A CLOSED - LOOP BIOREFINING SYSTEM TO CONVERT ORGANIC RESIDUES INTO FUELS By Rui Chen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Biosystems Engineering - Doctor of Philosophy 2015 ABSTRACT A CLOSED - LOOP BIOREFINING SYSTEM TO CONVERT ORGANIC RESIDUES INTO FUELS By Rui Chen Th is project deliver s an energy positive and water neutral , closed - loop biorefin ing system that converts organic wastes into renewable energy and reduces the overall impacts on the environment. The research consisted of three major stages: The first stage of this project was conducted in an anaerobic co - digestion system. Effects of the ratio of dairy manure - to - food waste as well as operating temperature were tested on the performance of the co - digestion system. Results illustrated an increase in biogas produc tivity with the increase of supplemental food waste; fiber analysis revealed similar chemical composition (cellulose, hemicellulose and lignin) of final solid digestate regardless their different initial feedstock blends and digestion conditions. The molecular genetic analyses demonstrated that anaerobic methanogenic microorganisms were able to adjust their community assemblage to maximize biogas production and produce homogenized solid digestate. The second stage utilized electrocoagulation (EC) pret reated liquid digestate from previous stage to culture freshwater algae. Kinetics study showed a s imilar maximum growth rate (0.2 01 - 0 .207 g TS day - 1 ) in both 2× and 5× dilutions of EC solution; however, the algal growth was inhibited in original EC solutio n (1 × ), possibly due to the high ammonia - to - phosphate ratio. Algal community assemblage changed drastically in different dilutions of EC solution after a 9 - day culture. The following semi - continuous culture in 2× and 5× EC media established steady biomass productivities and nitrogen removal rate s ; in addition, both conditions illustrated a phenomenon of phosphorus luxury uptake . Biomass composition analyses showed that algae cultured in medium containing higher nitrogen ( 2× EC medium) accumulated more protein but less carbohydrate and lipid than the 5 × EC medium. The last stage involved hydrolyzing the algal biomass cultured in anaerobic digestion effluent and analyzing the effects of the neutralized algal hydrolysate on the performance of enzymatic hyd rolysis of acid and alkali pretreated lignocelluosic substrates (poplar, corn stover, switchgrass, and solid fiber from anaerobic digestion). Results found that algal hydrolysate significantly improved the efficiency of enzymatic hydrolysis of lignin - rich, structurally recalcitrant biomass such as poplar and solid fiber from anaerobic digestion. This discovery broadened the potential application o f algal biomass besides direct use for biofuel production. iv ACKNOWLEDGEMENTS d like to express my deep gratitude and appreciation to my advisor, Dr. Wei Liao, for the generous guidance, mentorship and encouragement he provided to me all the way from when I just started participating graduate research for my m the PhD program in Biosystems Engineering, through to completion of my doctoral degree. Dr. feel truly fortunate to have had the opportunity to work with him. I would also like to thank my committee members, Drs. Yan Liu, Terence Marsh, and Ilsoon Lee for the meaningful guidance, inspiring suggestions and countless recognize D rs. Dana Kirk, R. Jan Stevenson, and Daniel Guyer for the contributions that each of them made to my academic and professional development during my years of study at the Michigan State University. Last but not least, I would like to dedicate my success to my parents, my husband, Jason, and people I love. v TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ............. viii LIST OF FIGURES ................................ ................................ ................................ ............... x Chapter 1 : Introduction ................................ ................................ ................................ ......... 1 1. Literature review ................................ ................................ ................................ ................ 1 1.1. Fundamentals of anaerobic digestion (AD) ................................ ................................ . 3 1.1.1. The biochemistry and microbiology of AD ................................ ........................... 3 1.1.2. Configuration and current technologies of AD ................................ ..................... 6 1.1.3. Molecular genetic analyses of microbial community in AD system ..................... 8 1.1.3.1. Terminal restriction fragment length polymorphism (T - RFLP) ..................... 9 1.1.3.2. Clone library ................................ ................................ ................................ . 10 1.1.3.3. Next Generation Sequencing (NSG) method: 454 pyrosequencing ............. 11 1.2. Fundamentals of algae ................................ ................................ ............................... 12 1.2.1. Algae: an up - and - coming star in chemical and environmental engineering ....... 12 1.2.2. Algal cultivation in pretreated AD effluent ................................ ......................... 14 1.3. Fundamentals of lignocellulose biorefining ................................ .............................. 