51 7-30 Austria, Belgium, Denmark, (Anderson et al., France, Germany, Greece, USA 2014) (Illinois), Latvia, Portugal, Spain, The Netherlands, UK Denmark, France, Japan, (Nsanganwiman M. sinensis 17-49 6-22 Slovakia, Switzerland a et al., 2014) (Borkowska and M. sacchariflorus 27-40 14-17 Poland Molas, 2013) (Huang et al., M. floridulus NA 28-38 Taiwan 2011) M. lutarioriparius* 10-30 NA Southern China (Liu et al., 2012) *Obtained by model calculation. NA: not available.
The perennial life cycle of Miscanthus involves the translocation of nutrients back to the rhizomes in fall. This process allows the recycling of nutrients, resulting in less requirement of fertilizers than other crops and allowing carbon fixation on the soil
(Dohleman et al., 2010; Pyter et al., 2009). Miscanthus has a very favorable energy balance because it does not need one planting activity every year, the fertilization inputs are very low, and the harvested material is fairly dry (25-40% moisture if late harvesting is done) (Jones and Walsh, 2001). Therefore, energy input to output ratio may be less than 0.2 (Acaroğlu and Şemi Aksoy, 2005; Angelini et al., 2009; Heaton et al., 2004).
Miscanthus photosynthesis follows a C4 pathway, which is a characteristic of tropical crops and very rare for a grass that can survive cold temperatures (Dohleman et al.,
16
2010). The C4 photosynthesis allows the plant to fix carbon in a 40% more efficient manner (Heaton et al., 2004), and to generate greater biomass yields. It also connotes better water and nitrogen use efficiency (Lewandowski and Clifton-Brown, 2000).
Miscanthus also uses sunlight more efficiency, in part because of the C4 photosynthesis, but also because of the long leaves canopy duration. The leaves grow very early compare to maize or soybean, due to the large reserves of nutrients present in the rhizomes, and because Miscanthus can be harvested late, the sunlight of the whole summer and the first part of fall are utilized (Dohleman et al., 2010). It has been shown that Miscanthus annual average leaf area can double that of maize in the Midwest (Dohleman and Long, 2009).
Miscanthus cultivation has some advantages for the environment. It gives resistance to soil erosion and loss of nutrients due to its highly widespread root system (Heaton et al., 2004; Neukirchen et al., 1999). Also, as a perennial crop, Miscanthus sequesters carbon to the soil, making it richer in organic matter (Clifton-Brown et al., 2007). It has been proved that it can harbor a broad variety of wildlife (Bellamy et al., 2009; Jones and
Walsh, 2001). Moreover, it can be grown on marginal land, avoiding competition with food and animal feed production (Sanderson and Adler, 2008) and allowing the spreading of sludge or effluent that cannot be applied to food crop land (Jones and Walsh, 2001).
2.2.2 Miscanthus production
2.2.2.1 Planting
The propagation of M. × giganteus, because of its sterility, is made mainly using rhizomes. It can also be performed using micro-plants, but this method is usually more
17 expensive and less successful (Atkinson, 2009). The rhizomes are harvested from mature plants at the end of the winter and planted between March and May, late enough to avoid severe frosts (Jones and Walsh, 2001). It is not convenient to store the rhizomes for long periods of time (Lewandowski and Clifton-Brown, 2000), and storage should be done at low temperatures to avoid root growth (Clifton-Brown et al., 2008). Some machinery has been developed to plant the rhizomes (Clifton-Brown et al., 2008), but standard machinery could be adapted (Jones and Walsh, 2001). Generally, this process is considered the least developed in the Miscanthus logistic chain. Compared with other crops, M. × giganteus can be difficult and expensive to establish. However, if this process is done successfully, the crop will prevail for at least 15 to 20 years (Atkinson,
2009).
The rhizomes should be planted at 20 cm to 30 cm depth, in a recommended density of 1 plant per square meter (Lewandowski and Clifton-Brown, 2000). Yield would be low the first two years, and the maximum yield would only be attained around the third to fifth year (Clifton-Brown et al., 2008). Weed control is also important during the first year, because it is very difficult for the Miscanthus to compete with them while establishing, but after that no weed control would be needed (Heaton et al., 2008b;
Lewandowski and Clifton-Brown, 2000).
2.2.2.2 Nutrients and water requirements
M. × giganteus can be grown in a wide variety of soils, and it grows better in soils with high water holding capacity (Lewandowski et al., 2003). In highly fertile soils,
18 establishment can occur faster (Pyter et al., 2009), but Miscanthus can grow even in marginal land (Dohleman et al., 2010). This option can be beneficial since Miscanthus can help to recover marginal land with its nutrient sequestration capacity (Ng et al., 2010)
As mentioned before, Miscanthus needs a very low fertilizer input (Dohleman et al.,
2010; Pyter et al., 2009). Some studies in Europe have shown N fertilization had no significant effect on Miscanthus growth during several years of cultivation (Christian et al., 2008; Danalatos et al., 2007; Himken et al., 1997), but others have determined that N fertilization is crucial (Ercoli et al., 1999). The need for N depends on the availability of this element in the soil (Lewandowski and Clifton-Brown, 2000). Generally, N, K, and P fertilizations are suggested to avoid soil depletion (Christian et al., 2008; Heaton et al.,
2004), but the applications are usually very low compared to other similar crops, and are normally determined by the nutrient concentration in the harvested biomass (Himken et al., 1997).
M. × giganteus yield increases with water availability (Pyter et al., 2009). Therefore, irrigation can be beneficial in some regions but generally the effect of irrigation on biomass yield is not significant (Jones and Walsh, 2001); in the Midwest of the United
States, the high precipitations, humidity, and water retaining soils are probably sufficient for not implementing irrigation (Pyter et al., 2009). Prolonged times of drought can be very harmful for M. × giganteus, to the limit of causing senescence (Clifton-Brown,
2000). The actual amount of water use for the biomass production can be greater in
Miscanthus than in other similar crops like maize, because although the water use
19 efficiency is higher, the yield, and therefore the total biomass, is usually higher too
(Dohleman et al., 2010).
Even though some pests have been reported to be able to affect Miscanthus
(Lewandowski and Clifton-Brown, 2000), none of them have been shown to have an effect on production. Thus, Miscanthus cultivation have been performed successfully without the use of pesticides in Europe (Christian et al., 2008; Lewandowski et al., 1995) and the United States (Heaton et al., 2008b).
2.2.2.3 Weather conditions
For a C4 plant, Miscanthus has high tolerance to low temperatures. In Illinois, M. × giganteus has survived several winters with temperatures below -20°C (Pyter et al.,
2009). The key to Miscanthus establishment is the survival of the first winter, which can be the most difficult task (Jones and Walsh, 2001).
Temperature has a driving effect in the growth cycle of the crop. First, rhizome planting should be done late enough to avoid spring hard frosts that can harm them (Jones and Walsh, 2001). When around 10°C is reached in the soil, the plant begins to grow
(Lewandowski and Clifton-Brown, 2000). Then, the first fall frost induces the senescence of the leaves, which is essential for the translocation of nutrients and the drying of the biomass (Pyter et al., 2009). The Miscanthus cycle is shown in Figure 2.3.
20
Figure 2.3 Miscanthus annual cycle. Source: adlib.everysite.co.uk
2.2.3 Feedstock logistics
2.2.3.1 Harvesting
Miscanthus can be harvested from senescence in fall until early spring; however, late harvesting is most widely used (Jones and Walsh, 2001). Before Miscanthus is harvested in spring, the crop remained in the field during the winter, producing a biomass with less moisture content and allowing the complete translocation of nutrients back to the rhizomes, but also generating less biomass yields since 30% to 50% of the dry matter can be lost, mainly because of the leaf fall (Heaton et al., 2004; Lewandowski and Clifton-
Brown, 2000). When harvested in fall, biomass losses can be reduced, but the product has higher moisture and nutrients contents (Heaton et al., 2004), increasing logistic expenses,
21 such as drying prior to storage, and fertilizer requirements for next season (Amougou et al., 2011).
Existing harvesting machinery can be used to collect Miscanthus (Clifton-Brown et al., 2008; Heaton et al., 2004). Harvesting can be done by two methods: multi-phase harvesting and single-phase harvesting, depending on the equipment available (Jones and
Walsh, 2001). Frequently, mowing, swathing, and then baling are executed, but chopping and bundling are also appropriate. The multi-phase method allows drying in a swath
(Figure 2.4) (Lewandowski and Clifton-Brown, 2000) and an easy pick up for the baler
(Jones and Walsh, 2001). The single-phase systems can be very productive, but usually need special machinery. Chopping can be performed with a forage harvester (Figure 2.5) and has been shown to be suitable for large-scale production systems in Europe (Smeets et al., 2009) as well as in North America (Dondini et al., 2009).
Figure 2.4 Miscanthus swathing. Source: www.energycrops.com
22
Figure 2.5 Miscanthus chopping. Source: www.quinns.ie
2.2.3.2 Drying
Having a biomass with low moisture content is important because it prevents microbiological activity that will result in loss of biomass and baling warm-up due to microbial activity. When harvested in spring, normally drying is not necessary. If moisture content is above 25%, drying in the field (arranged in swaths) can be useful
(Clifton-Brown et al., 2008). This technique allows a more uniform drying than leaving the crop without being harvested in the field, because in this last case the plant gets dried faster in the top and remains wet near the ground. The disadvantage of the swathing is that some soil is going to be picked-up when recovering the dry matter from the swath, diminishing the quality of the biomass (Jones and Walsh, 2001). Drying in storage or industrial installations can also be performed. For this operation, heated or unheated air can be used, depending on the moisture content of the biomass (Jones and Walsh, 2001).
2.2.3.3 Storage
Most of the biomass uses required availability during the whole year, thus appropriate storage is fundamental for Miscanthus industry. Storage can be done in the open air, with 23 or without covering (Smeets et al., 2009). The lack of covering would imply biomass losses in the external layer of the material (Jones and Walsh, 2001). The covering could be plastic, which can generate problems when covering bales, because of the sharp stem pieces can break the covering. Piles can also be covered with organic waste, but this requires availability of some type of organic waste close to the Miscanthus storage location (Lewandowski and Clifton-Brown, 2000).
Storage can also be performed under roof as bales, or chopped material can be ensilaged. For the bales, regular methods for transporting and storage straw or hay bales can be used (Lewandowski and Clifton-Brown, 2000). For the chopped biomass, standard methods for maize ensilage can be employed, arranging the biomass in piles and covering them to create anaerobic conditions that would stop most of the microbiological activity
(Jones and Walsh, 2001).
2.2.4 Miscanthus potential uses
Miscanthus biomass has several potential uses for energy production, but it is typically combusted to generate electricity and heat (Clifton-Brown et al., 2008). This process has been very well studied (Collura et al., 2006; Lewandowski and Kicherer,
1997; Ryu et al., 2006), and it has several advantages because of the good quality of biomass, namely, low moisture and minerals content. These qualities are enhanced when late harvesting is carried out (Lewandowski and Kicherer, 1997).
Several investigations have been conducted on the conversion of Miscanthus to liquid and gaseous biofuels through thermochemical conversion such as gasification and
24 pyrolysis (Collura et al., 2005; Couto et al., 2017; Dejong et al., 2003; Hodgson et al.,
2011; Khelfa et al., 2009; Michel et al., 2011, 2006; Porbatzki et al., 2011; Wafiq et al.,
2016; Yorgun and Simşek, 2008), to ethanol (Boakye-Boaten et al., 2016; Bomberg et al.,
2014; Ge et al., 2011; Han et al., 2011; Kang et al., 2013a; Scordia et al., 2013; Sørensen et al., 2008; Yeh et al., 2016), butanol (Agu et al., 2016a; Zhang and Ezeji, 2014), and hydrogen (de Vrije et al., 2009, 2002) via fermentation, and biogas (Baldini et al., 2017;
Frydendal-Nielsen et al., 2016; Jurado et al., 2013; Klimiuk et al., 2010; Mayer et al.,
2014; Menardo et al., 2013; Michalsk and Ledakowicz, 2014; Theuretzbacher et al.,
2014; Whittaker et al., 2016) via anaerobic digestion. For biological transformation of
Miscanthus into biofuels, most of the recent studies are focused on developing and optimizing biomass pretreatment processes (Cha et al., 2016; Kärcher et al., 2015; Li et al., 2016; Lou et al., 2016; Nkemka et al., 2016; Padmanabhan et al., 2016; Soares
Rodrigues et al., 2016; Yu et al., 2014; Zhou et al., 2017).
The production of alternative products can enhance the economic advantage of the biomass, as described by the biorefinery concept (Brosse et al., 2012). For the biofuels production pretreatment and hydrolysis would be needed (Brosse et al., 2009; Le Ngoc
Huyen et al., 2010; Vanderghem et al., 2012; Vintila et al., 2010) and some of them generate lignin as a byproduct. It has been shown that Miscanthus lignin has potential uses for fabrication of internal-grade panels for construction (Hage et al., 2011), or for extractives as vanillin, aromatics, quinones, sterols, among others (Serrano et al., 2010;
Villaverde et al., 2009). Furthermore, Miscanthus biomass can be used for the production of paper (Ye et al., 2005) and as fiber in polymers (Johnson et al., 2005).
25
2.3 Pretreatment of lignocellulosic biomass
The lignocellulosic biomass structure is usually recalcitrant to enzymatic digestion with cellulases and xylanases. As a result, pretreatment of the lignocellulosic feedstock, which improves sugar release from polysaccharides, is usually a necessary step prior to enzymatic hydrolysis in biorefineries. However, pretreatment is generally the most expensive process in the conversion of lignocellulosic biomass, because of high energy and chemical requirements (Mosier et al., 2005).
In order to make the fermentable sugars accessible, lignocellulosic biomass should be pretreated and then enzymatically hydrolyzed (McMillan, 1994). The enzymatic hydrolysis releases sugar monomers that can be then fermented. Enzymatic hydrolysis is performed using a cocktail of hydrolases that includes exo- and endo-glucanases, and β- glucosidases. Also, xylanases, mannanases and other hydrolases that target specific hemicellulose polymers are commonly added (Banerjee et al., 2010a; Van Dyk and
Pletschke, 2012). Depending on the feedstock, accessory enzymes such as pectinases, esterases, and proteases can also be included to degrade some less common structures in the plant cell wall and enhance the overall sugar yield (Banerjee et al., 2010b). However, enzymes are large protein molecules that cannot access most of the intricate structure of the lignocellulose. Thus, a pretreatment process that disrupts the lignocellulosic arrangement and exposes the carbohydrate polymers is required (Zhang and Shahbazi,
2011). The purpose of the pretreatment is to reduce the recalcitrance of the biomass and enhance the accessibility for the hydrolytic enzymes (Kim, 2013).
26
Considerable effort is being dedicated in optimizing the pretreatment of lignocellulosic biomass to increase its feasibility in large scale (Kim, 2013). A good pretreatment should allow the recovery of cellulose and hemicellulose in a less recalcitrant form and break down lignin selectively, with minimal sugar degradation and the least formation of fermentation inhibitors. Also, it must be able to be performed on a commercial scale, considering the technical and economic implications (Brosse et al.,
2009; Kumar et al., 2009; Mosier et al., 2005).
2.3.1 Types of pretreatment
There are several kinds of pretreatment, and new technologies are being developed to improve this key step in the biomass-to-biofuels process (Zhang and Shahbazi, 2011).
These technologies can be classified as physical, chemical, and biological (Alvira et al.,
2010). Physical pretreatments are usually mechanical disruptions that enhance the surface area of the material (Hendriks and Zeeman, 2009) or processes that use temperature and/or pressure to disrupt the material (Taherzadeh and Karimi, 2008; Zheng et al.,
2014). Chemical pretreatments use chemicals such as acids, bases, and ionic liquids, and are frequently conducted at high temperature and/or pressure (Amarasekara, 2013; da
Costa Sousa et al., 2009). Biological pretreatment employs microorganisms or enzymes that can degrade the lignocellulosic biomass, change its structure and make it more accessible to the hydrolytic enzymes (Chen et al., 2010; Wan and Li, 2012) (Table 2.3).
Some of the most studied pretreatments are dilute acid pretreatment, alkali pretreatment, and hot water pretreatment. Dilute acid pretreatment uses a strong acid
27
(usually sulfuric acid) in concentration lower than 3%, and high temperature (140-220°C) to disrupt lignin and hydrolyze hemicellulose (Kim, 2013; Zhang and Shahbazi, 2011).
This is the leading technology to pretreat biomass, but it also has several problems: it needs equipment resistant to corrosion, usually generates significant amounts of inhibitors of fermenting microorganism such as furans that need to be removed in an additional step prior to fermentation (detoxification), and requires neutralization (Mosier et al., 2005; Zhang and Shahbazi, 2011). Alkali pretreatment uses bases such as lime or sodium hydroxide to extract lignin and part of the hemicellulose (da Costa Sousa et al.,
2009; Zhang and Shahbazi, 2011). It can be performed at less severe temperature (25-
190°C), but it also needs neutralization and detoxification (Amarasekara, 2013; Kim,
2013). Hot water pretreatment uses pressure to maintain water in liquid phase at high temperatures (140-220°C) (Mosier et al., 2005). It does not use an external addition of any chemical, but organic acids are generated under those conditions, so the mechanism is similar to that of the dilute acid pretreatment (Kim, 2013). This method mainly solubilizes the hemicellulose and fractionates some parts of the lignin (Hendriks and
Zeeman, 2009; Kim, 2013). It is less corrosive and generates less inhibitors than the dilute acid pretreatment, and does not need to be neutralized (Mosier et al., 2005; Zhang and Shahbazi, 2011).
28
Table 2.3 Difference between physical, chemical and biological pretreatments
Pretreatment Physical Chemical Biological Do not use chemicals or Use of microorganisms or their Definition Use of chemicals. microorganisms. enzymes. Comminution (milling and Dilute acid, alkaline, grinding), steam explosion, catalyzed steam explosion Fungal, microbial consortium, Examples liquid hot water, extrusion, (AFEX, CO2, SO2), enzymatic. irradiation. organosolv, ionic liquids. Main Reduce particle size. Enhance Lignin and hemicellulose Alter the chemical structure. purposes surface area. degradation. Uses of microorganisms or Gridding/milling or use of Use of chemicals: acids, their enzymes to degrade changes in pressure like steam Mechanism bases, solvents, ionic liquids. biomass. Better when it explosion. degrades lignin preferentially. Alteration of physicochemical With white rot fungi: lignin, structure. Usually gives two hemicellulose, -and sometimes Enhance surface area. Reduce Main effect streams, one is reach in cellulose- degradation. crystallinity of cellulose. cellulose. Usually helps With brown rot fungi: mainly separate the lignin cellulose degradation. Effect in Degradation of non-crystalline Could be partially degraded. Reduce on crystallinity (all). cellulose cellulose. Some fungi leave it intact. Solubilized in most cases Solubilized under steam Effect in from 10-100%. Remains solid Partially degraded (10-50%). hemicellulose explosion and liquid hot water. in AFEX. Enzymatic oxidative Partially hydrolyzed and Solubilized in alkaline Effect in degradation (when white rot redistributed during steam pretreatments, organosolv and lignin fungi or their enzymes are explosion. some ionic liquids. used). Minutes to hours. Days Hours for enzymatic. Days to Minutes to hours. Time sometimes for alkaline. weeks for microbial. Ambient for grinding. 100-250°C Ambient-250°C. 50-300 psi 20-55°C. Atmospheric Conditions for others. 100-700 psi for steam for acid or AFEX. Others pressure. explosion and liquid hot water. usually atmospheric. Low for alkaline when mild High for comminution when conditions are used. Medium particle size is too small. High for acid and catalyzed steam Low for microbial. Medium to Relative cost for irradiation. Medium for explosion for energy used. high for enzymatic. steam explosion and liquid hot High for organosolv and ionic water for energy use. liquids for chemicals. None in comminution/extrusion. Secondary metabolites. Usually several products, Some sugar degradation products Biomass in case of microbial Byproducts some toxic for fermentation: and aromatics in steam explosion pretreatment. Usually no sugar degradation products generation and liquid hot water, but less adverse effect for hydrolysis and aromatic compounds. than chemical. and fermentation. High: increase of 100% at Comminution: 5-25% increase in Increase of 20-70% in least. Sometimes hydrolysis yield. hydrolysis yield. Lower than Efficiency degradability > 80% is Steam explosion: most of other pretreatments. achieved. Comminution is used most of the More developed. High variety Comments times, before other of effect depending on the Lower energy input. pretreatments. specific pretreatment. (With information from: da Costa Sousa et al., 2009; Hendriks and Zeeman, 2009; Kim, 2013; Kumar et al., 2009; McMillan, 1994; Mosier et al., 2005; Zhang and Shahbazi, 2011; Zheng et al., 2014).
29
The most common biological pretreatment uses fungi that grow on the lignocellulosic biomass, secreting enzymes that degrade cell wall components (da Costa Sousa et al.,
2009; Kumar et al., 2009). White rot fungi are the preferred type of microorganisms for biomass pretreatment, because of their exceptional capacity to degrade lignin with their enzymatic pool of phenol oxidases, mainly laccases and peroxidases (Chen et al., 2010;
Wan and Li, 2012). Pretreatment with white rot fungi is usually performed in a solid-state aerobic process, where the fungus grows in the material and secretes the ligninolytic enzymes (Wan and Li, 2012), while some cellulose and hemicellulose degradation also occurs (Chen et al., 2010). This process is longer than traditional pretreatment processes
– it takes several days or weeks, instead of minutes or hours. It is performed in mild conditions of temperature (close to ambient temperature) and does not produce liquid waste (Wan and Li, 2012, 2010a). The topic of biological pretreatment is developed in higher depth in section 2.3.
2.3.2 Pretreatment of Miscanthus
Miscanthus is highly recalcitrant to enzymatic digestion with hydrolyses such as cellulases and xylanases, due to their considerable lignin content (Table 2.1) and the complex structure formed by the cell wall components. Thus, pretreatment is a required step prior to enzymatic hydrolysis to enhance the availability of polysaccharides to the hydrolytic enzymes. Several pretreatments have been studied for Miscanthus (Table 2.4), and most are performed in the presence of chemicals (e.g., acids, bases, salts, or solvents) and at high temperatures (usually higher than 100°C). Some of the pretreatments reduce
30 the lignin content of the material, while others reduce the hemicellulose content or crystallinity of the cellulose, or simply disrupt the complex arrangement of the cell wall components, allowing the hydrolytic enzymes to access the cellulose.
Table 2.4 Pretreatment of Miscanthus
Pretreatment Conditions Results* Reference
Dilute sulfuric acid 1% H2SO4, 160°C, 10 min. 96% digestibility. (Khullar et al., 2013) Dilute acid pre-soaking: 0.75% of Dilute sulfuric acid + 61% glucose yield. 95% H SO , 100°C, 14 h. Wet (Sørensen et al., 2008) wet explosion 2 4 xylose yield. explosion: 5 min, 170°C. 150-190°C, 10-40 min, 2-8% Dilute oxalic acid 46-91% glucose yield. (Scordia et al., 2013) H2C2O4 (w/w). Hot water 200°C, 30 min. 93% digestibility. (Khullar et al., 2013)
49% glucose yield. 53-72% 5-30% NH OH, 150-160°C, 5-60 (Khullar et al., 2013; Aqueous ammonia 4 digestibility. 74% min. Z. Liu et al., 2013) delignification.
Ammonia fiber 2:1 NH OH to biomass, 160°C, 5 96% glucose yield. 81% 4 (Murnen et al., 2007) expansion (AFEX) min. xylose yield. 69% glucose yield. 55% sugar yield. 70% Extrusion-Alkaline 12% NaOH, 70°C, 4 h. (de Vrije et al., 2002) digestibility. 77% delignification. Alkaline hydrogen 0.2-2M NaOH and 0.3-6M H O , 2 2 80-85% digestibility. (Hideno et al., 2013) peroxide 30°C 70°C, 3-21 h.
Inorganic salt 2% NH4Cl or 0.5% FeCl3, 200°C, 37-43% glucose yield. 40- (Kang et al., 2013a, pretreatment 15 min. 72% digestibility. 2013b)
Cellulose solvent-based 84% H PO , 50°C, 45 min. Then lignocellulosic 3 4 82% glucose yield. (Ge et al., 2011) mix with acetone. fractionation (CSLF)
First stage: 1% of NH4OH, 120°C, 15 min. Second stage: 86% sugar yield. 95% (Boonmanumsin et al., Two-stage microwave 1.78% H3PO4, 140°C, 30 min. glucose yield. 2012) (No enzymatic hydrolysis). 64-93% glucose yield. 75- (Brosse et al., 2009; Solvents: ethanol, or formic and Organosolv 98% digestibility. 80% Vanderghem et al., acetic (40%). 107-180°C, 1-3 h. delignification. 2012) *Digestibility (or enzymatic digestibility) refers to the sugars liberated after enzymatic hydrolysis compared with the sugars present in the pretreated material. Sugar yield refers to the sugars liberated at the end of the whole process (including enzymatic hydrolysis, if used) compared with the sugars present in the untreated feedstock. It can be referred to glucan (or glucose), xylan (or xylose), or the total sugar content.
31
Among the most efficient pretreatments are ammonia fiber expansion (AFEX)
(Murnen et al., 2007), organosolv (Brosse et al., 2009; Vanderghem et al., 2012), and cellulose solvent-based fractionation (CSLF) (Ge et al., 2011), with some pretreatment processes achieving sugar yields greater than 90% (Brosse et al., 2009; Khullar et al.,
2013; Murnen et al., 2007; Scordia et al., 2013). Using dilute oxalic acid pretreatment at the same severity factor, the glucose yield obtained for Miscanthus was about 91%
(Scordia et al., 2013).
2.4 Biological pretreatment
Biological pretreatment is mediated by microorganism and/or enzymes. This pretreatment is performed under milder conditions (near ambient temperature and pressure), thus requiring low energy and chemical inputs with reduced production of inhibitors (Saritha et al., 2012a). However, some aspects of biological pretreatment, such as incubation time and overall efficiency (i.e., sugar yield), still require improvement to make it a suitable alternative to thermochemical pretreatments (Wan and Li, 2012).
Most biological pretreatments take advantage of the ligninolytic enzyme system: a group of oxidoreductases that are able to degrade lignin, improving the degradation of biomass. This system includes laccases and peroxidases with high redox potential that oxidize the lignin polymer directly, or by the use of low-molecular-weight organic compounds (mediators) that can diffuse into the cell wall pores and attack the lignin structure (Martínez et al., 2005). Biological pretreatment can be performed with microorganisms that secrete ligninolytic enzymes, or with the enzymes directly, to
32 degrade part of the lignin and disrupt its structure, allowing access to holocellulose by hydrolytic enzymes that are routinely used in the biorefinery process to liberate sugars
(Chen et al., 2010).
Besides increasing the enzymatic digestibility of the feedstock, biological pretreatment can have other applications in the context of the biorefinery. The enzymes produced can be isolated and used in other processes; the pretreatment can produce coproducts, such as lignin-derivatives and organic compounds, that can act as platform chemicals, adding value to the process (Chen et al., 2010). The conversion of non- fermentable components of lignocellulosic biomass into value-added products constitutes one of the missing links required for the success of the biorefinery concept (Axelsson et al., 2012). In general, ligninolytic microorganisms and ligninolytic enzyme extracts can be used in the biorefinery as biocatalyst for many reactions for the production of biofuels and bioproducts.
2.4.1 Microbial depolymerization of lignin
Because of the complex and heterogeneous structure of lignin, microorganisms have developed a highly unspecific system to degrade it, based on oxidative reactions catalyzed by extracellular enzymes. This system is referred to as “enzymatic combustion”
(Kirk and Farrell, 1987) because unlike most enzymatic reactions, which tend to be specific and have a defined metabolic pathway, it is a group of nonspecific reactions that lead to numerous intermediate products, which vary significantly with the substrate, strain, and ambient conditions.
33
Microbial depolymerization can be done by fungi or bacteria, but the main type of microorganisms responsible for this process are wood decay fungi (Hatakka, 2005). They are mostly basidiomycetes that can overcome obstacles to growth such as low nitrogen content and the presence of toxic compounds (Martínez et al., 2005). The most effective lignin degradation microorganisms are the white rot fungi (Reid, 1995).
Microbial degradation of lignin is important in the biogeochemical carbon cycle, since lignin contains one third of the total carbon fixed by photosynthesis, but also because its degradation allows the subsequent hydrolysis of other carbon sources, such as cellulose (Ruiz-Dueñas and Martínez, 2009). Microbial depolymerization has a high potential for application in many industries that utilize lignocellulosic biomass, such as paper and biofuels production (Martínez et al., 2009).
2.4.2 Ligninolytic enzyme system
Lignin is an aromatic insoluble polymer formed by the oxidative polymerization of the monolignols p-coumaryl, sinapyl, and coniferyl alcohols that are the precursors of the phenylpropanoid lignin units: p-hydroxyphenyl (H), syringyl (S), and guaiacyl (G)
(Figure 2.6) (Martínez et al., 2005). Lignin polymerization is based on radical coupling mediated by peroxidases, generating a three-dimensional network that can be cross- linked with carbohydrates and proteins (Calvo-Flores and Dobado, 2010). This complex structure has low accessibility for enzymes. Even though the monomers of lignin are phenolic compounds, 80-90% of the lignin structure is non-phenolic, because during polymerization, most of the bonds use this phenolic position (C4-ether). Hence, the most
34 common bond is the ether β-O-4 bond, but the α carbon also forms ether bonds. C-C bonds are formed mostly between the β carbon and other carbons such as the 1, 5 or another β, or between the carbon 5 of each monolignol (Figure 2.7) (Reid, 1995). These bonds are not hydrolysable, which is why the microbial lignin depolymerization is based on oxidative reactions. The non-phenolic lignin requires a higher redox potential to be oxidized, which makes it more difficult to be degraded (ten Have and Teunissen, 2001).
Figure 2.6 Monolignols: precursors of lignin
Lignin heterogeneity is not only due to the intrinsic variability generated by the randomness in the polymerization process, but it is also related to the plant tissue and the species. Lignin from hardwood is composed of G and S units, while lignin from softwood is composed mainly of G units, and lignin from herbaceous plants contains the three lignin units G, S, and H (Abdel-Hamid et al., 2013).
35
Figure 2.7 Common bonds in lignin structure
The only microorganisms capable of extensively mineralizing lignin to carbon dioxide and water are white rot fungi. Thus, most of the enzymology studies in lignin degradation are based on the enzymatic machinery of these fungi. Several enzymes have been identified as being involved in lignin degradation, and include laccases, high redox ligninolytic peroxidases (lignin peroxidases (LiP), manganese peroxidases (MnP), and versatile peroxidases (VP)), and dye-decolorizing peroxidases (DyP). Along with these enzymes, some accessory enzymes are part of the multienzymatic lignin degradation process, and are implicated in different processes such as hydrogen peroxide production
(Hatakka, 2005) (Table 2.5). 36
Table 2.5 Ligninolytic enzyme system
Enzyme type Description Copper-containing phenol oxidases, not dependent on hydrogen peroxide. Laccases Degrade phenolic lignin directly and non-phenolic lignin through the laccase- mediator system. Class II heme- Most studied group including lignin peroxidases (LiP), manganese peroxidases peroxidases (MnP), and their hybrid versatile peroxidases (VP). Dye-decolorizing Newly discovered superfamily of peroxidases, capable of degrading high- peroxidases (DyP) redox potential substrates. Newly discovered superfamily of peroxidases that include unspecific Heme-thiolate peroxidases (UPO), which have shown catalytic activity with non-phenolic peroxidases lignin model compounds. Including hydrogen peroxide-producing enzymes (aryl-alcohol oxidase and Auxiliary enzymes the glyoxal oxidase) and dehydrogenases that reduce lignin degradation intermediates (aryl-alcohol dehydrogenases and quinone reductases).
Ligninolytic enzymes have developed two strategies to overcome the structural difficulties in accessing lignin: (1) the reaction takes place on the surface of the enzyme instead of in a “channel type” active site as occurs for most peroxidases; (2) low- molecular-weight mediators that diffuse into the lignin are produced (Ruiz-Dueñas and
Martínez, 2009). The first strategy uses a surface amino acid that can form a stable radical (tryptophan, histidine, or tyrosine) and is also connected to the heme group that is present in all peroxidases (Ruiz-Dueñas and Martínez, 2009). This allows an easy coupling with lignin that is exposed to the enzyme. The second strategy uses mediators, or low-molecular-weight compounds, that form stable free radicals, which are produced by the enzyme and can diffuse into the lignocellulosic matrix to oxidize lignin
(Leonowicz et al., 1999). Most of the studied mediators are synthetic molecules that were evaluated in-vitro. Recently, some natural mediators have been discovered; most of them are actually lignin precursors or, even more interesting, lignin degradation products.
37
Thus, it has been hypothesized that lignin degradation acts as a chain reaction, using the degradation products as mediators to continue the oxidation (Christopher et al., 2014).
Peroxidases that degrade lignin are the most studied group of ligninolytic enzymes, because most of them have high redox potential. They contain a heme group and use a similar mechanism as other common peroxidases (ten Have and Teunissen, 2001). LiP,
MnP, and VP belong to class II of the superfamily of non-animal peroxidases, which are secreted by fungi. DyPs constitutes a different superfamily of peroxidases (Martínez et al., 2014) produced by fungi and bacteria. LiP and MnP were discovered in the 1980s
(Glenn et al., 1983; Kuwahara et al., 1984; Tien and Kirk, 1983), and the first VP was not discovered until the end of the 1990s (ten Have and Teunissen, 2001). The first DyP was described in 1999 and named after its ability to decolorize dyes (Kim and Shoda, 1999), but the role of this superfamily in lignin degradation has been recognized only recently
(Ahmad et al., 2011; Liers et al., 2010). Laccases are a group of phenoloxidases produced by fungi and bacteria (Leonowicz et al., 2001). Despite the fact that laccases have been known for many years, their role in lignin degradation was only recognized after the role of mediators was identified in the 1990s (Hatakka, 2005). Some accessory enzymes have been used to complement the lignin degradation process by producing hydrogen peroxide for the peroxidases or by reducing the radicals generated by the enzymes (Abdel-Hamid et al., 2013).
38
2.4.2.1 Laccase
Laccases have been studied since the 19th century, but their role in lignin degradation was not recognized until the 1990s (Baldrian, 2006). Laccases are a diverse group of phenol oxidases involved in many types of reactions, including lignin biodegradation, but also lignin polymerization (Munk et al., 2015). These enzymes have four copper ions that are fully oxidized on their active site. Laccases catalyze the oxidation of the substrate with one electron at a time, and the full catalytic cycle includes the transfer of four electrons to the four copper ions (Christopher et al., 2014), coinciding with the reduction of O2 to H2O (ten Have and Teunissen, 2001). Laccases can attack phenolic lignin forming a phenolxyl radical (Figure 2.8), but they cannot oxidize non-phenolic lignin directly because of their low redox potential. For this purpose, laccases can oxidize low- molecular-weight mediators, which act as electron carriers that diffuse into the insoluble lignin structure to oxidize it (Christopher et al., 2014), a mechanism that is represented in
Figure 2.9. In this way, laccases can depolymerize phenolic and non-phenolic lignin.
Some fungi produce laccases as their sole ligninolytic enzyme (Eggert et al., 1997).
Figure 2.8 Formation of phenoxyl radicals by laccases in phenolic lignin
39
Figure 2.9 Mechanism of lignin oxidation by laccase using mediators
Low-molecular-weight mediators are small chemical compounds oxidized by the laccases into relatively stable intermediates that can penetrate the plant cell wall pores.
These intermediates have a higher redox potential than laccases, and can oxidize lignin and then return to their original form. Phenolic lignin, which is less than 20% of the entire lignin structure, is more susceptible to degradation, so it probably gets degraded first, generating a set of phenolic residues that can act as natural mediators for the oxidation of non-phenolic lignin (Christopher et al., 2014). Some of those mediators can be syringaldehyde, acetosyngone, p-coumaric acid, vanillin, acetovanillin, and methyl vanillate, which are derived from the three phenylpropanoid lignin units (Camarero et al.,
2005). For in vitro applications of laccases, synthetic mediators have been studied, including 2,2’-azinobis(3-ethylbenzthiazoline-6-sulfonate) (ABTS) (Bourbonnais et al.,
1997; Bourbonnais and Paice, 1990) and 1-hydroxybenzotriazole (HBT) (Bourbonnais et
40 al., 1997; Call, 1994). For biorefinery purposes, the use of natural mediators has more potential than synthetic mediators, because they have a lower cost, less negative environmental impact, and are less hazardous (Baldrian, 2006).
Fungal laccases are usually produced as isoenzymes. Most of the fungal laccases are extracellular enzymes, because of the insoluble nature of their substrate (lignin). But some laccases have been found to be intracellular, which may play a role in oxidizing low-molecular compounds (Baldrian, 2006). Laccases are glycosylated enzymes, which improves their stability but hinders their production in heterologous systems (Baldrian,
2006).
Laccases do not need hydrogen peroxide to complete oxidation; instead they use atmospheric oxygen as the final electron acceptor (ten Have and Teunissen, 2001). Also, laccase production is not inhibited by high nutrient content, as it is for many peroxidases
(ten Have and Teunissen, 2001). These characteristics make them advantageous for commercial production and application. Even though commercial application of laccases is challenged by their low redox potential compared with peroxidases and issues of low stability and catalytic activity, these difficulties can be addressed by using directed evolution and hybrid genomics strategies (Alcalde, 2015; Bugg et al., 2011b).
2.4.2.2 Lignin peroxidase (LiP)
LiP was discovered in the white rot fungus Phanerochaete chrysosporium (Glenn et al., 1983; Tien and Kirk, 1983). It has a wide range of substrates, including phenolic and non-phenolic lignin. It can attack non-phenolic lignin directly because of its unusual high
41 redox potential, generating cation radicals (Figure 2.10) that can cleavage α-β carbon bonds and mediate ring opening reactions (Reid, 1995). LiPs have a low optimum pH and are usually secreted as a group of isoenzymes. While LiPs have been considered to play an important role in lignin degradation by fungi in nature, it has been discovered that some white rot fungi do not produce any LiP but still can degrade lignin significantly, showing the specificity and differentiation of organisms in lignin degradation systems
(Furukawa et al., 2014).
Figure 2.10 Formation of cation radicals by lignin peroxidases in non-phenolic lignin
LiP has a characteristic catalytic cycle typical of heme-containing peroxidases, as illustrated in Figure 2.11. The enzyme in its resting state (LiP, with Fe3+) is oxidized by a
4+ hydroperoxide (e.g., H2O2) into compound I (C-I), with Fe -oxo and a porphyrin cation radical remaining. Then, compound I is reduced in a two-step process by two substrate molecules (S), such as veratryl alcohol (VA) (Ruiz-Dueñas and Martínez, 2009).
42
Figure 2.11 Catalytic cycle of peroxidases
VA’s role in LiP activity is still a matter of controversy. It is clear that VA can be a substrate for LiP, generating the cation radical VA+•. Also, it is known that VA prevents hydrogen peroxide inactivation of LiP (ten Have and Teunissen, 2001), which is probably its main role (Sigoillot et al., 2012). Finally, VA has also been considered to be a mediator because it generates a radical that could oxidize lignin; but this function is not completely clear because of VA radicals’ short half-life in solution (Ruiz-Dueñas and
Martínez, 2009).
LiP has a heme channel where only small substrates can enter to transfer the electron directly to the heme group, but the lignin polymer cannot access that active site. For that purpose, the enzyme possesses a long-range electron transfer system, which uses a tryptophan on the surface of the enzyme, forming a free radical that interacts with the lignin polymer and then transfers the electron to the heme (Doyle et al., 1998).
43
2.4.2.3 Manganese peroxidase (MnP)
MnPs were also discovered in the white rot fungi P. chrysosporium (Kuwahara et al.,
1984). They are heme peroxidases usually secreted as isoenzymes, and are actually the most common type of ligninolytic enzymes in white rot fungi (Floudas et al., 2012;
Hatakka, 2005). MnPs convert Mn2+ into Mn3+, which diffuses away from the enzyme in complex with chelators such as oxalate or malonate. This Mn3+-chelator complex oxidizes only phenolic compounds, because of its low redox-potential (Figure 2.12)
(Wong, 2009). In the presence of unsaturated lipids or thiols, MnP can use their peroxidation radicals as mediators to oxidize non-phenolic lignin (ten Have and
Teunissen, 2001). The generation of Mn3+ follows the generic catalytic cycle of peroxidases (Figure 2.9) (Paliwal et al., 2012).
Figure 2.12 Formation of phenoxyl radicals by manganese peroxidase in phenolic lignin
2.4.2.4 Versatile peroxidase (VP)
VPs, first discovered in Pleurotus eryngii (Martínez et al., 1996), are heme peroxidases and are considered to be hybrid enzymes of LiP and MnP (Hatakka, 2005),
44 because they have both the ability to attack lignin directly like LiP and to oxidize Mn2+ like MnP, acting in both phenolic and non-phenolic substrates. VPs have surface amino acids with the potential to generate radicals that can attack bulky lignin directly and perform long-range electron transfer, as mentioned before for LiP. VPs are produced by a limited number of white rot fungi, with a small amount of isoenzymes (Floudas et al.,
2012). The potential for industrial application of VPs has been acknowledged based on their catalytic versatility, and modifications have been made to improve their stability towards temperature and pH and to modify their affinity to different substrates (García-
Ruiz et al., 2012). Also, other peroxidases (MnP and LiP) with better stability have been modified to emulate the functions of VPs (Fernández-Fueyo et al., 2014; Mester and
Tien, 2001).
2.4.2.4 Dye-decolorizing peroxidase (DyP)
Most of the DyPs identified today are bacterial and some have been found in fungi and archaea (Colpa et al., 2014). DyPs are a superfamily of heme peroxidases that show a particular dissimilarity in sequence and structure with the other peroxidases. They can oxidize most of the traditional substrates of the peroxidases, but also a wide variety of other organic compounds, including high redox potential dyes. They have shown potential to oxidize non-phenolic lignin model compounds (Liers et al., 2010), kraft lignin (Ahmad et al., 2011), and lignin in lignocellulosic substrates (Salvachúa et al.,
2013).
45
DyPs have certain characteristics that make them unique compared to other peroxidases: they work at very low pH, have a wide variety of substrates, have a heme group that is bound non-covalently, and some have shown hydrolytic activity in addition to the oxidative activity (Colpa et al., 2014; Sugano, 2009). Their physiological function in the microorganisms is still unclear, but it has been suggested that they can play a key role in lignin degradation by bacteria (Bugg et al., 2011b). Their mechanism of action is still not completely elucidated, but a tyrosine residue on the surface with the potential to form a catalytic site for long-range electron transfer has been found (Strittmatter et al.,
2013a, 2013b). It appears that they also use a mediator system, similar to class II peroxidases. Recently, some DyPs capable of oxidizing Mn2+ have been reported
(Fernández-Fueyo et al., 2015).
DyPs have a high potential for industrial applications because of their unique ability to convert a wide range of organic compounds, such as synthetic dyes and β-carotene, which have industrial interest. Also, bacterial DyPs present advantages in their heterologous expression and modification compared with fungal enzymes (Colpa et al.,
2014).
2.4.2.5 Accessory enzymes
Several enzymes have been identified as playing an important role in the lignin depolymerization pathway. The main group of enzymes that assist the lignin degradation by peroxidases are the hydrogen peroxide-producing enzymes, such as aryl-alcohol oxidase and the glyoxal oxidase (Alcalde, 2015). Hydrogen peroxide is indispensable for
46 the formation of compound I in the peroxidases. Some dehydrogenases reduce compounds derived from the lignin degradation, avoiding re-polymerization, and include aryl-alcohol dehydrogenases and quinone reductases (Dashtban et al., 2010; Guillén et al., 1997).
The role in lignin degradation of a group of enzymes that belong to another superfamily of peroxidases, the heme-thiolate peroxidases, has been explored recently.
These enzymes are known as unspecific peroxidases (UPOs) or aromatic peroxygenases
(APOs) and have a high redox potential. They are able to catalyze a wide range of oxidizing reactions, including the cleavage of ether bonds, which confers to them the ability to degrade non-phenolic lignin model compounds (Kinne et al., 2011). More than
300 substrates have been found for these enzymes, including several reactions of industrial interest such as hydroxylation of alkenes and epoxidation of olifens (Alcalde,
2015).
2.4.3 Fungal pretreatment
2.4.3.1 White rot fungi
As stated above, white rot fungi are the most studied microorganisms for the pretreatment of lignocellulosic biomass, due to their capacity for lignin degradation and their unique ability to extensively mineralize lignin to carbon dioxide and water (Kirk and Farrell, 1987), albeit their inability to use lignin as a sole source of carbon and energy
(Sánchez, 2009).
47
White rot fungi are a diverse group of wood rot basidiomycetes that can degrade all the components of the lignocellulosic biomass, producing a characteristic bleached and fibrous appearance in the material (Martínez et al., 2005). These fungi produce two different rotting patterns: selective delignification and simultaneous rot. Selective rotting fungi degrade lignin preferentially over cellulose; whereas, simultaneous rotting fungi degrade the holocellulose while simultaneously degrading lignin (Otjen and Blanchette,
1986). Selective degraders are preferred for pretreatment of biomass, given that holocellulose is the primary substrate for the production of biofuels and bioproducts in biorefineries (Wan and Li, 2012).
Lignin depolymerization is considered a secondary metabolic process for white rot fungi, not required for fungal growth (Paliwal et al., 2012). White rot fungi are more frequently found in hardwood than softwood and preferentially degrade S units over G units (Hatakka, 2005).
Not every species of white rot fungi contains all the enzyme types and even when they have similar enzymatic systems, the strategy of degradation can be different
(simultaneous rot or selective delignification), which implies that there is more than one approach to successfully degrade lignin (Reid, 1995).
The most studied white rot fungus is P. chrysosporium. This fungus has a strong ligninolytic system that follows a simultaneous rot pathway. As a result, it is not the most effective for biomass pretreatment. Many other fungi have also been studied and applied for fungal pretreatment of lignocellulosic biomass, including Pleurotus ostreatus,
48
Ceriporiopsis subvermispora, Irpex lacteus, and Trametes versicolor (Canam et al., 2011;
Taniguchi et al., 2005; Wan and Li, 2010a).
Numerous studies have focused on the screening of several fungi to pretreat a specific feedstock (Hatakka, 1983; Müller and Trösch, 1986). These screenings usually showed better results with selective lignin degraders, such as Pleurotus spp. (Hatakka, 1983;
Müller and Trösch, 1986). Salvachúa et al. evaluated 21 basidiomycetes to pretreat wheat straw and obtained the highest glucose yields of 69% and 66% with C. subvermispora and I. lacteus, respectively, after 21 days of fungal pretreatment. However, the lignin degradation of 30-35% that these fungi caused was lower than the 46% delignification obtained with Panus triginus and T. versicolor, but the cellulose degradation for the later was also high (about 23%), so the overall sugar yield was lower (Salvachúa et al., 2011).
Zhang et al. evaluated 34 white rot fungi for the pretreatment of bamboo culms and found that, after 120 days of fungal pretreatment with Echinodontium taxodii, 29% of lignin was degraded selectively over cellulose, and the sugar yield was increased by 6.7-fold
(Zhang et al., 2007).