15 2. Goal, scope and objectives ................................ ................................ .............................. 17 Chapter 2 : Responses of Anaerobic Microorganisms to Different Culture Conditions and Corresponding Effects on Biogas Product ion and Solid Digestate Quality ......................... 19 1. Introduction ................................ ................................ ................................ ..................... 20 2. Materials and methods ................................ ................................ ................................ ..... 21 2.1. Feedstock ................................ ................................ ................................ ................... 21 2.2. Anaerobic digestion systems ................................ ................................ ..................... 22 2.3. Analytical methods ................................ ................................ ................................ .... 23 2.4. Bacterial community analysis ................................ ................................ .................... 23 2.5. Archaeal community analysis ................................ ................................ .................... 24 2.6. Statistical analysis ................................ ................................ ................................ ..... 26 3. Results and discussion ................................ ................................ ................................ ..... 27 3.1. Characteristics of different feedstock ................................ ................................ ........ 27 3.2. Effects of feedstock composition and culture temperature on AD performance ...... 28 3.3. Effects of feedstock composition and culture temperature on anaerobic microbes .. 31 4. Conclusion ................................ ................................ ................................ ....................... 46 Chapter 3 : Using an environment - friendly system combining electrocoagulation process and algal cultivation to treat high strength wastewater ................................ ........................ 47 1. Introduction ................................ ................................ ................................ ..................... 48 2. Material and methods ................................ ................................ ................................ ...... 50 vi 2.1. EC treatment of liquid AD digestate ................................ ................................ ......... 50 2.2. Preparation of algal inoculum ................................ ................................ ................... 51 2.3. Kinetics study of algal culture in EC medium ................................ ........................... 52 2. 4. Semi - continuous algal culture in EC medium ................................ ........................... 53 2.5. Analytical methods ................................ ................................ ................................ .... 53 3. Results and Discussion ................................ ................................ ................................ .... 54 3.1. Kinetics study ................................ ................................ ................................ ............ 54 3.2. Eff ects of EC media on algal assemblage ................................ ................................ . 61 3.3. Semi - continuous algal culture ................................ ................................ ................... 62 4. Conclusion ................................ ................................ ................................ ....................... 67 Chapter 4 : Effects of algal hydrolysate as reaction medium on enzymatic hydrolysis of lignocelluloses ................................ ................................ ................................ ...................... 68 1. Introduction ................................ ................................ ................................ ..................... 69 2. Methods and Materials ................................ ................................ ................................ .... 70 2.1. Lignocellulosic feedstock ................................ ................................ .......................... 70 2.2. Algae ................................ ................................ ................................ .......................... 71 2.3. Pretreatment of feedstock and algal biomass ................................ ............................ 72 2.4. Enzymatic hydrolysis ................................ ................................ ................................ 73 2.5. Analytical methods ................................ ................................ ................................ .... 74 2.6. Statistical analysis ................................ ................................ ................................ ..... 74 3. Results and Discussion ................................ ................................ ................................ .... 75 3.1. Characteristics of fibers and algae ................................ ................................ ............. 75 3.2. Enzymat ic hydrolysis of lignocellulos ic substrates using algal hydrolysate as reaction medium ................................ ................................ ................................ ................ 77 3.3. Combined effect of pretreatment and reaction medium on the improvement of enzymatic hydrolysis ................................ ................................ ................................ ......... 82 4. Conclusion ................................ ................................ ................................ ....................... 84 Summary ................................ ................................ ................................ .............................. 85 APPENDICES ................................ ................................ ................................ ..................... 88 APPENDIX A: Procedure for analyzing structural carbohydrates and lignin content of lignocellulosic biomass ................................ ................................ ................................ ........ 89 APPENDIX B: Procedure for DNA extraction ................................ ................................ . 90 APPENDIX C: PCR procedure for 454 pyrosequencing of bacterial 16S rDNA ............ 92 APPENDIX D: PCR procedure for T - RFLP of archaeal 16S rDNA ................................ 94 APPENDIX E: Procedure for archaeal 16S rDNA cloning ................................ .............. 96 APPENDIX F: Statistical analysis for AD performance analysis ................................ .... 97 APPENDIX G: Control tests for algal growth (in DI water) and nutrient reduction (TN, TP, Iron) in batch kinetics study ................................ ................................ .......................... 99 APPENDIX H: Algal carbohydrate analysis procedure ................................ ................. 101 APPENDIX I: Algal protein analysis procedure (bicinchoninic acid assay) .................. 102 APPENDIX J: Alg al crude lipid extraction procedure ................................ ................... 103 vii APPENDIX K: Statistical analysis for algal growth and nutrient reduction in kinetics study ................................ ................................ ................................ ................................ ... 104 APPENDIX L: Statistical analysis for algae enhanced enzymatic hydrolysis of lignocelluloses ................................ ................................ ................................ .................... 1 06 REFERENCES ................................ ................................ ................................ ................... 111 viii LIST OF TABLES Table 2. 1: Characteristics of feedstock ................................ ................................ ................ 27 Table 2.2: Performance of anaerobic digestion ................................ ................................ .... 31 Table 2.3: The diversity and evenness of bacterial and archaeal communities calculated based on their 16S rRNA gene targeted sequencing. ................................ ........................... 34 Table 3.1: Chemical analysis of culture media before and after semi - continuous culture. . 65 Table 3.2: Algal biomass chemical composition. ................................ ................................ 65 Table 4.1: Pretreatment conditions for each feedstock ................................ ........................ 72 Table 4.2: Structural carbohydrate and lignin content of the raw lignocellulosic biomass . 76 Table 4.3: Structural carbohydrate and lignin of pretreated lignocellulosic biomass .......... 76 Table 4.4: Composition of Algal Biomass ................................ ................................ ........... 76 Table 4.5: Characteristics of algal hydrolysate ................................ ................................ .... 77 Table 4.6: Sugar concentrations and overall glucan and xylan conversions of diff erent lignocellulosic feedstock ................................ ................................ ................................ ...... 78 Table AP.F.1: Two - way ANOVA: Biogas (mL/L AD) versus Temp, Ratio ...................... 97 Table AP.F. 2 : Two - way ANOVA: TS reduced (%) versus Temp, Ratio ............................ 97 Table AP.F. 3 : Two - way ANOVA: Productivity (TS) versus Temp, Ratio ......................... 97 Table AP.F. 4 : Two - way ANOVA: Cellulose in residue versus Temp, Ratio ..................... 97 Table AP.F. 5 : Two - way ANOVA: Xylan in residue versus Temp, Ratio ........................... 9 8 Table AP.F.6: Two - way ANOVA: Lignin in residue versus Temp, Ratio .......................... 9 8 Table AP.F.7: Pairwise comparison: Biogas (mL/L AD) ................................ .................... 9 8 Table AP.F.8: Pairwise comparison: TS reduced (%) ................................ ......................... 9 8 Table AP.K.1: Two - way ANOVA: TS versus Dilution, Time ................................ .......... 104 Table AP.K.2: Two - wa y ANOVA: OD750 versus Dilution, Time ................................ ... 104 ix Table AP.K.3: Two - way ANOVA: Std. Biovolume versus Dilution, Time ...................... 104 Tabl e AP.K.4: Two - way ANOVA: TN versus Dilution , Time ................................ ......... 104 Table AP.K. 5: Two - way ANOVA: TP versus Dilution , Time ................................ .......... 105 Table AP.K. 6: Two - way ANOVA: Fe versus Dilution , Time ................................ ........... 