Table 2.6 summarizes some of the most recent results on the fungal pretreatment of lignocellulosic feedstocks with white rot fungi. It is noticeable that most of these studies were no longer performed with P. chrysosporium, probably due to the degradation of hollocellulose caused by the simultaneous rotting system. It is also noteworthy that, when fungi with selective delignification systems, such as C. subvermispora, I. lacteus, and
Trametes hirsuta were employed, sugar yields close to 70% were achieved (Song et al.,
2013a; Sun et al., 2011; Wan and Li, 2010b). Moreover, fungal pretreatment in
49 combination with mild pretreatments such as alkaline (López-Abelairas et al., 2013;
Salvachúa et al., 2011) or ethanol-based organosolv pretreatment (Canam et al., 2011) achieved similar sugar yields to those of thermochemical pretreatments, around 90-100%.
This suggests that the strategy of combining different pretreatments and/or adding mediators or other chemicals to enhance ligninolytic activity can improve the yields of the fungal pretreatment to a point where it can be economically feasible.
Most of the research reports the enzymatic digestibility (based on the sugars present in the pretreated material) instead of the overall sugar yield, and for some studies it is not clear what parameters are being reported. The overall sugar yield includes the effect of the sugars degraded during fungal pretreatment, which can be significant, especially when simultaneous rot fungi, such as P. chrysosporium or T. versicolor, are used.
Comparison of fungal pretreatments should be performed using overall sugar yield, and not enzymatic digestibility, unless the pretreatment is performed using selective degraders that produced insignificant cellulose degradation.
50
Table 2.6 Recent studies in fungal pretreatment of lignocellulosic biomass
Fungus Feedstock Conditions Results* Reference (Cianchetta et Corn stover, al., 2014; Ge et 18-48 days, switchgrass, al., 2015; Ceriporiopsis 28°C, 60-75% 24-40% lignin degradation, wheat straw, Salvachúa et al., subvermispora moisture 37-69% sugar yield Albizia 2011; Wan and
(hardwood) Li, 2011b, 2010b) Optimized Dichomitus liquid media, 58% enzymatic Rice straw (Bak et al., 2010) squalens 15 days, 29°C, digestibility 150 rpm (López-Abelairas 13-34% lignin degradation. Wheat et al., 2013; 14-28 days, 66-84% glucose yield. 61- Irpex lacteus straw, corn Salvachúa et al., 28°C** 100% enzymatic stover 2011; Song et digestibility. al., 2013a) (Zeng et al., Wheat 28-34% lignin degradation. Phanerochaete 12-21 days, 2011; Zhao et straw, 41-47% enzymatic chrysosporium 29-30°C al., 2012; Zhi cornstalk digestibility and Wang, 2014) 30% lignin reduction, 90% Pycnoporus sp. enzymatic digestibility Switchgrass 36 days, 30°C (Liu et al., 2013) SYBC-L3 (compared with 75% for untreated material) NA-71% lignin (Saritha et al., Trametes Paddy straw, 10-42 days, degradation. 53-74%*** 2012b; Sun et hirsuta corn stover 30°C enzymatic digestibility al., 2011) Poplar wood 47% lignin degradation. Trametes (Wei Wang et (Populus 84 days, 28°C 41% enzymatic orientalis al., 2014) tomentosa) digestibility Trametes 90% enzymatic versicolor Canola 84 days, 62% digestibility (compared (Canam et al., (cellobiose straw moisture**** with 45% for untreated 2011) dehydrogenase- material) deficient) Note: different conditions (cellulase concentration and time) were used for enzymatic hydrolysis of pretreated material. NA: not available *Enzymatic digestibility refers to the sugars liberated after enzymatic hydrolysis, compared with the sugars present in the pretreated material. Sugar yield refers to the sugars liberated at the end of the whole process (pretreatment and enzymatic hydrolysis) compared with the sugars present in the untreated feedstock. Both can be referred to as glucan (or glucose), xylan (or xylose), or the total sugar content. **Followed by an alkaline treatment (0.1% w/v NaOH, 50°C, 1 h) or adding 0.01 mM/g of Mn. ***Followed by an alkaline wash (0.1% NaOH). ****Followed by an organosolv pretreatment at 100°C with 60% acidify ethanol.
51
Fungal pretreatment can be performed in solid or liquid state, but the former is preferred, mostly because it simulates the natural environment, but also because it allows a higher substrate loading and does not generate liquid waste streams. Factors such as moisture content, temperature, and aeration have a significant effect on fungal growth and lignin depolymerization. The biomass particle size also influences the process, mostly because an adequate particle size allows the optimal area of exposure without blocking the air flux, which is indispensable because of the highly aerobic nature of the oxidative process. Decontamination of the feedstock is usually needed before fungal inoculation, because white rot fungi can be outcolonized easily by other microbes in the biomass (Wan and Li, 2012).
2.4.3.2 Ceriporiopsis subvermispora: a special white rot fungus
C. subvermispora is a white rot basidiomycetes naturally associated with white rot conifers and hardwood. It was first described by Albert Pilát as Poria subvermispora in
1940 (Gilbertson and Ryvarden, 1985). C. subvermispora is one of the most studied white rot fungi for its industry application potential in pulping and biomass pretreatment, since it degrades lignin selectively over cellulose (Wan and Li, 2012). This fungus is capable of mineralizing lignin, which is usually enhanced under conditions of nitrogen limitation (Ruttimann-Johnson et al., 1993). C. subvermispora has a robust ligninolytic system that is mainly composed of MnPs and laccases, and only recently, with the availability of the full genome, two enzymes with LiP capabilities have been discovered
(Fernández-Fueyo et al., 2012b). Compared with P. chrysosporium, C. subvermispora
52 has a stronger ligninolytic system that includes 13 MnPs and at least seven lacasses, plus the two new putative LiPs (or evolutionary intermediates between LiPs and VPs) and nine heme-thiolate peroxidases. In contrast, P. chrysosporium has no reported laccases, five MnPs, and 10 LiPs. However, P. chrysosporium has a stronger cellulosome that is up-regulated in the presence of lignocellulosic substrates, which helps explain the differences in degradation patterns (Fernández-Fueyo et al., 2012a).
C. subvermispora has been used to successfully pretreat lignocellulosic biomass, to enhance enzymatic hydrolysis for fermentation (Wan and Li, 2010a) and methane yield in anaerobic digestion (Zhao et al., 2014b), resulting in significant lignin degradation and enhanced enzymatic digestibility. Although some parts of the hemicellulose are also degraded, most of the cellulose remains intact (Wan and Li, 2011b). Glucose yields between 24% and 69% have been obtained for fungal pretreatment with C. subvermispora of feedstocks such as hardwood, corn stover, and switchgrass, but this pretreatment has been unsuccessful with other feedstocks such as soybean and wheat straw (Cianchetta et al., 2014; Ge et al., 2015; Salvachúa et al., 2011; Wan and Li, 2011b,
2010b). Fungal pretreatment of corn stover during wet-storage showed 2- to 3-fold increase in enzymatic degradability compared to silage, and similar results to alkaline pretreatment of corn stover (Cui et al., 2012a). Fungal pretreatment usually needs sterilization prior to inoculation with the fungus; however, Zhao et al. demonstrated that fungal pretreatment with C. subvermispora is possible without sterilization of yard waste, if material previously colonized with the fungus is used as inoculum instead of the pure mycelium (Zhao et al., 2014a).
53
2.4.3.3 Brown rot, soft rot, and other fungi
Brown rot fungi have also been used for microbial pretreatment of lignocellulosic biomass. These basidiomycetes degrade carbohydrates at a high rate, but they can modify lignin only to some extent (Eriksson et al., 1990). Because of the selective holocellulose degradation, the relative lignin content increases, but they still reduce the recalcitrance of the material to biodegradation and can enhance the enzymatic digestibility, probably by increasing porosity and sometimes reducing cellulose crystallinity and degree of polymerization. Most brown rot fungi degrade cellulose to such a high extent that it does not compensate for the enhanced enzymatic digestibility, resulting in a lower overall sugar yield. Some other brown rot fungi such as Gloeophyllum trabeum have been found to produce a higher overall sugar yield under some conditions. G. trabeum is considered a plant pathogen and it is the best understood brown rot fungus (Cohen et al., 2005), including its genome that has been completely sequenced (Floudas et al., 2012). It has been used to pretreat corn stover, achieving a sugar yield of about 45% after 20 days
(Gao et al., 2012). It was also used to pretreat softwood (Pinus radiata and Eucalyptus globulus), obtaining an enzymatic digestibility of about 14%, but the overall sugar yield based on the original amount of glucan was not calculated (Monrroy et al., 2011). G. trabeum was also used to pretreat spruce and pine, enhancing enzymatic digestibility to
25% for spruce and about 12% for pine but, after adjusting for carbohydrate mass loss, the overall sugar yields were actually about 12% and 8%, respectively (Schilling et al.,
2009). In the most successful pretreatment, a 51% overall sugar yield was achieved from aspen pretreated with G. trabeum for 2 weeks (Schilling et al., 2012).
54
Other brown rot fungi have also been successfully used for pretreatment of lignocellulosic biomass. Coniophora puteana and Postia placenta were used to pretreat softwood (pine), achieving sugar yields of around 20% after 35 days (Ray et al., 2010).
Similarly, the enzymatic digestibility of 6-week fungal pretreated pine was about 30% when C. puteana was used (Vaidya and Singh, 2012). Fomitopsis pinicola was used to pretreat pine and spruce, but only around 7% overall sugar yield was obtained (Schilling et al., 2009).
Soft rot fungi and other microfungi have also been studied for the potential to degrade lignin in lignocellulosic biomass. They can degrade hollocellulose and some have the ability to alter lignin structure or even mineralize it (Hatakka, 2005). Even though their ability to degrade lignin has been known for decades (Haider and Trojanowski, 1975), there is insufficient knowledge about their ligninolytic enzymes system and not much work has been done to apply these microorganisms for fungal pretreatment of lignocellulosic biomass. The mycromycete Myrothecium roridum was shown to extensively delignify paddy straw and the herbaceous weed Parthenium sp. (more than
50% lignin degradation), while leaving most of the cellulose intact and significantly enhancing the enzymatic digestibility in only 7 days of incubation (Tiwari et al., 2013).
2.4.4 Bacterial pretreatment
Some bacteria, most of which are present in soil, also have the ability to degrade lignin. They grow faster than most fungi and degrade the lignin into small water-soluble fragments that can be recovered and used as value-added products (Hatakka, 2005),
55 constituting a potential for enzyme production that can be applied in the context of the biorefinery.
One of the best characterized bacteria capable of lignin depolymerization is the α- proteobacteria Sphingobium sp. SYK-6. The degradation pathways for biaryls and monoaryls have been extensively studied, and were reviewed by Masai et al. (Masai et al., 2007), but the bacterial ligninolytic systems are still not well elucidated (Bugg et al.,
2011a). Other bacteria, such as Pseudomonas putida mt-2 and Rhodococcus jostii RHA1, have also been identified to degrade lignin in lignocellulosic biomass. Depolymerization of lignin by these strains is not dependent on hydrogen peroxide, which suggests that their ligninolytic system is probably composed of laccases or other non-peroxidases
(Ahmad et al., 2010).
Several bacteria from the genus Streptomyces have been shown to degrade around
15% of lignin (Hatakka, 2005). Bacterial pretreatment was performed on paddy straw for
10 days with Streptomyces griseorubens, achieving a very promising enzymatic digestibility of 97.8% (Saritha et al., 2013). Cellulomonas uda was found to have a potential for microbial pretreatment of sugarcane bagasse, according to lignin degradation and cellulose solubilization, but no enzymatic hydrolysis was performed in this study (Singh et al., 2007). As investigations on the use of ligninolytic bacteria in the context of biofuel production are limited (Saritha et al., 2013), most existing studies have not evaluated the effect of bacterial pretreatment on the enzymatic hydrolysis; however, research on understanding the mechanisms behind delignification by bacteria is increasing (Bugg et al., 2011b).
56
2.4.5 Enzymatic pretreatment
Instead of growing the microorganisms directly in the biomass, the enzymatic extracts can be added to the feedstock to depolymerize lignin. This approach can potentially reduce the pretreatment time and the carbohydrate degradation, while simplifying the processing (Chen et al., 2010). This technology is usually not as effective as the microbial pretreatment, mainly because of the difficulties associated with heterogeneous catalysis and the complexity of the ligninolytic enzyme system, which usually requires synergy of several enzymes and the presence of mediators for effective lignin degradation. Enzymatic pretreatment also requires an efficient process for the production of the enzymes. However, production of these enzymes usually depends on specific nutrient depletion (such as nitrogen) and is related to secondary metabolism, not primary growth. As a result, most of the lignin-degrader microorganisms naturally produce low amounts of ligninolytic enzymes and yields are unstable due to the complicated regulatory mechanisms (Singh and Chen, 2008). Also, even though ligninolytic enzymes are extracellular, their recovery represents a challenge. Recent developments in understanding ligninolytic systems (fungal genomes, secretomes, and biochemical pathways) (Camarero et al., 2014), plus technologies to produce the enzymes in heterologous systems and to improve the enzymes by direct evolution, (Alcalde, 2015;
Martínez et al., 2014) may help make enzymatic pretreatment a viable option for the biorefineries.
Laccase-mediator systems are probably the most promising for direct enzymatic pretreatment, because they do not required hydrogen peroxide and have been proved for
57 biopulping (Chen et al., 2010). Rico et al. were able to degrade up to 50% of the lignin from wood chips (Eucalyptus globulus) by adding a commercial fungal laccase from the recombinant Myceliophthora thermophila and a mediator for the enzyme (methyl syringate), followed by an alkaline peroxide extraction (1% of NaOH and 3% H2O2). An increase of 41% of glucose yield was obtained with this treatment, with respect to the untreated sample (Rico et al., 2014).
LiP and MnP from P. chrysosporium did not produced any changes in total mass and lignin content when used singly in aqueous media with lignin extracted from hardwood; however, when used together, these enzymes showed a synergistic effect producing 11% of mass loss and 5% reduction of lignin content (Thompson et al., 1998). Crude enzymatic extracts from P. chrysosporium and Coridus versicolor increased the sugar yield from hydrolysis of corn stover by 50.2%, with partial removal of the lignin (Wang et al., 2013). Enzymatic extracts from white rot fungi T. versicolor and Bjerkandera adusta were able to increase the enzymatic digestibility of wheat straw by 13%, although no changes in lignin content were observed (Rodrigues et al., 2008).
The use of enzymes for lignin degradation has the potential for combining two biorefinery steps: pretreatment and enzymatic hydrolysis. The main shortcoming of this approach is the observed potential of oxidative enzymes, such as laccases, to inactive hydrolases (Bendl et al., 2008; Wang et al., 2013). However, synergy was detected between ligninolytic enzymes and hydrolases when a crude extract of the fungus
Cladosporium cladosporioides, which included laccase and MnP activities, was mixed
58 with commercial xylanase, obtaining about a 5-fold increase in sugar release from bagasse compared with both types of enzymes used alone (Ji et al., 2014).
2.4.6 Challenges and perspectives of biological pretreatment
Biological pretreatment of lignocellulosic biomass has potential in the context of plant-based biorefineries. However, some challenges must be overcome to make this process commercially feasible. Pretreatment time is still one of the main concerns for biological pretreatment. Fungal growth is slow and several weeks are usually needed for an effective pretreatment (Wan and Li, 2012). Enzymatic pretreatment can be faster because it does not depend on the growth of the microorganisms on the biomass (Chen et al., 2010). Also, a combination of microbial pretreatment with other mild physicochemical pretreatments, such as alkaline or organosolv, can help reduce the total pretreatment time (Wan and Li, 2012). Because lignocellulosic biomass usually requires storage time due to material handling logistics, combining wet-storage with fungal pretreatment offers a solution for the time concern, and also may avoid drying of the biomass feedstocks (Wan and Li, 2010b).
Microbial pretreatment usually needs sterilization of the biomass prior to inoculation, because ligninolytic microorganisms can easily be outcolonized by indigenous microbes present in the feedstock. This process is usually energy intensive and requires an additional step in the process (Wan and Li, 2012). Zhao et al. showed that inoculating unsterilized yard waste with pre-colonized feedstock instead of a pure culture of the
59 white rot fungus resulted in a successful pretreatment that yielded equivalent results than the fungal pretreatment under sterile conditions (Zhao et al., 2014a).
Sugar loss during the microbial pretreatment is also a concern. Most microbes cannot use lignin as carbon and energy source, thus in order to grow in the biomass, they need to use part of the cellulose or hemicellulose. Ideal biological pretreatment should minimize this loss because of the high value of these polysaccharides for the biorefinery process.
To minimize the cellulose loss, selective degraders are usually chosen for fungal pretreatment, instead of simultaneous rotting fungi. The use of enzymes instead of microorganisms for the pretreatment can avoid the degradation of the holocellulose, and even allow combining the lignin degradation step with the enzymatic hydrolysis of the polysaccharides (Chen et al., 2010).
Recent advances in genomics, proteomics, and metabolic engineering have allowed the design of specific systems that enable not only the optimization of the existing systems but the development of new technologies for the integral utilization of the plant biomass. Modified VPs have been developed by site-directed mutagenesis to remove specific phenolic compounds (Camarero et al., 2014). Laccase and VP variants have been produced by directed molecular evolution, showing improved stability and secretion yield
(Camarero et al., 2012; García-Ruiz et al., 2012; Maté et al., 2010).
During the last two decades, a significant amount of knowledge about the natural ligninolytic system has been developed, including details about the functionality of the well-known class II heme peroxidase enzymes and the discovery of the new superfamilies DyP and heme-thiolate peroxidases (Martínez et al., 2014). The possibility
60 of heterologous expression of these enzymes and their modification to improve desired properties, such as catalytic activity and stability in industrial conditions, present opportunities for application in several operations in the biorefinery, including but not limited to pretreatment. Another interesting prospective is the control of the depolymerization process by engineering homogenous streams that allow efficient recovery of value-added chemicals derived from lignin (Fisher and Fong, 2014). These tailored systems can be used for detoxification after thermochemical treatment, in the production of value-added chemicals and organic compounds, and to improve exploitation of lignocellulosic biomass (Camarero et al., 2014). The next step is to develop consolidated bioprocessing microorganisms with ligninolytic and cellulolytic ability, capable of completely processing the biomass into fuels and chemicals (Alcalde,
2015). These microorganisms would have to include a complete ligninolytic system with the accessory enzymes, in addition to hydrolytic and fermentation machinery.
2.5 Anaerobic digestion
Anaerobic digestion (AD) is the microbial decomposition of organic matter in an oxygen-depleted environment. The main products of AD are biogas, which is a mixture of gases, mainly methane and carbon dioxide, and a nutrient rich effluent that can be used as fertilizer (Korres et al., 2013). AD has been used for decades as a waste stabilization process, but more recently it has been considered for its potential for bioenergy production (Sawatdeenarunat et al., 2016), since the biogas, composed of 40% to 70% of
61 methane, can be burned directly for heat and power, or can be upgraded for use as a transportation fuel (Korres et al., 2013).
AD is performed by a community of microorganisms that complete different tasks in the conversion phases of the organic matter into biogas. First, large organic molecules, such as carbohydrates, lipids, and proteins, are hydrolyzed into their respective monomers. Second, these monomers are converted into volatile fatty acids during the phase called acidogenesis. Third, in the acetogenesis phase volatile fatty acids are transformed into acetic acid, carbon dioxide, and hydrogen. Finally, methanogens convert the acetic acid and some of the hydrogen into methane and carbon dioxide (Mao et al.,
2015).
AD has several advantages over other waste treatment and/or bioenergy processes.
AD is ideally for feedstocks with high water content because it is performed in an aqueous environment, unlike combustion or gasification. It produces less effluent and removes microbial pathogens more effectively than aerobic treatments. Also, in comparison with leaving the organic matter untreated, AD reduce odors and greenhouse gases emissions (Abbasi et al., 2012), especially methane and nitrous oxide (Clemens et al., 2006), which have 25 and 298 times more global warming potential than carbon dioxide in a 100-year timescale. AD produces both a fuel (biogas) that burns cleaner than combusting the biomass directly, and a nutrient rich fertilizer, instead of just one or the other (KC et al., 2014).
AD is a mature technology mostly used for treatment of solid waste, such as the organic fraction of municipal solid waste and sewage sludge. It is also performed with
62 other substrates depending on location and availability, including animal manure, food and feed waste, and lignocellulosic biomass (agricultural residues and energy crops)
(Korres et al., 2013). Due to its abundance, the use of lignocellulosic biomass as a feedstock can enhance the AD ability to be a reliable source of renewable energy (Liew et al., 2012). This lignocellulosic biomass could be waste of the agroindustry, such as corn stover or wheat straw, but recently other feedstocks grown exclusively for energy purposes have being gaining attention, such as switchgrass, giant reed, and Miscanthus
(Klimiuk et al., 2010; Mayer et al., 2014; Menardo et al., 2013; Yang and Li, 2014).
Currently, the AD technology is commercially available in the U.S., particularly the liquid anaerobic digestion (L-AD), which is usually operated within 0.5-15% of total solids (TS). The solid state anaerobic digestion (SS-AD) systems generally operate within
15-40% of TS, which makes them more suitable for processing lignocellulosic biomass
(Li et al., 2011). Some of the advantages of the SS-AD over the L-AD systems are the higher volumetric biogas productivity, less energy required for heating and no required stirring, in addition to be easy operation and relatively inexpensive to construct. The SS-
AD end-product can be economically transported for use as a fertilizer or soil amendment due to its high TS (Li et al., 2011). Specifically, it can be applied to marginal land for
Miscanthus or other non-food energy crops’ production. The disadvantages of the SS-AD systems include the large amount of inoculum required, a much longer retention time
(twice as L-AD), and the need of nitrogen supplements when lignocellulosic biomass is used (Xu et al., 2013). To solve some of the disadvantages of the SS-AD system, an integrated anaerobic digestion (iADs) technology has been developed and patented. In
63 this system L-AD effluent is mixed with the lignocellulosic biomass and fed into the SS-
AD reactor. The L-AD effluent works as inoculum and nitrogen supplement, reducing retention time and effectively disposing any excess of L-AD effluent (Li et al., 2012).
Biogas yields in AD depend highly on the digestibility of the feedstock. Higher digestibility is usually preferred, but when in excess, it can generate accumulation of volatile fatty acids that could inhibit the methanogens and consequently the biogas production. Lignocellulosic biomass has a low digestibility and usually requires pretreatment, in order to make the holocellulose more available, and reducing the retention time in the digester (Zheng et al., 2014). Similar pretreatments to the traditional ones used prior to enzymatic hydrolysis ‒and previously discuss in section 2.2‒ can be used for lignocellulosic biomass that is going to be used for AD, but those pretreatments are generally high energy intensive and too expensive for this kind of application
(Sawatdeenarunat et al., 2015). Pretreatments such as steam explosion, liquid hot water, dilute acid, alkaline, and biological pretreatment have been used to improve the methane yield from different lignocellulosic resources, obtaining 1.2- to 6-fold in methane yield
(Yang et al., 2015). Fungal pretreatment has some advantages over other pretreatment methods for anaerobic digestion: it requires low energy inputs and no chemicals, so it is considered low cost, and it is usually performed is solid-state, so it generates a feedstock with high TS, idea for SS-AD. Fungal pretreatment of hardwood chips and yard trimmings with C. subvermispora increased the methane yield by 3.7- and 2-fold, respectively (Ge et al., 2015; Zhao et al., 2014b). However, fungal pretreatment of the energy crop giant reed with C. subvermispora was unsuccessful, and fungal pretreated
64 giant reed produced 8-33% lower methane yield than untreated giant reed (Liu et al.,
2016).
Harvest time of the lignocellulosic biomass has a high influence on the digestibility, as well as in the concentration of nutrients such as nitrogen, that could influence the AD performance. In general, it is suggested that energy crops should be harvested as early as possible, preferably before lignification. AD is also influenced by the carbon-to-nitrogen
(C/N) ratio of the feedstock. Ideally, C/N ratio should be around 20-30, to maximize biogas production. However, lignocellulosic biomass generally has a low nitrogen content and its C/N ratio is about 50-100, thus it needs to be co-digested with other substrates to balance the nutrients (Sawatdeenarunat et al., 2015; Yang et al., 2015). The nitrogen content of perennial grasses, such as Miscanthus, reduces after senesce, since these grasses translocate nutrients back to the soil, which justify the need for early harvest of grasses used for AD (Jones and Walsh, 2001).
Anaerobic digestion of Miscanthus has only been investigated in L-AD systems.
Klimiuk et al. determined methane yields of 190 L/kg volatile solids (VS) for M. sacchariflorus and 100 L/kg VS for M. × giganteus in mesophilic conditions after 60 days for Miscanthus harvested in October and ensiled (Klimiuk et al., 2010). A similar yield of 130 L/kg VS was obtained for M. × giganteus harvested in February
(Theuretzbacher et al., 2014). After evaluating several energy crops, Mayer et al. suggests M. × giganteus as the best substitute for maize for biogas production, obtaining a methane yield of 5,500 m3/ha, using Miscanthus harvested in fall and ensiled (Mayer et al., 2014). SS-AD has been performed with other perennial grasses. Giant reed produced
65 methane yields of 129.7 L/kg VS in SS-AD and 150.8 L/kg VS in L-AD (Yang and Li,
2014), while switchgrass generated methane yields of 116.9 L/kg VS in SS-AD and 111.0
L/kg VS in L-AD (Brown et al., 2012).
2.6 Biobutanol
Butanol (n-butanol) is a four-carbon flammable alcohol that has several uses in the plastics, polymers, and solvents industries, as well as a fuel (Green, 2011; Lee et al.,
2008). Butanol is produced catalytically from propylene derived from the petrochemical industry. It can also be produced by acetone-butanol-ethanol (ABE) fermentation from sugars, using microorganism generally referred as solventogenic clostridia. In the fermentation, acetone, butanol, and ethanol (ABE) are produced simultaneously, but butanol is considered the main product (Qureshi et al., 2013a).
As a biofuel, butanol has several advantages: it has higher energy content than ethanol – closer to that of gasoline (Wang et al., 2014). It also has a low vapor pressure and it is both less hygroscopic and less corrosive; this makes it safer to transport and allows its use in existing gasoline distribution infrastructures and engines, either in pure form or in any mixed proportion with gasoline (Qureshi et al., 2013a). It can also be mixed with diesel at relatively high ratios (Ezeji and Blaschek, 2010). Currently, biobutanol is not produced as a biofuel because there are still challenges related with productivity, butanol toxicity to fermenting microorganisms, and product recovery
(Dürre, 2011; Green, 2011). Low butanol titers causes high energy demand for downstream processing (Qureshi et al., 2013a).
66
The most studied microorganisms to produce butanol are the solventogenic clostridia: anaerobic bacteria that can use a wide variety of carbon sources, including five and six carbons sugars, such as the ones derived from cellulose and hemicellulose (Ezeji et al.,
2007a; Qureshi and Ezeji, 2008). The metabolism of these bacteria during butanol production follows a two-phase process: acidogenesis and solventogenesis. Acidogenesis occurs during the exponential growth phase and it consists in the transformation of pyruvate (produced from sugars) to acetate and butyrate. Solventogenesis takes place during the late exponential and stationary phases and the acids previously produced are re-assimilated to generate butanol, acetone, and ethanol (Qureshi et al., 2013a). The most studied species for butanol production are Clostridium acetobutylicum and Clostridium beijerinckii (Kumar and Gayen, 2011). These microorganisms have a low tolerance to the solvents produced, which limits the final concentration of butanol that can be achieved
(Ezeji et al., 2010). Total ABE concentration usually goes to 12-15 g/L although concentrations as high as 25 g/L have been reported (Qureshi and Ezeji, 2008; Ranjan and Moholkar, 2012). The best approach to solve the butanol toxicity problem is the use of continuous recovery of the solvents, to avoid their accumulation the fermentation broth, which in conjunction with a continuous process allows the production of more than
450 g/L of ABE (Ezeji et al., 2005).
It has been proved that butanol can be produced efficiently from lignocellulosic biomass (Qureshi et al., 2014). In order to use lignocellulose to produce butanol, the lignocellulosic biomass should generally be pretreated and then enzymatically hydrolyzed. However, most solventogenic clostridia are very sensitive to lignocellulose
67 derived microbial inhibitory compounds (LDMICs) produced during the chemical pretreatment processes; hence, usually an additional detoxification step is required to obtain efficient butanol production (Ezeji and Blaschek, 2010), which can increase the cost of the process (Qureshi et al., 2013a).
LDMICs are a set of molecules that are generated under the extreme conditions of temperature, pressure, and chemical concentration during the pretreatment process. They are mostly furans and weak acids derived from holocellulose, and aromatic compounds from lignin (Baral and Shah, 2014). Salts formed during neutralization are also potent inhibitory compounds (Qureshi et al., 2008b). LDMICs inhibit the microbial growth and/or solvents production, by different mechanisms that are not completely understood.
Some of these compounds can disrupt cell membranes, damage DNA, disturb redox balance, and inhibit enzymes (Baral and Shah, 2014). Synergistic effects of these
LDMICs have been reported; furthermore, low concentration of some of these compounds, such as hydroxymethylfurfural (HMF) and furfural, have been found to have an effect of fermentation stimulators under some conditions (Qureshi et al., 2013a).
Therefore, predicting a particular effect of a hydrolysate with a certain concentration of
LDMICs on microbial growth and butanol fermentation is still challenging.
Use of switchgrass hydrolysate as a substrate for butanol production using C. beijerinckii P260 showed poor results, even after detoxification by overliming. This process was only improved by dilution with water and glucose supplementation, achieving 14.6 g/L of ABE (Qureshi et al., 2010c). M. × giganteus, after pretreatment with hot water and hydrolysis, was used for butanol production with C.
68 beijerinckii NCIMB 8052. Significant concentration of fermentation inhibitors was present in the hydrolysate: there was no growth in the undiluted hydrolysate, and dilutions as low as 10% and 25%, were required to observe some growth, and certain level of inhibition was still detected compared to the control (Zhang and Ezeji, 2014).
2.7 Techno-economic analysis
Techno-economic analysis (TEA) has been performed extensively to evaluate different technologies for biofuels production in the last three decades. These studies help to determine the techno-economic feasibility of the technology, besides allowing to recognize the parts of the process that need more improvement. Most of the studies have been centered on biodiesel and bioethanol, the two major liquid biofuels (Tao and Aden,
2009; Wang, 2010), and one of the most studied processes is the production of cellulosic ethanol (Hamelinck et al., 2005; Kazi et al., 2010a, 2010b; Sassner et al., 2008; Wingren et al.; Yuzbashev et al., 2010). Recently, a lot of the focus has been diverged to the second generation biofuels that use lignocellulosic biomass as a feedstock, because of their potential to generate a renewable fuel with less environmental impact (Carriquiry et al., 2011; Gnansounou and Dauriat, 2010).
Several studies have compared the economics and environmental sustainability of different pretreatment methods for lignocellulosic biomass (Eggeman and Elander, 2005;
Kumar and Murthy, 2012, 2011a). Most of these pretreatments have been analyzed in the context of the ethanol production, where pretreatment is usually considered one of the steps with a stronger contribution to the cost (Yang and Wyman, 2008). Mostly chemical
69 and thermal pretreatments have been considered in these comparisons, including dilute acid, alkali, ammonia fiber expansion, and hot water pretreatment, among others
(Eggeman and Elander, 2005; Kumar and Murthy, 2012, 2011a). Ofori-Boateng and Lee compared three different biomass pretreatments, including steam explosion, organosolv, and microbial pretreatment, for oil palm fronds for ethanol production. They used exergy analysis to compare the pretreatments because this method provides the actual energy that is available for doing work and allows the systematic identification of inefficiencies. This work has very narrow system boundaries that only include the pretreatment process, and not the whole production of the biofuel. Also, it does not include production of raw materials or construction of facilities (Ofori-Boateng and Lee, 2013). It is important to understand that the type of pretreatment has influence on the performance of the following processes (hydrolysis, fermentation, product separation) and the characteristics of byproducts like stillage and lignin. For the microbial pretreatment, they used data from other researchers and even though these processes are performed in solid-state (Arora et al., 2002; Shi et al., 2008), they included a centrifuge step after the pretreatment. They found out that microbial pretreatment is the most exergetically efficient, and that the centrifuges and dryers were the most exergy draining equipment (Ofori-Boateng and Lee,
2013). They also suggest that the microbial pretreatment is the most cost effective, but no work has been published that proves that statement.
Most of the work that has been performed in TEA of butanol production by fermentation, is concentrated on first generation butanol, mainly from corn (Ranjan and
Moholkar, 2012; Wu et al., 2008, 2007). Economic studies of butanol production are
70 reported since 1980s. These first studies are based on processes with a variety of substrates like molasses, whey permeate, and corn (Lenz and Morelra, 1980; Marlatt and
Datta, 1986; Ranjan and Moholkar, 2012). Improvements in the process have been reflected on the more recent economic studies of butanol production, that include microorganisms with higher yields such as Clostridium beijerinckii BA10, diverse technologies for butanol recovery, and different uses for the byproducts (Mariano et al.,
2013; Qureshi and Blaschek, 2001, 2000; Tao and Aden, 2009; van der Merwe et al.,
2013). Most of them claim the importance that the substrate price has in the economics of the process.
A recent study has developed an economic evaluation of butanol from the lignocellulosic biomass wheat straw. This work assumes a plant capacity of 150 × 106 kg butanol/year. The process includes pretreatment with dilute acid and separated saccharification and fermentation, which is performed using C. beijerinckii P260 in batch mode. Two solvent separation methods were evaluated: distillation and pervaporation; acetone and ethanol were also separated and sold as byproducts, while the microbial biomass was sold as cattle feed. Because of high temperatures of their pretreatment method, the plant uses a considerable amount of cooling water (Qureshi et al., 2013b).
They determined that the separation and selling of the byproducts, including acetone and ethanol, is important for the economics of the process. The cost of the wheat straw is assumed to be between $24-75/ton, which is significantly lower than the breakeven price of Miscanthus reported by Jain et al. for Ohio ($65-101/ton) (Jain et al., 2010). Baral and
Shah also performed a TEA of a 30 million gallon/year butanol production plant from
71 corn stover pretreated with dilute acid. Under the current conditions of the technology, they estimated a cost of $1.5/L of butanol. To reduce this cost to about $0.6/L, significant improvements in sensitive parameters including butanol recovery, sugar utilization, feedstock cost, and heat recovery, are required (Baral and Shah, 2016a).
Kaparaju et al. conducted a study to compare the production of different biofuels such as bioethanol, biohydrogen, and biogas using wheat straw as substrate. Some of their scenarios included the direct use of wheat straw for biogas, and the use of wheat straw hydrolysates for ethanol production and its stillage for biogas through AD. For the latter, they used hydrothermal pretreatment and there is contradictory information about the hydrolysis and fermentation type (separated or simultaneous). They performed a significant amount of experimental work to determine the potential of using their feedstock in all the scenarios considered, including pretreatment, fermentation for bioethanol and biohydrogen, and determination of biological methane potential. AD experiments were performed at thermophilic temperature. They evaluated the energy outputs of each scenario, but not the inputs, i.e. they did not perform an energy balance.
As they claimed, this is a very simplified analysis, and it needs more data to draw conclusions that can be generalized. Their results showed that the highest energy output was obtained when the straw was pretreated and then subjected to AD. This scenario was only 10% higher than the same case but without pretreatment, and considering the energy needed for the hydrothermal pretreatment, it is probably not the most energetically efficient or cost effective. The third highest energy output was obtained from the scenario when the three biofuels are produced, including biogas from AD of stillage. They
72 claimed the latter scenario could be more economical, but a cost analysis is needed to confirm that conclusion (Kaparaju et al., 2009).
Anaerobic digestion (AD) is a well-developed technology to treat waste and produce renewable energy. Recently, an increasing interest in using energy crops as a source for
AD has been developed (Amon et al., 2007; Chynoweth et al., 2001; Mayer et al., 2014).
Most of the economic and sustainability studies for AD of energy crops are done in
Europe, where AD is more widely used (Li et al., 2011). One of the most important factors for the economics of the process is the yield of the crops (Amon et al., 2007;
Mayer et al., 2014). From the environmental perspective, it is usually more convenient to use waste as feedstock for AD, but the digestion of energy crops is still a source of renewable energy that can replace some fossil-fuel derived energy. Gerin et al. calculated a net energy and CO2 balance for the AD of maize and grass. They found out that both crops give a positive energy balance and generate less CO2 emission than the reference system (a gas-steam turbine). They did not take into account the heat and electricity required to operate the digester (Gerin et al., 2008), which could be significant. In their case, the biogas is used to produce electricity. Different results would be obtained if biogas is used for other purposes.
Miscanthus as an energy crop has been studied for several decades in Europe and there is an increasing interest in this crop in the United States (Heaton et al., 2008a; U.S.
DOE, 2016). Most of the research in TEA of Miscanthus as an energy crop aim to compare it with similar crops such as switchgrass (Clifton-Brown et al., 2007; Smeets et al., 2009; Styles and Jones, 2008), and to determine the farm-gate price of the crop (Jain
73 et al., 2010; Khanna et al., 2008; Vyn et al., 2012; S. Wang et al., 2012). In 2010, Jain et al. estimated the yields and economics of Miscanthus production in the Midwestern region of the United States. They provided a range for yield and farm-gate price for
Miscanthus by state, calculated as production cost plus opportunity cost of land, and concluded that the breakeven price for Miscanthus would be between $53/ton of dry matter to $153/ton of dry matter (Jain et al., 2010).
74
Chapter 3: Solid-state fungal pretreatment for non-sterile Miscanthus
Abstract
Miscanthus was pretreated with the fungus C. subvermispora under non-sterile conditions, using sterile Miscanthus that had been previously colonized with the fungus as the inoculum. Inoculum ratios equal or greater than 30% yielded a successful pretreatment, enhancing the enzymatic digestibility of Miscanthus by 3- to 4-fold over that of untreated Miscanthus, similarly to the results obtained with fungal pretreatment under sterile conditions. This enhanced digestibility was linearly correlated with lignin degradation. Although cellulose loss of up to 13% was observed for the successful non- sterile pretreatments, the final glucose yield was 3 to 4 times higher than that of untreated
Miscanthus and equivalent to that of the sterile pretreated Miscanthus. A time course study showed that maximum glucose yield can be achieved with a pretreatment time of
21 days.
3.1 Introduction
Sterilization of the feedstock, a costly step, is usually required for fungal pretreatment, because the indigenous microorganisms may outcolonized the introduced fungi. C. subvermispora has been reported to be specially sensitive to the presence of other microorganisms (Akin et al., 1995). A promising alternative strategy is to grow the
75 fungus on sterile feedstock first, and then use the colonized feedstock as an inoculum for fungal pretreatment of non-sterile feedstock (Zhao et al., 2014a). However, this strategy has only been studied in woody biomass, and it is not clear how it would behave in herbaceous crops, such as Miscanthus. White rot fungi naturally colonizes wood, and it has been shown that its growth in herbaceous biomass can be more difficult (Vasco-
Correa and Li, 2015; Wan and Li, 2011b), probably because of differences in the cell wall structure and composition. However, herbaceous biomass has more potential as a bioenergy feedstock because it is less recalcitrant to enzymatic hydrolysis and grows faster than wood, producing higher biomass yields (Wan and Li, 2011b).
It is unknown how much previously colonized feedstock would be needed as an inoculum for successful pretreatment of herbaceous biomass, how the moisture content of the feedstock-inoculum mixture would affect this process, and how this strategy would affect the pretreatment time. Therefore, systematic research on using this strategy for fungal pretreatment of non-sterile herbaceous crops was needed. The main hypothesis of this study was that fungal pretreatment of unsterilized Miscanthus can be performed successfully using Miscanthus colonized with C. subvermispora as inoculum, yielding similar results as the pretreatment under sterile conditions. Therefore, fungal pretreatment of non-sterile Miscanthus, using Miscanthus colonized with C. subvermispora as the inoculum, was evaluated. Effects of inoculum ratio and moisture content on glucose yield, enzymatic digestibility, lignin degradation, and sugar loss were investigated. Also, a time course study of the fungal pretreatment process was performed, monitoring the degradation of different components during 28 days and the effect on the glucose yield.
76
3.2 Materials and Methods
3.2.1 Feedstock collection and storage
Miscanthus × giganteus (hereafter Miscanthus) was harvested manually from a field in Ashtabula, OH, in spring 2013. The moisture content was around 6% (w/w) and its composition is shown in Table 3.1. The feedstock was milled to pass through a 12 mm screen using a hammer mill (The C.S Bell Co., Tiffin, OH, USA) and stored under dry conditions.
Table 3.1 Characteristics of Miscanthus
Inoculum (fungal Untreated colonized Miscanthus Miscanthus) Total solids (%) 93.8 ± 0.5 47.9 ± 0.5 a Extractives (%) 8.9 ± 0.2 15.3 ± 0.1 Cellulose a (%) 38.0 ± 0.2 43.6 ± 0.9 Hemicellulose a (%) 18.5 ± 0.4 17.4 ± 0.3
Lignin a (%) 20.9 ± 0.2 17.1 ± 0.2
a Based on total solids.
3.2.2 Inoculum preparation
Ceriporiopsis subversmipora (ATCC 96608) was obtained from the American Type
Culture Collection (Manassas, VA, USA) and kept on 2% (w/v) malt extract agar at 4°C.
Sterile Miscanthus colonized with C. subversmipora was used as the inoculum for the fungal pretreatment experiments. To prepare this inoculum, 130 g (dry basis) of untreated
Miscanthus was placed in 2 L reactors, which were supplemented with deionized water to adjust moisture to 60%, and then covered with cotton. Reactors were autoclaved (121°C,
77
15 min) and cooled down. C. subversmipora mycelium was grown in 50 ml of 2% (w/v) malt extract liquid medium (7 days, 28°C, static conditions), filter through cheesecloth, washed with deionized sterile water, homogenized with a manual blender, and added to each reactor, which were then incubated for 28 days at 28°C. At the end, the fungal- colonized Miscanthus was taken out of the flask and mixed well before using it as inoculum for the successive fungal pretreatment of non-sterile Miscanthus.
3.2.3 Fungal pretreatment experiments
Non-sterile Miscanthus and inoculum (fungal-colonized Miscanthus) were mixed in different proportions and added to 1 L reactors. Deionized water was added to adjust moisture content and reactors were covered with cotton and incubated at 28°C. First, a full factorial design with three replicates was performed, with the following factors: inoculum ratio (0%, 10%, 20%, 30%, 40%, and 50%, dry weight basis) and moisture content (60% and 75%, w/w). Reactors were incubated for 28 days. The inoculum ratio
0% (no inoculation) was considered to be the negative control. The positive control was sterile Miscanthus inoculated with mycelium from a liquid culture of C. subvermispora following the same procedure as the inoculum preparation, as described in section 3.2.2, and it was incubated along with the treatments. At the end of the 28 days, samples were taken out of the reactors and mixed well. Part of the fresh material was used for enzymatic hydrolysis and the rest was dried at 40°C in a convection oven for 24 h, then milled and passed through a 1 mm screen (Model 4 Wiley Mill, Thomas Scientific,
Swedesboro, NJ, USA) for cellulose, hemicellulose, and lignin content determination.
78
Second, an experiment to evaluate the degradation over time of components of
Miscanthus during fungal pretreatment was performed at 50% (dry weight basis) inoculum ratio and 60% (w/w) moisture content. Fifteen reactors were prepared in the same manner as in the first experiment and stopped at five different times (0, 7, 14, 21, and 28 days). For sampling at each time, three reactors were randomly selected and the whole material in each reactor was used for the analysis. Samples were processed as in the first experiment for subsequent enzymatic hydrolysis, and cellulose, hemicellulose, and lignin content determination.
3.2.4 Enzymatic hydrolysis
Enzymatic digestibility of untreated and pretreated Miscanthus was determined according to Selig et al., using the enzyme mix Cellic CTec2 (Novozymes, Bagsværd,
Denmark), at a cellulase concentration of 10 FPU/g dry substrate (Selig et al., 2008).
About 2.5 g (dry basis) of feedstock were added to 250 ml flasks along with citrate buffer
(0.05 M, pH 5.0), 1 ml of 2% (w/v) sodium azide solution, and the enzyme, for a total working volume of 100 ml. Enzymatic hydrolysis was performed at 50°C and a pH of 5.0 for 72 h, and then the hydrolysate was centrifuged (12,000 × g, 10 min), and the supernatant was passed through a 0.2 µm nylon filter into glass vials. Free sugars were analyzed by high-performance liquid chromatography (HPLC) (LC-20AB, Shimadzu,
Kyoto, Japan) with an Aminex HPX-87P column (Bio-Rad, Inc., Hercules, CA, USA) and a refractive index detector. Enzymatic digestibility and glucose yield were calculated as follow:
79
[퐺푙푢푐표푠푒] + 1.053[퐶푒푙푙표푏푖표푠푒] % 퐸푛푧푦푚푎푡푖푐 퐷푖푔푒푠푡푖푏푙푖푡푦 = × 100% 1.111푓푝[푝퐵푖표푚푎푠푠]
[퐺푙푢푐표푠푒] + 1.053[퐶푒푙푙표푏푖표푠푒] % 퐺푙푢푐표푠푒 푦푖푒푙푑 = × 100% 1.111푓푟[푟퐵푖표푚푎푠푠]
where [Glucose] and [Cellobiose] are the glucose and cellobiose released by the enzymatic hydrolysis (g), [pBiomass] and [rBiomass] are the dry pretreated biomass at the beginning of the enzymatic hydrolysis and the dry untreated biomass at the beginning of the pretreatment (g), respectively, and fp and fr are the fraction of cellulose in the pretreated and original (untreated) dry biomass (g/g), respectively (Wang et al., 2014).
Thus, the glucose yield accounted for the cellulose loss that could occur during the fungal pretreatment.
3.2.5 Analytical methods
Cellulose, hemicellulose, and lignin content were determined according to the protocol from the National Renewable Energy Laboratory (NREL) (Sluiter et al., 2012), using a two-step acid hydrolysis. Sugars liberated after the acid hydrolysis were quantified by HPLC (LC-20AB, Shimadzu, Kyoto, Japan) with an Aminex HPX-87P column (Bio-Rad, Inc., Hercules, CA, USA) and a refractive index detector. Acid soluble lignin was determined by UV-vis spectroscopy (BioMate 3, Thermo Fisher Scientific,
Inc., Waltham, MA, USA) and acid-insoluble lignin was quantified gravimetrically. Dry matter was determined gravimetrically, by drying triplicate samples of the Miscanthus at
80
105°C. Cellulose, hemicellulose, lignin, and dry matter degradation were calculated for pretreatment for the non-sterile Miscanthus, considering the initial content as the content of the mixture (untreated Miscanthus plus inoculum).
To measure the Miscanthus extractives, an Accelerated Solvent Extraction (ASE 350,
Thermo Fisher Scientific, Inc., Waltham, MA, USA) with water and ethanol was used, following the NREL analytical procedure (Sluiter et al., 2008). Then, the solvents were evaporated (Rocket Evaporator, Genevac, Ipswich, England), dried at 40°C, and the extractives determined gravimetrically.
3.2.6 Data analysis
Statistical significance was evaluated by analysis of variance (ANOVA) (α=0.05), and mean comparisons were performed by Tukey-Kramer test with a 95% significance, using the Software JMP® (SAS Institute Inc., Cary, NC, USA). Error bars in the figures represent the standard error.
3.3 Results and Discussion
3.3.1 Cellulose, hemicellulose, and dry matter loss
Cellulose, hemicellulose, and dry matter losses after 28 days of the fungal pretreatment of non-sterile Miscanthus strongly depended on the inoculum ratio, but not on the moisture content, for the conditions evaluated (Figure 3.1, Figure 3.2, and Figure
3.3). However, the interaction between inoculum ratio and moisture content had a significant effect on cellulose, hemicellulose, and dry matter loss (p<0.05). The losses of
81 cellulose (5-18%), hemicellulose (3-24%), and dry matter (4-18%) were similar with those obtained previously in fungal pretreatment of non-sterile yard trimmings (Zhao et al., 2014a).
Cellulose loss was between 7% and 13% for non-sterile pretreatments with inoculum ratios 30% to 50%, which was similar to that in the Miscanthus incubated without inoculation (5-9% cellulose loss). On the contrary, cellulose loss was negligible for the pretreated sterile Miscanthus (positive control) (Figure 3.1). This suggests that the cellulose loss in the non-sterile pretreatment was due to the presence of other microorganisms (contaminants) that prefer cellulose as carbon and energy source.
Cellulose is the main source of sugars for the subsequent processes and its degradation during the pretreatment is undesirable. However, the cellulose loss in the non-sterile pretreatment was still similar to those observed for other pretreatment methods. For example, alkaline pretreatment of Miscanthus with 1 M NaOH at 150°C for 60 min caused a cellulose loss of about 20% (Cha et al., 2014). Aqueous ammonia pretreatment showed a cellulose loss of 4% to 12% (Liu et al., 2013). Organosolv pretreatment with formic and acetic acid led to a cellulose loss of about 0% to 25%, depending on pretreatment temperature, time, and solvent concentration (Vanderghem et al., 2012).
82
40% Moisture 60% Moisture 75% 30%
20% Celluloseloss 10%
0% 0 10 20 30 40 50 + Inoculum ratio (%)
Figure 3.1 Cellulose loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from
liquid culture).
The hemicellulose loss was probably mainly caused by the growth of C. subvermispora, which uses these sugars as its main carbon source (Wan, 2011), but other microorganisms present in the reactor could also have consumed it at a small rate, as can be seen in the hemicellulose degradation in the Miscanthus incubated without inoculation
(negative control) (Figure 3.2). There was a significant difference between the hemicellulose loss in the pretreated sterilized Miscanthus (positive control) and non- sterilized Miscanthus with inoculum ratios of 30% to 50% (p<0.05). That difference was equivalent to the hemicellulose loss obtained in the Miscanthus without inoculation, which suggests that the higher degradation in the non-sterile treatments compared to sterilized Miscanthus was due to hemicellulose consumption by the other microorganisms present in the reactor. Hemicellulose loss can be beneficial to enhance the cellulose
83 digestibility, but if the subsequent fermentation processes use microorganisms that are able to utilize xylan, hemicellulose degradation is detrimental (Alvira et al., 2010).
Hemicellulose degradation for the successful non-sterile pretreatment was between 16% and 24%, while the hemicellulose loss in the sterile pretreatment was about 16-17%.
These values were still smaller than those obtained with other pretreatments. Alkaline pretreatment usually degrades a large percentage of the hemicellulose. For example,
Miscanthus treated with 1 M NaOH at 150°C for 60 min resulted in a hemicellulose loss of about 79% (Cha et al., 2014). Aqueous ammonia pretreatment of Miscanthus caused a hemicellulose loss of 23% to 61% (Liu et al., 2013). Organosolv pretreatment with formic/acetic acid produced a loss of hemicellulose of 9% to 79%, depending on the pretreatment temperature, time, and solvent concentration (Vanderghem et al., 2012).
Autohydrolysis of Miscanthus (pretreatment with water under pressure at 130-150°C) generated a hemicellulose loss of about 50%, with no cellulose or lignin degradation (El
Hage et al., 2010).
84
40% Moisture 60% Moisture 75% 30%
20%
Hemicellulose lossHemicellulose 10%
0% 0 10 20 30 40 50 + Inoculum ratio (%)
Figure 3.2 Hemicellulose loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from
liquid culture).
Dry matter loss was 13% to 18% for the non-sterile pretreatments with inoculum ratio
30% to 50%, which was significantly higher than that of the sterilized Miscanthus (7-9%) and the Miscanthus without inoculation (4-7%) (Figure 3.3). For the non-sterile pretreatments with 30% to 50% inoculum ratio, 40% to 50% of the dry matter loss was due to lignin degradation, and the rest was equally distributed between cellulose and hemicellulose loss. For the sterile pretreatment, about 70% to 80% of the dry matter loss was due to lignin degradation, and the rest was hemicellulose loss because, as mentioned previously, cellulose loss was negligible. For the Miscanthus without inoculation, 70% to
80% of the dry matter loss was due to loss in cellulose and the rest to hemicellulose loss, with practically no lignin degraded in this case. Therefore, the difference in dry matter loss between the non-sterile pretreatments with 30% to 50% inoculum ratio and the
85 sterile pretreatment (positive control) was due primarily to the loss in cellulose in the former. The main disparity between the non-sterile pretreatments with 30% to 50% inoculum ratio and the Miscanthus without inoculation (negative control) was due to the absence of lignin degradation in the latter. For the non-sterile pretreatments with 10% and
20% inoculum ratio, the dry matter loss proportions were similar to those of the
Miscanthus without inoculation, since the lignin degradation was very low in these cases
(0-5%).
40% Moisture 60% Moisture 75% 30%
20%
Dry matter lossmatter Dry 10%
0% 0 10 20 30 40 50 + Inoculum ratio (%)
Figure 3.3 Dry matter loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from
liquid culture).
3.3.2 Lignin degradation and enzymatic digestibility
Lignin loss was almost negligible for pretreatments with inoculum ratios of 10% and
20%, as well as for the Miscanthus without inoculation. On the other hand, lignin loss of
86 around 25% to 35% was achieved with inoculum ratios of 30% to 50%, as well as with the sterilized Miscanthus (Figure 3.4), which was significantly higher than that obtained using a similar pretreatment strategy for non-sterile yard trimmings (Zhao et al., 2014a).
An analysis of variance indicated that both inoculum ratio and moisture content had a significant effect (p<0.05) on lignin degradation; however, the interaction between these two factors was not significant (p>0.05) (Figure 3.4). At inoculum ratios between 30% and 50%, higher lignin degradation was observed at a moisture content of 75%.
40% Moisture 60% Moisture 75% 30%
20%
Lignindegradation 10%
0% 0 10 20 30 40 50 + Inoculum ratio (%) Figure 3.4 Lignin loss after 28 days of fungal pretreatment of Miscanthus. “+”: positive
control (sterilized Miscanthus inoculated with C. subvermispora mycelium from liquid
culture).
The enzymatic digestibility was significantly affected by the inoculum ratio (p<0.05), but not by the moisture content (p>0.05), and the interaction between these two factors was not significant (p>0.05), according to the analysis of variance (Figure 3.5). Moisture
87 content was evaluated at these two levels because they have been found to be optimum for fungal pretreatment with C. subvermispora of different feedstocks in previous studies
(Wan and Li, 2010a; Zhao et al., 2014b). The lack of significant difference between the two levels suggests that moisture content can vary between 60% and 75% without significantly affecting enzymatic digestibility, which is advantageous because moisture content can be a parameter difficult to control in solid-state fermentation (Robinson et al.,
2001).
60% Moisture 60% Moisture 75% 50%
40%
30%
20%
Enzymatic digestibility Enzymatic 10%
0% 0 10 20 30 40 50 + Inoculum ratio (%) Figure 3.5 Enzymatic digestibility of fungal pretreated Miscanthus after 28 days. “+”: positive control (sterilized Miscanthus inoculated with C. subvermispora mycelium from
liquid culture).
There was no significant difference between the enzymatic digestibility and lignin degradation of the reactors with inoculum ratios of 30%, 40%, and 50% and those of the sterilized Miscanthus (p>0.05), showing that at an inoculum ratio between 30% and 50%,
88 fungal pretreatment of non-sterile Miscanthus was feasible. There was also no significant difference between the enzymatic digestibility and lignin degradation of Miscanthus with inoculum ratios of 10% and 20% and those of the Miscanthus without inoculation (0% inoculum ratio) and the untreated Miscanthus (p>0.05), which had an enzymatic digestibility of about 12% (Figure 3.4 and Figure 3.5). This suggests that the pretreatment was unsuccessful for inoculum ratios less than 20%. When the inoculum ratio was increased to 30% to 50%, a 2-fold increase in enzymatic digestibility was obtained compared with the initial mixture, which was equivalent to a 3- to 4-fold compared to the untreated Miscanthus. These results were similar to those obtained by fungal pretreatment of sterile switchgrass in previous research (Wan and Li, 2011b). Enzymatic digestibility was also analogous to that obtained after a 28-days fungal pretreatment of non-sterile corn stover with Irpex lacteus, even though this feedstock underwent a disinfection process with lime prior fungal inoculation (Song et al., 2013b).
Strong correlation between lignin loss and enzymatic digestibility was observed
(Figure 3.6), suggesting that lignin was probably one of the main factors that reduced the accessibility of structural sugars to the enzyme, as described by other authors (Chen et al.,
2010; Wan, 2011), and that its removal positively and directly contributed to enhancement of enzymatic digestibility. This result confirmed that the main effect of fungal pretreatment on the lignocellulosic structure is lignin degradation, which makes the cellulose more accessible to hydrolytic enzymes (Saritha et al., 2012a). No significant correlation was found between enzymatic digestibility and cellulose, hemicellulose, or dry matter loss.
89
60%
50% R² = 0.9638 40%
30%
20%
Enzymatic digestibility Enzymatic 10%
0% 0% 10% 20% 30% 40% Lignin degradation
Figure 3.6 Correlation between lignin loss and enzymatic digestibility of fungal
pretreated Miscanthus after 28 days. “+”: positive control (sterilized Miscanthus
inoculated with C. subvermispora mycelium from liquid culture).
Glucose yield uses as reference the cellulose present in the original material, while enzymatic digestibility is based in the cellulose of the pretreated material. Therefore, glucose yield takes into account the glucose lost during the fungal pretreatment, and it is equivalent to enzymatic digestibility if there was no cellulose loss during pretreatment, or slightly lower on the contrary. Glucose yield for effective inoculum ratios (30%, 40%, and 50%) ranged between 35% and 48%, which represented an increase of 3- to 4-fold, compared with the untreated Miscanthus. Compared with the initial mixture of inoculum
(fungal colonized Miscanthus) and untreated Miscanthus, the glucose yield was increased by a factor of 2. The glucose yield for sterilized Miscanthus (positive control) was between 35% and 39%, which was similar to the effective pretreatments with inoculum ratios of 30% to 50%. These results suggests that even though cellulose degradation
90 during fungal pretreatment of non-sterile Miscanthus was higher than that of sterilized
Miscanthus, the overall glucose yield was not affected drastically. Fungal pretreatment of
Miscanthus could be performed under non-sterile conditions, yielding similar results as sterilized Miscanthus, using the process described in this study.
3.3.3 Time course of the non-sterile fungal pretreatment
During solid-state fungal pretreatment of non-sterile Miscanthus with 50% inoculum, the glucose yield increased significantly until day 21. There was no significant increment in the glucose yield from day 21 to day 28 (p>0.05) (Figure 3.7). There was also no significant cellulose loss between those dates; in fact, all the cellulose degradation seemed to occur at the beginning of the process (between day 0 and 7) and it did not increase after that. Therefore, the pretreatment could be reduced to 21 days without affecting the glucose yield, while enhancing the efficiency of the process. In contrast, there was continuous degradation of hemicellulose throughout the process, which suggests that the hemicellulose was acting as a carbon and energy source for fungal growth and maintenance. Glucose yield generally increased with the increase of lignin degradation, especially in the first 7 days when half of the lignin was degraded.
91
35% 45% Cellulose loss Hemicellulose loss Lignin loss 40% 30% Dry matter loss Glucose yield 35% 25% 30%
20% 25%
15% 20% Glucose yield (%) Glucose yield
Component lossComponent 15% 10% 10% 5% 5%
0% 0% 0 5 10 15 20 25 30 Time (days)
Figure 3.7 Time course study of fungal pretreatment of non-sterile Miscanthus during 28
days at 50% inoculum ratio and 60% moisture content. Components loss (cellulose, hemicellulose, lignin, and dry matter) is a percentage (in dry matter) of amount present in the original mixture (unsterilized Miscanthus plus inoculum). Glucose yield is calculated
as the amount of sugar released after enzymatic hydrolysis of the pretreated material,
over the glucose present as cellulose in the original Miscanthus.
Wan and Li (2010) studied the time course of fungal pretreatment of corn stover during 42 days of incubation under sterile conditions. In their case, most of the lignin degradation occurred during the first 15 days, and then the degradation rate decreased significantly, which was not seen in the present case. However, hemicellulose and dry matter loss occurred continuously during the pretreatment of sterile corn stover, similar to
92 the non-sterile Miscanthus. Glucose yield showed a similar tendency for sterile corn stover and non-sterile Miscanthus, with a high initial increase and a later stationary phase. The main difference was that the glucose yield curve for sterile corn stover showed a lag phase of about 7 days, which was not present in that of the non-sterile
Miscanthus. This difference was probably due to the inoculum, which was mycelium from a liquid culture in the case of the corn stover, and Miscanthus colonized with the fungus in the current study. In the present case, the inoculum probably had a higher concentration of the fungus and the fungus was better adapted to the feedstock, since it has been growing on it for 28 days, which could have shortened the lag phase.
3.4 Conclusions
Fungal pretreatment of non-sterile Miscanthus achieved 3- to 4-fold increases in enzymatic digestibility and glucose yield, compared to the untreated feedstock. This pretreatment can be performed in 21 days, using 30% to 50% fungal colonized
Miscanthus as the inoculum and a moisture content of 60% to 75%, to obtain a glucose yield equivalent to that of fungal pretreatment of sterile Miscanthus, despite having a higher cellulose loss. Sterilization was still required for preparing the inoculum that was used for this single batch pretreatment of non-sterile Miscanthus. Future research should explore consecutive batches of fungal pretreatment in order to further reduce the cost.
93
Chapter 4: Changes in composition, enzymatic digestibility and microbial communities during sequential solid-state fungal pretreatment of non-sterile Miscanthus
Abstract
A sequential fungal pretreatment of Miscanthus was conducted by mixing unsterile
Miscanthus with material previously colonized with the white rot fungus C. subvermispora, for three generations, each generation starting with inoculation by mixing unsterile fresh Miscanthus with end material from the previous generation, and ending after 28 days of incubation at 28˚C. The first generation produced an increase of 2-fold in enzymatic digestibility, compared to the unsterile Miscanthus, but the second and third generation showed no enhancements. Furthermore, high degradation of Miscanthus structural carbohydrates occurred during the first generation. A microbial community study showed that, even though the material previously colonized by C. subvermispora was composed mainly by this fungus (>99%), by the first generation its abundance was down to only 9%, and other fungi, mainly ascomycetes had outcolonized it. This reiterates the necessity of feedstock sterilization for the stability and reproducibility of fungal pretreatment of lignocellulosic biomass.
94
4.1 Introduction
Sterilization of the feedstock, a costly step, is usually required for fungal pretreatment, because the indigenous microorganisms may outcolonize the introduced white rot fungi. C. subvermispora has shown to be particularly sensitive to microbial contamination (Akin et al., 1995), creating a challenge for the scalability of the fungal pretreatment process. In Chapter 3, pretreatment of non-sterile Miscanthus was successfully performed by using Miscanthus previously colonized by the fungus C. subvermispora as inoculum, at inoculum ratios equal or higher than 30% (Vasco-Correa et al., 2016a). In order to further reduce the need for feedstock sterilization, the finished material from this fungal pretreatment of non-sterile Miscanthus could be potentially used as inoculum for subsequent generations of fungal pretreatment. If successful, this approach would reduce exponentially the need of sterilization, for each effective fungal pretreatment generation. However, the stability of the fungus and the propagation of microbial contamination throughout the generations are potential concerns that need to be studied.
It is hypothesized that fungal pretreatment of non-sterile Miscanthus can be successfully performed for at least three generations using finished material from the previous generation as inoculum, generating similar results in enzymatic digestibility than those obtained by fungal pretreatment of sterile Miscanthus inoculated with C. subvermispora mycelium from a liquid culture. In this study, fungal pretreatment of non- sterile Miscanthus was performed using, as inoculum, finished material from the previous fungal pretreatment generation, and the process was evaluated for three consecutive
95 generations. The effects of generation and moisture content on enzymatic digestibility, lignin degradation, and structural sugars loss were evaluated. Also, the changes of the fungal and bacterial communities throughout the fungal pretreatment generations were explored.
4.2 Materials and Methods
4.2.1 Feedstock collection and storage
Miscanthus was collected, milled, and stored as stated in section 3.2.1. The composition is shown in Table 3.1.
4.2.2 Inoculum preparation
Inoculum consisted of the fungal-colonized Miscanthus that was prepared as described in section 3.2.2, which included Miscanthus autoclaving, inoculation with mycelia of C. subvermispora, and incubation at the same conditions of the treatments.
Therefore, this was also considered to be the positive control and was performed in triplicates.
4.2.3 Fungal pretreatment experiments
First, the inoculum/positive control (fungal-colonized Miscanthus) was mixed at a
50% (w/w, dry matter) ratio with unsterile Miscanthus and added to 1 L flasks, to a total of 65 g of dry matter. Deionized water was added to adjust moisture content (45%, 60%, and 75%, w/w) and reactors were incubated at 28°C for 28 days. This experiment was
96 called “first non-sterile generation”. The experiment was performed in 12 replicates for each moisture content. After the incubation time, 3 replicates were randomly chosen for characterization of the pretreated material and processed as described below. The material of the rest of the flasks was combined, mixed thoroughly, and sampled for dry matter content analysis. This material was used as inoculum for a “second non-sterile generation”.
For the second and third non-sterile generations, the inoculum produced for the past generation was mixed at a 50% (w/w, dry matter) ratio with non-sterile Miscanthus, following the same procedure, conditions, and sample processing as in the first non- sterile generation (Figure 4.1).
97
Unsterile miscanthus
Autoclave 121°C 20 min
C. Sterile subvermispora miscanthus
28 days 28°C
Inoculum/ Positive control
Unsterile miscanthus
28 days 28°C First generation First gen non-sterile pretreated miscanthus
Unsterile miscanthus
28 days 28°C Second generation Second gen non-sterile pretreated miscanthus
Unsterile miscanthus
28 days Third generation 28°C Third gen non-sterile pretreated miscanthus
Figure 4.1 Experimental flowchart for the sequential fungal pretreatment of non-sterile
Miscanthus.
98
Untreated Miscanthus with no inoculum at the same moisture content levels was incubated in 1 L flasks (triplicates) along with the first generation experiments, and considered to be the negative control.
The material from the three reactors chosen for characterization from each generation and from the control flasks (positive and negative) was taken out of the flasks after the
28-days incubation time and mixed well. Samples were taken and used fresh for enzymatic hydrolysis and to determined dry matter content, another sample was frozen at
-80˚C for further DNA extraction, and the rest of the material was dried at 40°C in a convection oven for 24 h, milled and passed through a 1 mm screen (Model 4 Wiley Mill,
Thomas Scientific, Swedesboro, NJ, USA), and then used for cellulose, hemicellulose, and lignin content determination.
4.2.4 Enzymatic hydrolysis
Enzymatic digestibility of untreated and pretreated Miscanthus, and glucose yield were determined as described in section 3.2.4.
4.2.5 Microbial community analysis
Samples from 60% moisture content level were selected for microbial community analysis. Untreated and pretreated samples were freeze-dried and then ground using liquid nitrogen. Total DNA was extracted and purified using the PowerSoil® DNA
Isolation Kit (Mo Bio Laboratories, Carlsbad, CA, USA), and quantified using florescence with a Quant-iT dsDNA High-Sensitivity Assay Kit in a Qubit 2.0
99 fluorometer (Life Technologies, Carlsbad, CA, USA). DNA was diluted to a normalized concentration of 5 ng/μL. Library preparation and sequencing were performed in the
Molecular and Cellular Imaging Center (MCIC) in Wooster, OH. The V1-V3 hypervariable region of the bacterial 16S rRNA gene and the internal transcribed spacer
(ITS) region 1-2 of the fungal nuclear DNA were amplified using universal primers modified to include degenerate bases for maximal inclusiveness (Li et al., 2014; Smith et al., 2014).
Libraries were prepared in two rounds of PCR amplification. The first round amplified the locus of interest and added a portion of the Illumina adapter sequence, and the second round completed the Illumina adapter sequence, which contained a unique dual combination of the Nextera indices for individual tagging of each sample. Twenty nanograms of each genomic DNA was used as input for the first PCR reaction and 3 μl of the clean PCR 1 product was used as input for the second PCR reaction. PCR amplifications were carried as follows: initial denaturation at 96°C for 3 min, followed by
25 (PCR 1) or 8 (PCR 2) cycles each of 96°C for 30 s, 55°C for 30 s and 72°C for 30 s, and a final extension at 72°C for 5 min. The PCR products were purified after each PCR amplification using the Agencourt AMPure XP beads (Beckman Coulter Life Sciences,
Brea, CA, USA). All the steps for library preparation and cleaning were carried out on the epMotion5075 automated liquid handler (Eppendorf, Hamburg, Germany). The purified amplicon libraries were quantified and pooled at equimolar ratios before sequencing. A separate pool for each target was prepared. The final pools were also
100 purified using the Pippin Prep size selection system (Sage Science Inc., Beverly, MA,
USA) to discard the presence of any primer dimers.
The amplicon libraries were sequenced using the MiSeq sequencing platform
(Illumina, San Diego, CA, USA) at a final concentration of 14.3 pM. PhiX was mixed in with the pool of amplicon libraries for the sequencing run (expected at 20%). The run was clustered to a density of 665 k/mm2 and the libraries were sequenced using a 300PE
MiSeq sequencing kit with the standard Illumina sequencing primers. Image analysis, base calling, and data quality assessment were performed on the MiSeq instrument.
4.2.6 Analytical methods
Cellulose, hemicellulose, lignin, and dry matter degradation were calculated considering the initial content as the content of the mixture (untreated Miscanthus plus inoculum). Analytical methods for biomass compositional analysis were performed as described in section 3.2.5.
4.2.7 Data analysis
For the wet chemistry data, statistical significance was evaluated by analysis of variance (ANOVA) (α=0.05), and mean comparisons were performed by Tukey-Kramer test with a 95% significance, using the Software JMP® (SAS Institute Inc., Cary, NC,
USA). Error bars in the figures represent the standard error.
For the microbial community analysis, data processing and analysis was performed using QIIME v. 1.9.1 (Caporaso et al., 2010). A mapping file was created based on
101 barcoding sequences used in the sequencing run and sample numbers and variables.
Sequences were demultiplexed and filtered to remove sequences with Phred score below
21. Open references operational taxonomic unit (OTU) picking was performed against the UNITE fungal database for ITS sequences (Kõljalg et al., 2005) and the Greengenes database version 13_8 for 16S sequences (DeSantis et al., 2006). Additionally, the ITS sequence of C. subvermispora (ATCC 96608) was added to the local database to ensure its presence upon OTU taxonomical assignment. The identity threshold was set at 97% for each set of sequences. UCHIME2 was utilized to identify and remove chimeric sequences (Edgar et al., 2011). All sequences aligning with a non-target kingdom and all singleton sequences were filtered from the OTU tables. OTU tables were rarefied to the lowest number of sequences present in samples. Rarefaction to 8,204 reads and 22,400 reads was performed for 16S and ITS data, respectively. Alpha diversity metrics (number of observed species, chao1, Shannon, and Simpson diversity indices) were calculated for each sample. For beta diversity, principal component analysis (PCA) was performed for abundance-weighted Jaccard distance.
4.3 Results and Discussion
4.3.1 Enzymatic digestibility and lignin degradation
The enzymatic digestibility was significantly affected by the generation (p<0.05), but not by the moisture content (p>0.05), according to the analysis of variance (Figure 4.2).
This confirms that moisture content can vary within the evaluated range without significantly affecting enzymatic digestibility as stated in section 3.3.2. There was no
102 significant difference between the enzymatic digestibilities of the fungal pretreated
Miscanthus after the second and third generation non-sterile pretreatment, and the negative control (p>0.05), and these were also similar with the digestibility of the untreated Miscanthus, which was about 12%. Thus, the fungal pretreatment was unsuccessful for second and third non-sterile generations. Conversely, the first generation non-sterile fungal pretreatment produced material with enzymatic digestibilities of 19-
25%, which were significantly higher than those of the negative controls and the untreated material (p<0.05), but still lower than those of the positive controls (p<0.05), which were prepared by inoculating sterile Miscanthus with C. subvermispora mycelium from a liquid culture, and which reached enzymatic digestibility values between 27% and
35%. These values were slightly lower than those obtained previously for fungal pretreatment of Miscanthus in similar conditions (Vasco-Correa et al., 2016a), which can only be explained by the natural variability of the biological process, including changes in the microbial community of the feedstock, and variations in feedstock storage time.
103
45% 40% 45% 35% 60% 30% 75% 25% 20% 15%
Enzymatic digestiblity Enzymatic 10% 5% 0% Positive First Second Third Negative control generation generation generation control
Figure 4.2 Enzymatic digestibility of fungal pretreated Miscanthus after 28 days of
incubation at 45%, 60%, and 75% initial moisture content. Negative control: incubated
alown treatment without inoculation. Postive control: sterilized Miscanthus inoculated
with C. subvermispora mycelium from liquid culture.
The analysis of variance showed a significant interaction between moisture content and generation (p<0.05). This was evidenced by the differential effect of moisture content in the positive control and the first non-sterile generation. Enzymatic digestibility increased with moisture content in the positive control, consistently with previous observations (Wan and Li, 2010a). However, in the first non-sterile generation the highest enzymatic digestibility was obtained at 60% initial moisture content, and there was no significant difference between the enzymatic digestibilities at 45% and 75% moisture content (p>0.05), which were significantly lower than that at 60% moisture content (p<0.05) (Figure 4.2). This might be explained by a higher presence of
104 indigenous microorganisms at the 75% moisture level that could have been favored by the high moisture content and could have outcolonized the effect of C. subvermispora.
Microbial colonization had no effect in the positive control since, in that case, the
Miscanthus was sterilized prior fungal inoculation.
Similarly to the enzymatic digestibility, lignin degradation was affected by generation
(p<0.05), but not by moisture content (p>0.05), according to the analysis of variance
(Figure 4.3). However, in this case, there was no significant interaction between generation and moisture content (p>0.05). Lignin degradation was negligible (less than
5%) for the second and third generation non-sterile fungal pretreatment, and the negative control. Lignin degradation between 9% and 17% occurred after the first non-sterile generation, and it was significantly higher (22-30%) (p<0.05) for the positive control.
Lignin degradation results showed a similar pattern to the enzymatic digestibility, as demostrated by the high correlation between these two variables (Figure 4.4). This correlation suggested that lignin degradation was the main effect of fungal pretreatment on the recalcitrance reduction of Miscanthus and enzymatic digestibility increase, as proposed before (Saritha et al., 2012a; Vasco-Correa et al., 2016a). No significant correlation was found between enzymatic digestibility and cellulose, hemicellulose, or dry matter loss.
105
40%
35% 45% 60% 30% 75% 25%
20%
15%
Lignindegradation 10%
5%
0% Positive First Second Third Negative control generation generation generation control
Figure 4.3 Lignin degradation after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment
without inoculation. Postive control: sterilized Miscanthus inoculated with C.
subvermispora mycelium from liquid culture.
106
45%
40%
35% R² = 0.9544 30%
25%
20%
15%
10% Enzymatic digestibility Enzymatic
5%
0% -5% 0% 5% 10% 15% 20% 25% 30% 35% 40% Lignin degradation
Figure 4.4 Correlation between lignin degradation and enzymatic digestibility of fungal
pretreated Miscanthus.
4.3.2 Cellulose, hemicellulose, and dry matter loss
Similarly to lignin degradation and enzymatic digestibility, cellulose, hemicellulose, and dry matter losses after fungal pretreatment of Miscanthus were strongly influenced by generation (p<0.05), but not by moisture content (p>0.05), according to the analysis of variance. However, the interaction between moisture content and generation had a significant effect in cellulose, hemicellulose, and dry matter loss (p<0.05), as evidence by the differential trend in the effect of moisture content in each generation (Figure 4.5,
Figure 4.6 and Figure 4.7). Cellulose loss between 12% and 31% occurred in the first generation of non-sterile pretreatment, and it was significantly higher than that of the positive control (2-4%) (Figure 4.5), which could be caused by: (1) the presence of a microbial community that was feeding on the cellulose, or (2) the reach of an advance
107 part of the life cycle of C. subvermispora when it activated its cellulolytic system and started degrading cellulose, as has been observed after long incubation times (Tanaka et al., 2009). C. subvermispora has a complete cellulolytic system (Fernández-Fueyo et al.,
2012a); however, it has been shown that it only activates it after prolonged periods of incubation (Tanaka et al., 2009), probably when most of the easily degradable hemicelluloses have been significantly reduced, and cellulose has been exposed by the combined effect of lignin and hemicellulose degradation. The fact that the cellulose loss in the negative control (7-10%) was lower than that of the first non-sterile generation seems to favor the second hypothesis, since the cellulose loss in the negative control was the effect of only the indigenous microorganisms of the Miscanthus. However, the higher cellulose loss in the first generation could be also caused by a symbiotic effect of C. subvermispora and the contaminant microbial community, since the white rot fungus reduced the Miscanthus recalcitrance, leaving the cellulose more available to be consumed by the contaminant microorganisms. The high cellulose degradation caused a significant drop in the overall glucose yield for the first non-sterile generation, compared with the enzymatic digestibility, making it about 18% for 45% and 60% moisture content, and only 13% for 75% moisture content. Since cellulose loss was low in the positive controls, glucose yield values were very close to the reported enzymatic digestibilities
(Figure 4.2).
108
40%
35% 45% 30% 60%
25% 75%
20%
15% Celluloseloss 10%
5%
0% Positive First Second Third Negative control generation generation generation control
Figure 4.5 Cellulose loss after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment
without inoculation. Postive control: sterilized Miscanthus inoculated with C.
subvermispora mycelium from liquid culture.
Hemicellulose loss in the positive control (7-14%) was attributed to the consumption by C. subvermispora, since this has been considered its main carbon and energy source
(Wan, 2011). In the first non-sterile generation, hemicellulose loss was between 11% and
26%, which was higher than the positive control (p<0.05) (Figure 4.6), and probably caused by a combined effect of C. subvermispora and the indigenous microorganisms. In the first non-sterile generation, hemicellulose loss was significantly lower for the 45% moisture than for 60% and 75%. At 45% moisture content there was less increase in enzymatic digestibility and less lignin degradation, which suggested that less white rot fungus grew at this conditions, but there was also less cellulose and hemicellulose loss,
109 indicating that the low moisture content reduced the capacity of the indigenous microorganisms to grow and degrade extra sugars, with respect to 60% and 75% moisture content. Second and third non-sterile generation produced cellulose and hemicellulose losses close to those of the negative control, or slightly lower in some cases, which indicated that there was a limitation of these contaminant microorganisms to degrade the highly recalcitrant Miscanthus, providing more evidence for the hypothesis of a symbiotic effect in the first non-sterile generation between C. subvermispora and the contaminant microorganism.
30%
45% 25% 60% 20% 75%
15%
Hemicellulose lossHemicellulose 10%
5%
0% Positive First Second Third Negative control generation generation generation control
Figure 4.6 Hemicellulose loss after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment
without inoculation. Postive control: sterilized Miscanthus inoculated with C.
subvermispora mycelium from liquid culture.
110
For the first non-sterile generation, hemicellulose loss was similar to previous reports of C. subvermispora in non-sterile conditions with Miscanthus and hardwood (Vasco-
Correa et al., 2016a; Zhao et al., 2014a). However, cellulose loss was significantly higher in this case than in the aforementioned studies, suggesting that there was more dominance of the indigenous microbial community with higher cellulolytic capabilities than C. subvermispora.
Dry matter loss was 8-11% for the positive control, and 10-22% for first non-sterile generation (Figure 4.7). Dry matter loss in the positive control was mostly due to lignin and hemicellulose degradation by C. subvermispora. Moisture contents of 60% and 75% in the first non-sterile generation showed the highest dry matter loss of 21-22%, while
45% moisture content in the same generation only showed a 10% dry matter loss. This was probably caused by a limitation of the growth of some microorganisms at the low moisture content, due to the lack of water availability. The negative control showed a dry matter loss of 8-9%, showing that the indigenous microorganisms of the Miscanthus have a limited capacity of dry matter degradation of the biomass, and that the higher degradation caused in the first non-sterile generation was due to a synergistic effect of C. subvermispora with these microorganisms, which was also supported by the low dry matter loss observed in the second and third non-sterile generation of fungal pretreatment.
111
30% 45% 25% 60%
20% 75%
15%
10% Dry matter lossmatter Dry
5%
0% Positive First Second Third Negative control generation generation generation control
Figure 4.7 Dry matter loss after 28 days of fungal pretreatment of Miscanthus at 45%,
60%, and 75% initial moisture content. Negative control: incubated alown treatment
without inoculation. Postive control: sterilized Miscanthus inoculated with C.
subvermispora mycelium from liquid culture.
4.3.3 Microbial community analysis
There was a high diversity of fungal species in the untreated Miscanthus and in the negative control of the fungal pretreatment, which showed similar number of OTUs, richness, and evenness (Figure 4.8). The number of observed fungal species (OTUs) was similar to that found by similar methods in compost (Tian et al., 2017) and aboveground samples of poplar (Durand et al., 2017), considerable higher than that in wood fibers
(Montagne et al., 2017) and wheat (Hertz et al., 2016; Shi et al., 2017), and consistent with the fact that phyllosphere fungi is usually the most species-rich group of fungi
(Unterseher et al., 2011). The positive control had a low fungal diversity and low
112 evenness, which was expected because the material was autoclave prior fungal inoculation and end up being dominated by one fungal species. However, in the first generation of fungal pretreatment, even though half of the starting material consisted of untreated Miscanthus that was mixed with finished material from the positive control, increase in the number of observed species from the level of the positive control was minor. Furthermore, although an increase in number of OTUs and richness in the following generations was evident (Figure 4.8a and Figure 4.8b), levels similar to the untreated material were not reached. This could be an effect of the presence of C. subvermispora as a dominant species in the positive control used as inoculum for the first generation, which could influence the community structure of the other fungi favoring certain species. Similar results were observed in composting, where inoculation with particular fungi decreased the fungal diversity of the final compost compared with the non-inoculated control (Tian et al., 2017). This lack of diversity could also be caused by the structural and compositional changes in the material caused during the positive control incubation, which could favoring some fungal species or create conditions to inhibit others, since half of the material in the first generation was finished material from the positive control. Physical and chemical habitat modification is a well-documented form of indirect microbial interaction (Dighton and White, 2017).
113 a) b)
Untreated (n=3) Untreated (n=3) c) d)
Untreated (n=3) Untreated (n=3)
Figure 4.8 Alpha diversity of the fungal (ITS) data. a) Observed species (OTUs), b)
Chao1 richeness index, c) Shannon evenness index, and d) Simpson diversity index.
Fungal communities evolved significantly during the generations of fungal pretreatment. As stated above, there was a clear increase in the diversity through the generations, and it seemed like the community was evolving towards a community similar to the negative control (Figure 4.10). At the phylum level, positive control showed 99.7% of the species relative abundance in the phylum Basidiomycota (Figure
4.9), which was all from the genus Ceriporiopsis (Figure 4.10), as expected. Conversely the untreated material had 89.7% of the species relative abundance in the phylum
Ascomycota, and only 6.7% in the phylum Basidiomycota, and the proportion was
114 conserved in the negative control of the fungal pretreatment (89.1% Ascomycota and
10.5% Basidiomycota) (Figure 4.9). This proportion was similar to that found in different phyllosphere samples previously (Meiser et al., 2014), and differed significantly from fungal communities in soil, where basidiomycetes are usually more abundant (Tedersoo et al., 2014). First generation fungal pretreatment showed 23.6% of the species relative abundance in the phylum Basidiomycota, while the rest was in the phylum Ascomycota, and this value decreased to about 9-12% in the next two generations. At the class level, untreated material was composed primarily from Dothideomycetes (40.9%),
Eurotiomycetes (32.8%), and Sordariomycetes (6.3%). These classes were also prevalent in previous reports of phyllosphere, especially Dothideomycetes (Meiser et al., 2014), which was rare in other habitats (Zhang et al., 2013).
115
100%
80%
60% K-Fungi UD Basidiomycota
40% Ascomycota Relativeabundance
20%
0% Untreated Positive Gen1 Gen2 Gen3 Negative
Figure 4.9 Fungal community composition of untreated and fungal pretreated Miscanthus
expressed as relative abundance at the phylum level. UN: unidentified.
The abundance of C. subvermispora, which was predominant in the positive control
(99.7%), was reduced to only 9% by the first non-sterile generation, and was almost undetectable in the second and third generation. Apparently, the conditions of the non- sterile generation were unfavorable for C. subvermispora, and it was rapidly outcolonized. The decreased in its abundance suggested that other microorganisms might have used the fungal hyphae as substrate, as occurs in advance stages of wood decay in nature (van der Wal et al., 2015). Conversely, there was a strong, increasing presence of other fungi in the non-sterile fungal pretreatment. The abundance of an unidentified fungus of the order Myriangiales was 56-67% in first to third generation, and was only
12% in the untreated Miscanthus and 19% in the negative control. This order consist of
116 plant pathogens and entomopathogenic fungi, and includes endophytes and saprotrophs
(Fan et al., 2017; Samson et al., 1988). The other two dominant genera in the fungal pretreated non-sterilized samples were one unidentified from the order Helotiales (14-
17%) and Marchandiomyces (9-15%). The latter was also present in the untreated
Miscanthus (2%) and negative control (3%), but the former was almost undetectable in those samples (Figure 4.10). The order Heliotiales includes saprophytes, endophytes, and parasites, and although it has been encountered before in decaying wood, its role on decay communities in nature is not well understood (van der Wal et al., 2015). Since the abundance of the unidentified Heliotiales was insignificant in the untreated material and negative control, it is believed that its growth was particularity favored by the conditions in the fungal pretreated samples, possibly the feedstock composition changes caused by
C. subvermispora. In the case of Marchandiomyces (basidiomycetes), one common species of this genus is M. lignicola, a wood-inhabit lignicolous fungus (Depriest et al.,
2005). However, it is unclear if this was the fungal species detected, and how it could have affected the delignification during the fungal pretreatment, or its interaction with C. subvermispora.
Fungi from the genus Stachybotrys were present in the untreated Miscanthus, and can be observed significantly in the negative control, but their presence was significantly reduced in the fungal pretreatment samples, which seems to indicate that the conditions created by in the fungal pretreatment particularly inhibit these fungi. This taxa of widely spread molds that tend to inhabit cellulosic materials contains more than 50 species, including some indoor molds (Haugland et al., 2001). Similar phenomenon happened to
117 another genus of highly common mold, Aspergillus, a taxa with more than 300 species found in many different habitats (Samson et al., 2014).
Competitive interaction between fungal species is a very common occurrence in wood decay at different stages and it has a high influence in the fungal communities
(Dighton and White, 2017). White rot fungi are often primary colonizers in wood decay, which gives them a competition advantage or priority effect. However, they are usually eventually replaced by secondary colonizers, many of which take advantage of the increased availability of carbohydrate polymers or even from the fungal mycelium as carbon an energy source (Dighton and White, 2017). A similar effect seemed to have occurred in the fungal pretreatment of non-sterile Miscanthus. Fungi from the phylum
Ascomycota have been observed as the prevalent in fungal communities at advance stages of wood decay, in wood that decomposed faster at the beginning (van der Wal et al.,
2015), suggesting that ascomycetes have the ability of replacing basidiomycetes such as white rot fungi, after these have acted. In this study, C. subvermispora seemed to be outcolonized mostly by ascomycete fungi. Wood decay fungi also have low growth rates compared to other fungi, which gives them a disadvantage for competition in scenarios such as the non-sterile solid-state fungal pretreatment.
118
100% Ceriporiopsis O-Myriangiales UN Aspergillus O-Helotiales IS Stachybotrys 80% Marchandiomyces Trichoderma P-Acomycota UN Alternaria 60% Chaetomium Coprinopsis Phoma Talaromyces F-Phaeosphaeriaceae UN 40% K-Fungi UN
Cladophialophora Relative abundance Relativeabundance O-Pleosporales UN Aureobasidium Davidiella 20% Acremonium Neurospora Xylomelasma Wallemia Hypocrea 0% F-Nectriaceae UN Untreated Positive Gen1 Gen2 Gen3 Negative Other
Figure 4.10 Fungal community composition of untreated and fungal pretreated
Miscanthus expressed as relative abundance at the genus level. UN: unidentified. OTU
assignments with less than 1% abundance are represented as ‘other”.
PCA analysis showed clustering of the samples among different generations (Figure
4.11). There was a clear difference of the fungal pretreated samples (positive control and generation 1, 2 and 3) with the untreated Miscanthus and negative control, which was displayed mostly in the first principal component, which explained about 77% of the data. There seemed to be high similarities between the second and third generation, which was confirmed by the results of the changes in composition and digestibly reported 119 in Section 4.3.1 and 4.3.2. A progressive change from positive control to the third generation was expressed mostly in the second principal component, which explained about 15% of the total variation of the data. This showed that there were deeper differences between these two groups the fungal pretreated samples and the untreated
Miscanthus and negative control, than between generation in the fungal pretreatment, and that the structural and chemical changes caused by the presence of C. subvermispora in the positive control have a high influence on the rest of the fungal pretreatment generation, even if C. subvermispora was rapidly outcolonized and the following pretreatments not successful.
120
Figure 4.11 Principal component analysis (PCA) for the fungal community in the
untreated and fungal pretreated Miscanthus.
While fungal data showed an average of about 50,000 sequences per sample with a minimum of 22,438 sequences, bacterial sequences were only about 15,000 in average.
Also, bacterial data showed a very low diversity in samples, similar or lower than that of the background (buffer), which suggested that the majority of those OTUs were probably artifacts that were not eliminating during rarefication, since this process was performed at a lower number of reads compared to the fungal data, indicating that there was a very low presence of bacterial contamination in the samples (Figure 4.12a). Also, most of those
OTUs were unidentified, which together with the low relative diversity suggested that most of the amplified sequences were part of the MiSeq background and not actual 121 bacteria (Figure 4.12b). The samples incubated at an initial moisture content of 60% were the ones used for the microbial communities study, which could help explained the low bacterial presence compared with the fungal, since fungi are generally more resistant to water stress than bacteria (Manzoni et al., 2012). Also, most bacteria usually have less capabilities to degrade complex lignocellulosic materials, which made that environment unfavorable for bacterial growth. These results contrasted with previous findings where the white rot fungus P. chrysosporium inoculation in compost stimulated the increase in bacterial abundance (Zhang et al., 2013). This confirmed that fungi was the predominant type of microorganism in the fungal pretreated samples and the responsible for outcompeting C. subvermispora. From the identified OTUs, the core OTUs (over 250 read in 75% of the samples) were almost all Bacillus.
a) b) 100% Unassigned
Other 80% Acidobacteriales Sphingomonadales 60% Xanthomonadales Rhizobiales 40% Actinomycetales
Alteromonadales Relativeabundance Number of species (OTUs)species of Number 20% Streptophyta Oceanospirillales 0% Burkholderiales
Bacillales
Gen3 Gen2 Gen1
Untreated (n=3)
Positive
Negative Untreated
Figure 4.12 Bacterial communities in untreated and pretreated Miscanthus. a) Alpha
diversity of bacterial (16S) data. B) Bacterial community composition as relative
abundance at the order level.
122
4.4 Conclusions
Sequential fungal pretreatment of non-sterile Miscanthus increased the enzymatic digestibility of the feedstock by 2-fold compared to the untreated Miscanthus in the first generation, but it was unsuccessful in further generations, due to an outcolonization of the white rot fungi C. subvermispora by other fungi, mostly ascomycetes. The findings in this study confirmed the requirements of feedstock sterilization, or at least a significantly microbial reduction, for the stability and reliability of fungal pretreatment. The sensitivity of white rot fungi like C. subvermispora to fungal contamination is probably due to their slow growth rate, and/or the physical and chemical modifications they caused to the feedstock, making it then more favorable for other types of fungi.
123
Chapter 5: Changes in composition and structure during sequential fungal pretreatment of non-sterile feedstocks
Abstract
Fungal pretreatment with C. subvermispora was performed under sterile and non- sterile conditions for four different feedstocks: corn stover, Miscanthus, softwood, and hardwood, using fungal-colonized feedstock as inoculum. Significant differences in the effect of the fungal pretreatment were observed for different feedstocks with regards to changes in their structure, composition, and digestibility. Fungal pretreatment increased the enzymatic digestibility of hardwood, softwood, and Miscanthus by 2 to 4.5-folds, but was only successful for the first non-sterile generation. Fungal pretreatment of corn stover was not successful. Modifications in structure and composition were analyzed using microscopy, spectroscopy and thermogravimetric analysis, and showed that fungal pretreatment with C. subvermispora produced different effects that are feedstock- dependent.
5.1 Introduction
Obtaining the amount of biomass necessary to achieve the goals for biofuels production would only be possible by using a combination of resources including agricultural, forestry and waste resources, agricultural residues, and energy crops (U.S.
DOE, 2016), and depending on only few types of biomass would be unfeasible for the 124 bioeconomy. However, lignocellulosic biomass presents high variability in the proportions of the cell wall components, their physical arrangement, and their chemical interactions. Lignin especially, heterogeneous by nature, varies significantly depending on the plant species and clone, the development stage, and the environment. Therefore, transformation processes, particularly pretreatments, have been found to have differential effects on different types of lignocellulosic biomass, and they are usually tailored for one specific feedstock.
White rot fungi have a role of carbon recycling in nature, and they are adapted to certain type of feedstock, depending on their specific natural habitat. In general, most white rot fungi grow preferentially in wood. C. subvermispora, for example, has been usually observed and isolated from rotting conifers (softwood) and hardwoods.
Therefore, the effects of fungal pretreatment have been seen to vary significant across feedstocks. Wan and Li (2011) pretreated five different feedstocks with C. subvermispora. Pretreatment was effective for corn stover, switchgrass, and hardwood, but it was unsuccessful for wheat straw and soybean straw under the same conditions. In another study, C. subvermispora degraded distinctively different fractions of corn stover, showing a higher lignin, cellulose, and xylan degradations in corn stover leaves than in cobs and stalks (Cui et al., 2012b).
It is hypothesized that fungal pretreatment of diverse lignocellulosic feedstocks would produce different results with respect to changes in enzymatic digestibility, composition, structure, and physical-chemical properties. In order to further understand the particular impact of fungal pretreatment with C. subvermispora on different lignocellulosic
125 structures, four different types of lignocellulosic feedstocks were selected for the present study: corn stover, Miscanthus, softwood, and hardwood. Changes in enzymatic digestibility, composition, structure, and thermogravimetric properties during different stages of the sequential non-sterile fungal pretreatment of the feedstocks were studied, and compared to those of the fungal pretreatment of sterile feedstocks.
5.2 Materials and Methods
5.2.1 Feedstock collection and storage
Miscanthus × giganteus (hereafter Miscanthus) was harvested from a field in
Zanesville, OH. Corn stover, softwood (pine), and hardwood (white ash) were harvested in Wooster, OH. All feedstocks were obtained in December 2015. The feedstocks were dried at 40˚C in a convection oven for 24 h, milled to pass through a 12 mm screen using a hammer mill (The C.S Bell Co., Tiffin, OH, USA), and stored under dry conditions.