105 Table AP.K .7 : Two - way ANOVA: Turbidity versus Dilution, Time ................................ 105 Table AP. L .1 : Differences of Least Squares Means ................................ .......................... 106 Table AP. L.2 : Tests of Effect Slices ................................ ................................ .................. 110 x LIST OF FIGURES Figure 1.1. The key process stages of anaerobic digestion ................................ .................... 4 Figure 1.2: Flow chart of proposed closed - loop biorefining system ................................ ... 17 Figure 2.1: Effects of feedstock composition and culture temperature on AD performance ................................ ................................ ................................ ................................ .............. 30 Figure 2.2: Daily biogas productivity (daily biogas accumulation vs TS reduction) .......... 30 Figure 2.3: Rarefaction curves for 12 samples obtained by 454 pyrosequencing. .............. 33 Figure 2.4: Dendrograms and heat map of the microbial community. ................................ 36 Figure 2.5: Abundance of bacterial and archaeal communities. ................................ .......... 39 Figure 2.6: Nonmetric Multidimensional Scaling (NMDS) of bac teria and archaea .......... 43 Figure 3. 1: Sketch of column EC reactor . ................................ ................................ ............ 51 Figure 3.2: Algal growth kinetics in diluted EC media based on (a) biomass dry weight (g TS L - 1 ), (b) biomass density (OD 750nm ), and (c) standardized biovolume (mL biomas s per mL culture) . ................................ ................................ ................................ .......................... 55 Figure 3.3: Nutrients including (a) TN, (b) TP, (c) Fe, and (d) turbidity removal by algal growth in various EC media throughout 9 - day culture. ................................ ....................... 57 Figure 3.4: Algae community assembly on day - 0 and day - 9 within various culture media. ................................ ................................ ................................ ................................ .............. 62 Figure 3.5: Algal biomass productivity from two diff erent EC media in semi - continuous culture during steady growth phase. ................................ ................................ .................... 64 Figure 3.6: Comparing color and clarity of various fluids referred in this study. ................ 66 Figure 4.1: Enzymatic glucan conversion of differently treated feedstock using de - ionized water, citrate buffer, and algal hydrolysates as reaction medium. ................................ ....... 79 Figure 4.2: Effects of pretreatment and reaction medium on the improvement of enzymatic glucan conv ersion from different feedstock ................................ ................................ ......... 83 xi Figure AP.1 : Algal growth in DI water based on cell density (OD 750nm , unitless), standardized biovolume (100× dilution, unit: mL cell per mL culture), and biomass total solids (unit: g L - 1 ). ................................ ................................ ................................ ................ 99 Figure AP.2: T otal nitrogen reduction in control culture without algae inoculation.. ......... 99 Figure AP.3: T otal phosphorus reduction in control culture without algae inoculation.. .. 10 0 Figure AP.4: T otal iron reduction in control culture without algae inoculation.. .............. 10 0 1 Chapter 1 : Introduction 1. Literature review With the global gross domestic product rising by an average of 3.6% per year, world energy use is projected to grow from 524 quadrillion Btu in 2010 to 820 quadrillion Btu in 2040 (International Energy Outlook , 2013). C onsequently, the global climate change will be intensified due to the accumulation of greenhouse gases (GHG) tightly associated with the increasing energy consumption. H igh energy prices and environmental concerns of GHG emission have driven many countries to provide incen tives to support the development of alternative energy sources, which makes - growing energy source s (Annual Energy Outlook, 2012) . In the United States, t he comprehensive New Energy for America plan supported by Obama administ ration by 2035 ( Obama , 2011). Among various clean energy sources, not only biofuel s can reduce global warming and acidification emissions in comparison with petroleum fuels, but they also pos s es s advantages such as diversity, consistency, and potential of generating valuable by - products. While most conventional biofuel s manufacturers are still u s in g food crops (i.e. soybean, canola, corn) and animal fat as feedstock, more and more attention and investments are flowing towards new technologies that utilize non - food crops (i.e. algae, poplar, switchgrass) and organic wastes (i.e. corn stover, animal m anure, municipal waste water). In addition to the energy consumption , many human activit ies also generate excess waste s which severely harm the environment. For instance, approximately 54% of 2 municipal solid waste in the U.S. (typically containing 50 - 70% organic material) is directly disposed in the landfill s (U.S. EPA, 2008), and about 72 million tons of farm animal manure is generated and disposed in the