Composition of the feedstocks and total solid contents before and after drying are shown in Table 5.1.
126
Table 5.1 Composition and total solids of feedstocks
Miscanthus Corn stover Softwood Hardwood
Total solids – as 74.0 ± 1.1 91.4 ± 0.1 77.4 ± 0.5 65.7 ± 0.5 received (%) Total solids – 94.6 ± 0.0 93.9 ± 0.2 93.8 ± 0.1 94.3 ± 0.1 after drying (%)
Cellulose (%*) 37.2 ± 0.3 31.5 ± 0.5 29.4 ± 0.3 29.0 ± 0.1
Hemicellulose 17.3 ± 0.1 19.3 ± 0.3 16.2 ± 0.4 13.7 ± 0.2 (%*)
Lignin (%*) 18.1 ± 0.2 13.4 ± 0.2 25.4 ± 0.8 25.7 ± 0.4
Extractives (%*a) 8.3 ± 0.2 14.5 ± 0.6 16.8 ± 0.0 7.7 ± 0.2
Ash (%*) 3.4 ± 0.2 5.6 ± 0.1 0.5 ± 0.0 1.8 ± 0.0
Others (%*b) 15.7 ± 0.5 15.7 ± 0.9 11.7 ± 0.9 22.1 ± 0.5
*Based on dry weight aIncludes water and ethanol extractives bOther non-identified components of the biomass (i.e. acetic acid and other organic acids, proteins), calculated by difference.
5.2.2 Inoculum preparation
Sterile feedstock colonized with C. subversmipora was used as the inoculum for the fungal pretreatment experiments, as developed in Chapter 3. To prepare this inoculum, untreated feedstock was added in 2 L reactors (130 g for Miscanthus and corn stover, and
260 g for hardwood and softwood, dry basis), along with deionized water to adjust moisture to 60%. Reactors were covered with cotton and autoclaved (121°C, 15 min). C. subvermispora mycelium was grown in 250-ml flasks with 50 ml of 2% (w/v) malt extract liquid medium for 7 days at 28°C and static conditions. Mycelium was then
127 filtered through cheesecloth and homogenized. Autoclaved reactors were inoculated with the homogenized mycelium and then incubated for 28 days at 28°C. At the end, the fungal-colonized feedstocks were mixed thoroughly before using them as inoculum for the successive fungal pretreatment of non-sterile feedstocks.
5.2.3 Fungal pretreatment experiments
Non-sterile feedstocks and inocula (fungal-colonized feedstocks) were mixed in a ratio of 1:1 (dry weight basis) and added to 1 L reactors. Deionized water was added to adjust moisture content to 60% (w/w) and reactors were incubated at 28°C for 14 days.
Negative controls consisted in non-sterile feedstock adjusted to 60% (w/w) moisture content with deionized water and incubated along the treatments. A sterile pretreatment was also included for comparison, using sterile feedstock inoculated with mycelium (as described in section 2.2), which was incubated along with the treatments. At the end of the 14 days, samples were taken out of the reactors and mixed well. Part of the material was used as inoculum for a second generation of fungal pretreatment. This generation was set-up, as the first one, using a ratio of 1:1 (dry weight basis) non-sterile feedstock and fungal colonized material (which in this case was the product of the first generation), adjust to a moisture content of 60% (w/w) with deionized water, and incubated at 28˚C for 14 days. This procedure was repeated one more time, for a third generation of non- sterile fungal pretreatment.
At the end of the 14 days of each generation, samples were taken out of three of the reactors and mixed thoroughly. Part of the material was used fresh for enzymatic
128 hydrolysis and SEM, and the rest was dried at 40°C in a convection oven for 24 h, then milled and passed through a 1 mm screen (Model 4 Wiley Mill, Thomas Scientific,
Swedesboro, NJ, USA) for composition determination, Py-GC-MS, TGA, and FT-IR.
5.2.4 Enzymatic hydrolysis
Enzymatic digestibility of untreated and pretreated feedstocks was measured as described in section 3.2.4.
5.2.5 Analytical methods
Cellulose, hemicellulose, lignin, and dry matter degradation were calculated considering the initial content as the content of the mixture (untreated feedstock plus inoculum). Analytical methods were performed as described in section 3.2.5.
5.2.6 Scanning electron microscopy
Fresh untreated and pretreated samples were cut into pieces of 1-3 mm and incubated overnight in a structure fixative solution containing 3% glutaraldehyde and 2% paraformaldehyde in a 0.1 M phosphate buffer. Further, samples were washed with DI water and dehydrated with increasing concentrations of ethanol, until 100% ethanol was achieved. Then, samples were dry to the critical point (Autosamdri-814, Tousimis
Research Corporation, Rockville, MD, USA), cover in platinum with a 0.2 kÅ thickness
(Hummer 6.2 Sputtering System, Anatech Ltd, Battle Creek, MI, USA), and observed
129 under the SEM (Hitachi S-3500N Scanning Electron Microscope, Hitachi Ltd, Tokyo,
Japan).
5.2.7 Thermogravimetric analysis
Thermogravimetric (TG) analysis was carried out at a heating rate of 15˚C/min from
40˚C to 800˚C under nitrogen atmosphere using a Q50 thermogravimetric analyzer (TA
Instruments, New Castle, DE, USA). TG curves and its derivatives (DTG) were obtained using 5-10 mg of samples and analyzed with the Universal Analysis 2000 software version 4.5A (TA Instruments, New Castle, DE, USA).
5.2.8 Analytic pyrolysis
Pyrolysis GC-MS was performed using a multi-shot pyrolyzer (EGA/PY-3030D,
Frontier Laboratories, Fukushima, Japan) coupled with gas chromatography-mass spectrometry (GC-MS, GCMS-QP2010 SE, Shimadzu Corp, Kyoto, Japan) via a ZB-
5HT column (30 m, 0.25 mm I.D., 0.25 μm, Phenomenex, Torrance, CA, USA). The pyrolysis was carried out at 600˚C for 12 s. The chromatograph was programmed from
50˚C (1 min) to 280˚C at a rate of 5˚C/min. The final temperature was held for 6 min.
The mass spectrometer was operated in electron ionization (EI) mode with m/z scan range of 35-800, and its ion source temperature and the interface temperature were kept at 200˚C and 280˚C, respectively. The pyrolyzates were identified by comparing their mass spectra with those reported in the NIST library.
130
5.2.9 Fourier transform infrared spectroscopy
FT-IR spectra were recorded using a Spectrum Two FTIR spectrophotomer (Perkin
Elmer Inc., Waltham, MA, USA) equipped with a universal attenuated total reflectance accessory (UATR). FT-IR spectra were averaged over 16 scans from 4,000 to 450 cm-1 wavenumber with a resolution of 4 cm-1, baselined corrected and normalized using the software Spectrum 10 (Perkin Elmer Inc., Waltham, MA, USA).
5.2.10 Data analysis
Statistical analyses were performed using the Software JMP® (SAS Institute Inc.,
Cary, NC, USA). Statistical significance was evaluated by analysis of variance
(ANOVA) (α=0.05), and mean comparisons were performed by Tukey-Kramer test with a 95% significance. Principal component analysis (PCA) was performed to relative abundance data of the components detected by Py-GC-MS. Error bars in the figures represent the standard error.
5.3 Results and Discussion
5.3.1 Enzymatic digestibility
Enzymatic digestibility increased 2-, 3- and 4.5- fold after the first generation non- sterile fungal pretreatment of Miscanthus, softwood, and hardwood, respectively, compared with the untreated material (Figure 5.1). For Miscanthus, a similar increased was obtained with the sterile pretreatment (positive control) (p>0.05). However, for softwood and hardwood, the enzymatic digestibility after the non-sterile pretreatment was
131 considerable higher than the positive control, and similar to that of the inoculum, which was prepared under the same conditions as the positive control, but incubated for 28 days instead of 14 days. These times were selected because fungal pretreatment when inoculated with mycelium from a liquid culture showed a lag phase of about 7 days (Wan and Li, 2010a), while fungal pretreated inoculated with fungal colonized feedstock has shown significant increase in enzymatic digestibility at shorter times (Vasco-Correa et al., 2016a). This suggested that the fungal pretreatment of non-sterile feedstocks required less incubation time than the sterile pretreatment, which agreed with the fact that, in the former case, half of the material was already colonized with the fungus, the initial enzymatic digestibility was therefore higher, and the fungus was already adapted to the feedstock.
Conversely, corn stover showed a 38% reduction in enzymatic digestibility after non- sterile fungal pretreatment (Figure 5.1), which has been characteristic of unsuccessful fungal pretreatment (Liu et al., 2016; Vasco-Correa and Li, 2015), and was probably due to consumption of the easily accessible sugars by the indigenous microorganisms, leaving a material more recalcitrant to enzymatic hydrolysis. This reduction of enzymatic digestibility was also observed in the negative control (55% reduction), which consisted of non-sterile corn stover without fungal inoculation incubated under the same conditions of the treatments, and in the second generation non-sterile pretreatment (53% reduction).
The sterile pretreatments of corn stover showed a similar or slightly higher enzymatic digestibility than the untreated material. Also, fungal growth was not detectable to the naked eye for any of the corn stover treatments, but it was easily observable in the
132 successful fungal pretreatments (sterile and first non-sterile generation) of miscanthus, softwood, and hardwood. The negative results of the fungal pretreatment for corn stover could be caused by a number of factors, including, but not limited to, (1) a low water activity (measurement of water available for the fungal growth), since this material was more hygroscopic than the other feedstocks, (2) a higher compaction of the material that generated low oxygen availability, or (3) a faster-grower indigenous microbial community in corn stover compared with the other feedstocks. An additional experiment was performed for fungal pretreatment of corn stover at 75% initial moisture content and, after the first non-sterile generation, the enzymatic digestibility of the material was 12.5 ±
1.3% and the lignin degradation negligible, which were not significantly different from the results at 60% moisture content (p>0.05), suggesting that low water availability was likely not the main factor causing the negative results in the fungal pretreatment of corn stover. The results for fungal pretreatment of corn stover were similar to those obtained by Saha et al. (2016) with C. subvermispora incubated for 30 days in sterile corn stover, where there was no lignin degradation and no significant increase in sugar yield (Saha et al., 2016). However, other authors have been able to successfully pretreat corn stover with C. subvermispora and other white rot fungi. Wan and Li (2010) obtained a glucose yield of about 57% after 18 days of incubation of C. subvermispora in sterile corn stover at 75% moisture content (Wan and Li, 2010a), and equivalent results were obtained in other studies (Wan and Li, 2011b, 2010b). Enzymatic digestibility of about 47% was obtained with incubation of P. chrysosporium and T. hirsuta yj9 for 2 weeks in sterile corn stover (Sun et al., 2011; Zhao et al., 2012).
133
40% UntreatedRaw feedstockfeedstock 35% Negative control -14 days Sterile pretreatment -14 days (Positive control) Sterile pretreatment - 28 days (Inoculum) 30% First generation -14 days Second generation -14 days Third generation -14 days 25%
20%
15% Enzymatic digestibility Enzymatic 10%
5%
0% Miscanthus Corn stover Softwood Hardwood
Figure 5.1 Enzymatic digestibility of untreated and fungal pretreated feedstocks.
Negative control: unsterile uninoculated feedstocks incubated along treatments.
Both feedstock type and generation had a significant effect on enzymatic digestibility
(p<0.05). Fungal pretreatment enhanced the enzymatic digestibility of wood more effectively than the other feedstocks, as wood is white rot fungi’s natural feedstock
(Eriksson et al., 1990). Higher increase in enzymatic digestibility was obtained with hardwood than softwood, given that most white rot fungi have a preferential degradation for hardwood than softwood (Tuor et al., 1995), softwood is usually more recalcitrant to enzymatic hydrolysis and more naturally resistant to microbial decay (Scheffer and
Cowling, 1966), and softwood typically has a higher content of antimicrobial compounds in the extractives that can inhibit fungal growth (Himejima et al., 1992). Yu et al. (2009) 134 compared the effect of fungal pretreatment with the white for fungus E. taxodii in hardwood and softwood, and obtained values of enzymatic digestibility of about 33% and
17%, respectively, which were remarkably similar to the ones obtained in the present study (Figure 5.1). However, in that study the incubation time of the fungal pretreatment was 120 days (Yu et al., 2009). Fungal pretreatment results in this work contrasted with those obtained by Wan and Li (2011) using C. subvermispora for 18 days, where the highest glucose yield (56%) was obtained with corn stover, while pretreated hardwood only produced about 24% glucose yield, even though that hardwood had a lower lignin content than the one used in this study, exposing the intrinsic variability of the fungal pretreatment process (Wan and Li, 2011b).
For the most part, second and third generation fungal pretreatment of non-sterile feedstocks were unsuccessful, suggesting that the indigenous microbial community propagates through the generations for all feedstocks, outcolonizing C. subvermispora by the second generation. Only the second generation fungal pretreated softwood had a significantly higher (2-fold) enzymatic digestibility than the untreated material, but it was lower than that of the first generation fungal pretreated softwood, and by the third generation, the enzymatic digestibility was only about 50% higher than the untreated softwood.
5.3.2 Components degradation during fungal pretreatment
Cellulose loss was minimal (between 0% and 5%) during sterile pretreatment (Table
5.2), which is consistent with previous reports (Wan and Li, 2011b), since C.
135 subvermispora activates its cellulolytic machinery after longer incubation times (Tanaka et al., 2009), and it is less strong than in other white rot fungi (Fernández-Fueyo et al.,
2012a). After the first generation non-sterile fungal pretreatment, cellulose loss was still low (5-8%), except for corn stover where it reached 27%, since the contaminant microorganisms were likely feeding on it. Hemicellulose loss was 10-13% in the successful fungal pretreatments of hardwood, probably because this has been considered to be the main carbon source for C. subvermispora (Wan and Li, 2012). However, hemicellulose loss was not as high in other successful pretreatments, such as the ones for softwood where hemicellulose loss was about 2%. This suggest that fungal biomass growth varied with the feedstock, and that the structural differences of the hemicelluloses from different feedstocks are degraded distinctively by C. subvermispora. Remarkably, cellulose and hemicellulose loss in the negative control of corn stover was 18% and 19%, respectively, which was significantly higher than for the other feedstocks (p<0.05). That suggests that corn stover microbial community grew faster than that of the other feedstocks, and easily outcolonized C. subvermispora from the first generation non- sterile pretreatment. Dry matter loss of over 15% occurred in most treatments of corn stover, except for the positive control, and was higher than the dry matter loss after other unsuccessful pretreatments, such as second and third generation fungal pretreated
Miscanthus, softwood, and hardwood, further supporting the presence of a faster-growing indigenous microbial community in corn stover.
136
Table 5.2 Components degradation during fungal pretreatment
Miscanthus Corn stover Softwood Hardwood Negative control 0.46 ± 1.09 1.23 ± 2.25 3.34 ± 1.34 2.97 ± 1.62 Positive Lignin loss control 5.75 ± 1.08 4.64 ± 0.59 4.06 ± 0.39 26.30 ± 0.97 First (%) generation 6.82 ± 0.41 -0.27 ± 1.11 4.32 ± 0.63 26.97 ± 1.45 Second generation 4.23 ± 0.93 19.28 ± 1.67 7.22 ± 0.43 10.56 ± 0.57 Third generation 0.50 ± 2.70 N/A 0.83 ± 1.14 3.11 ± 0.59 Negative control 6.20 ± 0.75 17.94 ± 1.05 0.40 ± 1.09 -0.94 ± 0.36 Positive Cellulose loss control 2.96 ± 0.73 0.94 ± 1.13 0.02 ± 0.37 5.10 ± 1.02 First (%) generation 5.06 ± 1.07 27.59 ± 1.68 -0.87 ± 1.80 8.18 ± 0.82 Second generation 7.14 ± 1.09 24.93 ± 0.20 5.28 ± 0.32 22.87 ± 2.29 Third generation 10.65 ± 1.56 N/A 1.58 ± 0.98 3.94 ± 1.99 Negative control 4.59 ± 0.98 18.69 ± 0.75 4.75 ± 0.09 4.94 ± 1.27 Positive Hemicellulose control 11.39 ± 0.71 4.83 ± 1.27 2.30 ± 1.19 12.68 ± 1.00 First loss (%) generation 4.14 ± 1.37 26.67 ± 0.89 1.76 ± 0.96 10.45 ± 1.52 Second generation 7.90 ± 1.46 24.43 ± 0.33 4.72 ± 1.87 19.02 ± 1.52 Third generation 7.93 ± 1.77 N/A -0.84 ± 3.02 1.33 ± 1.44 Negative control 2.57 ± 0.68 16.29 ± 0.56 2.75 ± 0.55 2.77 ± 0.61 Positive Dry matter control 3.61 ± 0.78 3.22 ± 0.75 2.65 ± 0.70 12.45 ± 0.17 First loss (%) generation 4.05 ± 1.11 21.23 ± 0.46 1.86 ± 0.02 15.50 ± 0.59 Second generation 2.18 ± 1.07 15.88 ± 0.83 2.63 ± 0.35 12.43 ± 0.11 Third generation 6.38 ± 1.49 N/A 1.12 ± 0.14 1.85 ± 0.53 N/A: not applicable (third generation corn stover was not performed)
137
Lignin degradation of over 26% occurred during fungal pretreatment of sterile and first generation non-sterile hardwood, and it was similar to that obtained for other feedstocks in similar conditions (Wan and Li, 2011b). However, lignin degradation was only around 4-7% for Miscanthus and softwood. There was a significant correlation between lignin degradation and increase in enzymatic digestibility (Figure 5.2), which indicated that lignin is one of the main factors affecting biomass digestibility, as has been shown previously (Chen et al., 2010; Vasco-Correa et al., 2016a).
6
5 R² = 0.8093 4
3 (folds) 2
1
Increase in enzymatic digestibility enzymatic in Increase 0 -5% 5% 15% 25% 35% Lignin degradation
Figure 5.2 Correlation between lignin degradation and increase in enzymatic digestibility
in fungal pretreated feedstocks.
5.3.3 Structural changes in biomass
Microscopy studies showed fungal colonization and changes in cell wall structure in all the feedstocks (Figure 5.3 and Figure 5.4). The pretreated feedstocks from the positive
138 controls were all well-colonized by what it is expected to be C. subvermispora hyphae, since the feedstocks were autoclave prior fungal inoculation. This shows that there was indeed some growth of C. subvermispora in the corn stover (Figure 5.3e), even though there was no increase in the enzymatic digestibility. It is possible than the ligninolytic system was not activated in this case to the same extend, which could be caused be the availability of nutrients (mainly nitrogen) and water (Linnard et al., 2013), or by oxygen depletion due to the compaction of the material, which is required for the oxidative reactions that depolymerize lignin (Reid and Seifert, 1982). Corn stover first-generation fungal pretreated material showed presence of several types of microorganisms, including fungi with different morphology than C. subvermispora and bacteria, which confirms the presence of indigenous microorganisms in this pretreatment (Figure 5.3f).
For Miscanthus, an evident increase in porosity and cell wall degradation is observed in the positive control (Figure 5.3b) compared with the smooth surface of the untreated material (Figure 5.3a), and by the first non-sterile generation of fungal pretreatment, the cell wall was almost completely disrupted, and significant degradation can be observed, not only in the surface, but in deeper cells layers (Figure 5.3c). The pattern of degradation for the latter case had characteristics of that caused by fungi that simultaneous degrades lignin and cellulose during the rotting process, and not that of selectively delignifiers such as C. subvermispora, since there was an extensive degradation of the cell wall
(Akhtar et al., 1997). This was probably caused by the presence of microbial contaminants that could feed on the cellulose exposed after the delignification.
139
Even though abundant colonization was observed in fungal pretreated softwood and hardwood, the degree of cell wall deconstruction was significantly higher for the latter.
Fungal hyphae penetrated the vascular system of the hardwood and by the first non- sterile generation, the cell wall on that area was almost completely disintegrated (Figure
5.4e and Figure 5.4f). These observations correlated well with the enzymatic digestibility of the materials (section 5.3.1). Also, degradation of the middle lamella in the cell corners was observed in hardwood (arrowhead, Figure 5.4e), which is characteristic of the selective delignification expected from C. subvermispora (Srebotnik and Messner, 1994).
There was also evident presence of deposits of unknown material in the surface after most treatments, similarly to previous findings (Xu et al., 2010), probably due to migration of structural carbohydrates and lignin, product of the disruption of the cell wall.
140
Untreated material Sterile pretreatment (positive control) First generation non-sterile pretreatment
a) b) c)
Miscanthus
d) e) f)
141
Cornstover
Figure 5.3 SEM images of untreated and fungal pretreated Miscanthus and corn stover
141
Untreated material Sterile pretreatment (positive control) First generation non-sterile pretreatment
a) b) c)
Softwood
d) e) f)
142
Hardwood
Figure 5.4 SEM images of untreated and fungal pretreated softwood and hardwood
142
5.3.4 Thermogravimetric analysis of untreated and pretreated feedstocks
Untreated feedstocks showed different TG degradation patterns, according to their composition (Figure 5.5). Woods showed a higher temperature at the maximum degradation rate (Tpeak) than Miscanthus and corn stover (Table 5.3), which means that a higher activation energy was necessary to degrade woody biomass, due to its higher lignin content (Burhenne et al., 2013). The thermal decomposition of the feedstocks occurred in three phases, including dehydration below 150˚C, active pyrolysis from about
150˚C to 600˚C, and passive pyrolysis and char formation after that, similar to that reported by other authors (Ma et al., 2013). The main peak in the differential thermogravimetric (DTG) curves was mostly due to volatilization of cellulose, while the left shoulder present in some of the curves was typically due to hemicellulose degradation. Lignin degradation occurs during a wider range of temperatures, and it could not be separated from the degradation peaks of the holocellulose (Yang et al., 2007).
However, the small peak on the right of the Miscanthus and corn stover DTG curves was likely part of the lignin degradation (Burhenne et al., 2013). A significantly lower amount of residue was produced by the thermal degradation of Miscanthus than the other feedstocks (Table 5.3), and it was very close to the amount of ash in the feedstock (Table
5.1), showing that almost all of the carbon was volatilized in Miscanthus, but not in the other untreated feedstocks.
Hemicellulose volatilization in softwood caused a shoulder at 330˚C. However, there was an additional peak with maximum at 238˚C, which was low for hemicellulose. This
143 peak was probably due to the volatilization of resins and other extractives, which are characteristic of pine wood (Grønli et al., 2002).
Figure 5.5 TG and DTG curves for untreated feedstocks (TG: dashed lines, DTG: solids
lines).
Fungal pretreated Miscanthus and hardwood showed more defined left shoulders
(Figure 5.6, Figure 5.7, and Figure 5.8), similarly to previous results obtained with fungal pretreatment of corn stover with different fungi (Ma et al., 2013; Zeng et al., 2011), and it was attributed to the modification/degradation of the hemicellulose. Particularly, there was more definition of the left shoulder in the positive control, where there was a higher
144 hemicellulose degradation, compared with the first generation non-sterile pretreatment
(Table 5.2). The small peak on the right disappeared in the fungal pretreated Miscanthus
(Figure 5.8), probably due to the lignin modification caused by the fungus. Also, fungal pretreated Miscanthus had higher maximum volatilization rates (DTGmax) than the untreated Miscanthus (Table 5.3), probably due to the effects of fungal pretreatment in recalcitrance reduction, and previously observed in other fungal pretreatments (Zeng et al., 2011). However, this phenomenon was not observed for hardwood and was only slightly noticeable in softwood, showing that the effects of fungal pretreatment on the TG properties of the material were feedstock-dependent. Fungal pretreated Miscanthus produced more char residue than untreated Miscanthus (Table 5.3). However, this pattern was not observed consistently for the other feedstocks. Previous research showed both increase (Ma et al., 2013) and decrease (Zeng et al., 2011) in char formation after fungal pretreatment. Char formation is influenced by the lignin content (Yang et al., 2007), but also by the type of lignin residues, and by the presence of inorganic compounds
(Nowakowski et al., 2008), that could be altered by the fungal growth and metabolism.
145
Figure 5.6 TG and DTG curves for first generation non-sterile fungal pretreated feedstock
(TG: dashed lines, DTG: solid lines).
The resin peak of the softwood (at 238˚C) was reduced in the first generation non- sterile fungal pretreatment (Figure 5.6) and almost completely disappeared in the fungal pretreatment positive control (Figure 5.7). Previous studies have shown that some microorganisms are able to degrade these type of extractives from pine wood (Burnes et al., 2000), including C. subvermispora (Fischer et al., 1996) and other white rot fungi
(Martínez-Inigo et al., 1999). Untreated Miscanthus also showed a small shoulder at a similar temperature probably attributed to extractives, that smoothed down completely after fungal pretreatment (Figure 5.8).
146
Figure 5.7 TG and DTG curves for positive control fungal pretreated feedstocks (TG:
dashed lines, DTG: solid lines).
147
untreated
Figure 5.8 TG and DTG curves for untreated and fungal pretreated Miscanthus.
Table 5.3 Thermogravimetric parameters of untreated and fungal pretreated feedstocks
a b c DTGmax (%/˚C) Tpeak (˚C) Final residue (%) Miscanthus 0.7927 339 4.9 Untreated Corn stover 0.6780 316 11.3 feedstock Softwood 0.7609 373 13.4 Hardwood 0.8115 362 16.4 Miscanthus 0.9095 346 19.4 First generation Corn stover 0.8713 330 24.4 non-sterile Softwood 0.7928 380 14.2 pretreatment Hardwood 0.8116 352 5.8 Miscanthus 0.9206 351 18.8 Corn stover 0.7847 323 4.5 Positive control Softwood 0.7749 373 15.0 Hardwood 0.8049 360 17.2 a DTG main peak height b Temperature at the DTG main peak c Percentage of weight of char residue at the end of the TGA under nitrogen atmosphere
148
5.3.5 Analytic pyrolysis of untreated and pretreated feedstocks
Fast pyrolysis of biomass broke the structural carbohydrates into a set of compounds that provided a deeper understanding of the composition of the feedstocks. The pyrograms for untreated and fungal pretreated feedstocks (first generation non-sterile pretreatment) are shown in Figure 5.9 and Figure 5.10. The list of the identified compounds and their relative abundances (peak area of the compound as a percentage of the peaks’ total area) are included in Table 5.4. Significant differences were observed among the pyrolysis products produced by different feedstocks. The fungal pretreated feedstocks produced similar compounds as their corresponding untreated feedstocks, but changes in the relative abundances were observed.
Acetic acid, a prominent peak at the beginning of the chromatogram, was not detectable in fungal pretreated Miscanthus and hardwood. This compound was mostly derived from the acetyl units of hemicellulose, but lignin could also be acetylated in certain feedstocks, such as hardwood (Karimi, 2015). Previous reports have shown the decrease of acetic acid after fungal pretreatment (Yang et al., 2011). Conversely, in softwood, the acetic acid relative abundance was significantly lower, consistent with reports that hemicellulose in this feedstock has lower acetyl units than in hardwood, mostly attached to the galactoglucomannan (Karimi, 2015). There was no significant reduction of those units after the fungal pretreatment of softwood, in accordance with the low hemicellulose degradation observed for this feedstock. Acetic acid reduction by fungal pretreatment is an added advantage if the feedstock is going to be used for fast pyrolysis, since acetic acid in one of the main reasons for the high acidity in bio-oils
149 derived from biomass pyrolysis, which causes corrosives and imposes needs for upgrading before using bio-oil as a fuel (Zhang et al., 2007).
Higher content of S-derived compounds was observed in hardwood compared to the other feedstocks. This included 1,2-benzenediol, 3-methoxy- (peak #45), syringol (peak
#49), phenol, 3,4-dimethoxy- (peak #51), 1,2,3-trimethoxybenzene (peak #55), 4-methyl-
2,6-dimethoxybenzaldehyde (peak #59), synringaldehyde (peak #62), and acetosyringone
(peak #63). Conversely, softwood showed a higher content of G-derived compounds such as guaiacol (peak #3), creosol (peak #40), 4-vinylguaiacol (peak #48), isoeugenol (peak
#50), eugenol (peak #56), and coniferol (peak #64). Coniferol showed in the pyrogram as a set of peaks in the eluted at retention time of 27-28 minutes.
Coumaran (2,3-dihydrobenzofuran, peak #43) was considered to be derived from H lignin, as reported previously (Yan et al., 2016; Yu et al., 2013). However, this polycyclic compounds have also been produced by pyrolysis of syringol and guaiacol before
(Kojima, 2015). This compound produced a prominent peak for corn stover and
Miscanthus, but it was significantly lower for woods (Table 5.4). Fungal pretreatment caused the reduction of this peak in Miscanthus and hardwood, and it was not detected in fungal pretreated softwood, but it increased in corn stover, suggesting that the reduction effect was due to the action of C. subvermispora.
Levoglucosan (peak #57) is the main pyrolysis product of cellulose, and it showed as a broad peak in the pyrograms. The relative abundance of this peak decreased in all fungal pretreated materials, compared with the respective untreated material, due to cellulose modifications produced by the fungal pretreatment, mostly because of to the
150 microbial contaminants that should have higher cellulolytic power than C. subvermispora.
After fungal pretreatment, the content of oxazolidine, 2,2-diethyl-3-methyl- (peak
#23) increased 27% in hardwood and 10% in Miscanthus. This compounds is derived from proteins, so this increase was believed to be related with fungal growth. Previous reports have shown a similar trend (Yang et al., 2011).
Interestingly, a set of terpene compounds was detected in untreated softwood, but not in any other feedstock, and most of those components were also absent in the pretreated softwood. These compounds (carene, terpinene, o-cymene, p-cymene, and D-limonene, peaks #24-28) have been identified as part of the pine resin (Paine et al., 1987), and were likely part of those extractives that produced a small peak in the DTG curve (section
5.3.4) at 238˚C in the untreated softwood, since it has been found that resins from pine wood are volatilized at similar temperature during TGA (Missio et al., 2015). These type of molecules have shown antifungal properties in previous studies (Himejima et al.,
1992), and could have contributed to the low performance of fungal pretreatment on softwood. However, the absence of most of these terpenes after fungal pretreatment suggested that they were degraded by C. subvermispora. Other fungi have been to be able to degrade terpenes in resins from pine, and this capacity could be utilized not only in biomass pretreatment, but in other biotechnological processes, such as bioremediation
(Vilanova et al., 2014).
151
1 3 (a) 48 11,12 43
9 14 33 2 40 49 56 7 15 19 23 37 52 18 22 29 31 32 41 46 57 64 4 13 16 21 44 45 47 50 53 54 60 6 17 20 34 35 39
(b) 1. 10. 20. 30.
0 0 0 0
(c) 1. 10. 20. 30.
0 0 0 0
(d) 1. 10. 20. 30.
0 0 0 0
1.0 10.0 20.0 (min)
Figure 5.9 Py-GC-MS of untreated Miscanthus (a), fungal pretreated Miscanthus (b),
untreated corn stover (c), and fungal pretreated corn stover (d).
152
(a)
(b) 1. 10. 20. 30.
0 0 0 0
(c) 1. 10. 20. 30.
0 0 0 0
1. 10. 20. (d30.)
0 0 0 0
1.0 10.0 20.0 (min)
Figure 5.10 Py-GC-MS of untreated softwood (a), fungal pretreated softwood (b),
untreated hardwood (c), and fungal pretreated hardwood (d).
153
Table 5.4 Relative peak area of components identified in Py-GC-MS of untreated and fungal pretreated feedstocks
Peak* Corn stover Miscanthus Hardwood Softwood Compound # Untreated Pretreated Untreated Pretreated Untreated Pretreated Untreated Pretreated 1 C Acetic acid 5.45 2.68 6.47 - 7.63 - 2.52 2.55 2 C 2-Butenal 0.52 0.45 0.39 0.33 0.50 - - - 3 C Acetic acid, methyl ester 3.31 5.76 3.21 3.59 2.21 - 2.43 2.40 4 C 3-Penten-2-one 0.39 0.04 0.22 0.03 0.29 - 0.19 0.43 5 C 2,3-Pentanedione 0.23 0.25 0.23 0.13 - - - 0.17 6 C 3-Penten-2-one, (E)- 0.23 0.19 0.20 0.26 - - - 0.18 7 C 1-Penten-3-one 0.54 0.44 0.85 0.83 0.63 0.84 0.40 0.57 Cyclopropane, 1,1-dimethyl-2- 8 C 0.21 0.11 ------methylene- 9 C 2-Propanone, 1-hydroxy- 2.13 2.19 2.37 2.26 2.18 2.55 1.41 1.57 C 154 10 2-Butenal, 2-methyl- 0.25 0.32 0.37 0.27 0.31 0.24 - 0.23 11 C Succindialdehyde 1.10 1.58 1.69 1.45 1.00 1.36 0.50 0.71
12 C Piruvic acid methyl ester 1.20 1.10 0.79 1.35 1.61 1.01 1.52 1.64 5-(Hydroxymethyl)-2(5H)- 13 C 0.38 0.34 0.55 0.47 0.51 0.34 - - furanone 14 C Furfural 1.78 2.05 1.85 2.08 1.70 2.31 1.11 1.19 15 C 2-Hexanone, 6-hydroxy- 0.69 0.89 0.97 0.96 0.70 1.02 0.58 0.58 16 C 2-Propanone, 1-(acetyloxy)- 0.75 0.86 0.43 0.58 0.61 0.61 0.42 0.42 17 C 2-Cyclopenten-1-one, 2-methyl- 0.12 0.26 0.23 0.08 0.16 0.25 - - 18 C 2(5H)-Furanone 0.97 1.20 1.33 1.34 0.38 0.38 0.30 0.27 19 C 1,2-Cyclopentanedione 1.02 1.76 0.92 1.64 0.98 1.02 1.01 0.78 20 C 2,5-Furandione, 3-methyl- 0.16 0.11 0.11 0.20 0.12 0.21 0.12 0.03 21 C 2,4-Dimethylfuran 0.35 0.36 0.34 - 0.35 0.42 0.34 - 22 H Phenol 1.28 1.15 0.58 1.10 0.36 0.16 0.62 0.37 Continued 154
Table 5.4 Continued
Corn stover Miscanthus Hardwood Softwood Peak # Compound Untreated Pretreated Untreated Pretreated Untreated Pretreated Untreated Pretreated 23 P Oxazolidine, 2,2-diethyl-3-methyl- 1.14 0.66 1.66 1.83 1.54 1.96 1.08 1.01 24 R Carene ------0.12 - 25 R Terpinene ------0.10 - 26 R o-Cymene ------0.06 - 27 R p-Cymene ------0.17 - 28 R D-Limonene ------0.28 0.46 29 C 1,2-Cyclopentanedione, 3-methyl- 1.06 1.29 1.15 0.89 0.79 1.19 0.36 0.20 2-Cyclopenten-1-one, 2,3- 30 C 0.28 0.30 0.35 0.38 0.24 0.30 - - dimethyl- 31 H o-Cresol 0.73 0.84 0.79 0.80 0.47 0.61 0.48 0.40 H 155 32 p-Cresol 0.82 0.83 0.44 0.84 0.65 0.79 0.67 0.71 33 G Guaiacol 0.96 1.49 1.29 1.34 0.96 1.27 1.44 1.81
34 H Phenol, 2,6-dimethyl- 0.17 0.22 0.23 0.16 - - 0.08 - 2-Cyclopenten-1-one, 3-ethyl-2- 35 C 0.40 0.53 0.45 0.71 0.25 0.43 - - hydroxy- 2,4(3H,5H)-Furandione, 3- 36 C 0.34 0.29 0.28 - 0.29 0.32 0.30 0.30 methyl- 37 H Phenol, 2,4-dimethyl- 0.27 0.31 0.30 - 0.61 0.55 0.58 0.43 38 H Phenol, 4-ethyl- 0.41 0.37 0.23 0.32 - - 0.26 0.15 39 H Phenol, 3,4-dimethyl- - - 0.21 0.17 - - 0.23 0.15 40 G Creosol 0.59 0.68 0.83 0.89 0.98 1.15 2.33 2.41 41 H Catechol 0.63 0.96 1.09 0.93 1.05 0.79 0.94 0.89 1,4:3,6-Dianhydro-.alpha.-d- 42 C 0.34 0.34 - - 0.29 0.35 - - glucopyranose 43 H Coumaran 5.01 6.03 5.79 3.44 0.27 0.24 0.34 - 155 Continued
Table 5.4 Continued
Corn stover Miscanthus Hardwood Softwood Peak # Compound Untreated Pretreated Untreated Pretreated Untreated Pretreated Untreated Pretreated 44 C 5-Hydroxymethylfurfural 0.48 0.20 0.26 0.62 0.29 0.58 0.25 - 45 S 1,2-Benzenediol, 3-methoxy- 0.49 0.92 0.90 0.51 1.10 1.34 0.51 0.54 46 G 4-Ethylguaiacol 0.50 0.66 0.69 0.56 0.76 0.95 0.74 0.73 47 H 4-Methylcatechol - - 0.45 0.81 - - 0.80 0.76 48 G 4-Vinylguaiacol 3.18 3.25 4.11 3.22 1.56 1.56 2.55 2.13 49 S Syringol 1.03 1.45 1.48 0.97 2.13 2.46 - - 50 G Eugenol 0.25 0.44 0.34 0.87 1.00 0.95 2.71 2.81 51 S Phenol, 3,4-dimethoxy- - - 0.28 0.15 0.44 0.47 - - 52 G 4-Propylguaiacol - - 0.31 - - - 0.35 - 53 G Vanillin 0.23 0.63 0.20 0.26 0.43 0.47 0.41 0.57 156 54 G Methyleugenol - - 0.17 - - - - -
55 S 1,2,3-Trimethoxybenzene 0.23 0.37 - - 1.64 1.49 - - 56 G Isoeugenol 0.37 0.21 0.55 0.39 0.55 0.49 0.82 0.88 57 C Levoglucosan 3.52 2.53 4.42 3.27 4.17 3.68 1.66 1.50 58 G Guaiacylacetone 0.12 0.20 - - - - 0.36 - 4-Methyl-2,6- 59 S 0.56 1.04 0.67 - 2.71 2.98 - - dimethoxybenzaldehyde 60 G Methoxyeugenol 0.62 1.05 0.68 0.58 1.86 2.36 - - 61 G Homovanillic acid ------0.51 0.26 62 S Synringaldehyde 0.11 0.19 0.30 0.12 0.52 0.53 - - 63 S Acetosyringone 0.23 0.38 - - 0.69 0.68 - - 64 G Coniferol 0.36 0.61 1.64 0.33 3.37 2.51 4.48 3.51 *Derived from carbohydrates (C), hydroxyphenyl units (H), syringil units (S), guaiacyl units (G), protein (P), or resin (R)
156
Lignin to carbohydrate (L/C) ratio decreased by 23%, 6%, and 14% for Miscanthus, hardwood, and softwood, respectively, after first generation non-sterile fungal pretreatment. However, L/C ratio slightly increased for corn stover (Table 5.5). The trends of these results were consistent with the data from the wet chemistry analysis, since the L/C ratios were higher for the woods than for corn stover and Miscanthus, there was a reduction in L/C ratios after the successful fungal pretreatments, and an increased for the non-sterile fungal pretreated corn stover, since there was a high holocellulose degradation in this treatment (Table 5.2). However, the values calculated with the Py-
GC-MS data overestimated the L/C ration by about 2-fold. The reduction of the L/C ratio confirmed the selective degradation of C. subvermispora, in accordance with previous reports (Del Río et al., 2001a), even in the presence of contaminant microorganisms that degrade holocellulose.
Table 5.5 Lignin-to-carbohydrate and syringil-to-guaiacil ratios based on Py-GC-MS of
untreated and fungal pretreated feedstocks
L/C S/G Untreated 0.84 0.37 Corn stover Pretreated 0.86 0.47 Untreated 1.02 0.34 Miscanthus Pretreated 0.79 0.20 Untreated 1.35 0.80 Hardwood Pretreated 1.27 0.85 Untreated 1.44 ND Softwood Pretreated 1.24 ND ND: not determined
With respect to the preferential degradation of certain types of lignin subunits, literature has conflicted reports on the effect of C. subvermispora: a reduction of S/G 157 ratio was observed in eucalypt (hardwood) pretreated with this and other white rot fungi
(Del Río et al., 2001a, 2001b) and a slight, but non-consistent decreased was observed in fungal pretreated aspen wood (Choi et al., 2006). However, an increased in S/G ratio was detected in fungal pretreatment with C. subvermispora of Bermuda grass (Akin et al.,
1995) and poplar (Skyba et al., 2013). This last study also proved that wood with higher content of S was more resistant to fungal decay with a wide range of wood decay fungi, including C. subvermispora (Skyba et al., 2013). In the current study, S/G ratio decreased for Miscanthus, and increased for corn stover and hardwood, after first generation non- sterile fungal pretreatment, compared with the untreated feedstock (Table 5.5). The S/G ratio was not calculated for softwood, since practically no presence of S units was detected, as expected for this type of wood (Wagner et al., 2015). This showed that C. subvermispora had differential degradation patterns depending on the feedstock and possibly on other factors that could cause the inconsistencies in the results reported in literature, such as distribution of the subunits in the cell wall, nutrients concentration that can trigger or inhibit different enzymes, and mediators available for lignin depolymerization. Additionally, a significant number of the identified compounds were considered to be derived from H lignin (Table 5.4). However, H content in most types of lignin is usually considerably lower than G and S contents, which was not the case in the present study. It has been shown before that pyrolysis of protein yielded components such as phenol and cresol (Choi et al., 2001), and it is expected that the concentration of protein changed significantly with the fungal pretreatment, due to fungal growth.
Therefore, the H content obtained in this study should be analyzed carefully.
158
Additionally, demethoxylation of lignin units is possible by the effect of pyrolysis
(Harman-Ware et al., 2013), and it can be enhanced by catalytic effect of even small amounts of inorganic salts (Kleen and Gellerstedt, 1995). Demethoxylation is also possible by the effect of ligninolytic enzymes (Wan and Li, 2012), which could explain the fact that H units increased, while S and G units decreased after fungal pretreatment of
Miscanthus (Table 5.4).
Principal components analysis (PCA) was performed on the Py-GC-MS compounds’ relative abundance data. The first five components were significant and described over
90% of variance of the original data. The first three components, shown in Figure 5.11, explained 74% of variance of the original data. Clustering of the samples according to the type of feedstocks was evident, and it was independent of the pretreatment (Figure 5.11).
This showed that the modifications caused by the fungal pretreatment were not as radical as the differences in structure and composition between feedstocks.
159
Figure 5.11 PCA score plot of untreated and fungal pretreated feedstocks
characterized by Py-GC-MS.
5.3.6 FT-IR analysis
FT-IR analysis was performed for untreated and fungal pretreated samples. The changes in selected peaks’ intensities were calculated for the fungal pretreated samples with respect to the corresponding untreated feedstock. The relative changes for the positive control (sterile) and the first generation non-sterile pretreatment are shown in
Table 5.6, and, as an example, the spectra for untreated and fungal pretreated hardwood is shown in Figure 5.12. Most peaks showed reduction in intensity after fungal
160
pretreatment, and reduction was generally more severe after the first generation non-
sterile pretreatment than the positive control.
Sample 093 By Administrator Date Thursday, March 03 2016 03 March Thursday, Date Administrator By 093 Sample 22 Hardwood_1_1
Untreated hardwood
Sample 114 By Administrator Date Thursday, March 03 2016 03 March Thursday, Date Administrator By 114 Sample 20 17-pc-Hardwood_1_1
Fungal pretreated hardwood (positive control)
Sample 105 By Administrator Date Thursday, March 03 2016 03 March Thursday, Date Administrator By 105 Sample
47-1st fungal-Hardwood_1_1 18 47-1st
First generation non-sterile pretreated hardwood
Description
Name 16
cm-1
14 450 1000 1500 2000 2500 3000 3500 4000 12 -1
A 10 -0
8 1160 3340
6 1600
2 1370 830 4 1510 900
2 830
1510 4 900 -0
-1
4000 3500 3000 2500 2000 1500 1000 450
1370 cm-1
Name Description 1600 47-1st fungal-Hardwood_1_16 Sample 105 By Administrator Date Thursday, March 03 2016 17-pc-Hardwood_1_1 Sample 114 By Administrator Date Thursday, March 03 2016
Hardwood_1_1Figure 5.Sample12 FT093 By -AdministratorIR spectra Date Thursday, of March untreated 03 2016 and fungal pretreated hardwood. 3340
1160
8
Peaks related10 with the stretching of the aromatic ring showed a marked decreased in A
woods after fungal pretreatment, especially the one at 1510 cm-1. Fungal pretreated 12 Miscanthus showed similar trends, except with the peak at 1600 cm-1 for the positive
control, which14 decreased. These decrements are probably related with lignin modifications16 caused by the pretreatment. Corn stover showed a marked increase in these
peaks, probably related with the negative results of its fungal pretreatment. Syringil lignin 18
peak highly decreased for Miscanthus, corn stover, and hardwood, for all conditions
expect non-sterile20 fungal pretreated corn stover. This results agrees with the fact that the latter was an22 unsuccessful fungal pretreatment, according to the enzymatic digestibility
161
(Figure 5.1). However, the correlation was not seen with the data from the Py-GC-MS
(Table 5.4), showing that the relative abundances of the compounds determined by this method were dependent of other factors related with the pyrolysis. As expected, this peak was not detected in softwood, since softwood lignin does not contain a significant amount of S units, as explained in section 5.3.5.
Peaks relates with cellulose degradation showed a significant decrease in intensity for all non-sterile fungal pretreatments, product of the cellulose degradation by microbial contaminants. For the non-sterile pretreatment of corn stover, there was a high decreased in the peak for amorphous cellulose, but not for the crystalline cellulose, since the former is easier to digest by microorganisms. However, this phenomena was not observed for the other feedstocks, where there was a decrease in the intensity of the crystalline cellulose peak. Peaks in this spectra could be the product of more than one type of component, since the individual spectra of cellulose, hemicellulose, and lignin have peaks at similar positions (Yang et al., 2007), which can generate confounded effects.
162
Table 5.6 Relative changes (%) in FT-IR spectra after fungal pretreatment
Miscanthus Corn stover Softwood Harwood Band Component Non- Non- Non- Non- (cm-1) Sterile Sterile Sterile Sterile sterile sterile sterile sterile Syringyl lignin (C-H out plane 830 -40 ± 14 -69 ± 17 -36 ± 9 0 ± 7 ND ND -75 ± 4 -74 ± 4 bending) Amorphous cellulose (C-O-C 900 8 ± 6 -29 ± 6 -2 ± 8 -41 ± 8 -5 ± 6 -30 ± 7 -20 ± 1 -29 ± 8 vibration at β-(1-4)-glycosidic) Cellulose and hemicellulose (C-O- 1160 -9 ± 4 -36 ± 5 13 ± 1 -3 ± 9 -8 ± 6 -29 ± 7 -25 ± 2 -26 ± 7 C asymmetrical stretching) Crystalline cellulose (CH2 1370 11 ± 5 -33 ± 8 19 ± 3 6 ± 8 -17 ± 8 -31 ± 7 -22 ± 5 -20 ± 8 vibrations) Lignin (C=C stretching of aromatic 1510 -25 ± 4 -59 ± 12 45 ± 5 102 ± 23 -26 ± 6 -48 ± 1 -41 ± 9 -38 ± 3 ring) Lignin (C=C stretching of aromatic
163 1600 18 ± 3 -19 ± 12 58 ± 1 58 ± 15 -10 ± 8 -19 ± 6 -8 ± 2 -11 ± 2 ring)
3340 Cellulose (O-H stretching) 24 ± 3 -30 ± 8 4 ± 9 -27 ± 8 -14 ± 4 -31 ± 4 -13 ± 7 -17 ± 9 *Changes in intensity relative to untreated material (%). Negative number indicates reduction. Sterile refers to positive control. Non-sterile refers to first generation non-sterile fungal pretreatment.
163
5.4 Conclusions
Fungal pretreatment of non-sterile Miscanthus, softwood, and hardwood was successful for the first generation, using previously colonized material as inoculum, and achieving a 2- to 4.5-fold in enzymatic digestibility. This pretreatment was performed with 14 days of incubation time, yielding similar results than the fungal pretreatment of sterile feedstocks after 28 days. However, fungal pretreatment was unsuccessful for subsequent generations using non-sterile feedstocks. Additionally, fungal pretreatment was ineffective for non-sterile corn stover, probably due to the presence of a faster- grower indigenous microbial community that outcolonized the white rot fungus. In general, fungal pretreatment was more effective in woods than herbaceous feedstocks, and it generated evident modifications in composition and structure of the feedstocks, but no specific pattern for these changes was identified, as these changes were feedstock- dependent. It was also observed that C. subvermispora was able to degrade terpenes from the resin of pine wood (softwood) during the fungal pretreatment.
164
Chapter 6: Solid-state anaerobic digestion of fungal pretreated Miscanthus harvested in two different seasons
Abstract
Solid-state anaerobic digestion of Miscanthus sinensis harvested in fall and spring was compared under different total solids contents and feedstock-to-inoculum ratios. The highest specific methane yields reached 170-175 L CH4/kg volatile solids for both harvest seasons. Miscanthus harvested in fall generated a 6% higher methane yield in average than Miscanthus harvested in spring. Fungal pretreatment performed with Ceriporiopsis subvermispora decreased the lignin content of Miscanthus harvested in spring by 25.7%, but there was no significant delignification of Miscanthus harvested in fall. Fungal pretreatment of Miscanthus harvested in spring increased the specific methane yield 25%, but there was a slight methane yield reduction for Miscanthus harvested in fall. Methane yields for Miscanthus were similar to those from other energy crops.
6.1 Introduction
Miscanthus is typically combusted to generate electricity and heat (Jones and Walsh,
2001), but it can also be used as a feedstock for biological conversion such as anaerobic digestion (AD) for biogas production. For the transformation of lignocellulosic biomass such as Miscanthus, solid-state anaerobic digestion (SS-AD) is recommended. SS-AD operates with a TS content of 15% or more, and therefore requires a smaller reactor 165 volume for the same solids loading, less energy for heating and mixing, and generates an easier-to-handle digestate with a lower moisture content (Li et al., 2011). However, SS-
AD usually has lower methane yields and needs a nitrogen supplement because the lignocellulosic feedstocks usually have a carbon-to-nitrogen (C/N) ratio higher than that recommended for anaerobic digestion (20-30) (Li et al., 2011).
Previous studies have employed L-AD to digest Miscanthus in mesophilic conditions:
Klimiuk et al. (2010) obtained a methane yield of 100 L/kg volatile solids (VS) for M. × giganteus harvested in October and ensiled, while Theuretzbacher et al. (2014) attained a methane yield of 84 L/kg VS for M. × giganteus harvested in February, but increased it up to 345 L/kg VS by pretreating the Miscanthus with steam explosion. A smaller increase of 25-27% methane yield was obtained with aqueous ammonia soaking pretreatment (Jurado et al., 2013). AD has been performed with other perennial grasses similar to Miscanthus; giant reed produced methane yields of 129.7 and 150.8 L/kg VS
(Yang and Li, 2014), while switchgrass generated methane yields of 116.9 and 111.0
L/kg VS (Brown et al., 2012), with SS-AD and L-AD, respectively.
Several researchers have studied the effect of harvest date on methane yield for energy crops other than Miscanthus, sometimes showing inconsistent results (Lehtomäki et al., 2008; Seppälä et al., 2009). Moreover, most of the studies focused on harvest dates during the summer and fall, and excluded harvest in the following spring, because unlike
Miscanthus, most crops do not remain standing after winter. Even though there are several studies of the effect of Miscanthus harvest date on composition and transformation (Hayes, 2013; Le Ngoc Huyen et al., 2010; Lewandowski and Heinz,
166
2003), to date there are no studies to our knowledge of the effect of harvest date on the anaerobic digestion of Miscanthus or on how fungal pretreatment can affect this process.
It is well known that the harvest date of grasses can affect biochemical processes such as AD or solid-state fermentation with fungi (Lehtomäki et al., 2008), but it is unclear how the harvest date of M. sinensis will affect these processes, especially because of its unconventional harvest window between late fall and early spring. It is hypothesized that fungal pretreatment can be performed successfully for M. sinensis harvested in both fall and spring, increasing the digestibility of the feedstock and the methane yield obtained by
SS-AD. This work studied the SS-AD of M. sinensis harvested in two different seasons: fall and spring. Also, it evaluated the fungal pretreatment of M. sinensis harvested in two seasons and its effect on biogas production by SS-AD.
6.2 Materials and Methods
6.2.1 Feedstock collection and preparation
Miscanthus sinensis was harvested from a field located in The Ohio State University campus in Columbus, OH, USA. Harvesting was performed manually in November 2013
(fall harvest) and April 2014 (spring harvest). The initial TS of the Miscanthus harvested in fall and spring were 47.4% and 90.1%, respectively. Because of the low TS content, the former was first air dried, reaching a TS content of 80.2%; the latter was used directly because the TS content was high enough. The whole dry plants (leaves and stems) were milled and passed through a 12 mm screen using a hammer mill (The C.S Bell Co.,
Tiffin, OH, USA) and stored at 4°C.
167
6.2.2 Solid-state anaerobic digestion
The inoculum for anaerobic digestion was the effluent from an operating mesophilic anaerobic digester fed with sewage sludge (BK BioEnergy, Inc., Akron, OH, USA). Part of this inoculum was dewatered by centrifugation (Sorvall RC 6+ Centrifuge, Thermo
Fisher Scientific, Inc, Waltham, MA, USA) to achieve the high TS values desired for the treatments. The inocula were kept at 4°C and then incubated at 37°C for 7 days to activate the microbial population before use.
SS-AD tests were performed following a full factorial design with two feedstock-to- inoculum (F/I) ratios (2 and 4, based on VS), three TS contents (20%, 25% and 30%), and two harvest times for the feedstock (fall and spring), with two replicates of each combination. The inoculum and feedstock were mixed using a hand-mixer (Black &
Decker, New Britain, CT, USA) and incorporated into 1 L flasks, leaving a uniform head space. Flasks were closed with rubber stoppers, connected to Tedlar® bags (CEL-
Scientific, Cerritos, CA, USA) and incubated at 37°C for 60 days. A control reactor was set-up with only centrifuged effluent to estimate the amount of biogas produced by the effluent. The biogas content and volume were measured every three days.
6.2.3 Fungal pretreatment
C. subvermispora (ATCC 96608) was obtained and preserved as stated in section
3.2.2. For the inoculum preparation, one disc of agar (5 mm diameter) colonized with mycelium was transferred to a new malt extract agar plate and grown for 5 days at 28°C.
Then, one disc of this agar was transferred to 50 ml of liquid malt extract sterile medium
168 in a 500 ml Erlenmeyer flask and incubated under no agitation conditions at 28°C for 7 days. At the end of the incubation time, fungal mycelium was harvested by filtering the liquid medium and re-suspended in deionized water. The mixture of mycelium and water was gently homogenized with a hand blender (Hamilton Beach, Inc., Richmond, VA,
USA) and used as inoculum for the fungal pretreatment.
About 65 g (dry matter) of Miscanthus samples was added to 1 L bottles (reactors) with water to adjust the moisture to 60% (w/w). Reactors were closed with cotton and sterilized by autoclaving (121°C, 15 min). After cool down, reactors were inoculated with the prepared mycelia suspension and incubated at 28°C under 50% humidity for 28 days.
At the end, samples were taken out of the reactor and well mixed to allow homogenous sampling. Part of the finished material was dried at 40°C in a convection oven for 24 h, milled and passed through a 1 mm screen (Model 4 Wiley Mill, Thomas Scientific,
Swedesboro, NJ, USA), and then analyzed. The rest was used fresh for enzymatic hydrolysis and an SS-AD experiment using the best conditions of F/I ratio and TS content encountered in the experiment of section 6.2.2, for both the Miscanthus harvested in fall and spring. Controls were set-up with non-pretreated material.
6.2.4 Enzymatic hydrolysis
Enzymatic hydrolysis of untreated and pretreated Miscanthus was performed as described in section 3.2.4. Enzymatic digestibility was calculated using the equation detailed in that same section.
169
6.2.5 Analytical methods
TS, VS, pH, and alkalinity were determined according to the Standard Methods for
Examination of Water and Wastewater (APHA et al., 2005). Total carbon and nitrogen were measured using an Elemental Analyzer (Vario MAX CNS, Elementar
Analysensysteme, Hanau, Germany). Ammonium nitrogen was determined by distillation and titration (ISO, 1984) in a AutoKjeldahl Unit K-370 (Buchi Labortechnik AG,
Switzerland). Volatile fatty acids were measured by gas chromatography (GC-2010 Plus,
Shimadzu, Kyoto, Japan) with a Stabilwax polar phase column. Crude protein in the feedstock was determined using the Kjeldahl method (Persson, 2008). Briefly, the ammonium nitrogen previously obtained was subtracted from the total Kjeldahl nitrogen, which was determined by Kjeldahl digestion (Tecator Digester, FOSS, Eden Prairie, MN,
USA), and the crude protein was calculated using a factor of 6.25. To measure the extractives, samples were extracted with water and ethanol using the Accelerated Solvent
Extraction (ASE 350, Thermo Fisher Scientific, Inc., Waltham, MA, USA) following the
NREL analytical procedure (Sluiter et al., 2008). The water and ethanol fractions of each sample were mixed, and the solvents were evaporated (Rocket Evaporator, Genevac,
Ipswich, England), dried at 40°C, and weighed. The extractives-free samples were analyzed for structural sugars and lignin following the NREL method (Sluiter et al.,
2012). Sugars were determined by HPLC (LC-20AB, Shimadzu, Kyoto, Japan) with an
Aminex HPX-87P column (Bio-Rad, Inc., Hercules, CA, USA) and a refractive index detector. Acid insoluble lignin was determined gravimetrically and acid-insoluble lignin
170 was quantified by UV-vis spectroscopy (BioMate 3, Thermo Fisher Scientific, Inc.,
Waltham, MA, USA).
Biogas volume was measured by a drum-type gas meter (Ritter TG 5, Bochum,
Germany). Biogas composition (CH4, CO2, O2 and N2) was determine by gas chromatography (HP 6890, Agilent Technologies, Wilmington, DE, USA) with an alumina/KCl deactivation column and a thermal conductivity detector, using helium as a carrier at a flow rate of 5.2 mL/min. Temperatures of the injector, column, and detector were 150°C, 40°C, and 200°C, respectively.
6.2.6 Statistical analysis
Data was analyzed using the software JPM® (SAS Institute Inc., Cary, NC, USA).
Statistical significance was determined by analysis of variance (ANOVA) with a significance level of 0.05. Results were presented as average values ± standard deviation, represented by error bars in the figures.
6.3 Results and Discussion
6.3.1 Characteristics of Miscanthus and anaerobic digestion inoculum
Miscanthus harvested in fall had significantly higher extractives content and significantly lower total carbon, cellulose, hemicellulose, and lignin contents than
Miscanthus harvested in spring (p<0.05) (Table 6.1). Presence or absence of leaves is one of the possible reasons for these differences in composition as Miscanthus harvested in the fall had the majority of its leaves, while most of the leaves were lost by spring.
171
Leaves usually have higher extractives content, but lower structural carbohydrate and lignin contents than stems (Le Ngoc Huyen et al., 2010). Extractives are mainly composed of free sugars, organic acids, and oligomers (Liew et al., 2012), but can contain a variety of other compounds that depend on the particular composition of the source.
Table 6.1 Characteristics of Miscanthus and inoculum for AD
Miscanthus Inoculum Parameters Fall harvest Spring harvest Originalb Centrifugedb Total solids (%) 80.2 ± 0.3 90.1 ± 0.0 7.8 ± 0.0 18.4 ± 1.2 Volatile solids (%) 75.8 ± 0.3 87.2 ± 0.2 4.6 ± 0.0 10.1 ± 0.8 Total carbon (%) 37.6 ± 0.3 42.5 ± 0.2 3.1 ± 0.1 7.5 ± 0.0 Total nitrogen (%) 0.8 ± 0.0 0.5 ± 0.0 0.5 ± 0.0 0.9 ± 0.0 C/N ratio 45.7 ± 0.9 87. 5 ± 1.0 6.7 ± 0.3 8.2 ± 0.0 pH ND ND 8.2 ± 0.0 8.6 ± 0.1 Alkalinity (gCaCO3/kg) ND ND 12.3 ± 0.6 14.3 ± 0.2 Total volatile fatty acids (g/kg) ND ND 0.2 ± 0.0 0.3 ± 0.0 Ammonium nitrogen (gN/kg) 2.7 ± 0.1 0.3 ± 0.0 3.6 ± 0.1 4.6 ± 0.0 Extractives a (%) 16.6 ± 0.2 8.7 ± 0.2 14.0 ± 0.2 14.0 ± 1.1 a Cellulose (%) 30.2 ± 0.3 35.1 ± 0.3 1.5 ± 0.2 1.5 ± 0.1 Hemicellulose a (%) 17.5 ± 0.1 19.2 ± 0.1 2.8 ± 0.6 2.4 ± 0.3 Lignin a (%) 15.6 ± 0.2 16.5 ± 0.2 ND ND Crude protein a (%) 6.8 ± 0.1 3.4 ± 0.0 ND ND a Based on TS of sample; the others are based on wet weight of sample; ND=not determined. b Characteristics of inoculum were determined after 7 days of activation at 37ºC.
Miscanthus harvested in fall had significantly higher nitrogen and crude protein contents (p<0.05) than that harvested in spring (Table 6.1), suggesting that the nutrient translocation, which starts with senescence, was uncompleted by the time of the fall
172 harvest. The translocation, along with leaf fall, facilitates nutrient recycling. Incomplete nutrient translocation can have detrimental repercussions including increased necessity of fertilizers and the deterioration of soil quality and long-term sustainability of the crop
(Amougou et al., 2011).
6.3.2 Effect of harvest season, TS, and F/I ratio on the SS-AD of Miscanthus
Specific methane yields obtained were between 108-175 L CH4/kg VS (Figure 6.1,
Figure 6.2), which are consistent with the reported Miscanthus methane potential of 200
L CH4/kg VS (Murphy and Power, 2009). The effluent control produced about 23 L
CH4/kg VS. An analysis of variance (ANOVA) of these results indicated that all three factors (TS, F/I ratio and harvest season) had a significant effect on methane yield
(p<0.05). Miscanthus harvested in fall generated a 6% higher methane yield on average than Miscanthus harvested in spring. The difference in methane yield between harvest dates was significant (p<0.05) for an F/I ratio of 4 (Figure 6.2), but not for an F/I ratio of
2 (p>0.05) (Figure 6.1), when making the comparison of each treatment by a t-test. Thus,
Miscanthus that will be used for SS-AD at an F/I ratio of 2, can be harvested during a 5-6 month harvest window without a large influence on the specific methane yield. This flexibility in harvest date might reduce the need for storage.
173
200
VS) Fall - 180 160 Spring 140
120 yiled L/kg yiled 4 100 80 60 40
20 Cumulative CH Cumulative 0 20 25 30 TS (%)
Figure 6.1 Cumulative methane yield obtained during 60 days of SS-AD of fall and
spring harvested Miscanthus at F/I ratio of 2.
200 VS)
- 180 Fall 160 Spring 140
120 yiled L/kg yiled 4 100 80 60 40
20 Cumulative CH Cumulative 0 20 25 30 TS (%)
Figure 6.2 Cumulative methane yield obtained during 60 days of SS-AD of fall and
spring harvested Miscanthus at F/I ratio of 4.
Increasing the F/I ratio from 2 to 4 or the TS content from 20% to 30% reduced the specific methane yield by 13% or 29% on average, respectively. This reduction is similar
174 to that of previous studies (Liew et al., 2012; Yang and Li, 2014). The TS increase probably generated a barrier for mass diffusion of hydrolyzed sugars to microorganisms that produce biogas, thus resulting in an accumulation of these sugars and, as a consequence, inhibiting the hydrolysis step of the AD process (Xu et al., 2014).
Increasing the F/I ratio implies an increase in the organic loading, which can generate an initial accumulation of volatile fatty acids that can inhibit the methanogenic process (Park and Li, 2012). Also, the liquid AD effluent acts as a buffer to organic acids produced at the beginning (Shi et al., 2014), reducing their inhibitory effect. At the lower F/I ratio, more AD effluent was added to the reactor, therefore, it had more buffer capacity.
The effect model parameters suggest that the TS content is the factor with higher effect on the methane yield, for the range studied, follow by the F/I ratio. Also, there is a significant interaction (p<0.05) between the F/I ratio with both the harvest season and the
TS content. That is reflected in the differential effect of those two factors at the two different F/I ratios, suggested by the differences between Figure 6.3 and Figure 6.4.
The daily methane yield had two peaks during the 60 days of digestion for most of the treatments (Figure 6.3, Figure 6.4). One possible reason for this is the difference in degradation kinetics for different components: the first peak was possibly due to the conversion of easy-to-digest components, such as extractives or hemicellulose, while the second peak might correspond to the degradation of more recalcitrant components, such as crystalline cellulose and protein. The first peak occurred around day 4 for an F/I ratio of 2 (Figure 6.3) and it was delayed until day 7-10 for an F/I ratio of 4 (Figure 6.4). For the former case, more inoculum was added, so more active microorganisms were present
175 at the beginning, allowing a faster start of the digestion. The highest daily methane yield of 8.9 L CH4/kg VS was obtained for an F/I ratio of 2 and a TS of 20% and it was reduced to 6.9 L CH4/kg VS for an F/I ratio of 4 at the same TS content. These values were similar to those obtained for giant reed (Yang and Li, 2014) and switchgrass
(Brown et al., 2012).
10
9 TS=20% Fall 8 TS=25% Fall
TS=30% Fall VS.d) - 7 TS=20% Spring TS=25% Spring 6 TS=30% Spring 5 4 3
2 Daily methane yield (L/kg yield methane Daily 1 0 0 10 20 30 40 50 60 Time (days)
Figure 6.3 Daily methane yield during 60 days of SS-AD of fall and spring harvested
Miscanthus at F/I ratio of 2.
176
10
9 TS=20% Fall 8 TS=25% Fall
TS=30% Fall VS.d) - 7 TS=20% Spring TS=25% Spring 6 TS=30% Spring 5
4 3
2 Daily methane yield (L/kg yield methane Daily 1
0 0 10 20 30 40 50 60 Time (days)
Figure 6.4 Daily methane yield during 60 days of SS-AD of fall and spring harvested
Miscanthus at F/I ratio of 4.
6.3.3 Effect of harvest season on fungal pretreatment of Miscanthus
After 28 days of fungal pretreatment, lignin degradation for Miscanthus harvested in fall was insignificant (p>0.05) and the enzymatic digestibility increased by only 10%, while for Miscanthus harvested in spring, lignin degradation was over 25% and there was a 140% increase in enzymatic digestibility (Figure 6.5 and Table 6.2). Along with that significant lignin degradation, a significant (p<0.05) reduction of hemicellulose occurred, which is one of the preferred carbon sources for C. subvermispora. The component degradation percentages obtained for spring Miscanthus were similar to those obtained for switchgrass after 18 days of fungal pretreatment under similar conditions (Wan and
Li, 2011b). Also, there was practically no cellulose degradation in Miscanthus harvested
177 in both seasons, which agreed with previous results (Wan and Li, 2011b) and was probably due to the weak cellulosome of C. subvermispora (Fernández-Fueyo et al.,
2012a). Fungal pretreatment can be considered effective for Miscanthus harvested in spring, but not for Miscanthus harvested in fall.
35% RawUntreated 30% Pretreated 25%
20%
15%
Enzymatic digestiblity Enzymatic 10%
5%
0% Fall Spring Miscanthus harvest season
Figure 6.5 Enzymatic digestibility of untreated and fungal pretreated Miscanthus.
Table 6.2 Degradation of Miscanthus components after 28 days of fungal pretreatment
Harvest time
Fall Spring Cellulose degradation (%) 1.62 ± 2.71 2.36 ± 1.01 Hemicellulose degradation (%) 7.77 ± 2.54 13.71 ± 0.90 Lignin degradation (%) 0.25 ± 0.80 25.69 ± 0.77 Dry matter loss (%) 3.65 ± 2.65 7.64 ± 0.96
178
It is not clear if the failure of the fungal pretreatment for the Miscanthus harvested in fall was due to inhibition of fungal growth and/or ligninolytic activity. The significant degradation of hemicellulose suggests that there was some fungal growth, although it was barely visible, because hemicellulose constitutes one of the preferred substrates for C. subvermispora. There was no significant difference (p>0.05) between the enzymatic digestibility of untreated Miscanthus harvested in fall and spring (Figure 6.5), suggesting that the unsuccessful pretreatment was probably not due to differences in the recalcitrance of the materials. Thus, the reason might be the difference in composition such as nitrogen content (or C/N ratio) (Table 6.1), which has been shown to influence the enzymatic activity of ligninolytic enzymes (Tekere et al., 2001), or the presence of antifungal compounds in the leaves, as previous studies have shown that some leaf compounds can have antifungal activity against wood-decay fungi (Bento et al., 2014;
Wang et al., 2005); Bento, et al, 2013).
6.3.4 Effect of fungal pretreatment on SS-AD of Miscanthus
There was a 25% increase in methane yield for the fungal pretreated Miscanthus harvested in spring, compared to the untreated material, by the end of the 60 days of SS-
AD (Figure 6.6). This result is in accordance with the significant lignin degradation
(Table 6.2) and the increased enzymatic digestibility (Figure 6.5), which reduced recalcitrance and facilitated the hydrolytic step of AD. On the contrary, fungal pretreatment did not significantly (p>0.05) affect the methane yield for the first 30 days of AD for Miscanthus harvested in fall (Figure 6.6). Because there was no lignin
179 degradation and only a slight increase in the enzymatic degradability, differences in the methane yield before and after pretreatment were not expected. By the end of the 60 days, the methane yield of the pretreated Miscanthus harvested in fall was slightly lower than the non-pretreated materials (Figure 6.6). This can be due to the natural variability of the process, but some reports showed that unsuccessful pretreatment decreased the methane yield compared to the untreated material (Zhao et al., 2014a). If there was some incipient fungal growth, it probably consumed some free sugars and, as seen in Table 6.2, some hemicellulose, which are easily digestible components that were then not available for AD, making the material slightly more recalcitrant to biodegradability. This cannot be reflected in the enzymatic digestibility because this measurement only took into account the cellulose digestibility, and does not include the whole complexity of the hydrolytic step during AD.
250 VS) - 200
150
yield (L/kg yield 4
100
Fall harvest pretreated 50 Spring harvest pretreated
Cumulative CH Cumulative Fall harvest untreatedraw Spring harvest untreatedraw 0 0 10 20 30 40 50 60 Time (days)
Figure 6.6 Cumulative methane yield during 60 days of SS-AD of untreated and
pretreated Miscanthus. 180
Degradation of the main components for fungal pretreated Miscanthus harvested in spring was statistically different (p<0.05) from all the other conditions, for every component evaluated (Figure 6.7). This result was due to the effectiveness of this pretreatment, which reduced the recalcitrance and increased the digestibility of the feedstock. In particular, cellulose and hemicellulose, which were the main components that contributed to methane production, showed a larger degradation for fungal pretreated
Miscanthus harvested in spring, reaching degradations of more than 55%, for both cases.
As expected, TS and VS degradation were also larger for this treatment. But extractives degradation was somehow smaller for this case. As stated, extractives are a mixture of components, and their composition can change significantly with pretreatment and AD.
Probably, most of the easily digestible extractives, such as free sugars, were used for fungal growth, and the remaining extractives were not as digestible for the SS-AD process.
181
70% Fall harvest pretreated Spring harvest pretreated 60% Fall harvest untreatedraw Spring harvest untreatedraw 50%
40%
30% Degradation
20%
10%
0% Cellulose (%) Hemicellulose (%) Extractives (%) TS (%) VS (%)
Figure 6.7 Degradation of components of untreated and fungal pretreated Miscanthus
after 60 days of SS-AD.
6.3.5 Methane potential per hectare
3 3 The methane yield was around 5,100 m CH4/ha and 2,500 m CH4/ha for untreated
Miscanthus harvested in fall and spring, respectively. Miscanthus harvested in fall produced a yield of 29.6 Mg dry matter/ha, but that yield was reduced by about 50% when Miscanthus was harvested in spring. For fungal pretreated Miscanthus, the methane
3 3 yield was around 4,000 m CH4/ha and 2,900 m CH4/ha for Miscanthus harvested in fall and spring, respectively. These values were in the range reported previously for
3 Miscanthus of 1,432-5,450 m CH4/ha (Braun et al., 2008). They were also close to the
3 methane yields of 2,900-5,400 m CH4/ha obtained for other energy crops: Jerusalem artichoke, timothy-clover grass, and reed canary grass (Lehtomäki et al., 2008). Thus,
182 even though Miscanthus is considered to have a low methane potential compared to other energy crops (Murphy and Power, 2009), its methane yield per hectare was similar to most of them (Braun et al., 2008), because it was compensated by the high biomass yields.
These results might suggest that harvesting Miscanthus in fall would be the best option to maximize the biogas obtained from its AD. But they do not take into consideration that early harvest can be detrimental for the soil, increase the necessity of fertilizer, and increase logistics costs because this material has a high moisture content
(about 50%) and would need to be dried prior to storage. A more complete study that includes the impact of harvest date on Miscanthus nutrient requirements and logistics is needed to determine the ideal harvest time for this grass.
6.4 Conclusion
SS-AD of M. sinensis produced specific methane yields up to 175 L CH4/kg VS (with an F/I ratio of 2 and TS content of 20%), and was similar for Miscanthus harvested in fall and spring. Fungal pretreatment was successful for Miscanthus harvested in spring, but not for that harvested in fall. The successful fungal pretreatment increased the specific methane yield from Miscanthus harvested in spring by 25%. To determine the ideal harvest time for Miscanthus that is being used for AD, knowledge of the agronomy (soil quality, nutrient requirements) and logistics (drying, transportation, and storage) should be also taken into consideration.
183
Chapter 7: Effect of fungal pretreatment of Miscanthus on microbial inhibitors’ generation and butanol fermentation
Abstract
Harsh conditions of temperature and added chemicals from traditional pretreatments promote the generation of molecules derived from the lignocellulosic biomass that can inhibit subsequent fermentation. Alternatively, fungal pretreatment is performed at mild temperature with no added chemicals, which suggest that the production of these compounds should be lower, producing less inhibitory effect on subsequent butanol fermentation. To prove this hypothesis, Miscanthus was pretreated with the white rot fungus Ceriporiopsis subvermispora, and two traditional pretreatment methods (liquid hot water and alkaline pretreatment), for comparison of enzymatic saccharification yields, production of selected lignocellulose derived microbial inhibitory compounds (LDMICs), and butanol fermentation with Clostridium beijerinckii. Fungal pretreatment did not generate a significant concentration of LDMICs and produced 17.7-19.4 g/L total ABE, while a control medium with equivalent amount of sugars produced 15.9 g/L. Liquid hot water pretreated Miscanthus inhibited C. beijerinckii growth, and alkaline pretreated material required washing prior enzymatic hydrolysis to avoid fermentation inhibition.
Nevertheless, the total sugar yield with fungal pretreatment was 17-24% lower than that with the traditional pretreatments.
184
7.1 Introduction
Butanol is an alternative biofuel and platform chemical with better characteristics than ethanol, including higher energy content, low vapor pressure, and less hygroscopicity and corrosiveness, which make it safer to transport in existing infrastructure and allows it to be mixed with gasoline in any proportion (Qureshi et al.,
2013a; Wang et al., 2014). Butanol can be produced by some anaerobic bacteria, such as solventogenic clostridia, via fermentation of sugars (Ezeji and Blaschek, 2010), which can be generated from lignocellulosic biomass with potentially lower cost. However, lignocellulosic biomass is highly recalcitrant to biodegradation and requires pretreatment to facilitate the subsequent enzymatic saccharification. Traditional pretreatments use high temperature and pressure, and harsh chemicals that reduce the recalcitrance of the biomass, but these conditions also promote the formation of molecules, such as furans, weak acids, and aromatic compounds, that are degradation products of the holocellulose and lignin (Baral and Shah, 2014). At certain concentrations, these molecules have the capacity to inhibit the growth of the fermentative microorganisms and/or the production of butanol, so they are known as lignocellulose derived microbial inhibitory compounds
(LDMICs) (Baral and Shah, 2014; Ujor et al., 2015). Since fungal pretreatment is performed at mild conditions of temperature, atmospheric pressure, and without added chemicals, this pretreatment has been considered to produce low amounts of LDMICs
(Keller et al., 2003; Wan and Li, 2010a). However, the main mechanism of fungal pretreatment is the degradation of lignin by oxidative enzymes (Vasco-Correa et al.,
2016b). If a high concentration of soluble lignin derived compounds is present at the end
185 of the fungal pretreatment, it could have a detrimental effect on the butanol fermentation, since solventogenic clostridia are especially sensitive to these compounds (Cho et al.,
2009; Li et al., 2017).
The most commonly studied pretreatment for lignocellulosic biomass is the dilute acid pretreatment, which uses low concentrations of acids at high temperatures and pressures, and its main effect is the removal of hemicellulose (Mosier et al., 2005). This pretreatment has been found to generate significant amounts of LDMICs (Larsson et al.,
1999) and it is known to inhibit the production of butanol from certain substrates
(Qureshi et al., 2008a). Alternatively, liquid hot water (LHW) pretreatment avoids the use of added acids, and pretreats the material at similar or slightly higher temperature and pressure, and it has been found to be effective for recalcitrance reduction of lignocellulosic biomass (Alvira et al., 2010). The mechanism of LHW pretreatment is similar to that of the diluted acid pretreatment, since the pretreatment conditions favor the release of organic acids from biomass, but LHW pretreatment is milder and generates less
LDMICs (Brodeur et al., 2011). However, previous studies showed that Clostridium beijerinkii growth was severely inhibited when a hydrolysate from LHW pretreatment of
Miscanthus was used for fermentation (Zhang and Ezeji, 2014). Alkaline pretreatment, which uses bases such as sodium hydroxide or lime, has been reported to produce less amounts of LDMICs than other pretreatments (Jönsson and Martín, 2016). The main effect of this pretreatment is lignin degradation, and it is effective at lower temperatures, including room conditions, and it can be performed at a higher total solids content, using less water (Modenbach and Nokes, 2012). However, the pretreated material is usually
186 washed to eliminate most of the base. Alternatively, to avoid the use of washing water, the pretreated biomass can be neutralized with a strong acid prior to the enzymatic hydrolysis (Cheng et al., 2010), but this impedes the recovery of the base and generates salts that have also been found to inhibit butanol fermentation (Qureshi et al., 2008b).
It has been hypothesized that fungal pretreatment does not generate significant levels of LDMICs that could inhibit butanol fermentation. To test this hypothesis, this study explores the effect of fungal pretreatment of Miscanthus in the production of LDMICs and butanol fermentation using C. beijerinkii, and compares it with two conventional pretreatment methods: liquid hot water and low-temperature alkaline pretreatment. These two methods have been considered “green”, the former because it does not use chemicals, and the latter because of its low energy requirement and water usage (Gu, 2013).
7.2 Materials and Methods
7.2.1 Feedstock collection and storage
Miscanthus was collected, milled, and stored as stated in section 5.2.1. The composition of the feedstock is summarized in Table 5.1.
7.2.2 Microorganisms and inoculum preparation
C. subvermispora (ATCC 96608) was obtained and preserved as stated in section
3.2.2. C. beijerinkii ATCC 51743 (strain NCIMB 8052) was obtained from the American
Type Culture Collection (Manassas, VA, USA), and maintained as spore suspension in sterile distilled water at 4˚C. For the inoculum preparation, 200 μl of the spore suspension
187 was heat-socked at 75˚C for 10 min, cooled in ice for 2 min, added to 10 ml of sterile anioxic tryptone-glucose-yeast (TGY) medium, and incubated for 12 h at 35˚C under anaerobic conditions. Then, the culture was transferred into 90 ml of fresh TGY and incubated for 3 h. This culture was used as inoculum for subsequent fermentations.
7.2.3 Pretreatment of Miscanthus
Miscanthus was fungal pretreated in triplicate under sterile and non-sterile conditions, according to the procedures described in sections 3.2.2 and 3.2.3, respectively. After the
28-days incubation time, samples were dried at 40˚C for compositional analysis, and the rest of the pretreated material was used fresh for enzymatic hydrolysis.
For comparison purposes, Miscanthus was also pretreated using alkaline and liquid hot water (LHW) pretreatment. For the LHW pretreatment, 25 g (dry matter) of
Miscanthus along with DI water to adjust to a 15% (w/v) solids loading were treated in a
Parr reactor (Parr Instrument Company, Moline, IL, USA) at 190˚C, for 20 min, agitated at 400 rpm by a double Rushton turbine impeller. Then, material was taken out of the reactor, weighed, and used directly for enzymatic hydrolysis or put to dry at 40˚C for compositional analysis. For the alkaline pretreatment, 50 g (dry matter) of Miscanthus were added to a 2 L flask with a NaOH solution to achieve a loading of 12 g NaOH/g
Miscanthus (dry matter), and a total solids loading of 21% (w/v). Flasks were covered with Parafilm and incubated at 25˚C for 24 h. At the end of the incubation time, the material of half of the flasks was washed with 500 ml of DI water, four times, until pH of about 8 was reached. Then, concentrated hydrochloric acid was used to adjust pH to 5.0.
188
For the other half of the alkaline treatment, 100 ml of a hydrochloric acid solution (0.73
M) was added, to adjust pH to 5.0 directly, without previous washing. Samples were taken from both the washed and non-washed alkaline pretreatment and dried at 40˚C for compositional analysis, and fresh material was used directly for enzymatic hydrolysis.
7.2.4 Enzymatic hydrolysis
Hydrolysates were prepared from the pretreated Miscanthus using Cellic CTec2 and
HTec2 (Novozymes, Bagsværd, Denmark), at a cellulase concentration of 60 FPU/g dry substrate. About 50 g (dry matter) of the fresh pretreated material was added to a 3 L bioreactor (New Brunswick Bioflo 110, Eppendorf, Hamburg, Germany). DI water was added if necessary to adjust to a 15% solid loading, and pH was adjusted to 5.0 with either ammonium hydroxide or hydrochloric acid. Saccharification was performed for
144 h, at 50˚C, and 800 rpm of agitation speed. At the end of the reaction, solid were eliminated by filtering through cheesecloth, following centrifugation (10,000 × g, 10 min,
4˚C). Finally the hydrolysates were sterilized by filtration (Corning vacuum filter, 0.22
µm, Corning Inc., Corning, NY, USA), and stored at -20˚C until use. Samples of 2 ml were taken from the hydrolysate, boiled for 15 min to inactivate enzymes, and passed through a 0.2 µm nylon filter into glass vials, to analyze the sugars concentration by
HPLC as explained in section 3.2.4. Cellobiose, glucose, and xylose were determined; concentrations of other sugars were below the detection limits. Sugar yields were calculated as follow:
189
[퐺푙푢푐표푠푒] + 1.053[퐶푒푙푙표푏푖표푠푒] % 퐺푙푢푐표푠푒 푦푖푒푙푑 = × 100% 1.111푓푔[푟퐵푖표푚푎푠푠]
[푋푦푙표푠푒] + 1.053[퐶푒푙푙표푏푖표푠푒] % 푋푦푙표푠푒 푦푖푒푙푑 = × 100% 1.14푓푥[푟퐵푖표푚푎푠푠]
[퐺푙푢푐표푠푒] + 1.053[퐶푒푙푙표푏푖표푠푒] + [푋푦푙표푠푒] % 푇표푡푎푙 푠푢푔푎푟 푦푖푒푙푑 = × 100% (1.111푓푔 + 1.14푓푥)[푟퐵푖표푚푎푠푠]
where [Glucose], [Cellobiose] and [Xylose] are the amount of glucose, cellobiose and xylose released by the enzymatic hydrolysis (g), respectively, [rBiomass] is the mass of dry untreated biomass at the beginning of the pretreatment (g), and fg and fx are the fraction of glucan and xylan in the original (untreated) dry biomass (g/g), respectively.
Thus, the sugar yield accounted for the cellulose loss that could occur during the fungal pretreatment.
7.2.5 Fermentation
Batch butanol fermentations of the hydrolysates and control were performed in triplicate in 100-ml Pyrex bottles with loose caps and 50 mL of total fermenation media, in anaerobic chamber (Coy Laboratory Products Inc., Ann Arbor, MI, USA) at 35˚C. To allow comparison, hydrolysates were supplemented with sugars to adjust them to the same sugar concentrations. Thus, the higher concentrations of cellobiose, glucose, and xylose were choosen, and all the other hydrolysates were adjusted to those levels (Table
190
7.2). Fermentation media for the hydrolysates consisted of 41 mL of hydrolysates, 5 mL of P2 medium stock containing vitamins, minerals, and MES buffer, 1 ml of yeast extract stock (50 g/L), and 3 mL of C. beijerinkii inoculum. Control was prepared similarly, but a mixed sugar media was used instead of the hydrolysates. pH was adjusted to 6.6-6.8 with ammonium hydroxyde. The final concentrations of the sugars in the the fermentation media for all the hydrolysates and the control were 12.3 g/L cellobiose, 41 g/L glucose, and 16.4 g/L xylose. The media were sterilized by filtration (Corning vacuum filter, 0.22 µm, Corning Inc., Corning, NY, USA) and equilibrated in the anaerobic chamber for at least 12 h before inoculation. Fermentation was performed for
72 h, and 2-mL samples were taken every 12 h for analysis of cell concentration, pH, fermentation products, and residual sugars.
7.2.7 Analytical methods
Cellulose (measured as glucan), xylan, arabinan, lignin, and dry matter content, sugar concentration, and extractives were determined as described in section 3.2.5. Cell concentrations in fermentations were monitored by optical density at 600 nm in a DU spectrophotometer (Beckman Coulter Inc., Brea, CA, USA). Fermentation products
(butanol, ethanol, acetone, acetic acid, and butyric acid) were quantified in a 7890A
Agilent GC-FID (Agilent Technologies, Santa Clara, CA, USA) with a 30 m (length) ×
320 μm (internal diameter) × 0.50 μm (HP-INNOWax film) J × W 19091 N-213 capillary column as previously described (Ujor et al., 2015). Acetone-butanol-ethanol (ABE) yield was calculated as the total production (g) of these three products per gram of total utilized
191 sugars. Concentrations of selected LDMICs (hydroxymethylfurfural, furfural, vanillic acid, syringic acid, vanillin, syringaldehyde, and p-coumaric acid) were determined by
HPLC in a Waters 2796 Bioseparation Module (Waters, Milford, MA, USA) with a photodiode array (PDA) detector and a 3.5 μm Xbridge C18 column, 150 mm × 4.6 mm
(Waters, Milford, MA, USA) as previously described (Zhang et al., 2012).
7.2.8 Data analysis
Statistical significance was evaluated by analysis of variance (ANOVA) (α=0.05), and mean comparisons were performed by Tukey-Kramer test with a 95% significance, using the Software JMP® (SAS Institute Inc., Cary, NC, USA). Error bars in the figures represent the standard error of three replicates.
7.3 Results and Discussion
7.3.1 Pretreatment and enzymatic hydrolysis
Solids recoveries between 71% and 89% were obtained after pretreatment of
Miscanthus (Table 7.1). Fungal pretreatment showed the highest solids recovery, and there was not significant difference between the solids recovery from the two fungal pretreatments and the alkaline non-washed (p>0.05). Conversely, liquid hot water had the lowest solids recovery, and there was no significant difference between this and the solids recovery of the washed alkaline pretreatment (p>0.05). However, the solids recovery of liquid hot water and washed alkaline pretreatment were significantly lower than that of the other pretreatments (p<0.05). Similar recoveries were found for LHW and alkaline
192 pretreatment of giant reed (Jiang et al., 2016), and for fungal pretreatment of Miscanthus
(Vasco-Correa et al., 2016a), corn stover, and switchgrass (Wan and Li, 2011b). The solids recovery after alkaline pretreatment was higher than that obtained from Miscanthus previously (de Vrije et al., 2009).
Table 7.1 Comparison of fungal, liquid hot water, and alkaline pretreatments on solids
recovery and changes in Miscanthus composition
Solids recovery (%) Cellulose Xylan Arabinan Lignin Pretreatment removal removal removal removal Total Solids Liquid* (%) (%) (%) (%) Fungal pretreatment sterile 88.8 88.8 - -0.1 9.8 37.5 22.3 Fungal pretreatment non- 83.5 83.5 - 14.2 18.1 36.1 21.0 sterile LHW pretreatment 70.9 63.8 7.1 3.3 86.8 100.0 50.3 Alkaline pretreatment 72.3 72.3 - 11.9 22.1 5.9 41.7 washed Alkaline pretreatment non- 81.3 72.4 8.9 13.5 28.6 12.0 42.2 washed *Soluble solids recovered in the liquor or free liquid. Only included if free liquid was present at the end of pretreatment, and if it was not washed away.
Significant lignin degradation, between 21% and 50%, was observed after pretreatments. However, fungal pretreatment showed the lowest degree of lignin degradation compared to LHW and alkaline pretreatment (p<0.05), and it was similar to previous results in Miscanthus (Vasco-Correa et al., 2016a), but slightly lower than that obtained for switchgrass and corn stover with C. subvermispora (Wan and Li, 2011b).
Lignin degradation for the LHW pretreatment was slightly lower than that obtained for
Miscanthus in similar conditions (Boakye-Boaten et al., 2015) and similar to the one attained for giant reed (Jiang et al., 2016). Alkaline pretreatment showed a lignin
193 degradation of about 42%, which was lower than that previously obtained for Miscanthus using NaOH under higher temperature (de Vrije et al., 2009), but close to that achieved for giant reed (Jiang et al., 2016) and corn stover (Zhu et al., 2010) under similar conditions to the present study.
The main differences between pretreatments were observed in the degradation of structural carbohydrates. LHW removed the totality of arabinan and the majority of xylan, leaving most of the cellulose intact, as it has been observed for wheat straw (Pérez et al., 2008) and giant reed (Jiang et al., 2016), and it has been considered to be the main effect of this pretreatment in biomass (Mosier et al., 2005). Fungal pretreatment removed some of the xylan (10-18%) and arabinan (26-38%), probably by using it as a carbon and energy source for C. subvermispora (and other microorganisms in the case of the non- sterile pretreatment), as it has been reported before (Vasco-Correa et al., 2016a; Wan and
Li, 2012). Alkaline pretreatment showed a significant removal of xylan (22-28%), which was slightly higher than that obtained for corn stover and giant reed in similar conditions
(Jiang et al., 2016; Zhu et al., 2010). Cellulose recovery was high in all pretreatments
(>85%), especially the sterile fungal pretreatment and the LHW pretreatment. The former was due to the relatively weak cellulosome system of C. subvermispora, which has been known to be activated only after longer rotting time (Tanaka et al., 2009). Cellulose degradation was higher in the non-sterile than in the sterile fungal pretreatment (p<0.05), probably due to an effect of contaminant microorganisms, and not to the white rot fungi
(Chapter 4).
194
Sugar yields were highest for the LHW pretreatment, and lowest for both fungal pretreatment (Table 7.2). For the LHW pretreatment, concentration of xylose in the hydrolysate was close to that of the study of Zhang and Ezeji (2014) using Miscanthus under similar conditions, but the glucose and cellobiose concentrations were significantly higher in the current results (Zhang and Ezeji, 2014), probably due to differences in the conditions of the enzymatic hydrolysis, including the enzymes used and the rheology of the process due to differences in equipment. Glucose and xylose yields obtained from the
LHW pretreatment were similar to those obtained by Pérez et al. (2008) from wheat straw
(Pérez et al., 2008) and by Khullar et al. (2013) from Miscanthus, but in their case, a significantly lower particle size was used, which tends to favor the enzymatic hydrolysis
(Khullar et al., 2013). For the alkaline pretreatment, glucose yields were higher than those obtained for Miscanthus previously by De Vrije et al. (2002), even though their lignin removal was higher (de Vrije et al., 2002). Non-washed alkaline pretreatment produced lower sugar yields than the washed pretreated Miscanthus, possibly due to the presence of soluble enzyme inhibitors that were removed by washing, as has been reported before (Cheng et al., 2010; Gao and Rehmann, 2014). Sugar yield for the fungal pretreated sterile Miscanthus was similar to that reported previously for corn stover and switchgrass pretreated with C. subvermispora (Wan and Li, 2011b). However, sugar yields were particularly low for the non-sterile fungal pretreatment, probably due to the presence of contaminant microorganisms that degraded the most of the available sugars.
195
Table 7.2 Sugar concentrations and yields of fungal, liquid hot water, and alkaline
pretreated Miscanthus after enzymatic hydrolysis
Concentration of sugars in Sugar yield (%) hydrolysate (g/L) Pretreatment Total Glucose Xylose sugars Cellobiose Glucose Xylose yielda yieldb yieldc Fungal pretreatment sterile 76.3 40.9 66.2 11.4 35.8 10.1 Fungal pretreatment non-sterile 53.1 19.3 43.5 9.5 23.3 4.8 LHW pretreatment 94.4 59.3 84.4 11.2 47.4 14.8 Alkaline pretreatment washed 89.8 78.8 86.7 13.2 42.5 19.6 Alkaline pretreatment non-washed 83.8 68.9 79.5 14.8 36.9 17.2 a Glucose in hydrolysate as a percentage of total glucose in form of cellulose in untreated Miscanthus b Xylose in hydrolysate as a percentage of total xylose in form of xylan in untreated Miscanthus c Total sugars (glucose and xylose) as a percentage of total sugars in form of cellulose and xylan in untreated Miscanthus
7.3.2 Fermentation
To study the effect of the LDMICs that could be present in the hydrolysates, the hydrolysates were supplemented with pure sugars, to achieve a concentration of 14.8 g/L of cellobiose, 47.4 g/L of glucose, and 19.6 g/L of xylose (Table 7.2). By adding the rest of the components of the fermentation medium (buffer, vitamins, minerals, and yeast extract), a dilution effect occurred and the concentration of sugars of the starting media was 12.3 g/L cellobiose, 41 g/L glucose, and 16.4 g/L xylose. Therefore, the differences in the fermentation profiles were mainly due to the presence of LDMICs and other components of the hydrolysates, and not to the differences in sugar yield from the pretreatments.
C. beijerinkii showed similar growth in the hydrolysates from fungal pretreated
Miscanthus and the control medium, and a similar level of maximum growth was obtained with the washed alkaline pretreated Miscanthus, but a lag phase of about 12
196 hours was observed (Figure 7.1a). In contrast, slight growth was detected in the hydrolysate from the non-washed alkaline pretreatment, and complete growth inhibition was observed in the hydrolysate from the LHW pretreatment. A previous study showed similar results for the hydrolysates of LHW pretreated Miscanthus (Zhang and Ezeji,
2014). LHW and non-washed alkaline pretreated Miscanthus showed total inhibition of butanol production. However, the inhibition mechanism was clearly different among them, since the unwashed alkaline pretreatment went into “acid crash” evidenced by the decrease in pH (Figure 7.1b) (Maddox et al., 2000), while LHW pretreatment did not show a significant change from the initial pH, or from any of the other parameters measured during the fermentation (Figure 7.1). Fungal pretreatments and washed alkaline pretreatment showed typical profiles of no-inhibited butanol production (Figure 7.1c).
Maximum butanol produced by fungal pretreatment hydrolysates was 16-18% higher than the control. Alkaline washed pretreated Miscanthus showed a maximum butanol concentration that was 36% higher than the control (Table 7.3). ABE yields were similar to those reported previously for Miscanthus (Agu et al., 2016b) and other lignocellulosic feedstocks (Ezeji and Blaschek, 2010). ABE yield obtained with the LHW pretreatment was similar to that obtained with other pretreatments, such as fungal pretreatment, even though the total ABE production with LHW pretreatment was considerably lower, because the sugar consumption was also low (Table 7.3).
197 a) b) 12 Fungal pret sterile 7 Fungal pret non-sterile 10 LHW pret Alkaline pret washed 6.5 Alkaline pret non-washed 8 Control (P2 medium) 6
6 pH 5.5
4 OD (600 OD nm) (600
2 5
0 4.5 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Time (h) Time (h) c) d) 16 25 14 20 12
10 15 8
6 10 Butanol (g/L) 4 TotalABE (g/L) 5 2 0 0 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Time (h) Time (h) e) f) 14 8
12 7 6 10 5 8 4 6
3 Acetate(g/L)
4 2 Butyric Butyric acid (g/L) 2 1 0 0 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Time (h) Time (h)
Figure 7.1 Fermentation profile of Miscanthus hydrolysates. a) Cell growth of C.
beijerinckii estimated by optical density at 600 nm, b) pH of fermentation media, c)
butanol concentration, d) total ABE concentration, e) acetate concentration, f) butyric
acid concentration.
198
Table 7.3 ABE fermentation results: final product concentrations, total sugars consumed
and yield
Final product concentration (g/L) Total sugar ABE Sample consumed yield Acetone Ethanol Butanol Total ABE (g/L) (g/g) Fungal pretreatment sterile 4.7 1.9 11.6 18.1 54.3 0.33 Fungal pretreatment non-sterile 5.5 1.8 12.1 19.4 57.1 0.34 LHW pretreatment 0 1.4 0 1.4 3.1 0.34 Alkaline pretreatment washed 5.5 1.8 14.0 21.2 55.8 0.38 Alkaline pretreatment non-washed 0 1.6 0 1.6 10.3 0.16 Control 4.0 1.8 10.1 15.9 53.0 0.30 Total ABE might not correspond with the sum of individual components because of rounding. Total sugar comprises cellobiose, glucose and xylose.
LHW pretreatment showed similar LDMICs production to that obtained previously under similar conditions from Miscanthus × giganteus (Zhang and Ezeji, 2014) (Table
7.4). In that case, the hydrolysate also inhibited cell growth and butanol fermentation.
Another study showed considerable higher LDMICs concentrations from Miscanthus × giganteus by using dilute acid pretreatment, since that pretreatment has higher severity, and fermentation was also inhibited (Agu et al., 2016a). Similar results for HMF and furfural were also obtained with LHW pretreatment of Miscanthus lutatioriparious (Q. Q.
Wang et al., 2012). Even though the concentrations of furfural and HMF in the fermentation medium from LHW pretreatment were below the limits of inhibition found previously (Ezeji et al., 2007b), it is believed that synergistic effects of these molecules and the possible presence of other unidentified compounds could cause the inhibition, since only a small amount of LDMICs have been identified, but there are possible hundreds of them (Liu and Blaschek, 2010; Zhang and Ezeji, 2014).
199
Table 7.4 Concentration of selected LDMICs in the fermentation media at the beginning of the fermentation (T = 0)
Pretreatment HMF Furfural Vanillic acid Syringic acid Vanillin Syringaldehyde p-coumaric acid (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) Fungal - sterile 6.8 ± 0.6 14.6 ± 0.9 16.5 ± 3.9 0.3 ± 0.0 10.0 ± 0.5 ND 1.7 ± 0.2 Fungal - non-sterile 5.5 ± 0.3 16.9 ± 0.5 7.6 ± 2.0 1.0 ± 0.3 10.2 ± 2.7 ND 0.9 ± 0.3
LHW 350.2 ± 6.1 1852.2 ± 0.6 113.8 ± 17.8 6.6 ± 1.7 25.5 ± 0.8 18.8 ± 0.8 6.5 ± 0.6
Alkaline - washed 2.3 ± 0.3 2.9 ± 0.1 2.1 ± 0.6 5.0 ± 1.4 21.4 ± 2.5 8.8 ± 0.8 5.2 ± 0.2
Alkaline - non-washed 2.2 ± 0.1 1.5 ± 0.1 61.2 ± 1.3 4.0 ± 0.4 6.9 ± 0.4 3.8 ± 0.4 0.6 ± 0.3 ND: not detected
2
00
200
Washing after alkaline pretreatment was an indispensable step to allow the fermentation of this hydrolysate. The neutralization of the high content of NaOH generated salts that inhibited the butanol fermentation as shown previously, probably due to osmotic pressure and other mechanisms that limit cell growth and ABE production
(Qureshi et al., 2008b). However, after washing, the alkaline pretreatment produced higher amounts of butanol and total ABE than the control medium (Table 7.3). Similar results have been reported for NaOH-pretreated corn cob, where the production of ABE was about 20% higher than the control (Gao and Rehmann, 2014). The pH profile in this fermentation was considerably different than the rest (Figure 7.1b). The typical biphasic pattern of ABE fermentation was observed, since the pH decreased in the first 24 hours during the exponential growth phase, and then increased after that. However, the initial drop was less drastic than in the control and fungal pretreatment media, and then the pH was more stable and remained at a higher value, close to 6.5. This was likely due to a buffer effect produced by the components of the hydrolysate, including the remaining salts and the acetate. pH control has been shown to enhance the production of ABE (Guo et al., 2012; Jiang et al., 2014), and therefore this buffer effect of the hydrolysate stimulated the butanol production. Additionally, higher initial acetate concentration was detected in this hydrolysate, and it also showed a higher uptake rate during the fermentation (Figure 7.1e). This could also contribute to the higher production of ABE, since acetate has been found to be a fermentation stimulator (Ezeji et al., 2007b; Qureshi et al., 2013a). The lag phase observed on this hydrolysate could have been produced by an initial uptake of some fermentation stimulators, including acetate, as has been
201 proposed previously (Zhang and Ezeji, 2014). Interestingly, concentration of the measured LDMICs in alkaline pretreatment media was considerably low (Table 7.4), including that of phenolic compounds that are expected to be derived from lignin, which is considerably degraded during this type of pretreatment (Table 7.1). The low production of LDMICs is likely attributable to the low temperature used for the alkaline pretreatment in this study. Possibly, most of the phenolic compounds derived from the lignin degradation were left as molecules of higher molecular size and not monomers, which were the ones measured in this study.
In the case of fungal pretreatment, the higher concentration of ABE produced compared to the control could also be explained in part by the higher acetate concentration at the beginning of the fermentation (Figure 7.1e). Additionally, pH of the fungal pretreatment media also showed a clear biphasic behavior, but this was not observed in the control (Figure 7.1b), which could explain the differences in ABE production. Other studies have shown better ABE titers for hydrolysates from pretreated lignocellulosic biomass than for control medium, in some cases after detoxification
(Qureshi et al., 2010a, 2010b, 2007), and it is suggested that some of those hydrolysates could have fermentation stimulators, which are chemicals that in low concentration could stimulate the production of ABE (Qureshi et al., 2013a). In the case of the fungal pretreatment, the presence of remaining components of the fungal biomass or residual fungal enzymes could act as fermentation stimulators, but more studies are needed to corroborate this. More important, there was no significant concentration of LDMICs produced by fungal pretreatment, including phenolic compounds derived from lignin
202 degradation (Table 7.4). This, along with the fermentation results, proved the hypothesis that fungal pretreatment produce low concentrations of fermentation inhibitors, including
LDMICs and salts, and that as a result, it does not affect the growth of C. beijerinkii and the ABE fermentation.
7.4 Conclusions
Fungal pretreatment under the studied conditions did not produce a significant concentration of LDMICs that would negatively affect butanol fermentation by
Clostridium beijerinckii. Thus, hydrolysates produced from fungal pretreated Miscanthus can be used directly for butanol fermentation without dilution or detoxification to obtain equivalent or higher ABE titers and yields than that of pure sugars’ control. However, the sugar yields obtained by enzymatic hydrolysis from the fungal pretreated material were significantly lower than those obtained from other pretreatment methods, and the fungal pretreatment under non-sterile conditions showed particularly low sugar yields. Alkaline pretreated Miscanthus under high solid content at low temperature showed great promise as a pretreatment strategy for butanol production, but a washing after pretreatment was required.
203
Chapter 8: Techno-economic analysis of solid-state fungal pretreatment of Miscanthus for the production of fermentable sugars
Abstract
Fungal pretreatment has been proposed as an alternative to conventional pretreatment technologies of lignocellulosic biomass for the production of fermentable sugars in the cellulosic biorefineries. Because fungal pretreatment is performed at low temperature and without added chemicals, it has been presumed to be low cost. This study asses the techno-economics of a fungal pretreatment facility for the production of fermentable sugars at the scale of a commercial cellulosic biorefinery of 30 million gallons of biofuel per year. An estimated sugars production cost of $2.36 per kg of fermentable sugars was obtained, which was significantly higher than previous estimates for other pretreatment technologies. Facility-related cost contributed 70% of the total sugars production cost, which was caused by high requirements in equipment size and quantity, due to low sugar yields, high bulk density of feedstock, and long retention times. Therefore, fungal pretreatment in the current state of the technology is not feasible at cellulosic biorefinery scale.
8.1 Introduction
Cellulosic biorefineries are projected to produce a collection of biofuels and bioproducts from lignocellulosic biomass (Cherubini, 2010). Most of these products 204 would be generated by fermentation of sugars released from the feedstocks by enzymatic saccharification. However, lignocellulosic biomass requires pretreatment prior enzymatic hydrolysis, to increase feedstock digestibility. Pretreatment has been considered one of the main contributors to the cost of final products in cellulosic biorefineries (Yang and
Wyman, 2008). Traditional pretreatments are performed at high temperature and pressure, use harsh chemicals such as acids or bases, and require high amounts of water
(Mosier et al., 2005). Also, they usually require an additional detoxification step, because their harsh conditions promote the generation of compounds derived from lignocellulose, such as furans, weak acids and aromatic compounds, which could inhibit fermentation microorganisms (Baral and Shah, 2014).
Fungal pretreatment has been presented as an alternative to conventional pretreatments with potential lower costs, since it is performed close to room temperature, at atmospheric pressure, in solid-state with low water use and no wastewater generation, and without added chemicals that would need disposal or recovery and can be corrosive
(Wan and Li, 2012). Particularly, fungal pretreatment with C. subvermispora degrades lignin preferentially over cellulose, minimizing sugar losses during the pretreatment
(Vasco-Correa et al., 2016a; Wan and Li, 2011b). However, fungal pretreatment has some potential disadvantages compared to traditional pretreatments, including high retention times (weeks compared to days or hours), lower sugar yields, and feedstock sterilization requirements (Vasco-Correa et al., 2016b). Baral and Shah (2017) studied the techno-economics of four different pretreatment technologies in a butanol biorefinery, including steam explosion, dilute acid, AFEX, and biological pretreatment. They found
205 that biological pretreatment required higher capital investment, and therefore, it had the higher sugars production cost. However, some important steps for the biological pretreatment, such as feedstock sterilization, were not included in the analysis (Baral and
Shah, 2017).
Pretreatment of lignocellulosic biomass is expected to be performed for a cellulosic biorefinery with an estimated scale of 20-30 million gallons of biofuel per year (Shah and
Darr, 2016). Higher scales for cellulosic biorefineries process are preferred because it tends to lower the cost per production unit, due to economies of scale (Lynd et al., 2005).
However, limitations of scale are imposed by feedstocks logistics (Shah and Darr, 2016).
Because of the relative simplicity of the process and the long retention times, fungal pretreatment has been proposed to be performed at a smaller scale, on-farm or during wet-storage of the feedstocks (Cui et al., 2012a).
It is hypothesized that fungal pretreatment of the lignocellulosic feedstock
Miscanthus with C. subvermispora is a low cost technology due to its mild processing conditions, and that is therefore feasible at cellulosic biorefinery scale, while compared to other traditional pretreatments. The main objective of the present study is to estimate the techno-economic feasibility of a commercial-scale fungal pretreatment process facility for the production of fermentable sugars from Miscanthus. Total cost of fermentable sugars is estimated, and the effect of process parameters on this cost is evaluated.
206
8.2 Materials and Methods
8.2.1 Modeling overview
A techno-economic process model for the transformation of Miscanthus into fermentable sugars using fungal pretreatment at solid-state with C. subvermispora and enzymatic hydrolysis with commercial hydrolases was performed for a cellulosic biorefinery with a capacity of 30 million gallons/year (113.5 million liters/year), since most of the current lignocellulosic biorefineries have capacities of 20-30 million gallon/year (Shah and Darr, 2016). To achieve this capacity, a sugar requirement of
181,787 tons/year was calculate previously (Baral and Shah, 2016b). The process model
(Figure 8.1) included storage, sterilization by autoclaving, solid-state fungal pretreatment, and enzymatic hydrolysis. A process flow diagram was completed using SuperPro
Designer software v.9.5 (Intelligen Inc., Scotch Plains, NJ, USA), and the data and expressions necessary to perform material and energy balances, equipment capacity limitations, purchase price, and installation and maintenance expenses were included in the model. The material and energy balance were used to determine the required equipment size and quantity, raw materials, utilities, as well as other direct and indirect costs.
207
Figure 8.1 Simplified overview of the fungal pretreatment process including major equipment. Fungal pretreatment is performed in pack bed bioreactor. Fungal inoculum is prepared in air-lift seed fermentor. Feedstock is sterilized by autoclaving. Miscanthus is
used as feedstock and sugars solution is the main product.
8.2.2 Miscanthus preparation unit
Composition of Miscanthus used for the modeling is depicted in Table 5.1, and bulk density was assumed to be 150 kg/m3 (Caslin et al., 2010). Based on data from Chapter 7, to produce the amount of sugars needed for a cellulosic biorefinery of 30 million gallons/year, 530,628 ton of dry Miscanthus would be needed annually. An average moisture content of 15% was assumed, so a total of 624,628 of Miscanthus was required.
208
Miscanthus logistics was not modeled in the present study, and therefore price at biorefinery gate of $0.110/kg was based in previous studies for Miscanthus and other lignocellulosic feedstocks (Shah and Darr, 2016; U.S. DOE, 2016). Miscanthus was assumed to be harvested using a forage chopper that produced a particle size in the range validated previously for fungal pretreatment of Miscanthus (Caslin et al., 2010; Vasco-
Correa et al., 2016a); thus, no milling in the facility was required. Miscanthus was temporary stored in concrete tanks before being transported using a bucket elevator and a belt conveyor into horizontal batch autoclaving units for sterilization with superheated steam for 40 minutes. Water to adjust moisture content to 60% was added before autoclaving. After that, material was transported out of the autoclave by screw conveyors.
8.2.3 Pretreatment unit
Pretreatment units consisted of pack bed bioreactors, which have been one of the most recommended options for solid-state aerobic fermentation with microorganisms that are sensitive to shear stress (Mitchell et al., 2006), such as white rot fungi (Wan and Li,
2012). This reactor was inoculated from the top with C. subvermispora mycelium grown in liquid culture in a seed air-lift fermenter, and aerated from the bottom with compressed air to achieve an aeration rate of 0.01 vvm. The reactor was kept at 28˚C using chilling water, since the metabolism of the white rot fungi would likely increase the temperature, and the incubation time was 28 days. Degradation rates of structural carbohydrates and lignin were as reported in Chapter 7 (Table 7.1).
209
8.2.4 Enzymatic hydrolysis unit
After pretreatment, Miscanthus was transported using screw conveyors into stirred tanks reactors for enzymatic hydrolysis. Enzymatic saccharification was performed using commercial hydrolases at an enzyme loading of 20 mg of protein/g of cellulose (Selig et al., 2008). The price of the enzymes was estimated according to previous studies
(Humbird et al., 2011). Enzymatic saccharification was performed at 50˚C during 72 h, at a 15% solid loading, and hydrolysis conversion rates were as described in Chapter 7
(Table 7.2).
8.2.5 Economic analysis
Parameters to estimate capital and operating cost were determined according to previous studies (Aden et al., 2002; Baral and Shah, 2017; Brown, 2003; Humbird et al.,
2011; Ulrich, 1984), and are shown in Table 8.1. The facility was estimated to operated
330 days per year, during 24 h per day (Baral and Shah, 2017; Humbird et al., 2011).
Equipment size and quantity were calculated based on the material and energy balance using the “Design Mode” of the SuperPro Design software (Intelligent, 2014). The expressions used to estimate the purchasing price of equipment were based in the literature (Aden et al., 2002; Baral and Shah, 2017, 2016a, 2016b; Humbird et al., 2011) and on the built-in data from the modeling software (Intelligent, 2014). Equipment installation expenses were also calculated by the modeling software (Intelligent, 2014).
210
Table 8.1 Economic evaluation parameters used in this study
Time parameters Value Financing parameters Value Construction Plan Value Analysis year 2017 Equity (%) 40 1st year ( % of DFC) 30 Year construction starts 2017 Loan term (years) 12 years 2nd year ( % of DFC) 40 Construction period (months) 18 Loan interest (%) 8 3rd year ( % of DFC) 30 Startup period (months) 12 Depreciation method Straight line Plant direct cost parameters: Project life (years) 30 Depreciation period (years) 15 Process piping (% of PC) 35 Inflation rate (%) 2 Income tax rate (%) 40 Instrumentation (% of PC) 40 Fixed cost parameters: Operating parameters: Insulation (% of PC) 5 Auxiliary facilities (% of PC) 40 Annual operating time (h) 7920 Electrical (% of PC) 10 Engineering (% of TPDC) 20 Salvage factor (% of DFC)** 5 Buildings (% of PC) 45 Construction (% of TPDC) 20 Startup cost (% of DFC) 5 Yard improvement (% of PC) 15 21 Contractor's fee (% of TPC) 5
1 Contingency (% of TPC) 10 Based on: Aden et al., 2002; Baral and Shah, 2017; Brown, 2003; Humbird et al., 2011; Ulrich, 1984. PC: purchase cost of equipment; TPDC: total plant direct cost (physical cost); TPC: total plant cost (includes TPDC and indirect cost,); DFC: direct fixed capital.
211
Total capital investment (TCI) was determined as the sum of direct fixed capital
(DFC), working capital, and start-up cost. DFC included total plant direct cost (TPDC), total plant indirect cost (TPIC), and contractor’s fee and contingency. TPDC comprised equipment cost, installation, piping, instrumentation, insulation, electrical, buildings, yard improvement, and auxiliary facilities. TPIC was estimated as the cost of the engineering and construction. Working capital was calculated by the modelling software and included cost of raw materials, labor and utilities needed to run the plant for 30 days. Start-up cost was estimated as 5% of the DFC. Annual operating cost included cost of raw materials, labor, utilities, and facility-dependent, and laboratory and quality control related costs
(Intelligent, 2014).
8.2.6 Sensitivity analysis and evaluation of strategies to reduce sugars production cost
Sensitivity analysis was performed to study the effect of process parameters changes in the capital investment and annual operating cost. Parameters evaluated were glucose and xylose yield during enzymatic hydrolysis, fungal pretreatment incubation time, autoclaving time, feedstock and enzyme costs, feedstock bulk density, and plant scale.
Changes in the sugars production cost were evaluated by variating the values set-up of these parameters ± 20%.
Plant scale effect on sugars production cost was also evaluated for smaller scale plants. Values chosen were sugars necessary to produce 4.6, 0.4, and 0.04 million gallons of cellulosic biofuel per year.
212
Potential strategies to reduce the sugars production cost in the fungal pretreatment facility were evaluated by changing sensitive parameters one-at-a-time into potentially achievable values for this parameters, according to experimental data and literature, as explained in the results.
8.3 Results and Discussion
8.3.1 Material balance
Figure 8.2 summarized the material balances to produce 1 ton of fermentable sugars
(glucose and xylose) in the fungal pretreatment facility. Complete mass recovery after autoclaving was assumed, as shown in previous experimental studies (Wan and Li,
2011b). Almost 3 tons of dry Miscanthus were needed to produce 1 ton of fermentable sugars, and more than 2 tons of solid material remained at the end of the enzymatic hydrolysis. Those solids contained lignin (about 420 kg/ton of fermentable sugars), which is usually burned for heat and power, but they also contained significant amounts of high value structural carbohydrates (about 670 kg/ton of fermentable sugars). The recovery and potential economic benefits of these streams were not included in the model.
Baral and Shah (2017) reported that biological pretreatment needed a considerable amount of water, compared to conventional pretreatments. However, in this study the total water used per ton of fermentable sugars was significantly lower, because (1) less feedstock was used, since better yields were assumed based on experimental data, and (2) moisture content of 60% instead of 75% was established for the fungal pretreatment, since no significant difference in sugar yields have been observed with these two
213 moisture content levels (Vasco-Correa et al., 2016a). Noticeably, a significant amount of carbon dioxide was produced during the fungal pretreatment, due to the metabolism of the white rot fungus and the lignin mineralization, which is usually not discussed as an impact of this pretreatment. However, all the lignin degraded during pretreatment was considered to be mineralized to carbon dioxide and water, but actually some part would probably remained as intermediate soluble compounds that have to be measured experimentally to further improve the model. Therefore, this model probably overestimated the amount of carbon dioxide produced.
214
Inoculum: 293 Water: 3656 Water: 297
Carbon dioxide: 5420 Miscanthus (raw-dry weight): 2984 Miscanthus (sterile-dry weight): 2984 Water (moisture): 527 Water (moisture): 2984 Fungal Autoclave pretreatment
Miscanthus (pretreated-dry weight): 2617 Water (moisture): 4555
21
5
Enzyme: 490 Enzymatic Water: 9771 hydrolysis
Glucose: 814 Xylose: 186 Other solids: 2209 Water: 14222
Figure 8.2 Material balance requirement to produce 1 ton of fermentable sugars in a fungal pretreatment facility (values in kg).
215
8.3.2 Economic analysis
Estimated production cost of fermentable sugars in the fungal pretreatment facility was $2.36/kg sugars. This value was 4- to 10-fold over previous reports of conventional pretreatments (Baral and Shah, 2017; Humbird et al., 2011). Facility related costs comprised 70% of that cost, while materials constituted 22%, and utilities 7% (Figure
8.3). This contrasted with most economic studies of cellulosic biorefineries, which have found that feedstock cost was the main contributor to the final cost (Aden et al., 2002;
Baral and Shah, 2017, 2016b; Humbird et al., 2011). However, this result was similar to that obtained by Baral and Shah (2017) in biological pretreatment (Baral and Shah,
2017), and is due to the high capital investment required (Table 8.2). Total capital investment was 14- to 18-fold over that estimated previously for conventional pretreatments at similar scale (Baral and Shah, 2017). High capital investment was due mostly to the large size of equipment needed, and the requirement of several units for each key process, such as fungal pretreatment and autoclaving (Table A. 2), which was caused by the large residence time of the fungal pretreatment process and the low bulk density of the feedstock.
216
2.50
2.00
1.50
1.00
0.50 Sugar production cost($/kg) production Sugar
0.00 Materials Facility Labor Lab/QC/QA Utilities TOTAL
Figure 8.3 Fermentable sugars production cost from Miscanthus with fungal pretreatment
process at biorefinery scale.
Table 8.2 Capital investment for fungal pretreatment facility producing fermentable
sugars from Miscanthus for a 30 million gallon/year cellulosic biorefinery
Cost Item Annual operating cost ($) Direct fixed capital 1,854,921,170 Working capital 11,379,921 Startup cost 92,746,058 TOTAL 1,959,047,149
Feedstock cost constituted 75% of the total cost of the materials required, and enzymes cost was 23% (Table 8.3). Even though significant quantities of superheated steam were used for sterilization of Miscanthus (Table A. 4), the total utilities only accounted for 6% of the annual operating cost. Therefore, the higher impact associated
217 with autoclaving was due to the capital investment in the equipment, and not to the actual energy used for the operation. Additionally, there was heat generated in the bioreactors, which could be recovered for heating other units, such as the enzymatic hydrolysis, but this was not accounted in the present study.
Table 8.3 Annual operating cost of fungal pretreatment facility producing fermentable
sugars from Miscanthus for a 30 million gallon/year cellulosic biorefinery
Cost Item Annual operating cost ($) Miscanthus 68,669,477 Enzyme 20,930,425 Other material 1,611,607 Labor 6,021,496 Facility-dependent 294,243,134 Laboratory/QC/QA 903,224 Utilities 27,946,125 TOTAL 420,325,489
8.3.2 Sensitivity analysis
Glucose yield during the enzymatic hydrolysis was the parameter with most influence in
in the fermentable sugars cost, followed by pretreatment time and xylose yield (
Figure 8.4). Glucose yield was also the most sensitive parameter in Baral and Shah
(2017) study of biological pretreatment, while feedstock cost is usually the most sensitive parameter for most conventional pretreatments (Baral and Shah, 2017; Humbird et al.,
2011; Kumar and Murthy, 2011b). Glucose yields affected the size of equipment required, thus affecting the facility-related portion of the cost, which was the most substantial fraction, while feedstock cost affected only the material-related share of the
218 cost, which was significantly lower than facility-related costs, in the case of fungal pretreatment in the present study. Glucose yield had more influence than xylose yield on the fermentable sugars cost, because the assumed glucose yield was significantly higher than the xylose yield (Table 7.2); therefore, a constant percentage change in the glucose yield had a higher net effect. Also, glucose and xylose yields influenced both the amount of fermentable sugars produced and the feedstock required. Fungal pretreatment residence time is massive compared to conventional pretreatments, and therefore, changes in this time had high influence in the total sugars cost. The major effect in the reduction of the pretreatment residence time was the decrease in the number of packed bed bioreactors needed for the fungal pretreatment, thus reducing substantially the capital cost.
Sugar production cost ($/kg) 1.80 2.00 2.20 2.40 2.60 2.80 3.00
Glucose yield
Bulk density
Pretreatment time
Xylose yield
Feedstock cost
Enzyme cost
Autoclave time Decrease Plant size Increase
Figure 8.4 Sensitivity analysis of sugars production cost in fungal pretreatment
facility from Miscanthus for a 30 million gallon/year cellulosic biorefinery 219
Feedstock and enzyme costs were not as sensitive as they have been found with conventional pretreatments (Baral and Shah, 2017; Humbird et al., 2011; Kumar and
Murthy, 2011b), because in fungal pretreatment the capital cost comprised the majority of the cost share, and changes in this factor had a more significant impact in the sugars cost.
Autoclaving time and plat size were the most insensitive of the parameters evaluated.
Reduction in autoclaving time impacted the number of units required, but not substantially enough to have a significant impact in the capital cost. It also reduced the amount of high pressure steam used, but as stated previously, utilities had a small share in the sugars cost.
8.3.3 Effect of plant size
Fungal pretreatment has been proposed to be performed at small-scale, during storage of the feedstock. That would reduce the capital cost required for the facility. However, because of the requirement to keep the feedstock sterile, autoclaving units and bioreactors are still necessary. Additionally, unit cost usually decreases with increase on plant size, because of economies of scale (Lynd et al., 2005). Figure 8.5 shows the effect of plant size on the fermentable sugars production cost. Decreasing the size of the fungal pretreatment facility increased the sugars production cost following a power law.
Therefore, performing fungal pretreatment at facilities of smaller scale than cellulosic biorefineries would not favor the economics of the process.
220
40
35 792, $34.98
30
25 y = 354.1x-0.405 20 R² = 0.8718
15 7920, $5.97 10 624270.24, 95040, $2.47
5 $2.36 Fermentable sugars cost ($/kg) cost sugars Fermentable
0 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000
Biorefinery size (kg of input wet Miscanthus per year)
Figure 8.5 Effect of plant size on fermentable sugars production cost from Miscanthus in
a fungal pretreatment facility.
8.3.4 Potential strategies to reduce sugars production cost
The sensitivity analysis provided information about the process parameters that have higher influence over the sugars production cost (Figure 8.4). However, the variation and optimization of these parameters depend on the technical constrains of the process.
Figure 8.6 shows some strategies to reduce the sugars production cost and their impact.
First, enhancing the enzymatic hydrolysis solids loading from 15% to 30% is highly likely, since NREL and others have experimentally demonstrated this before (Modenbach and Nokes, 2013; Roche et al., 2009). Feedstock cost was considered $110/ton according to studies for corn stover (Shah and Darr, 2016). However, estimates from the U.S
Department of Energy predicted a much lower cost for lignocellulosic feedstocks such as
221
Miscanthus (U.S. DOE, 2016). Enzymes cost has been considered before at a lower price, so this effect was also evaluated in the present study (Baral and Shah, 2016b). To increase the bulk density, a change on the feedstock should be considered, because densification strategies are not suitable for fungal pretreatment, since the process is highly aerobic. A woody feedstock such as poplar or willow could be used instead of
Miscanthus, since they have a considerably higher bulk density (Adrián and Ladislav,
2014) and could potentially generate better results with fungal pretreatment, as has been shown before for other woody feedstocks (Chapter 5). To increase the sugar yields and reduce the pretreatment time, more advance strategies for optimization of fungal pretreatment would be necessary, such as combination with other low severity pretreatments, genetic modification of fungal strains, or developing of new strains.
By combining these strategies, an estimated cost of $0.47/kg of fermentable sugars could potentially be achieved. Previous reports have estimated a cost of about $0.25/kg of fermentable sugars for cellulosic biorefineries (Baral and Shah, 2017; Humbird et al.,
2011) .
222
Sugar production cost ($/kg)
0.00 0.50 1.00 1.50 2.00 2.50
Base case Enz. hyd. solid loading (15% --> 30%) Feedstock cost ($110/ton --> $60/ton) Enzyme cost ($0.24/kg --> $0.12/kg) Bulk density (150 kg/m3 --> 300 kg/m3) Glucose yield (76% --> 90%) Xylose yield (41% --> 80%) Pretreatment time (28 d --> 7 d) Eliminate autoclaving $0.47/kg
Figure 8.6 Potential strategies and impacts for the reduction of fermentable sugars
production cost in a fungal pretreatment facility.
8.4 Conclusions
The fungal pretreatment facility under the conditions studied was not feasible at full cellulosic biorefinery scale, due to low sugar yields, long pretreatment residence time, and low bulk density of the feedstock, which generated high requirements in equipment size and number of units for key processes, such as fungal pretreatment and autoclaving.
This had a direct impact on the capital cost and thus, the fermentable sugars production cost. Optimization of sensitive parameters such as sugar yield and pretreatment time could reduce the sugars production cost to about $0.47/ kg of fermentable sugars.
Performing the fungal pretreatment at facilities of smaller scales would not benefit the economics of the process.
223
Chapter 9: Conclusions and recommendations for future research
9.1 Conclusions
Solid-state fungal pretreatment of Miscanthus with the white rot fungus Ceriporiopsis subvermispora performed at low temperature, with no added chemicals, and no wastewater generation enhances the enzymatic digestibility and sugar yield of the feedstock. The main effect of fungal pretreatment with C. subvermispora is lignin degradation, which directly impacts the recalcitrance of the feedstock to enzymatic hydrolysis. C. subvermispora uses hemicellulose as the main carbon and energy source, leaving most of the cellulose unaltered. Fungal pretreatment of non-sterile Miscanthus was performed by using fungal colonized Miscanthus as inoculum, yielding similar results as the pretreatment of sterile Miscanthus. However, this process could not be completed in a sequential or semi-continuous mode, since the fungal community of
Miscanthus easily outcolonized the white rot fungus. Therefore, sterilization of feedstocks, or at least a significant microbial reduction, is an indispensable step for the reliability and reproducibility of fungal pretreatment. Fungal pretreatment with C. subvermispora was effective for Miscanthus harvested in winter or spring, but not for the green Miscanthus harvested in fall. Changes over a wide range of moisture contents showed no significant effect during fungal pretreatment of Miscanthus, which is
224 beneficial for the scale-up of this process, since moisture content is usually a parameter difficult to control in solid-state fermentation.
Fungal pretreatment produces specific changes in the chemical and physical structure of the lignocellulosic biomass that are feedstock-dependent. In general, fungal pretreatment is more effective for woods than herbaceous feedstocks. C. subvermispora has the capacity of making specific changes to components of lignocellulosic biomass, including reduction of acetyl units and degradation of constituents of wood resin, such as terpenes.
Fungal pretreated Miscanthus can be used for biogas generation via anaerobic digestion or for fermentable sugars production by enzymatic hydrolysis. Fungal pretreatment increases the specific methane yield in solid-state anaerobic digestion of
Miscanthus, because it reduces the feedstock recalcitrance, enhancing its digestibility while preserving a significant amount of the sugars that are the direct substrate for biogas production. Fermentable sugars produced by enzymatic hydrolysis of the fungal pretreated feedstock using commercial hydrolases can be used for the production of biofuels and bioproducts. This work showed that fungal pretreatment does not generate a significant amount of lignocellulosic derived microbial inhibitors. Therefore, hydrolysates from fungal pretreatment can be used directly for fermentation into biofuels such as butanol, without detoxification, dilution or washing. However, the sugar yields of fungal pretreatment are still lower than other leading pretreatment technologies. Finally, techno-economic analysis showed that fungal pretreatment of Miscanthus with the current state of the technology is not feasible at cellulosic biorefinery scale, because the
225 long residence time, low sugar yields, and low bulk density impose high requirements in capital investment.
9.2 Recommendations for future research
In order to overcome some of the main disadvantages of the fungal pretreatment process, research is still needed in different areas. In the short term, studies of fungal pretreatment at pilot-scale are necessary. This would allow to identify the best bioreactor design and the effects of scale on mass transfer, oxygen availability, temperature increase due to fungal metabolism, and compaction and bulk density during and after the pretreatment process, among others. Also, in order to increase the sugar yields to levels that are commercially feasible (usually estimated around 90% or higher), combination of other low severity pretreatments with fungal pretreatment could be considered.
This study showed that using non-sterilized feedstocks is not sustainable for a stable fungal pretreatment process, since indigenous fungi easily outcolonize the white rot fungus. However, alternative methods of microbial reduction could be evaluated to replace the energy-intensive autoclaving, such as pasteurization, gamma radiation, or chemical treatments.
The significant recent developments in understanding the microbial degradation of lignocellulosic biomass, with studies in genomics, transcriptomics, metabolomics, microbial ecology, and enzymatic technology, are the foundation of future work in the design of better systems for lignocellulosic biomass utilization. In the long term, this knowledge could allow the development of designed systems such as genetic engineering
226 microorganisms that could perform the pretreatment with higher yields and in shorter times. For example, a faster-growing microorganism, such as yeast or bacteria, could be engineered with the ligninolytic system from white rot fungi, including the accessory enzymes that would allow the synergistic effect that is seen in the original fungi. Also, the direct use of the ligninolytic enzymes for the pretreatment, instead of the microorganisms, could allow the reduction of pretreatment time and increase efficiencies.
Further studies in the use of fungal pretreated feedstocks for biofuels production could explore conditions that better simulate real life conditions, such as using the hydrolysates without filtration of the solids and replacing expensive additives of fermentation media with lower cost components. Finally, life cycle assessment of the fungal pretreatment of biomass should be performed to evaluate the potential environmental advantages of this process compared to the conventional thermo-chemical pretreatments, and assess the long-term sustainability of this process.
227
References
Abbasi, T., Tauseef, S.M., Abbasi, S.A., 2012. Anaerobic digestion for global warming control and energy generation - An overview. Renew. Sustain. Energy Rev. 16, 3228–3242.
Abdel-Hamid, A.M., Solbiati, J.O., Cann, I.K.O., 2013. Insights into Lignin Degradation and its Potential Industrial Applications, Advances in Applied Microbiology. Elsevier.
Acaroğlu, M., Şemi Aksoy, a., 2005. The cultivation and energy balance of Miscanthus×giganteus production in Turkey. Biomass and Bioenergy 29, 42–48.
Aden, A., Ruth, M., Ibsen, K., Jechura, J., Neeves, K., Sheehan, J., Wallace, B., Montague, L., Slayton, A., Lukas, J., 2002. Lignocellulosic Biomass to Ethanol Process Design and Economics Utilizing Co-Current Dilute Acid Prehydrolysis and Enzymatic Hydrolysis for Corn Stover, National Renewable Energy Laboratory- NREL.
Adrián, B., Ladislav, D., 2014. Density and bulk density of green wood chips from dendromass of short rotation coppice grown on plantations. Acta Fac. xylologiae Zvolen.
Agu, C.V., Ujor, V., Gopalan, V., Ezeji, T.C., 2016a. Use of Cupriavidus basilensis-aided bioabatement to enhance fermentation of acid-pretreated biomass hydrolysates by Clostridium beijerinckii. J. Ind. Microbiol. Biotechnol. 43, 1215–1226.
Agu, C.V., Ujor, V., Gopalan, V., Ezeji, T.C., 2016b. Use of Cupriavidus basilensis- aided bioabatement to enhance fermentation of acid-pretreated biomass hydrolysates by Clostridium beijerinckii. J. Ind. Microbiol. Biotechnol. 43, 1215–1226.
Ahmad, M., Roberts, J.N., Hardiman, E.M., Singh, R., Eltis, L.D., Bugg, T.D.H., 2011. Identification of DypB from Rhodococcus jostii RHA1 as a lignin peroxidase. Biochemistry 50, 5096–107.
Ahmad, M., Taylor, C.R., Pink, D., Burton, K., Eastwood, D.C., Bending, G.D., Bugg, T.D.H., 2010. Development of novel assays for lignin degradation: comparative analysis of bacterial and fungal lignin degraders. Mol. Biosyst. 6, 815–21. 228
Akhtar, M., Blanchette, R.A., Kent Kirk, T., 1997. Fungal delignification and biomechanical pulping of wood. In: Advances in Biochemical Engineering/Biotechnology Vol. 57. Springer Berlin Heidelberg, pp. 159–195.
Akin, D.E., Rigsby, L.L., Sethuraman, A., Morrison, W.H., Gamble, G.R., Eriksson, K.- E.L., 1995. Alterations in structure, chemistry, and biodegradability of grass lignocellulose treated with the white rot fungi Ceriporiopsis subvermispora and Cyathus stercoreus. Appl. Envir. Microbiol. 61, 1591–1598.
Alcalde, M., 2015. Engineering the ligninolytic enzyme consortium. Trends Biotechnol. 33, 1–8.
Alvira, P., Tomás-Pejó, E., Ballesteros, M., Negro, M.J., 2010. Pretreatment technologies for an efficient bioethanol production process based on enzymatic hydrolysis: A review. Bioresour. Technol. 101, 4851–61.
Amarasekara, A.S., 2013. Pretreatment of Lignocellulosic Biomass. In: Handbook of Cellulosic Ethanol. John Wiley & Sons, Inc., pp. 147–217.
Amon, T., Amon, B., Kryvoruchko, V., Machmüller, A., Hopfner-Sixt, K., Bodiroza, V., Hrbek, R., Friedel, J., Pötsch, E., Wagentristl, H., Schreiner, M., Zollitsch, W., 2007. Methane production through anaerobic digestion of various energy crops grown in sustainable crop rotations. Bioresour. Technol. 98, 3204–12.
Amougou, N., Bertrand, I., Machet, J.-M., Recous, S., 2011. Quality and decomposition in soil of rhizome, root and senescent leaf from Miscanthus x giganteus, as affected by harvest date and N fertilization. Plant Soil 338, 83–97.
Anderson, E., Arundale, R., Maughan, M., Oladeinde, A., Wycislo, A., Voigt, T., 2014. Growth and agronomy of Miscanthus × giganteus for biomass production. Biofuels.
Angelini, L.G., Ceccarini, L., Nassi o Di Nasso, N., Bonari, E., 2009. Comparison of Arundo donax L. and Miscanthus x giganteus in a long-term field experiment in Central Italy: Analysis of productive characteristics and energy balance. Biomass and Bioenergy 33, 635–643.
APHA, AWWA, WEF, 2005. Standard Methods for the Examination of Water & Wastewater, 21st ed, Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC.
Arora, D.S., Chander, M., Gill, P.K., 2002. Involvement of lignin peroxidase, manganese peroxidase and laccase in degradation and selective ligninolysis of wheat straw. Int. Biodeterior. Biodegradation 50, 115–120.
Arundale, R.A., Dohleman, F.G., Heaton, E.A., Mcgrath, J.M., Voigt, T.B., Long, S.P.,
229
2014. Yields of Miscanthus x giganteus and Panicum virgatum decline with stand age in the Midwestern USA. GCB Bioenergy 6, 1–13.
Atkinson, C.J., 2009. Establishing perennial grass energy crops in the UK: A review of current propagation options for Miscanthus. Biomass and Bioenergy 33, 752–759.
Axelsson, L., Franzén, M., Ostwald, M., Berndes, G., Lakshmi, G., Ravindranath, N.H., 2012. Perspective: Jatropha cultivation in southern India: Assessing farmers’ experiences. Biofuels, Bioprod. Biorefining 6, 246–256.
Bak, J.S., Kim, M.D., Choi, I.G., Kim, K.H., 2010. Biological pretreatment of rice straw by fermenting with Dichomitus squalens. N. Biotechnol. 27, 424–434.
Baldini, M., da Borso, F., Ferfuia, C., Zuliani, F., Danuso, F., 2017. Ensilage suitability and bio-methane yield of Arundo donax and Miscanthus×giganteus. Ind. Crops Prod. 95, 264–275.
Baldrian, P., 2006. Fungal laccases - occurrence and properties. FEMS Microbiol. Rev. 30, 215–242.
Banerjee, G., Car, S., Scott-Craig, J.S., Borrusch, M.S., Bongers, M., Walton, J.D., 2010a. Synthetic multi-component enzyme mixtures for deconstruction of lignocellulosic biomass. Bioresour. Technol. 101, 9097–9105.
Banerjee, G., Scott-Craig, J.S., Walton, J.D., 2010b. Improving enzymes for biomass conversion: A basic research perspective. BioEnergy Res. 3, 82–92.
Baral, N.R., Shah, A., 2014. Microbial inhibitors: formation and effects on acetone- butanol-ethanol fermentation of lignocellulosic biomass. Appl. Microbiol. Biotechnol.
Baral, N.R., Shah, A., 2016a. Techno-Economic Analysis of Cellulosic Butanol Production from Corn Stover through Acetone-Butanol-Ethanol Fermentation. Energy and Fuels 30, 5779–5790.
Baral, N.R., Shah, A., 2016b. Techno-economic analysis of cellulose dissolving ionic liquid pretreatment of lignocellulosic biomass for fermentable sugars production. Biofuels, Bioprod. Biorefining 10, 70–88.
Baral, N.R., Shah, A., 2017. Comparative techno-economic analysis of steam explosion, dilute sulfuric acid, ammonia fiber explosion and biological pretreatments of corn stover. Bioresour. Technol. 232, 331–343.
Bellamy, P.E., Croxton, P.J., Heard, M.S., Hinsley, S.A., Hulmes, L., Hulmes, S., Nuttall, P., Pywell, R.F., Rothery, P., 2009. The impact of growing miscanthus for biomass on farmland bird populations. Biomass and Bioenergy 33, 191–199. 230
Bendl, R.F., Kandel, J.M., Amodeo, K.D., Ryder, A.M., Woolridge, E.M., 2008. Characterization of the oxidative inactivation of xylanase by laccase and a redox mediator. Enzyme Microb. Technol. 43, 149–156.
Bento, T.S., Torres, L.M.B., Fialho, M.B., Bononi, V.L.R., 2014. Growth inhibition and antioxidative response of wood decay fungi exposed to plant extracts of Casearia species. Lett. Appl. Microbiol. 58, 79–86.
Boakye-Boaten, N.A., Xiu, S., Shahbazi, A., Fabish, J., 2015. Liquid Hot Water Pretreatment of Miscanthus X giganteus for the Sustainable Production of Bioethanol. BioResources 10, 5890–5905.
Boakye-Boaten, N.A., Xiu, S., Shahbazi, A., Wang, L., Li, R., Mims, M., Schimmel, K., 2016. Effects of fertilizer application and dry/wet processing of Miscanthus x giganteus on bioethanol production. Bioresour. Technol. 204, 98–105.
Bomberg, M., Sanchez, D.L., Lipman, T.E., 2014. Optimizing fermentation process miscanthus-to-ethanol biorefinery scale under uncertain conditions. Environ. Res. Lett. 9, 64018.
Boonmanumsin, P., Treeboobpha, S., Jeamjumnunja, K., Luengnaruemitchai, A., Chaisuwan, T., Wongkasemjit, S., 2012. Release of monomeric sugars from Miscanthus sinensis by microwave-assisted ammonia and phosphoric acid treatments. Bioresour. Technol. 103, 425–431.
Borkowska, H., Molas, R., 2013. Yield comparison of four lignocellulosic perennial energy crop species. Biomass and Bioenergy 51, 145–153.
Bourbonnais, R., Paice, M.G., 1990. Oxidation of non-phenolic substrates. FEBS Lett. 267, 99–102.
Bourbonnais, R., Paice, M.G., Freiermuth, B., Bodie, E., Borneman, S., 1997. Reactivities of various mediators and laccases with kraft pulp and lignin model compounds. Appl. Environ. Microbiol. 63, 4627–32.
Braun, R., Weiland, P., Wellinger, A., 2008. Biogas from energy crop digestion [WWW Document]. IEA Bioenergy Task 37. URL http://biogasmax.co.uk/media/iea_1_biogas_energy_crop__007962900_1434_30032 010.pdf (accessed 11.16.14).
Brodeur, G., Yau, E., Badal, K., Collier, J., Ramachandran, K.B., Ramakrishnan, S., 2011. Chemical and physicochemical pretreatment of lignocellulosic biomass: a review. Enzyme Res. 2011, 787532.
Brosse, N., Dufour, A., Meng, X., 2012. Miscanthus: a fast •growing crop for biofuels
231
and chemicals production. Biofuels, Bioprod. Biorefining 580–598.
Brosse, N., Sannigrahi, P., Ragauskas, A., 2009. Pretreatment of Miscanthus x giganteus using the ethanol organosolv process for ethanol production. Ind. Eng. Chem. Res. 48, 8328–8334.
Brown, D., Shi, J., Li, Y., 2012. Comparison of solid-state to liquid anaerobic digestion of lignocellulosic feedstocks for biogas production. Bioresour. Technol. 124, 379– 86.
Brown, R.C., 2003. Biorenewable Resources: Engineering New Products from Agriculture. Wiley.
Bugg, T.D.H., Ahmad, M., Hardiman, E.M., Rahmanpour, R., 2011a. Pathways for degradation of lignin in bacteria and fungi. Nat. Prod. Rep. 28, 1883–96.
Bugg, T.D.H., Ahmad, M., Hardiman, E.M., Singh, R., 2011b. The emerging role for bacteria in lignin degradation and bio-product formation. Curr. Opin. Biotechnol. 22, 394–400.
Burhenne, L., Messmer, J., Aicher, T., Laborie, M.-P., 2013. The effect of the biomass components lignin, cellulose and hemicellulose on TGA and fixed bed pyrolysis. J. Anal. Appl. Pyrolysis 101, 177–184.
Burnes, T.A., Blanchette, R.A., Farrell, R.L., 2000. Bacterial biodegradation of extractives and patterns of bordered pit membrane attack in pine wood. Appl. Environ. Microbiol. 66, 5201–5205.
Call, H.P., 1994. Process for modifying, breaking down or bleaching ligin, materials containing ligin or like substances. CA2182182 A1.
Calvo-Flores, F.G., Dobado, J. a, 2010. Lignin as renewable raw material. ChemSusChem 3, 1227–35.
Camarero, S., Ibarra, D., Martínez, M.J., Martínez, A.T., 2005. Lignin-derived compounds as efficient laccase mediators for decolorization of different types of recalcitrant dyes. Appl. Environ. Microbiol. 71, 1775–84.
Camarero, S., Martínez, M.J., Martínez, A.T., 2014. Understanding lignin biodegradation for the improved utilization of plant biomass in modern biorefineries. Biofuels, Bioprod. Biorefining 8, 615–625.
Camarero, S., Pardo, I., Cañas, A.I., Molina, P., Record, E., Martínez, A.T., Martínez, M.J., Alcalde, M., 2012. Engineering platforms for directed evolution of Laccase from Pycnoporus cinnabarinus. Appl. Environ. Microbiol. 78, 1370–84.
232
Canam, T., Town, J.R., Tsang, A., McAllister, T.A., Dumonceaux, T.J., 2011. Biological pretreatment with a cellobiose dehydrogenase-deficient strain of Trametes versicolor enhances the biofuel potential of canola straw. Bioresour. Technol. 102, 10020–7.
Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Fierer, N., Pe?a, A.G., Goodrich, J.K., Gordon, J.I., Huttley, G.A., Kelley, S.T., Knights, D., Koenig, J.E., Ley, R.E., Lozupone, C.A., McDonald, D., Muegge, B.D., Pirrung, M., Reeder, J., Sevinsky, J.R., Turnbaugh, P.J., Walters, W.A., Widmann, J., Yatsunenko, T., Zaneveld, J., Knight, R., 2010. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7, 335–336.
Carriquiry, M. a., Du, X., Timilsina, G.R., 2011. Second generation biofuels: Economics and policies. Energy Policy 39, 4222–4234.
Caslin, B., Finnan, J., Easson, L., 2010. Miscanthus Best Practice Guidelines, Agri-Food and Biosciences Institute.
Cha, Y.-L., Yang, J., Seo, S., An, G.H., Moon, Y.-H., You, G.-D., Lee, J.-E., Ahn, J.-W., Lee, K.-B., 2016. Alkaline twin-screw extrusion pretreatment of Miscanthus with recycled black liquor at the pilot scale. Fuel 164, 322–328.
Cha, Y.L., An, G., Park, Y., Yang, J., An, J., 2014. Partial Simultaneous Saccharification and Fermentation at High Solids Loadings of Alkaline-pretreated Miscanthus for Bioethanol Production. BioResources 9, 6899–6913.
Chen, S., Zhang, X., Singh, D., Yu, H., Yang, X., 2010. Biological pretreatment of lignocellulosics: potential, progress and challenges. Biofuels 1, 177–199.
Cheng, Y.-S., Zheng, Y., Yu, C.W., Dooley, T.M., Jenkins, B.M., VanderGheynst, J.S., 2010. Evaluation of High Solids Alkaline Pretreatment of Rice Straw. Appl. Biochem. Biotechnol. 162, 1768–1784.
Cherubini, F., 2010. The biorefinery concept: Using biomass instead of oil for producing energy and chemicals. Energy Convers. Manag. 51, 1412–1421.
Cho, D.H., Lee, Y.J., Um, Y., Sang, B.-I., Kim, Y.H., 2009. Detoxification of model phenolic compounds in lignocellulosic hydrolysates with peroxidase for butanol production from Clostridium beijerinckii. Appl. Microbiol. Biotechnol. 83, 1035– 1043.
Choi, J.-W., Choi, D.-H., Ahn, S.-H., Lee, S.-S., Kim, M.-K., Meier, D., Faix, O., Scott, G.M., 2006. Characterization of trembling aspen wood (Populus tremuloides L.) degraded with the white rot fungus Ceriporiopsis subvermispora and MWLs isolated thereof. Holz als Roh- und Werkst. 64, 415–422.
233
Choi, J.-W., Faix, O., Meier, D., 2001. Characterization of Residual Lignins from Chemical Pulps of Spruce (Picea abies L.) and Beech (Fagus sylvatica L.) by Analytical Pyrolysis–Gas Chromatography/Mass Spectrometry. Holzforschung 55, 185–192.
Christian, D.G., Riche, a. B., Yates, N.E., 2008. Growth, yield and mineral content of Miscanthus×giganteus grown as a biofuel for 14 successive harvests. Ind. Crops Prod. 28, 320–327.
Christian, D.G., Yates, N.E., Riche, a. B., 2005. Establishing Miscanthus sinensis from seed using conventional sowing methods. Ind. Crops Prod. 21, 109–111.
Christopher, L.P., Yao, B., Ji, Y., 2014. Lignin Biodegradation with Laccase-Mediator Systems. Front. Energy Res. 2, 1–13.
Chynoweth, D.P., Owens, J.M., Legrand, R., 2001. Renewable methane from anaerobic digestion of biomass. Renew. Energy 22, 1–8.
Cianchetta, S., Di Maggio, B., Burzi, P.L., Galletti, S., 2014. Evaluation of selected white-rot fungal isolates for improving the sugar yield from wheat straw. Appl. Biochem. Biotechnol. 173, 609–23.
Clemens, J., Trimborn, M., Weiland, P., Amon, B., 2006. Mitigation of greenhouse gas emissions by anaerobic digestion of cattle slurry. Agric. Ecosyst. Environ. 112, 171– 177.
Clifton-Brown, J.C., 2000. Water Use Efficiency and Biomass Partitioning of Three Different Miscanthus Genotypes with Limited and Unlimited Water Supply. Ann. Bot. 86, 191–200.
Clifton-Brown, J.C., Breuer, J., Jones, M.B., 2007. Carbon mitigation by the energy crop, Miscanthus. Glob. Chang. Biol. 13, 2296–2307.
Clifton-Brown, J.C., Chiang, Y., Hodkinson, T.R., 2008. Miscanthus: genetic resources and breeding potential to enhance bioenergy production. In: Genetic Improvement of Bioenergy Crops. Springer, pp. 273–294.
Cohen, R., Suzuki, M.R., Hammel, K.E., 2005. Processive endoglucanase active in crystalline cellulose hydrolysis by the brown rot basidiomycete Gloeophyllum trabeum. Appl. Environ. Microbiol. 71, 2412–7.
Collura, S., Azambre, B., Finqueneisel, G., Zimny, T., Weber, J.-V., 2006. Miscanthus × Giganteus straw and pellets as sustainable fuels. Environ. Chem. Lett. 4, 75–78.
Collura, S., Azambre, B., Weber, J.-V., 2005. Kinetic modelling of the pyrolysis of Miscanthus × Giganteus from the thermogravimetric analysis of its fractionated 234
components. Environ. Chem. Lett. 3, 95–99.
Colpa, D.I., Fraaije, M.W., Van Bloois, E., 2014. DyP-type peroxidases: A promising and versatile class of enzymes. J. Ind. Microbiol. Biotechnol. 41, 1–7.
Couto, N.D., Silva, V.B., Monteiro, E., Rouboa, A., Brito, P., 2017. An experimental and numerical study on the Miscanthus gasification by using a pilot scale gasifier. Renew. Energy 109, 248–261.
Cui, Z., Shi, J., Wan, C., Li, Y., 2012a. Comparison of alkaline- and fungi-assisted wet- storage of corn stover. Bioresour. Technol. 109, 98–104.
Cui, Z., Wan, C., Shi, J., Sykes, R.W., Li, Y., 2012b. Enzymatic digestibility of corn stover fractions in response to fungal pretreatment. Ind. Eng. Chem. Res. 51, 7153– 7159. da Costa Sousa, L., Chundawat, S.P.S., Balan, V., Dale, B.E., 2009. “Cradle-to-grave” assessment of existing lignocellulose pretreatment technologies. Curr. Opin. Biotechnol. 20, 339–47.
Danalatos, N., Archontoulis, S., Mitsios, I., 2007. Potential growth and biomass productivity of Miscanthus×giganteus as affected by plant density and N- fertilization in central Greece. Biomass and Bioenergy 31, 145–152.
Dashtban, M., Schraft, H., Syed, T. a., Qin, W., 2010. Fungal biodegradation and enzymatic modification of lignin. Int. J. Biochem. Mol. Biol. 1, 36–50. de Vrije, T.D., Bakker, R.R., Budde, M.A., Lai, M.H., Mars, A.E., Claassen, P.A., 2009. Efficient hydrogen production from the lignocellulosic energy crop Miscanthus by the extreme thermophilic bacteria Caldicellulosiruptor saccharolyticus and Thermotoga neapolitana. Biotechnol. Biofuels 2, 12. de Vrije, T.D., de Haas, G.G., Tan, G.B., Keijsers, E.R.P., Claassen, P.A.M., 2002. Pretreatment of Miscanthus for hydrogen production by Thermotoga elfii. Int. J. Hydrogen Energy 27, 1381–1390.
Dejong, W., Pirone, a, Wojtowicz, M., 2003. Pyrolysis of Miscanthus Giganteus and wood pellets: TG-FTIR analysis and reaction kinetics. Fuel 82, 1139–1147.
Del Río, J.C., Gutiérrez, A., Martínez, M.J., Martínez, A.T., 2001a. Py–GC/MS study of Eucalyptus globulus wood treated with different fungi. J. Anal. Appl. Pyrolysis 58– 59, 441–452.
Del Río, J.C., Gutiérrez, A., Romero, J., Martínez, M.J., Martínez, A.T., 2001b. Identification of residual lignin markers in eucalypt kraft pulps by Py–GC/MS. J. Anal. Appl. Pyrolysis 58–59, 425–439. 235
Depriest, P.T., Sikaroodi, M., Lawrey, J.D., Diederich, P., 2005. Marchandiomyces lignicola sp. nov. shows recent and repeated transition between a lignicolous and a lichenicolous habit. Mycol. Res. 109, 57–70.
DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Huber, T., Dalevi, D., Hu, P., Andersen, G.L., 2006. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl. Environ. Microbiol. 72, 5069–72.
Dighton, J., White, J.F., 2017. The Fungal Community: Its Organization and Role in the Ecosystem, Fourth Edition, Mycology. CRC Press.
Dohleman, F.G., Heaton, E.A., Long, S.P., 2010. Handbook of Bioenergy Economics and Policy. Springer New York, New York, NY.
Dohleman, F.G., Long, S.P., 2009. More productive than maize in the Midwest: How does Miscanthus do it? Plant Physiol. 150, 2104–15.
Dondini, M., Hastings, A., Saiz, G., Jones, M.B., Smith, P., 2009. The potential of Miscanthus to sequester carbon in soils: Comparing field measurements in Carlow, Ireland to model predictions. GCB Bioenergy 1, 413–425.
Doyle, W.A., Blodig, W., Veitch, N.C., Piontek, K., Smith, A.T., 1998. Two substrate interaction sites in lignin peroxidase revealed by site-directed mutagenesis. Biochemistry 37, 15097–105.
Durand, A., Maillard, F., Foulon, J., Gweon, H.S., Valot, B., Chalot, M., 2017. Environmental Metabarcoding Reveals Contrasting Belowground and Aboveground Fungal Communities from Poplar at a Hg Phytomanagement Site. Microb. Ecol. 1– 15.
Dürre, P., 2011. Fermentative production of butanol-the academic perspective. Curr. Opin. Biotechnol. 22, 331–336.
Edgar, R.C., Haas, B.J., Clemente, J.C., Quince, C., Knight, R., 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194–2200.
Eggeman, T., Elander, R.T., 2005. Process and economic analysis of pretreatment technologies. Bioresour. Technol. 96, 2019–25.
Eggert, C., Temp, U., Eriksson, K.-E.L., 1997. Laccase is essential for lignin degradation by the white-rot fungus Pycnoporus cinnabarinus. FEBS Lett. 407, 89–92.
El Hage, R., Chrusciel, L., Desharnais, L., Brosse, N., 2010. Effect of autohydrolysis of Miscanthus x giganteus on lignin structure and organosolv delignification. Bioresour. Technol. 101, 9321–9. 236
Ercoli, L., Mariotti, M., Masoni, A., Bonari, E., 1999. Effect of irrigation and nitrogen fertilization on biomass yield and efficiency of energy use in crop production of Miscanthus. F. Crop. Res. 63.
Eriksson, K.-E.L., Blanchette, R.A., Ander, P., 1990. Microbial and enzymatic degradation of wood and wood components. Springer-Verlag.
Ezeji, T.C., Blaschek, H.P., 2010. Butanol production from lignocellulosic biomass. In: Blaschek HP, Ezeji TC, Scheffran, J. (Ed.), Biofuels from Agricultural Wastes and Byproducts. Wiley-Blackwell. Ames, Iowa, USA, pp. 19–37.
Ezeji, T.C., Milne, C., Price, N.D., Blaschek, H.P., 2010. Achievements and perspectives to overcome the poor solvent resistance in acetone and butanol-producing microorganisms. Appl. Microbiol. Biotechnol. 85, 1697–712.
Ezeji, T.C., Qureshi, N., Blaschek, H.P., 2005. Process for continuous solvent production. US 20050089979.
Ezeji, T.C., Qureshi, N., Blaschek, H.P., 2007a. Bioproduction of butanol from biomass: from genes to bioreactors. Curr. Opin. Biotechnol. 18, 220–7.
Ezeji, T.C., Qureshi, N., Blaschek, H.P., 2007b. Butanol production from agricultural residues: Impact of degradation products onClostridium beijerinckii growth and butanol fermentation. Biotechnol. Bioeng. 97, 1460–1469.
Faik, A., 2004. “Plant Cell Wall Structure-Pretreatment” the Critical Relationship in Biomass. In: Gu, T. (Ed.), Green Biomass Pretreatment for Biofuels. Springer, New York, NY.
Fan, X.L., Barreto, R.W., Groenewald, J.Z., Bezerra, J.D.P., Pereira, O.L., Cheewangkoon, R., Mostert, L., Tian, C.M., Crous, P.W., 2017. Phylogeny and taxonomy of the scab and spot anthracnose fungus Elsinoë (Myriangiales, Dothideomycetes). Stud. Mycol. 87, 1–41.
Feng, Q., Chaubey, I., Engel, B., Cibin, R., Sudheer, K.P., Volenec, J., 2017. Marginal land suitability for switchgrass, Miscanthus and hybrid poplar in the Upper Mississippi River Basin (UMRB). Environ. Model. Softw. 93, 356–365.
Fernández-Fueyo, E., Linde, D., Almendral, D., López-Lucendo, M.F., Ruiz-Dueñas, F.J., Martínez, A.T., 2015. Description of the first fungal dye-decolorizing peroxidase oxidizing manganese(II). Appl. Microbiol. Biotechnol.
Fernández-Fueyo, E., Ruiz-Dueñas, F.J., Ferreira, P., Floudas, D., Hibbett, D.S., Canessa, P., Larrondo, L.F., James, T.Y., Seelenfreund, D., Lobos, S., Polanco, R., Tello, M., Honda, Y., Watanabe, T., Watanabe, T., Ryu, J.S., San, R.J., Kubicek, C.P.,
237
Schmoll, M., Gaskell, J., Hammel, K.E., St John, F.J., Vanden Wymelenberg, A., Sabat, G., Splinter BonDurant, S., Syed, K., Yadav, J.S., Doddapaneni, H., Subramanian, V., Lavín, J.L., Oguiza, J.A., Perez, G., Pisabarro, A.G., Ramirez, L., Santoyo, F., Master, E., Coutinho, P.M., Henrissat, B., Lombard, V., Magnuson, J.K., Kües, U., Hori, C., Igarashi, K., Samejima, M., Held, B.W., Barry, K.W., LaButti, K.M., Lapidus, A., Lindquist, E.A., Lucas, S.M., Riley, R., Salamov, A.A., Hoffmeister, D., Schwenk, D., Hadar, Y., Yarden, O., de Vries, R.P., Wiebenga, A., Stenlid, J., Eastwood, D.C., Grigoriev, I. V, Berka, R.M., Blanchette, R.A., Kersten, P., Martínez, A.T., Vicuña, R., Cullen, D., 2012a. Comparative genomics of Ceriporiopsis subvermispora and Phanerochaete chrysosporium provide insight into selective ligninolysis. Proc. Natl. Acad. Sci. U. S. A. 109, 5458–63.
Fernández-Fueyo, E., Ruiz-Dueñas, F.J., Martínez, A.T., 2014. Engineering a fungal peroxidase that degrades lignin at very acidic pH. Biotechnol. Biofuels 7, 114.
Fernández-Fueyo, E., Ruiz-Dueñas, F.J., Miki, Y., Martínez, M.J., Hammel, K.E., Martínez, A.T., 2012b. Lignin-degrading Peroxidases from Genome of Selective Ligninolytic Fungus Ceriporiopsis subvermispora. J. Biol. Chem. 287, 16903– 16916.
Ferraz, A., Guerra, A., Mendonça, R., Masarin, F., Vicentim, M.P., Aguiar, A., Pavan, P.C., 2008. Technological advances and mechanistic basis for fungal biopulping. Enzyme Microb. Technol. 43, 178–185.
Fischer, K., Akhtar, M., Messner, K., Blanchette, R.A., Kirk, T.K., 1996. Pitch reduction with the white-rot fungus Ceriporiopsis subvermispora. In: Proceedings of the 6th International Conference on Biotechnology in the Pulp and Paper Industry: Advances in Applied and Fundamental Research. pp. 193–198.
Fisher, A.B., Fong, S.S., 2014. Lignin biodegradation and industrial implications. AIMS Environ. Sci. 1, 92–112.
FitzPatrick, M., Champagne, P., Cunningham, M.F., Whitney, R.A., 2010. A biorefinery processing perspective: treatment of lignocellulosic materials for the production of value-added products. Bioresour. Technol. 101, 8915–22.
Floudas, D., Binder, M., Riley, R., Barry, K.W., Blanchette, R.A., Henrissat, B., Martínez, A.T., Otillar, R., Spatafora, J.W., Yadav, J.S., Aerts, A., Benoit, I., Boyd, A., Carlson, A., Copeland, A., Coutinho, P.M., de Vries, R.P., Ferreira, P., Findley, K., Foster, B., Gaskell, J., Glotzer, D., Górecki, P., Heitman, J., Hesse, C., Hori, C., Igarashi, K., Jurgens, J.A., Kallen, N., Kersten, P., Kohler, A., Kües, U., Kumar, T.K.A., Kuo, A., LaButti, K.M., Larrondo, L.F., Lindquist, E.A., Ling, A., Lombard, V., Lucas, S.M., Lundell, T.K., Martin, R., McLaughlin, D.J., Morgenstern, I., Morin, E., Murat, C., Nagy, L.G., Nolan, M., Ohm, R.A., Patyshakuliyeva, A., Rokas, A., Ruiz-Dueñas, F.J., Sabat, G., Salamov, A.A., Samejima, M., Schmutz, J., 238
Slot, J.C., St John, F.J., Stenlid, J., Sun, H., Sun, S., Syed, K., Tsang, A., Wiebenga, A., Young, D., Pisabarro, A.G., Eastwood, D.C., Martin, F., Cullen, D., Grigoriev, I. V, Hibbett, D.S., 2012. The Paleozoic origin of enzymatic lignin decomposition reconstructed from 31 fungal genomes. Science 336, 1715–9.
Frydendal-Nielsen, S., Hjorth, M., Baby, S., Felby, C., Jørgensen, U., Gislum, R., 2016. The effect of harvest time, dry matter content and mechanical pretreatments on anaerobic digestion and enzymatic hydrolysis of miscanthus. Bioresour. Technol. 218, 1008–1015.
Furukawa, T., Bello, F.O., Horsfall, L., 2014. Microbial enzyme systems for lignin degradation and their transcriptional regulation. Front. Biol. (Beijing). 9, 448–471.
Gao, K., Rehmann, L., 2014. ABE fermentation from enzymatic hydrolysate of NaOH- pretreated corncobs. Biomass and Bioenergy 66, 110–115.
Gao, Z., Mori, T., Kondo, R., 2012. The pretreatment of corn stover with Gloeophyllum trabeum KU-41 for enzymatic hydrolysis. Biotechnol. Biofuels 5, 28.
García-Ruiz, E., Gonzalez-Perez, D., Ruiz-Dueñas, F.J., Martínez, A.T., Alcalde, M., 2012. Directed evolution of a temperature-, peroxide- and alkaline pH-tolerant versatile peroxidase. Biochem. J. 441, 487–98.
Ge, X., Burner, D.M., Xu, J., Phillips, G.C., Sivakumar, G., 2011. Bioethanol production from dedicated energy crops and residues in Arkansas, USA. Biotechnol. J. 6, 66– 73.
Ge, X., Matsumoto, T., Keith, L., Li, Y., 2015. Fungal Pretreatment of Albizia Chips for Enhanced Biogas Production by Solid-State Anaerobic Digestion. Energy & Fuels 29, 200–204.
Ge, X., Xu, F., Vasco-Correa, J., Li, Y., 2016. Giant reed: A competitive energy crop in comparison with miscanthus. Renew. Sustain. Energy Rev. 54, 350–362.
Gerin, P.A., Vliegen, F., Jossart, J.-M., 2008. Energy and CO2 balance of maize and grass as energy crops for anaerobic digestion. Bioresour. Technol. 99, 2620–7.
Gilbertson, R.L., Ryvarden, L., 1985. Some new combinations in the Polyporaceae. Mycotaxon.
Giles, R.L., Zackeru, J.C., Elliott, G.D., Parrow, M.W., 2012. Fungal growth necessary but not sufficient for effective biopulping of wood for lignocellulosic ethanol applications. Int. Biodeterior. Biodegrad. 67, 1–7.
Glenn, J.K., Morgan, M.A., Mayfield, M.B., Kuwahara, M., Gold, M.H., 1983. An extracellular H2O2-requiring enzyme preparation involved in lignin biodegradation 239
by the white rot basidiomycete Phanerochaete chrysosporium. Biochem. Biophys. Res. Commun. 114, 1077–1083.
Gnansounou, E., Dauriat, A., 2010. Techno-economic analysis of lignocellulosic ethanol: A review. Bioresour. Technol. 101, 4980–91.
Green, E.M., 2011. Fermentative production of butanol—the industrial perspective. Curr. Opin. Biotechnol. 22, 337–343.
Grønli, M.G., Várhegyi, G., Di Blasi, C., 2002. Thermogravimetric Analysis and Devolatilization Kinetics of Wood. Ind. Eng. Chem. Res. 41, 4201–4208.
Gu, T., 2013. Green Biomass Pretreatment for Biofuels.
Guillén, F., Martínez, M.J., Muñoz, C., Martínez, A.T., 1997. Quinone redox cycling in the ligninolytic fungus Pleurotus eryngii leading to extracellular production of superoxide anion radical. Arch. Biochem. Biophys. 339, 190–9.
Guo, T., Sun, B., Jiang, M., Wu, H., Du, T., Tang, Y., Wei, P., Ouyang, P., 2012. Enhancement of butanol production and reducing power using a two-stage controlled-pH strategy in batch culture of Clostridium acetobutylicum XY16. World J. Microbiol. Biotechnol. 28, 2551–2558.
Hage, R. El, Brosse, N., Navarrete, P., Pizzi, A., 2011. Extraction, Characterization and Utilization of Organosolv Miscanthus Lignin for the Conception of Environmentally Friendly Mixed Tannin/Lignin Wood Resins. J. Adhes. Sci. Technol. 25, 1549– 1560.
Haider, K., Trojanowski, J., 1975. Decomposition of specifically 14C-labelled phenols and dehydropolymers of coniferyl alcohol as models for lignin degradation by soft and white rot fungi. Arch. Microbiol. 105, 33–41.
Hamelinck, C.N., Hooijdonk, G. van, Faaij, A.P.C., 2005. Ethanol from lignocellulosic biomass: techno-economic performance in short-, middle- and long-term. Biomass and Bioenergy 28, 384–410.
Han, M., Choi, G.W., Kim, Y., Koo, B., 2011. Bioethanol production by Miscanthus as a lignocellulosic biomass: Focus on high efficiency conversion to glucose and ethanol. BioResources 6, 1939–1953.
Harman-Ware, A.E., Crocker, M., Kaur, A.P., Meier, M.S., Kato, D., Lynn, B., 2013. Pyrolysis–GC/MS of sinapyl and coniferyl alcohol. J. Anal. Appl. Pyrolysis 99, 161–169.
Hatakka, A.I., 1983. Pretreatment of wheat straw by white-rot fungi for enzymic saccharification of cellulose. Eur. J. Appl. Microbiol. Biotechnol. 18, 350–357. 240
Hatakka, A.I., 2005. Biodegradation of Lignin. In: Biopolymers Online. Wiley-VCH Verlag GmbH & Co. KGaA, pp. 129–145.
Haugland, R.A., Vesper, S.J., Harmon, S.M., 2001. Phylogenetic Relationships of Memnoniella and Stachybotrys Species and Evaluation of Morphological Features for Memnoniella Species Identification. Mycologia 93, 54.
Hayes, D.J.M., 2013. Mass and compositional changes, relevant to biorefining, in Miscanthus x giganteus plants over the harvest window. Bioresour. Technol. 142, 591–602.
Heaton, E.A., Dohleman, F.G., Juvik, J.A., Lozovaya, V., Miguez, A.F., Juvik, J.A., Lozovaya, V., Widholm, J., Zabotina, O.A., McIsaac, G.F., David, M.B., Voigt, T.B., Boersma, N.N., Long, S.P., 2010. Miscanthus. A Promising Biomass Crop. Adv. Bot. Res. 56, 76–137.
Heaton, E.A., Dohleman, F.G., Long, S.P., 2008a. Meeting US biofuel goals with less land: the potential of Miscanthus. Glob. Chang. Biol. 14, 2000–2014.
Heaton, E.A., Flavell, R.B., Mascia, P.N., Thomas, S.R., Dohleman, F.G., Long, S.P., 2008b. Herbaceous energy crop development: recent progress and future prospects. Curr. Opin. Biotechnol. 19, 202–9.
Heaton, E.A., Long, S.P., Voigt, T.B., Jones, M.B., Clifton-Brown, J.C., 2004. Miscanthus for renewable energy generation: European Union experience and projections for Illinois. Mitig. Adapt. Strateg. Glob. Chang. 9, 433–451.
Hendriks, A.T.W.M., Zeeman, G., 2009. Pretreatments to enhance the digestibility of lignocellulosic biomass. Bioresour. Technol. 100, 10–18.
Hertz, M., Jensen, I.R., Jensen, L. ?stergaard, Thomsen, S.N., Winde, J., Dueholm, M.S., S?rensen, L.H., Wollenberg, R.D., S?rensen, H.O., Sondergaard, T.E., S?rensen, J.L., 2016. The fungal community changes over time in developing wheat heads. Int. J. Food Microbiol. 222, 30–39.
Hideno, A., Kawashima, A., Anzoua, K.G., Yamada, T., 2013. Comparison of the enzymatic digestibility of physically and chemically pretreated selected line of diploid-Miscanthus sinensis Shiozuka and triploid-M.×giganteus. Bioresour. Technol. 146, 393–399.
Himejima, M., Hobson, K.R., Otsuka, T., Wood, D.L., Kubo, I., 1992. Antimicrobial terpenes from oleoresin of ponderosa pine treePinus ponderosa: A defense mechanism against microbial invasion. J. Chem. Ecol. 18, 1809–1818.
Himken, M., Lammel, J., Neukirchen, D., Czypionka-Krause, U., Olfs, H.-W., 1997.
241
Cultivation of Miscanthus under West European conditions: Seasonal changes in dry matter production, nutrient uptake and remobilization. Plant Soil 189, 117–126.
Hodgson, E.M., Nowakowski, D.J., Shield, I., Riche, A.B., Bridgwater, a V, Clifton- Brown, J.C., Donnison, I.S., 2011. Variation in Miscanthus chemical composition and implications for conversion by pyrolysis and thermo-chemical bio-refining for fuels and chemicals. Bioresour. Technol. 102, 3411–8.
Huang, C.L., Liao, W.C., Lai, Y.C., 2011. Cultivation studies of Taiwanese native Miscanthus floridulus lines. Biomass and Bioenergy 35, 1873–1877.
Humbird, D., Davis, R., Tao, L., Kinchin, C., Hsu, D., Aden, A., Schoen, P., Lukas, J., Olthof, B., Worley, M., Sexton, D., Dudgeon, D., 2011. Process Design and Economics for Biochemical Conversion of Lignocellulosic Biomass to Ethanol, National Renewable Energy Laboratory-NREL.
Intelligent, 2014. SuperPro Designer - User’s Guide.
ISO, 1984. ISO 5664:1984 Water quality- Determination of ammonium- Distillation and titration method.
Jain, A.K., Khanna, M., Erickson, M., Huang, H., 2010. An integrated biogeochemical and economic analysis of bioenergy crops in the Midwestern United States. GCB Bioenergy 2, 217–234.
Ji, L., Yang, J., Fan, H., Yang, Y., Li, B., Yu, X., Zhu, N., Yuan, H., 2014. Synergy of crude enzyme cocktail from cold-adapted Cladosporium cladosporioides Ch2-2 with commercial xylanase achieving high sugars yield at low cost. Biotechnol. Biofuels 7, 130.
Jiang, D., Ge, X., Zhang, Q., Li, Y., 2016. Comparison of liquid hot water and alkaline pretreatments of giant reed for improved enzymatic digestibility and biogas energy production. Bioresour. Technol. 216, 60–68.
Jiang, M., Chen, J., He, A., Wu, H., Kong, X., Liu, J., Yin, C., Chen, W., Chen, P., 2014. Enhanced acetone/butanol/ethanol production by Clostridium beijerinckii IB4 using pH control strategy. Process Biochem. 49, 1238–1244.
Johnson, M., Tucker, N., Barnes, S., Kirwan, K., 2005. Improvement of the impact performance of a starch based biopolymer via the incorporation of Miscanthus giganteus fibres. Ind. Crops Prod. 22, 175–186.
Jones, M.B., Walsh, M., 2001. Miscanthus: For Energy And Fibre. Earthscan LLC.
Jönsson, L.J., Martín, C., 2016. Pretreatment of lignocellulose: Formation of inhibitory by-products and strategies for minimizing their effects. Bioresour. Technol. 199, 242
103–112.
Jørgensen, U., 1997. Genotypic variation in dry matter accumulation and content of N, K and Cl in Miscanthus in Denmark. Biomass and Bioenergy 12, 155–169.
Jurado, E., Gavala, H.N., Skiadas, I. V, 2013. Enhancement of methane yield from wheat straw, miscanthus and willow using aqueous ammonia soaking. Environ. Technol. 34, 2069–2075.
Kang, K.E., Park, D.H., Jeong, G.T., 2013a. Effects of NH4Cl and MgCl2 on pretreatment and xylan hydrolysis of miscanthus straw. Carbohydr. Polym. 92, 1321–1326.
Kang, K.E., Park, D.H., Jeong, G.T., 2013b. Effects of inorganic salts on pretreatment of Miscanthus straw. Bioresour. Technol. 132, 160–165.
Kaparaju, P., Serrano, M., Thomsen, A.B., Kongjan, P., Angelidaki, I., 2009. Bioethanol, biohydrogen and biogas production from wheat straw in a biorefinery concept. Bioresour. Technol. 100, 2562–8.
Kärcher, M.A., Iqbal, Y., Lewandowski, I., Senn, T., 2015. Comparing the performance of Miscanthus x giganteus and wheat straw biomass in sulfuric acid based pretreatment, Bioresource Technology.
Karimi, K., 2015. Lignocellulose-Based Bioproducts. Volume 1 of Biofuel and Biorefinery Technologies, Biofuel and Biorefinery Technologies. Springer International Publishing.
Kazi, F.K., Fortman, J.A., Anex, R., Hsu, D.D., Aden, A., Dutta, A., Kothandaraman, G., 2010a. Techno-economic comparison of process technologies for biochemical ethanol production from corn stover. Fuel 89, S20–S28.
Kazi, F.K., Fortman, J.A., Anex, R., Kothandaraman, G., 2010b. Techno-economic analysis of biochemical scenarios for production of cellulosic ethanol. Golden, Colorado.
KC, S., Takara, D., Hashimoto, A.G., Khanal, S.K., 2014. Biogas as a sustainable energy source for developing countries: Opportunities and challenges. Renew. Sustain. Energy Rev. 31, 846–859.
Keller, F.A., Hamilton, J.E., Nguyen, Q.A., 2003. Microbial pretreatment of biomass. Appl. Biochem. Biotechnol. 105, 27–41.
Khanna, M., Dhungana, B., Clifton-Brown, J.C., 2008. Costs of producing miscanthus and switchgrass for bioenergy in Illinois. Biomass and Bioenergy 32, 482–493.
243
Khelfa, A., Sharypov, V., Finqueneisel, G., Weber, J.-V., 2009. Catalytic pyrolysis and gasification of Miscanthus Giganteus: Haematite (Fe2O3) a versatile catalyst. J. Anal. Appl. Pyrolysis 84, 84–88.
Khullar, E., Dien, B.S., Rausch, K.D., Tumbleson, M.E., Singh, V., 2013. Effect of particle size on enzymatic hydrolysis of pretreated Miscanthus. Ind. Crops Prod. 44, 11–17.
Kim, S.J., Shoda, M., 1999. Purification and Characterization of a Novel Peroxidase from Geotrichum candidum Dec 1 Involved in Decolorization of Dyes. Appl. Envir. Microbiol. 65, 1029–1035.
Kim, T.H., 2013. Pretreatment of Lignocellulosic Biomass. In: Yang, S.-T.; El-Enshasy, H. A.; Thongchul, N. (Ed.), Bioprocessing Technologies in Biorefinery for Sustainable Production of Fuels, Chemicals, and Polymers. John Wiley & Sons, Inc., Hoboken, NJ, pp. 91–110.
Kinne, M., Poraj-Kobielska, M., Ullrich, R., Nousiainen, P., Sipilä, J., Scheibner, K., Hammel, K.E., Hofrichter, M., 2011. Oxidative cleavage of non-phenolic β-O-4 lignin model dimers by an extracellular aromatic peroxygenase. Holzforschung 65.
Kirk, T.K., Farrell, R.L., 1987. Enzymatic “combustion”: the microbial degradation of lignin. Annu. Rev. Microbiol. 41, 465–505.
Kleen, M., Gellerstedt, G., 1995. Influence of inorganic species on the formation of polysaccharide and lignin degradation products in the analytical pyrolysis of pulps. J. Anal. Appl. Pyrolysis 35, 15–41.
Klimiuk, E., Pokój, T., Budzyński, W., Dubis, B., 2010. Theoretical and observed biogas production from plant biomass of different fibre contents. Bioresour. Technol. 101, 9527–9535.
Kojima, Y., 2015. Formation of Polycyclic Compounds from Phenols by Fast Pyrolysis. EC Agric. 1, 67–85.
Kõljalg, U., Larsson, K.-H., Abarenkov, K., Nilsson, R.H., Alexander, I.J., Eberhardt, U., Erland, S., Høiland, K., Kjøller, R., Larsson, E., Pennanen, T., Sen, R., Taylor, A.F.S., Tedersoo, L., Vrålstad, T., 2005. UNITE: a database providing web-based methods for the molecular identification of ectomycorrhizal fungi. New Phytol. 166, 1063–1068.
Korres, N., O’Kiely, P., Benzie, J.A.H., West, J.S., 2013. Bioenergy production by anaerobic digestion: using agricultural biomass and organic wastes. Routledge.
Kumar, D., Murthy, G.S., 2011a. Impact of pretreatment and downstream processing
244
technologies on economics and energy in cellulosic ethanol production. Biotechnol. Biofuels 4, 27.
Kumar, D., Murthy, G.S., 2011b. Impact of pretreatment and downstream processing technologies on economics and energy in cellulosic ethanol production. Biotechnol. Biofuels 4, 27.
Kumar, D., Murthy, G.S., 2012. Life cycle assessment of energy and GHG emissions during ethanol production from grass straws using various pretreatment processes. Int. J. Life Cycle Assess. 17, 388–401.
Kumar, M., Gayen, K., 2011. Developments in biobutanol production: New insights. Appl. Energy 88, 1999–2012.
Kumar, P., Barrett, D.M., Delwiche, M.J., Stroeve, P., 2009. Methods for Pretreatment of Lignocellulosic Biomass for Efficient Hydrolysis and Biofuel Production. Ind. Eng. Chem. Res. 48, 3713–3729.
Kumar, R., Singh, S., Singh, O., 2008. Bioconversion of lignocellulosic biomass: biochemical and molecular perspectives. J. Ind. Microbiol. Biotechnol. 35, 377–391.
Kuwahara, M., Glenn, J.K., Morgan, M.A., Gold, M.H., 1984. Separation and characterization of two extracelluar H2O2-dependent oxidases from ligninolytic cultures of Phanerochaete chrysosporium. FEBS Lett. 169, 247–250.
Larsson, S., Palmqvist, E., Hahn-H?gerdal, B., Tengborg, C., Stenberg, K., Zacchi, G., Nilvebrant, N.-O., 1999. The generation of fermentation inhibitors during dilute acid hydrolysis of softwood. Enzyme Microb. Technol. 24, 151–159.
Le Ngoc Huyen, T., Rémond, C., Dheilly, R.M., Chabbert, B., 2010. Effect of harvesting date on the composition and saccharification of Miscanthus x giganteus. Bioresour. Technol. 101, 8224–31.
LeBauer, D., Kooper, R., Mulrooney, P., Rohde, S., Wang, D., Long, S.P., Dietze, M.C., 2017. betydb: a yield, trait, and ecosystem service database applied to second- generation bioenergy feedstock production. GCB Bioenergy.
Lee, S.Y., Park, J.H., Jang, S.H., Nielsen, L.K., Kim, J., Jung, K.S., 2008. Fermentative butanol production by Clostridia. Biotechnol. Bioeng. 101, 209–28.
Lehtomäki, A., Viinikainen, T. a., Rintala, J., 2008. Screening boreal energy crops and crop residues for methane biofuel production. Biomass and Bioenergy 32, 541–550.
Lenz, T.G., Morelra, A.R., 1980. Economic Evaluation of the Acetone-Butanol Fermentation. Ind. Eng. Chem. Prod. Res. Dev. 19, 478–483.
245
Leonowicz, A., Cho, N.S., Luterek, J., Wilkolazka, A., Wojtaś-Wasilewska, M., Matuszewska, A., Hofrichter, M., Wesenberg, D., Rogalski, J., 2001. Fungal laccase: properties and activity on lignin. J. Basic Microbiol. 41, 185–227.
Leonowicz, A., Matuszewska, a, Luterek, J., Ziegenhagen, D., Wojtaś-Wasilewska, M., Cho, N.S., Hofrichter, M., Rogalski, J., 1999. Biodegradation of lignin by white rot fungi. Fungal Genet. Biol. 27, 175–85.
Lewandowski, I.M., Clifton-Brown, J.C., 2000. Miscanthus: European experience with a novel energy crop. Biomass and Bioenergynergy 19, 209–227.
Lewandowski, I.M., Heinz, A., 2003. Delayed harvest of miscanthus—influences on biomass quantity and quality and environmental impacts of energy production. Eur. J. Agron. 19, 45–63.
Lewandowski, I.M., Kicherer, A., 1997. Combustion quality of biomass: practical relevance and experiments to modify the biomass quality of Miscanthus x giganteus. Eur. J. Agron. 6.
Lewandowski, I.M., Kicherer, A., Vonier, P., 1995. CO 2-balance for the cultivation and combustion of Miscanthus. Biomass and Bioenergy 8, 81–90.
Lewandowski, I.M., Scurlock, J.M.O., Lindvall, E., Christou, M., 2003. The development and current status of perennial rhizomatous grasses as energy crops in the US and Europe. Biomass and Bioenergy 25, 335–361.
Li, C., Liu, G., Nges, I.A., Liu, J., 2016. Enhanced biomethane production from Miscanthus lutarioriparius using steam explosion pretreatment. Fuel 179, 267–273.
Li, J., Shi, S., Adhikari, S., Tu, M., 2017. Inhibition effect of aromatic aldehydes on butanol fermentation by Clostridium acetobutylicum. RSC Adv. 7, 1241–1250.
Li, Y.-F., Chen, P.-H., Yu, Z., 2014. Spatial and temporal variations of microbial community in a mixed plug-flow loop reactor fed with dairy manure. Microb. Biotechnol. 7, 332–346.
Li, Y., Park, S.Y., Zhu, J., 2011. Solid-state anaerobic digestion for methane production from organic waste. Renew. Sustain. Energy Rev. 15, 821–826.
Li, Y., Zhu, J., Wan, C., 2012. Combined liquid to solid-phase anaerobic digestion for biogas production from municipal and agricultural wastes. US Pat. 20,120,231,494. US 8,771,980 B2.
Liers, C., Bobeth, C., Pecyna, M., Ullrich, R., Hofrichter, M., 2010. DyP-like peroxidases of the jelly fungus Auricularia auricula-judae oxidize nonphenolic lignin model compounds and high-redox potential dyes. Appl. Microbiol. Biotechnol. 85, 1869– 246
79.
Liew, L.N., Shi, J., Li, Y., 2012. Methane production from solid-state anaerobic digestion of lignocellulosic biomass. Biomass and Bioenergy 46, 125–132.
Linde Laursen, I., 1993. Cytogenetic Analysis of Miscanthus“Giganteus”, an Interspecific Hybrid. Hereditas 119, 297–300.
Linnard, W., Schwarze, F.W.M.R., Engels, J., Mattheck, C., 2013. Fungal Strategies of Wood Decay in Trees. Springer Berlin Heidelberg.
Liu, J., Wang, M.L., Tonnis, B., Habteselassie, M., Liao, X., Huang, Q., 2013. Fungal pretreatment of switchgrass for improved saccharification and simultaneous enzyme production. Bioresour. Technol. 135, 39–45.
Liu, S., Xu, F., Ge, X., Li, Y., 2016. Comparison between ensilage and fungal pretreatment for storage of giant reed and subsequent methane production. Bioresour. Technol. 209, 246–253.
Liu, W., Yan, J., Li, J., Sang, T., 2012. Yield potential of Miscanthus energy crops in the Loess Plateau of China. GCB Bioenergy 4, 545–554.
Liu, Z., Padmanabhan, S., Cheng, K., Schwyter, P., Pauly, M., Bell, A.T., Prausnitz, J.M., 2013. Aqueous-ammonia delignification of miscanthus followed by enzymatic hydrolysis to sugars. Bioresour. Technol. 135, 23–29.
Liu, Z.L., Blaschek, H.P., 2010. Biomass conversion inhibitors and in situ detoxification. Biomass to biofuels Strateg. Glob. Ind. 233–259.
López-Abelairas, M., Álvarez Pallín, M., Salvachúa, D., Lú-Chau, T., Martínez, M.J., Lema, J.M., 2013. Optimisation of the biological pretreatment of wheat straw with white-rot fungi for ethanol production. Bioprocess Biosyst. Eng. 36, 1251–1260.
Lou, H., Hu, Q., Qiu, X., Li, X., Lin, X., 2016. Pretreatment of Miscanthus by NaOH/Urea Solution at Room Temperature for Enhancing Enzymatic Hydrolysis. BioEnergy Res. 9, 335–343.
Lynd, L.R., Wyman, C., Laser, M., College, D., 2005. Strategic Biorefinery Analysis: Analysis of Biorefineries.
Ma, F., Zeng, Y., Wang, J., Yang, Y., Yang, X., Zhang, X., 2013. Thermogravimetric study and kinetic analysis of fungal pretreated corn stover using the distributed activation energy model. Bioresour. Technol. 128, 417–422.
Maddox, I.S., Steiner, E., Hirsch, S., Wessner, S., Gutierrez, N.A., Gapes, J.R., Schuster, K.C., 2000. The Cause of " Acid Crash " and " Acidogenic 247
Fermentations " During the Batch Acetone-Butanol- Ethanol (ABE-) Fermentation Process Fermentation Symposium JMMB Research Article. J. Mol. Microbiol. Biotechnol 2, 95–100.
Manzoni, S., Schimel, J.P., Porporato, A., 2012. Responses of soil microbial communities to water stress: results from a meta-analysis. Ecology 93, 930–938.
Mao, C., Feng, Y., Wang, X., Ren, G., 2015. Review on research achievements of biogas from anaerobic digestion. Renew. Sustain. Energy Rev. 45, 540–555.
Mariano, A.P., Dias, M.O.S., Junqueira, T.L., Cunha, M.P., Bonomi, A., Filho, R.M., 2013. Utilization of pentoses from sugarcane biomass: techno-economics of biogas vs. butanol production. Bioresour. Technol. 142, 390–9.
Marlatt, J. a, Datta, R., 1986. Acetone-butanol fermentation process development and economic evaluation. Biotechnol. Prog. 2, 23–8.
Martínez-Inigo, M.J., Immerzeel, P., Gutierrez, A., Río, J.C. del, Sierra-Alvarez, R., 1999. Biodegradability of Extractives in Sapwood and Heartwood from Scots Pine by Sapstain and White-Rot Fungi. Holzforschung 53, 247–252.
Martínez, A.T., Ruiz-Dueñas, F.J., Gutiérrez, A., Del Río, J.C., Alcalde, M., Liers, C., Ullrich, R., Hofrichter, M., Scheibner, K., Kalum, L., Vind, J., Lund, H., 2014. Search, engineering, and applications of new oxidative biocatalysts. Biofuels, Bioprod. Biorefining 8, 819–835.
Martínez, A.T., Ruiz-Dueñas, F.J., Martínez, M.J., Del Río, J.C., Gutiérrez, A., 2009. Enzymatic delignification of plant cell wall: from nature to mill. Curr. Opin. Biotechnol. 20, 348–57.
Martínez, A.T., Speranza, M., Ruiz-Dueñas, F.J., Ferreira, P., Camarero, S., Guillén, F., Martínez, M.J., Gutiérrez, A., del Río, J.C., 2005. Biodegradation of lignocellulosics: microbial, chemical, and enzymatic aspects of the fungal attack of lignin. Int. Microbiol. 8, 195–204.
Martínez, M.J., Ruiz-Dueñas, F.J., Guillén, F., Martínez, A.T., 1996. Purification and catalytic properties of two manganese peroxidase isoenzymes from Pleurotus eryngii. Eur. J. Biochem. 237, 424–32.
Masai, E., Katayama, Y., Fukuda, M., 2007. Genetic and biochemical investigations on bacterial catabolic pathways for lignin-derived aromatic compounds. Biosci. Biotechnol. Biochem. 71, 1–15.
Maté, D., García-Burgos, C., García-Ruiz, E., Ballesteros, A.O., Camarero, S., Alcalde, M., 2010. Laboratory evolution of high-redox potential laccases. Chem. Biol. 17,
248
1030–41.
Mayer, F., Gerin, P.A., Noo, A., Lemaigre, S., Stilmant, D., Schmit, T., Leclech, N., Ruelle, L., Gennen, J., von Francken-Welz, H., Foucart, G., Flammang, J., Weyland, M., Delfosse, P., 2014. Assessment of energy crops alternative to maize for biogas production in the Greater Region. Bioresour. Technol. 166, 358–367.
McKendry, P., 2002. Energy production from biomass (part 1): overview of biomass. Bioresour. Technol. 83, 37–46.
McMillan, J.D., 1994. Pretreatment of lignocellulosic biomass. In: ACS Symposium Series. ACS Publications, pp. 292–324.
Meiser, A., B?lint, M., Schmitt, I., 2014. Meta-analysis of deep-sequenced fungal communities indicates limited taxon sharing between studies and the presence of biogeographic patterns. New Phytol. 201, 623–635.
Menardo, S., Bauer, A., Theuretzbacher, F., Piringer, G., Nilsen, P.J., Balsari, P., Pavliska, O., Amon, T., 2013. Biogas Production from Steam-Exploded Miscanthus and Utilization of Biogas Energy and CO2 in Greenhouses. BioEnergy Res. 6, 620– 630.
Mester, T., Tien, M., 2001. Engineering of a manganese-binding site in lignin peroxidase isozyme H8 from Phanerochaete chrysosporium. Biochem. Biophys. Res. Commun. 284, 723–8.
Michalsk, K., Ledakowicz, S., 2014. Alkaline hydrogen peroxide pretreatment of energy crops for biogas production. Chem. Pap. 68, 913–922.
Michel, R., Mischler, N., Azambre, B., Finqueneisel, G., Machnikowski, J., Rutkowski, P., Zimny, T., Weber, J.-V., 2006. Miscanthus × Giganteus straw and pellets as sustainable fuels and raw material for activated carbon. Environ. Chem. Lett. 4, 185–189.
Michel, R., Rapagnà, S., Di Marcello, M., Burg, P., Matt, M., Courson, C., Gruber, R., 2011. Catalytic steam gasification of Miscanthus X giganteus in fluidised bed reactor on olivine based catalysts. Fuel Process. Technol. 92, 1169–1177.
Missio, A.L., Mattos, B.D., Cademartori, P.H.G., Lourençon, T.V., Labidi, J., Gatto, D.A., 2015. The effect of oleoresin tapping on physical and chemical properties of Pinus elliottii wood. Sci. For. 43, 721–732.
Mitchell, D.A., Berovič, M., Krieger, N. (Eds.), 2006. Solid-State Fermentation Bioreactors. Springer Berlin Heidelberg, Berlin, Heidelberg.
Modenbach, A.A., Nokes, S.E., 2012. The use of high-solids loadings in biomass 249
pretreatment-a review. Biotechnol. Bioeng. 109, 1430–1442.
Modenbach, A.A., Nokes, S.E., 2013. Enzymatic hydrolysis of biomass at high-solids loadings – A review. Biomass and Bioenergy 56, 526–544.
Monrroy, M., Ortega, I., Ramírez, M., Baeza, J., Freer, J., 2011. Structural change in wood by brown rot fungi and effect on enzymatic hydrolysis. Enzyme Microb. Technol. 49, 472–7.
Montagne, V., Capiaux, H., Barret, M., Cannavo, P., Charpentier, S., Grosbellet, C., Lebeau, T., 2017. Bacterial and fungal communities vary with the type of organic substrate: implications for biocontrol of soilless crops. Environ. Chem. Lett. 1–9.
Mosier, N.S., Wyman, C.E., Dale, B.E., Elander, R.T., Lee, Y.Y., Holtzapple, M., Ladisch, M.R., 2005. Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresour. Technol. 96, 673–686.
Müller, H.W., Trösch, W., 1986. Screening of white-rot fungi for biological pretreatment of wheat straw for biogas production. Appl. Microbiol. Biotechnol. 24, 180–185.
Munk, L., Sitarz, A.K., Kalyani, D.C., Mikkelsen, J.D., Meyer, A.S., 2015. Can laccases catalyze bond cleavage in lignin ? Biotechnol. Adv. 33, 13–24.
Murnen, H.K., Balan, V., Chundawat, S.P.S., Bals, B., Sousa, L.D.C., Dale, B.E., 2007. Optimization of Ammonia Fiber Expansion (AFEX) pretreatment and enzymatic hydrolysis of Miscanthus x giganteus to fermentable sugars. Biotechnol. Prog. 23, 846–850.
Murphy, J.D., Power, N.M., 2009. An argument for using biomethane generated from grass as a biofuel in Ireland. Biomass and Bioenergy 33, 504–512.
Nazarpour, F., Abdullah, D.K., Abdullah, N., Zamiri, R., 2013. Evaluation of biological pretreatment of rubberwood with white rot fungi for enzymatic hydrolysis. Materials (Basel). 6, 2059–2073.
Neukirchen, D., Himken, M., Lammel, J., Czypionka-Krause, U., Olfs, H.-W., 1999. Spatial and temporal distribution of the root system and root nutrient content of an established Miscanthus crop. Eur. J. Agron. 11, 301–309.
Ng, T.L., Eheart, J.W., Cai, X., Miguez, F., 2010. Modeling miscanthus in the Soil and Water Assessment Tool (SWAT) to simulate its water quality effects as a bioenergy crop. Environ. Sci. Technol. 44, 7138–7144.
Nkemka, V.N., Li, Y., Hao, X., 2016. Effect of thermal and alkaline pretreatment of giant miscanthus and Chinese fountaingrass on biogas production. Water Sci. Technol. 73.
250
Nowakowski, D.J., Woodbridge, C.R., Jones, J.M., 2008. Phosphorus catalysis in the pyrolysis behaviour of biomass. J. Anal. Appl. Pyrolysis 83, 197–204.
Nsanganwimana, F., Pourrut, B., Mench, M., Douay, F., 2014. Suitability of Miscanthus species for managing inorganic and organic contaminated land and restoring ecosystem services. A review. J. Environ. Manage. 143, 123–134.
Ofori-Boateng, C., Lee, K.T., 2013. Comparative thermodynamic sustainability assessment of lignocellulosic pretreatment methods for bioethanol production via exergy analysis. Chem. Eng. J. 228, 162–171.
Otjen, L., Blanchette, R.A., 1986. A discussion of microstructural changes in wood during decomposition by white rot basidiomycetes. Can. J. Bot. 64, 905–911.
Padmanabhan, S., Schwyter, P., Liu, Z., Poon, G., Bell, A.T., Prausnitz, J.M., 2016. Delignification of miscanthus using ethylenediamine (EDA) with or without ammonia and subsequent enzymatic hydrolysis to sugars. 3 Biotech 6, 23.
Paine, T.D., Blanche, C.A., Nebeker, T.E., Stephen, F.M., 1987. Composition of loblolly pine resin defenses: comparison of monoterpenes from induced lesion and sapwood resin. Can. J. For. Res. 17, 1202–1206.
Paliwal, R., Rawat, a P., Rawat, M., Rai, J.P.N., 2012. Bioligninolysis: recent updates for biotechnological solution. Appl. Biochem. Biotechnol. 167, 1865–89.
Park, S.Y., Li, Y., 2012. Evaluation of methane production and macronutrient degradation in the anaerobic co-digestion of algae biomass residue and lipid waste. Bioresour. Technol. 111, 42–8.
Pérez, J.A., Ballesteros, I., Ballesteros, M., Sáez, F., Negro, M.J., Manzanares, P., 2008. Optimizing Liquid Hot Water pretreatment conditions to enhance sugar recovery from wheat straw for fuel-ethanol production. Fuel 87, 3640–3647.
Persson, J.Å., 2008. Handbook for Kjeldahl Digestion: A Recent Review of the Classical Method with Improvements Developed by Foss, 4th ed. Foss, Hilleroed, Denmark.
Porbatzki, D., Stemmler, M., Müller, M., 2011. Release of inorganic trace elements during gasification of wood, straw, and miscanthus. Biomass and Bioenergy 35, S79–S86.
Potumarhi, R., Baadhe, R.R., Bhattacharya, S., 2013. Fermentable Sugars from Lignocellulosic Biomass: Technical Challenges. In: Gupta, V.K., Tuohy, M.G. (Eds.), Biofuel Technologies. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 3– 27.
Pyter, R., Heaton, E.A., Dohleman, F.G., Voigt, T.B., Long, S.P., 2009. Agronomic 251
Experiences with Miscanthus x giganteus in Illinois, USA. In: Mielenz, J.R. (Ed.), Biofuels: Methods in Molecular Biology, Methods in Molecular Biology. Humana Press, Totowa, NJ, pp. 41–52.
Qureshi, N., Blaschek, H.P., 2000. Economics of Butanol Fermentation using Hyper- Butanol Producing Clostridium Beijerinckii BA101. Trans IChemE 78, 139–144.
Qureshi, N., Blaschek, H.P., 2001. ABE production from corn: a recent economic evaluation. J. Ind. Microbiol. Biotechnol. 27, 292–297.
Qureshi, N., Ezeji, T.C., 2008. Butanol,“a superior biofuel” production from agricultural residues (renewable biomass): recent progress in technology. Biofuels, Bioprod. Biorefining 2, 319–330.
Qureshi, N., Ezeji, T.C., Ebener, J., Dien, B.S., Cotta, M.A., Blaschek, H.P., 2008a. Butanol production by Clostridium beijerinckii. Part I: Use of acid and enzyme hydrolyzed corn fiber. Bioresour. Technol. 99, 5915–5922.
Qureshi, N., Liu, S.-W., Ezeji, T.C., 2013a. Cellulosic Butanol Production from Agricultural Biomass and Residues: Recent Advances in Technology. In: Lee, J.W. (Ed.), Advanced Biofuels and Bioproducts. Springer, New York, pp. 247–265.
Qureshi, N., Saha, B.C., Cotta, M.A., 2007. Butanol production from wheat straw hydrolysate using Clostridium beijerinckii. Bioprocess Biosyst. Eng. 30, 419–427.
Qureshi, N., Saha, B.C., Cotta, M.A., Singh, V., 2013b. An economic evaluation of biological conversion of wheat straw to butanol: A biofuel. Energy Convers. Manag. 65, 456–462.
Qureshi, N., Saha, B.C., Dien, B., Hector, R.E., Cotta, M.A., 2010a. Production of butanol (a biofuel) from agricultural residues: Part I ? Use of barley straw hydrolysate? Biomass and Bioenergy 34, 559–565.
Qureshi, N., Saha, B.C., Hector, R.E., Cotta, M.A., 2008b. Removal of fermentation inhibitors from alkaline peroxide pretreated and enzymatically hydrolyzed wheat straw: Production of butanol from hydrolysate using Clostridium beijerinckii in batch reactors. Biomass and Bioenergy 32, 1353–1358.
Qureshi, N., Saha, B.C., Hector, R.E., Dien, B., Hughes, S., Liu, S., Iten, L., Bowman, M.J., Sarath, G., Cotta, M.A., 2010b. Production of butanol (a biofuel) from agricultural residues: Part II ? Use of corn stover and switchgrass hydrolysates? Biomass and Bioenergy 34, 566–571.
Qureshi, N., Saha, B.C., Hector, R.E., Dien, B.S., Hughes, S., Liu, S., Iten, L., Bowman, M.J., Sarath, G., Cotta, M.A., 2010c. Production of butanol (a biofuel) from
252
agricultural residues: Part II – Use of corn stover and switchgrass hydrolysates☆. Biomass and Bioenergy 34, 566–571.
Qureshi, N., Singh, V., Liu, S., Ezeji, T.C., Saha, B.C., Cotta, M.A., 2014. Process integration for simultaneous saccharification, fermentation, and recovery (SSFR): production of butanol from corn stover using Clostridium beijerinckii P260. Bioresour. Technol. 154, 222–8.
Ranjan, A., Moholkar, V.S., 2012. Biobutanol: science, engineering, and economics. Int. J. Energy Res. 36, 277–323.
Ray, M.J., Leak, D.J., Spanu, P.D., Murphy, R.J., 2010. Brown rot fungal early stage decay mechanism as a biological pretreatment for softwood biomass in biofuel production. Biomass and Bioenergy 34, 1257–1262.
Reid, I.D., 1995. Biodegradation of lignin. Can. J. Bot. 73, 1011–1018.
Reid, I.D., Seifert, K.A., 1982. Effect of an atmosphere of oxygen on growth, respiration, and lignin degradation by white-rot fungi. Can. J. Bot. 60, 252–260.
REN21, 2016. Renewables 2016 Global Status report. Paris.
Rico, A., Rencoret, J., Río, J.C. del, Martínez, A.T., Gutiérrez, A., 2014. Pretreatment with laccase and a phenolic mediator degrades lignin and enhances saccharification of Eucalyptus feedstock. Biotechnol. Biofuels 7.
Robinson, T., Singh, D., Nigam, P., 2001. Solid-state fermentation: a promising microbial technology for secondary metabolite production. Appl. Microbiol. Biotechnol. 55, 284–289.
Roche, C.M., Dibble, C.J., Stickel, J.J., 2009. Laboratory-scale method for enzymatic saccharification of lignocellulosic biomass at high-solids loadings. Biotechnol. Biofuels 2, 28.
Rodrigues, M. a M., Pinto, P.A., Bezerra, R.M.F., Dias, A.A., Guedes, C. V., Cardoso, V.M.G., Cone, J.W., Ferreira, L.M.M., Colaço, J., Sequeira, C. a., 2008. Effect of enzyme extracts isolated from white-rot fungi on chemical composition and in vitro digestibility of wheat straw. Anim. Feed Sci. Technol. 141, 326–338.
Ruiz-Dueñas, F.J., Martínez, A.T., 2009. Microbial degradation of lignin: how a bulky recalcitrant polymer is efficiently recycled in nature and how we can take advantage of this. Microb. Biotechnol. 2, 164–77.
Ruttimann-Johnson, C., Salas, L., Vicuna, R., Kirk, T.K., 1993. Extracellular Enzyme Production and Synthetic Lignin Mineralization by Ceriporiopsis subvermispora.
253
Appl. Enviromental Microbiol. 59, 1792–1797.
Ryu, C., Yang, Y. Bin, Khor, A., Yates, N.E., Sharifi, V.N., Swithenbank, J., 2006. Effect of fuel properties on biomass combustion: Part I. Experiments—fuel type, equivalence ratio and particle size. Fuel 85, 1039–1046.
Saha, B.C., Qureshi, N., Kennedy, G.J., Cotta, M.A., 2016. Biological pretreatment of corn stover with white-rot fungus for improved enzymatic hydrolysis. Int. Biodeterior. Biodegradation 109, 29–35.
Salvachúa, D., Prieto, A., López-Abelairas, M., Lu-Chau, T., Martínez, A.T., Martínez, M.J., 2011. Fungal pretreatment: An alternative in second-generation ethanol from wheat straw. Bioresour. Technol. 102, 7500–7506.
Salvachúa, D., Prieto, A., Martínez, A.T., Martínez, M.J., 2013. Characterization of a novel dye-decolorizing peroxidase (DyP)-type enzyme from Irpex lacteus and its application in enzymatic hydrolysis of wheat straw. Appl. Environ. Microbiol. 79, 4316–24.
Samson, R.A., Evans, H.C., Latgé, J.-P., 1988. Atlas of Entomopathogenic Fungi. Springer Berlin Heidelberg, Berlin, Heidelberg.
Samson, R.A., Visagie, C.M., Houbraken, J., Hong, S.-B., Hubka, V., Klaassen, C.H.W., Perrone, G., Seifert, K.A., Susca, A., Tanney, J.B., Varga, J., Kocsub?, S., Szigeti, G., Yaguchi, T., Frisvad, J.C., 2014. Phylogeny, identification and nomenclature of the genus Aspergillus. Stud. Mycol. 78, 141–173.
Sánchez, C., 2009. Lignocellulosic residues: biodegradation and bioconversion by fungi. Biotechnol. Adv. 27, 185–94.
Sanderson, M.A., Adler, P.R., 2008. Perennial forages as second generation bioenergy crops. Int. J. Mol. Sci. 9, 768–88.
Saritha, M., Arora, A., Lata, 2012a. Biological Pretreatment of Lignocellulosic Substrates for Enhanced Delignification and Enzymatic Digestibility. Indian J. Microbiol. 52, 122–130.
Saritha, M., Arora, A., Nain, L., 2012b. Pretreatment of paddy straw with Trametes hirsuta for improved enzymatic saccharification. Bioresour. Technol. 104, 459–465.
Saritha, M., Arora, A., Singh, S., Nain, L., 2013. Streptomyces griseorubens mediated delignification of paddy straw for improved enzymatic saccharification yields. Bioresour. Technol. 135, 12–7.
Sassner, P., Galbe, M., Zacchi, G., 2008. Techno-economic evaluation of bioethanol production from three different lignocellulosic materials. Biomass and Bioenergy 254
32, 422–430.
Sawatdeenarunat, C., KC, S., Takara, D., Oechsner, H., Khanal, S.K., 2015. Anaerobic digestion of lignocellulosic biomass: Challenges and opportunities. Bioresour. Technol. 178, 178–186.
Sawatdeenarunat, C., Nguyen, D., KC, S., Shrestha, S., Rajendran, K., Oechsner, H., Xie, L., Khanal, S.K., 2016. Anaerobic biorefinery: Current status, challenges, and opportunities. Bioresour. Technol. 215, 304–313.
Scheffer, T.C., Cowling, E.B., 1966. Natural Resistance of Wood to Microbial Deterioration. Annu. Rev. Phytopathol. 4, 147–168.
Schilling, J.S., Ai, J., Blanchette, R.A., Duncan, S.M., Filley, T.R., Tschirner, U.W., 2012. Lignocellulose modifications by brown rot fungi and their effects, as pretreatments, on cellulolysis. Bioresour. Technol. 116, 147–54.
Schilling, J.S., Tewalt, J.P., Duncan, S.M., 2009. Synergy between pretreatment lignocellulose modifications and saccharification efficiency in two brown rot fungal systems. Appl. Microbiol. Biotechnol. 84, 465–75.
Scordia, D., Cosentino, S.L., Jeffries, T.W., 2013. Effectiveness of dilute oxalic acid pretreatment of Miscanthus × giganteus biomass for ethanol production. Biomass and Bioenergy 59, 540–548.
Scurlock, J.M.O., 1999. Miscanthus: A review of European experience with a novel energy crop. Oak Ridge National Lab., TN (US).
Selig, M., Weiss, N., Ji, Y., 2008. Enzymatic Saccharification of Lignocellulosic Biomass: Laboratory Analytical Procedure ( LAP ). Golden, Colorado.
Seppälä, M., Paavola, T., Lehtomäki, A., Rintala, J., 2009. Biogas production from boreal herbaceous grasses--specific methane yield and methane yield per hectare. Bioresour. Technol. 100, 2952–8.
Serrano, L., Egües, I., Alriols, M.G., Llano-Ponte, R., Labidi, J., 2010. Miscanthus sinensis fractionation by different reagents. Chem. Eng. J. 156, 49–55.
Shah, A., Darr, M., 2016. A techno-economic analysis of the corn stover feedstock supply system for cellulosic biorefineries. Biofuels, Bioprod. Biorefining 10, 542– 559.
Shi, J., Chinn, M.S., Sharma-Shivappa, R.R., 2008. Microbial pretreatment of cotton stalks by solid state cultivation of Phanerochaete chrysosporium. Bioresour. Technol. 99, 6556–64.
255
Shi, J., Xu, F., Wang, Z., Stiverson, J.A., Yu, Z., Li, Y., 2014. Effects of microbial and non-microbial factors of liquid anaerobic digestion effluent as inoculum on solid- state anaerobic digestion of corn stover. Bioresour. Technol. 157, 188–96.
Shi, Y., Cheng, Y., Wang, Y., Zhang, G., Gao, R., Xiang, C., Feng, J., Lou, D., Liu, Y., 2017. Investigation of the fungal community structures of imported wheat using high-throughput sequencing technology. PLoS One 12, e0171894.
Sigoillot, J.C., Berrin, J.G., Bey, M., Lesage-Meessen, L., Levasseur, A., Lomascolo, A., Record, E., Uzan-Boukhris, E., 2012. Fungal Strategies for Lignin Degradation, 1st ed, Advances in Botanical Research. Elsevier Ltd.
Singh, D., Chen, S., 2008. The white-rot fungus Phanerochaete chrysosporium: conditions for the production of lignin-degrading enzymes. Appl. Microbiol. Biotechnol. 81, 399–417.
Singh, P., Suman, A., Tiwari, P., Arya, N., Gaur, A., Shrivastava, A.K., 2007. Biological pretreatment of sugarcane trash for its conversion to fermentable sugars. World J. Microbiol. Biotechnol. 24, 667–673.
Skyba, O., Douglas, C.J., Mansfield, S.D., 2013. Syringyl-rich lignin renders poplars more resistant to degradation by wood decay fungi. Appl. Environ. Microbiol. 79, 2560–71.
Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D., 2012. Determination of structural carbohydrates and lignin in biomass, Laboratory Analytical Procedure (LAP). Golden, Colorado.
Sluiter, A., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., 2008. Determination of Extractives in Biomass, Laboratory Analytical Procedure (LAP). Golden, Colorado.
Smeets, E.M.W., Lewandowski, I.M., Faaij, A.P.C., 2009. The economical and environmental performance of miscanthus and switchgrass production and supply chains in a European setting. Renew. Sustain. Energy Rev. 13, 1230–1245.
Smith, D.P., Peay, K.G., Palmer, M., Gillikin, C., Keefe, D., 2014. Sequence Depth, Not PCR Replication, Improves Ecological Inference from Next Generation DNA Sequencing. PLoS One 9, e90234.
Soares Rodrigues, C.I., Jackson, J.J., Montross, M.D., 2016. A molar basis comparison of calcium hydroxide, sodium hydroxide, and potassium hydroxide on the pretreatment of switchgrass and miscanthus under high solids conditions. Ind. Crops Prod. 92, 165–173.
Song, L., Ma, F., Zeng, Y., Zhang, X., Yu, H., 2013a. The promoting effects of
256
manganese on biological pretreatment with Irpex lacteus and enzymatic hydrolysis of corn stover. Bioresour. Technol. 135, 89–92.
Song, L., Yu, H., Ma, F., Zhang, X., 2013b. Biological Pretreatment under Non-sterile Conditions for Enzymatic Hydrolysis of Corn Stover. Bioresources 8, 3802–3816.
Sørensen, A., Teller, P.J., Hilstrøm, T., Ahring, B.K., 2008. Hydrolysis of Miscanthus for bioethanol production using dilute acid presoaking combined with wet explosion pre-treatment and enzymatic treatment. Bioresour. Technol. 99, 6602–6607.
Srebotnik, E., Messner, K., 1994. A simple method that uses differential staining and light microscopy to assess the selectivity of wood delignification by white rot fungi. Appl. Environ. Microbiol. 60, 1383–1386.
Strittmatter, E., Liers, C., Ullrich, R., Wachter, S., Hofrichter, M., Plattner, D.A., Piontek, K., 2013a. First crystal structure of a fungal high-redox potential dye-decolorizing peroxidase: substrate interaction sites and long-range electron transfer. J. Biol. Chem. 288, 4095–102.
Strittmatter, E., Wachter, S., Liers, C., Ullrich, R., Hofrichter, M., Plattner, D.A., Piontek, K., 2013b. Radical formation on a conserved tyrosine residue is crucial for DyP activity. Arch. Biochem. Biophys. 537, 161–7.
Styles, D., Jones, M.B., 2008. Life-cycle environmental and economic impacts of energy- crop fuel-chains: an integrated assessment of potential GHG avoidance in Ireland. Environ. Sci. Policy 11, 294–306.
Sugano, Y., 2009. DyP-type peroxidases comprise a novel heme peroxidase family. Cell. Mol. Life Sci. 66, 1387–1403.
Sun, F.H., Li, J., Yuan, Y.X., Yan, Z.Y., Liu, X.F., 2011. Effect of biological pretreatment with Trametes hirsuta yj9 on enzymatic hydrolysis of corn stover. Int. Biodeterior. Biodegrad. 65, 931–938.
Taherzadeh, M.J., Karimi, K., 2008. Pretreatment of lignocellulosic wastes to improve ethanol and biogas production: a review., International journal of molecular sciences.
Tanaka, H., Koike, K., Itakura, S., Enoki, A., 2009. Degradation of wood and enzyme production by Ceriporiopsis subvermispora. Enzyme Microb. Technol. 45, 384–390.
Taniguchi, M., Suzuki, H., Watanabe, D., Sakai, K., Hoshino, K., Tanaka, T., 2005. Evaluation of pretreatment with Pleurotus ostreatus for enzymatic hydrolysis of rice straw. J. Biosci. Bioeng. 100, 637–43.
Tao, L., Aden, A., 2009. The economics of current and future biofuels. Vitr. Cell. Dev. 257
Biol. - Plant 45, 199–217.
Tedersoo, L., Bahram, M., Polme, S., Koljalg, U., Yorou, N.S., Wijesundera, R., Ruiz, L.V., Vasco-Palacios, A.M., Thu, P.Q., Suija, A., Smith, M.E., Sharp, C., Saluveer, E., Saitta, A., Rosas, M., Riit, T., Ratkowsky, D., Pritsch, K., Poldmaa, K., Piepenbring, M., Phosri, C., Peterson, M., Parts, K., Partel, K., Otsing, E., Nouhra, E., Njouonkou, A.L., Nilsson, R.H., Morgado, L.N., Mayor, J., May, T.W., Majuakim, L., Lodge, D.J., Lee, S.S., Larsson, K.-H., Kohout, P., Hosaka, K., Hiiesalu, I., Henkel, T.W., Harend, H., Guo, L. -d., Greslebin, A., Grelet, G., Geml, J., Gates, G., Dunstan, W., Dunk, C., Drenkhan, R., Dearnaley, J., De Kesel, A., Dang, T., Chen, X., Buegger, F., Brearley, F.Q., Bonito, G., Anslan, S., Abell, S., Abarenkov, K., 2014. Global diversity and geography of soil fungi. Science (80-. ). 346, 1256688–1256688.
Tekere, M., Zvauya, R., Read, J.S., 2001. Ligninolytic enzyme production in selected sub-tropical white rot fungi under different culture conditions. J. Basic Microbiol. 41, 115–29. ten Have, R., Teunissen, P.J.M., 2001. Oxidative Mechanisms Involved in Lignin Degradation by White-Rot Fungi. Chem. Rev. 101, 3397–3414.
Theuretzbacher, F., Lizasoain, J., Menardo, S., Nilsen, P.J., Gronauer, A., Bauer, A., Theuretzbacher, F., Lizasoain, J., Menardo, S., Nilsen, P.J., Gronauer, A., Bauer, A., 2014. Effect of Steam Explosion Pretreatment on the Specific Methane Yield of Miscanthus x giganteus. Agric. Conspec. Sci. 79, 19–22.
Thompson, D.N., Hames, B.R., Reddy, C.A., Grethlein, H.E., 1998. In vitro degradation of natural insoluble lignin in aqueous media by the extracellular peroxidases ofPhanerochaete chrysosporium. Biotechnol. Bioeng. 57, 704–717.
Tian, X., Yang, T., He, J., Chu, Q., Jia, X., Huang, J., 2017. Fungal community and cellulose-degrading genes in the composting process of Chinese medicinal herbal residues. Bioresour. Technol. 241, 374–383.
Tien, M., Kirk, T.K., 1983. Lignin-Degrading Enzyme from the Hymenomycete Phanerochaete chrysosporium Burds. Science 221, 661–3.
Tiwari, R., Rana, S., Singh, S., Arora, A., Kaushik, R., Agrawal, V.V., Saxena, A.K., Nain, L., 2013. Biological delignification of paddy straw and Parthenium sp. using a novel micromycete Myrothecium roridum LG7 for enhanced saccharification. Bioresour. Technol. 135, 7–11.
Tuor, U., Winterhalter, K., Fiechter, 1995. Enzymes of white rot fungi involved in lignin degradation and ecological determinants for wood decay. J. Biotechnol. 41, 1–17.
258
U.S. DOE, 2005. Biomass as feedstock for a bioenergy and bioproducts industry: the technical feasibility of a billion-ton annual supply. DTIC Document.
U.S. DOE, 2011. U.S. Billion-Ton Update: Biomass Supply for a Bioenergy and Bioproducts Industry. Oak Ridge National Laboratory, Oak Ridge.
U.S. DOE, 2016. 2016 Billion-Ton Report: Advancing Domestic Resources for a Thriving Bioeconomy, Volume 1: Economic Availability of Feedstocks. Oak Ridge, TN.
Ujor, V., Agu, C.V., Gopalan, V., Ezeji, T.C., 2015. Allopurinol-mediated lignocellulose- derived microbial inhibitor tolerance by Clostridium beijerinckii during acetone– butanol–ethanol (ABE) fermentation. Appl. Microbiol. Biotechnol. 99, 3729–3740.
Ulrich, G.D., 1984. A guide to chemical engineering process design and economics. Wiley, New York.
Unterseher, M., Jumpponen, A., Öpik, M., Tedersoo, L., Moora, M., Dormann, C.F., Schnittler, M., 2011. Species abundance distributions and richness estimations in fungal metagenomics - Lessons learned from community ecology. Mol. Ecol. 20, 275–285.
Vaidya, A., Singh, T., 2012. Pre-treatment of Pinus radiata substrates by basidiomycetes fungi to enhance enzymatic hydrolysis. Biotechnol. Lett. 34, 1263–7. van der Merwe, a. B., Cheng, H., Görgens, J.F., Knoetze, J.H., 2013. Comparison of energy efficiency and economics of process designs for biobutanol production from sugarcane molasses. Fuel 105, 451–458. van der Wal, A., Ottosson, E., de Boer, W., 2015. Neglected role of fungal community composition in explaining variation in wood decay rates. Ecology 96, 124–133.
Van Dyk, J.S., Pletschke, B.I., 2012. A review of lignocellulose bioconversion using enzymatic hydrolysis and synergistic cooperation between enzymes—Factors affecting enzymes, conversion and synergy. Biotechnol. Adv. 30, 1458–1480.
Vanderghem, C., Brostaux, Y., Jacquet, N., Blecker, C., Paquot, M., 2012. Optimization of formic/acetic acid delignification of Miscanthus × giganteus for enzymatic hydrolysis using response surface methodology. Ind. Crops Prod. 35, 280–286.
Vasco-Correa, J., Ge, X., Li, Y., 2016a. Fungal pretreatment of non-sterile miscanthus for enhanced enzymatic hydrolysis. Bioresour. Technol. 203, 118–23.
Vasco-Correa, J., Ge, X., Li, Y., 2016b. Biomass Fractionation Technologies for a Lignocellulosic Feedstock Based Biorefinery, Biomass Fractionation Technologies for a Lignocellulosic Feedstock Based Biorefinery. Elsevier. 259
Vasco-Correa, J., Li, Y., 2015. Solid-state anaerobic digestion of fungal pretreated Miscanthus sinensis harvested in two different seasons. Bioresour. Technol. 185, 211–217.
Vilanova, C., Marín, M., Baixeras, J., Latorre, A., Porcar, M., 2014. Selecting Microbial Strains from Pine Tree Resin: Biotechnological Applications from a Terpene World. PLoS One 9, e100740.
Villaverde, J.J., Domingues, R.M.A., Freire, C.S.R., Silvestre, a J.D., Neto, C.P., Ligero, P., Vega, A., 2009. Miscanthus x giganteus extractives: a source of valuable phenolic compounds and sterols. J. Agric. Food Chem. 57, 3626–31.
Vintila, T., Dragomirescu, M., Croitotiu, V., Vintila, C., Barbu, H., Sand, C., 2010. Saccharification of lignocellulose-with reference to Miscanthus-using different cellulases. Rom. Biotechnol. Lett. 15, 5499.
Vyn, R.J., Virani, T., Deen, B., 2012. Examining the economic feasibility of miscanthus in Ontario: An application to the greenhouse industry. Energy Policy 50, 669–676.
Wafiq, A., Reichel, D., Hanafy, M., 2016. Pressure influence on pyrolysis product properties of raw and torrefied Miscanthus: Role of particle structure. Fuel 179, 156–167.
Wagner, A., Tobimatsu, Y., Phillips, L., Flint, H., Geddes, B., Lu, F., Ralph, J., 2015. Syringyl lignin production in conifers: Proof of concept in a Pine tracheary element system. Proc. Natl. Acad. Sci. U. S. A. 112, 6218–23.
Wan, C., 2011. Microbial Pretreatment of Lignocellulosic Biomass with Ceriporiopsis Subvermispora for Enzymatic Hydrolysis and Ethanol Production. The Ohio State University.
Wan, C., Li, Y., 2010a. Microbial pretreatment of corn stover with Ceriporiopsis subvermispora for enzymatic hydrolysis and ethanol production. Bioresour. Technol. 101, 6398–403.
Wan, C., Li, Y., 2010b. Microbial delignification of corn stover by Ceriporiopsis subvermispora for improving cellulose digestibility. Enzyme Microb. Technol. 47, 31–36.
Wan, C., Li, Y., 2011a. Effect of hot water extraction and liquid hot water pretreatment on the fungal degradation of biomass feedstocks. Bioresour. Technol. 102, 9788–93.
Wan, C., Li, Y., 2011b. Effectiveness of microbial pretreatment by Ceriporiopsis subvermispora on different biomass feedstocks. Bioresour. Technol. 102, 7507– 7512.
260
Wan, C., Li, Y., 2012. Fungal pretreatment of lignocellulosic biomass. Biotechnol. Adv. 30, 1447–57.
Wang, F.Q., Xie, H., Chen, W., Wang, E.T., Du, F.G., Song, A.D., 2013. Biological pretreatment of corn stover with ligninolytic enzyme for high efficient enzymatic hydrolysis. Bioresour. Technol. 144, 572–578.
Wang, M., 2010. Life-Cycle Analysis of Biofuel. In: Mascia, P.N., Scheffran, J., Widholm, J.M. (Eds.), Plant Biotechnology for Sustainable Production of Energy and Co-Products, Biotechnology in Agriculture and Forestry. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 385–407.
Wang, Q.Q., He, Z., Zhu, Z., Zhang, Y.-H.P., Ni, Y., Luo, X.L., Zhu, J.Y., Ingram, L., Zhou, H., Zhu, W., 2012. Evaluations of cellulose accessibilities of lignocelluloses by solute exclusion and protein adsorption techniques. Biotechnol. Bioeng. 109, 381–389.
Wang, S.-Y., Chen, P.-F., Chang, S.-T., 2005. Antifungal activities of essential oils and their constituents from indigenous cinnamon (Cinnamomum osmophloeum) leaves against wood decay fungi. Bioresour. Technol. 96, 813–8.
Wang, S., Wang, S., Hastings, A., Pogson, M., Smith, P., 2012. Economic and greenhouse gas costs of Miscanthus supply chains in the United Kingdom. GCB Bioenergy 4, 358–363.
Wang, W., Meng, X., Min, D., Song, J., Jin, Y., 2014. Effects of Green Liquor Pretreatment on the Chemical Composition and Enzymatic Hydrolysis of Several Lignocellulosic Biomasses. BioResources 10, 709–720.
Wang, W., Yuan, T., Cui, B., 2014. Biological Pretreatment with White Rot Fungi and Their Co-Culture to Overcome Lignocellulosic Recalcitrance for Improved Enzymatic Digestion. BioResources 9, 3968–3976.
Wang, Y., Janssen, H., Blaschek, H.P., 2014. Fermentative Biobutanol Production: An Old Topic with Remarkable Recent Advances. In: Bisaria, V.S., Kondo, A. (Eds.), Bioprocessing of Renewable Resources to Commodity Bioproducts. John Wiley & Sons, Inc., Hoboken, NJ, pp. 227–260.
Whittaker, C., Hunt, J., Misselbrook, T., Shield, I., 2016. How well does Miscanthus ensile for use in an anaerobic digestion plant? Biomass and Bioenergy 88, 24–34.
Wingren, A., Galbe, M., Zacchi, G.,. Techno-economic evaluation of producing ethanol from softwood: comparison of SSF and SHF and identification of bottlenecks. Biotechnol. Prog. 19, 1109–17.
261
Wong, D.W.S., 2009. Structure and action mechanism of ligninolytic enzymes. Appl. Biochem. Biotechnol. 157, 174–209.
Wu, M., Wang, M., Liu, J., Huo, H., 2007. Life-cycle assessment of corn-based butanol as a potential transportation fuel. Argonne, IL.
Wu, M., Wang, M., Liu, J., Huo, H., 2008. Assessment of potential life‐cycle energy and greenhouse gas emission effects from using corn‐based butanol as a transportation fuel. Biotechnol. Prog. 24, 1204–1214.
Wu, X., McLaren, J., Madl, R., Wang, D., 2010. Biofuels from Lignocellulosic Biomass. In: Singh, O., Harvey, S. (Eds.), Sustainable Biotechnology: Sources of Renewable Energy. Springer, New York, NY.
Xu, C., Ma, F., Zhang, X., Chen, S., 2010. Biological pretreatment of corn stover by Irpex lacteus for enzymatic hydrolysis. J. Agric. Food Chem. 58, 10893–8.
Xu, F., Shi, J., Lv, W., Yu, Z., Li, Y., 2013. Comparison of different liquid anaerobic digestion effluents as inocula and nitrogen sources for solid-state batch anaerobic digestion of corn stover. Waste Manag. 33, 26–32.
Xu, F., Wang, Z.-W., Tang, L., Li, Y., 2014. A mass diffusion-based interpretation of the effect of total solids content on solid-state anaerobic digestion of cellulosic biomass. Bioresour. Technol. 167, 178–85.
Yaghoubi, K., Pazouki, M., Shojaosadati, S.A., 2008. Variable optimization for biopulping of agricultural residues by Ceriporiopsis subvermispora. Bioresour. Technol. 99, 4321–4328.
Yan, K., Liu, F., Chen, Q., Ke, M., Huang, X., Hu, W., Zhou, B., Zhang, X., Yu, H., 2016. Pyrolysis characteristics and kinetics of lignin derived from enzymatic hydrolysis residue of bamboo pretreated with white-rot fungus. Biotechnol. Biofuels 9, 76.
Yang, B., Wyman, C.E., 2008. Pretreatment: the key to unlocking low-cost cellulosic ethanol. Biofuels, Bioprod. Biorefining 2, 26–40.
Yang, H., Yan, R., Chen, H., Lee, D.H., Zheng, C., 2007. Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel 86, 1781–1788.
Yang, L., Li, Y., 2014. Anaerobic digestion of giant reed for methane production. Bioresour. Technol. 171, 233–9.
Yang, L., Xu, F., Ge, X., Li, Y., 2015. Challenges and strategies for solid-state anaerobic digestion of lignocellulosic biomass. Renew. Sustain. Energy Rev. 44, 824–834.
262
Yang, X., Ma, F., Yu, H., Zhang, X., Chen, S., 2011. Effects of biopretreatment of corn stover with white-rot fungus on low-temperature pyrolysis products. Bioresour. Technol. 102, 3498–3503.
Ye, D., Montane, D., Farriol, X., 2005. Preparation and characterisation of methylcelluloses from. Carbohydr. Polym. 62, 258–266.
Yeh, R.-H., Lin, Y.-S., Wang, T.-H., Kuan, W.-C., Lee, W.-C., 2016. Bioethanol production from pretreated Miscanthus floridulus biomass by simultaneous saccharification and fermentation. Biomass and Bioenergy 94, 110–116.
Yorgun, S., Simşek, Y.E., 2008. Catalytic pyrolysis of Miscanthus x giganteus over activated alumina. Bioresour. Technol. 99, 8095–100.
Yu, G., Afzal, W., Yang, F., Padmanabhan, S., Liu, Z., Xie, H., Shafy, M.A., Bell, A.T., Prausnitz, J.M., 2014. Pretreatment of Miscanthus×giganteus using aqueous ammonia with hydrogen peroxide to increase enzymatic hydrolysis to sugars. J. Chem. Technol. Biotechnol. 89, 698–706.
Yu, H., Guo, G., Zhang, X., Yan, K., Xu, C., 2009. The effect of biological pretreatment with the selective white-rot fungus Echinodontium taxodii on enzymatic hydrolysis of softwoods and hardwoods. Bioresour. Technol. 100, 5170–5.
Yu, Y., Zeng, Y., Zuo, J., Ma, F., Yang, X., Zhang, X., Wang, Y., 2013. Improving the conversion of biomass in catalytic fast pyrolysis via white-rot fungal pretreatment. Bioresour. Technol. 134, 198–203.
Yuzbashev, T. V, Yuzbasheva, E.Y., Sobolevskaya, T.I., Laptev, I. a, Vybornaya, T. V, Larina, A.S., Matsui, K., Fukui, K., Sineoky, S.P., 2010. Production of succinic acid at low pH by a recombinant strain of the aerobic yeast Yarrowia lipolytica. Biotechnol. Bioeng. 107, 673–82.
Zeng, J., Singh, D., Chen, S., 2011. Biological pretreatment of wheat straw by Phanerochaete chrysosporium supplemented with inorganic salts. Bioresour. Technol. 102, 3206–3214.
Zeng, Y., Yang, X., Yu, H., Zhang, X., Ma, F., 2011. Comparative studies on thermochemical characterization of corn stover pretreated by white-rot and brown- rot fungi. J. Agric. Food Chem. 59, 9965–9971.
Zeng, Y., Zhao, S., Yang, S., Ding, S.-Y., 2014. Lignin plays a negative role in the biochemical process for producing lignocellulosic biofuels. Curr. Opin. Biotechnol. 27, 38–45.
Zhang, B., Shahbazi, A., 2011. Recent developments in pretreatment technologies for
263
production of lignocellulosic biofuels. J. Pet. Environ. Biotechnol.
Zhang, J., Zeng, G., Chen, Y., Yu, M., Huang, H., Fan, C., Zhu, Y., Li, H., Liu, Z., Chen, M., Jiang, M., 2013. Impact of Phanerochaete chrysosporium inoculation on indigenous bacterial communities during agricultural waste composting. Appl. Microbiol. Biotechnol. 97, 3159–3169.
Zhang, Q., Chang, J., Wang, T., Xu, Y., 2007. Review of biomass pyrolysis oil properties and upgrading research. Energy Convers. Manag. 48, 87–92.
Zhang, X., Yu, H., Huang, H., Liu, Y., 2007. Evaluation of biological pretreatment with white rot fungi for the enzymatic hydrolysis of bamboo culms. Int. Biodeterior. Biodegradation 60, 159–164.
Zhang, Y., Ezeji, T.C., 2014. Elucidating and alleviating impacts of lignocellulose- derived microbial inhibitors on Clostridium beijerinckii during fermentation of Miscanthus giganteus to butanol. J. Ind. Microbiol. Biotechnol. 41, 1505–1516.
Zhang, Y., Han, B., Ezeji, T.C., 2012. Biotransformation of furfural and 5- hydroxymethyl furfural (HMF) by Clostridium acetobutylicum ATCC 824 during butanol fermentation. N. Biotechnol. 29, 345–351.
Zhao, J., Ge, X., Vasco-Correa, J., Li, Y., 2014a. Fungal pretreatment of unsterilized yard trimmings for enhanced methane production by solid-state anaerobic digestion. Bioresour. Technol. 158, 248–52.
Zhao, J., Zheng, Y., Li, Y., 2014b. Fungal pretreatment of yard trimmings for enhancement of methane yield from solid-state anaerobic digestion. Bioresour. Technol. 156, 176–81.
Zhao, L., Cao, G.-L.L., Wang, A.-J.J., Ren, H.-Y.Y., Dong, D., Liu, Z.-N.N., Guan, X.- Y.Y., Xu, C.-J.J., Ren, N.-Q.Q., 2012. Fungal pretreatment of cornstalk with Phanerochaete chrysosporium for enhancing enzymatic saccharification and hydrogen production. Bioresour. Technol. 114, 365–369.
Zheng, Y., Zhao, J., Xu, F., Li, Y., 2014. Pretreatment of lignocellulosic biomass for enhanced biogas production. Prog. Energy Combust. Sci. 42, 35–53.
Zhi, Z., Wang, H., 2014. White-rot fungal pretreatment of wheat straw with Phanerochaete chrysosporium for biohydrogen production: simultaneous saccharification and fermentation. Bioprocess Biosyst. Eng. 37, 1447–58.
Zhou, X., Li, Q., Zhang, Y., Gu, Y., 2017. Effect of hydrothermal pretreatment on Miscanthus anaerobic digestion. Bioresour. Technol. 224, 721–726.
Zhu, J., Wan, C., Li, Y., 2010. Enhanced solid-state anaerobic digestion of corn stover by 264 alkaline pretreatment. Bioresour. Technol. 101, 7523–7528.
265
Appendix: Supplemental data for Chapter 8
Table A. 1 Direct fixed capital estimated summary (prices in $)
Total Plant Direct Cost (TPDC) (physical cost) 1. Equipment Purchase Cost 354,639,000 2. Installation 123,673,000 3. Process Piping 124,124,000 4. Instrumentation 141,856,000 5. Insulation 17,732,000 6. Electrical 35,464,000 7. Buildings 159,587,000 8. Yard Improvement 53,196,000 9. Auxiliary Facilities 141,856,000 TPDC 1,152,125,000 Total Plant Indirect Cost (TPIC) 10. Engineering 230,425,000 11. Construction 230,425,000 TPIC 460,850,000 Total Plant Cost (TPC = TPDC+TPIC) TPC 1,612,975,000 Contractor's Fee & Contingency (CFC) 12. Contractor's Fee 80.649,000 13. Contingency 161.297,000 CFC = 12+13 241,946,000 Direct Fixed Capital Cost (DFC = TPC+CFC) DFC 1,854,924,000
266
Table A. 2 Cost of major equipment for fungal pretreatment facility producing
fermentable sugars for a 30 million gallon/year cellulosic biorefinery
Unit rate Purchasing Equipment name Equipment sizing Quantity ($) price ($) Receiver tank - concrete Vessel volume: 892.23 m3 8 246,000 1,968,000 Autoclave Vessel volume: 99.81 m3 81 260,000 21,060,000 Packed bed bioreactor Vessel volume: 3993.49 m3 194 1,089,000 211,266,000 Stirred reactor Vessel volume: 3383.42 m3 30 916,000 27,480,000 Air-Lift seed fermenter Vessel volume: 1040.85 m3 1 1,233,000 1,233,000 Air compressor Power: 2419.34 kW 5 2,614,000 13,070,000 Air filter Throughput: 14 m3/h 15 52,000 780,000 Screw conveyor Pipe length: 15 m 6 620,000 3,720,000 Screw conveyor Pipe length: 10 m 7 416,000 2,912,000 Belt conveyor Belt length: 50 m 1 144,000 144,000 Bucket elevator Elevator length: 10 m 1 20,000 20,000 Centrifugal pump Pump power: 0.27 kW 1 10,000 10,000 Centrifugal pump Pump power: 6.19 kW 1 48,000 48,000 Unlisted equipment - - - 70,928,000 TOTAL 354,639,000
Table A. 3 Feedstock and other materials cost
Material Price ($/kg) Miscanthus (feedstock) 0.11 Enzyme (~45 g protein/liter) 0.24 Medium inoculum (malt extract) 1.00 Water 0.0002
267
Table A. 4 Labor and utilities requirement for fungal pretreatment facility producing
fermentable sugars for a 30 million gallon/year cellulosic biorefinery
Parameter Annual amount Annual cost ($) Labor (h) 87,268 6,021,496 Electricity (kWh) 139,482,199 13,948,220 Steam (heating agent) (kg) 119,994 1,439,926 Steam (high pressure –autoclaving) (kg) 592,849 11,856,985 Cooling water (kg) 7,535,315 376,766 Chilled water (kg) 810,573 324,229
268