open - air compost piles or lagoons every year nationwide (Obour, 2015); the decomposition process of these organic materials can contr ibute significant amount of GHG ( such as methane , carbon dioxide , and nitrous oxide) to t he atmosphere. Another example w ould be inappropriate treatment of waste water and sludge ; these waste streams contain high organic matter (i.e. animal manure and food waste) which can cause eutrophication to freshwater bodies and their watersheds . Moreover, odor and pathogens carried and spread by these wastes also endanger the entire community that relies on such eco system. In order to minimize these environmental impacts, anaerobic digestion (AD) provides full containment of aforementioned potential pollutants and has become a renewed global interest. Historically, AD is one of the oldest processing technologies practiced by mankind . D uring the digestion, a complex anaerobic microbial consortium also converts high concentration of organic matter into biogas , a carbon neutral and renewable energy source , while odor emission and pathogen count are both well - controlled . Although AD shows its advantages in significantly decomposing organic materials and generating clean energy , the process does not reduce or recover nutrients such as phosphorus and nitrogen. Therefore, the liquid AD effluent is required to be further treated in order to alleviate its environmental impacts . Several methods have been frequently applied by the municipal wastewater treatment plants to remove nutrients and reclaim water, which include sedimentation, active carbon adsorption, coagulation, and flocculation (Dean, 1991; Tyagi et al., 2010). However, due to the high chemical and 3 energy consumptions, these treatments become less efficient for large - scale facilities. R ecent studies have prop osed combining sustainable biological techniques and anaerobic digestion might bring wastewater treatment process benefits such as no secondary pollutant generated, possible in situ bioremediation, and economic viability (Vijayaraghavan and Yun, 2008; Pach eco et al., 2015) . Therefore, algal cultivation has emerged as an option for treating wastewater containing high nutrients content, toxins and heavy metals (Mulbry and Wilkie, 2001; McHenry, 2009; Zhang et al., 2012). S ince most algae strains used in nutrient removal are autotrophic and excess insoluble solids in wastewater might weaken the light availability , several recent studies reported more efficient nutrient removal using pretreated wastewater, such as effluent fr om primary decantation, oxidation ditch, and electrocoagulation (Chen et al., 2012a, Ruiz - Martinez et al., 2012; Liu et al., 2015). Integrating AD process with algae culture can also reduce the GHG emission from biogas combustion and accumulate algal bioma ss for potential biofuel and value - added bio - based chemicals. 1.1. Fundamentals of anaerobic digestion (AD) 1.1.1. The biochemistry and microbiology of AD AD is a fermentation process using organic matter in an oxygen free en vironment to produce biogas. I t is a highly cooperative process carried out by series of facultative and obligate anaerobic microorganisms with distinct responsibilities (Bryant et al., 1967) , and the overall process can be described as: C 6 H 12 O 6 3CO 2 + 3CH 4 . 4 Figure 1 . 1 . The key process stages of anaerobic digestion There are four key stages of anaerob ic digestion process: hydrolysis, fermentation (acidogenesis), acetogenesis, and methanogenesis (Figure 1 .1 ). Undigested biomass is usually made of large organic polymers such as carbohydrates, fats and proteins. In order for microbes to carry out the anaerobic digestion and produce biogas, these large polymers must be broken down into smaller constituent monom ers first. Some of t hese monomers (i.e. simple sugars, fatty acids and amino acids) are directly converted into acetate and hydrogen, which will be utilized by methanogenic archaea; the rest monomers need to go 5 through several fermentative stages to furthe r break down the intermediate volatile fatty acids (VFAs) (i.e. propionate, butyrate, succinate, and alcohols) into acetate and hydrogen. And eventually, acetoclastic and hydrogenotrophic methanogenic archaea produce methane, carbon dioxide, and water from acetate and hydrogen, respectively. Because of four different stages in AD process, the anaerobic microorganisms can also be categorized into four groups: hydrolytic bacteria, fermentative bacteria, acetic acid - forming bacteria, and methane - forming archae a. Genera such as Clostridium and Bacillus were isolated and found to be able to produce extracellular enzymes that facilitate the P hyla Firmicutes , Bacteroidetes , Chloroflexi , Thermotogae , and Proteobacteria all contain species that are known for fermentation/acidogenesis (Balk et al., 2002; Ueki et al., 2006; Dong et al., 2000; Yamada et al., 2005). Some typical acetogenesis microorganisms such as genera Syntrophobacter and Syntrophomonas can convert VFAs (i.e. propionate and butyrate) to acetate (Boone and Bryant, 1980; Zhang, 2004), while other acetogens such as Clostridium and Acetobacterium go through a unique Wood Ljungdahl pathway to utilize carbon d ioxide as electron acceptor and hydrogen gas as electron donor to produce acetate, with the acetyl - CoA synthase as the key enzyme (Balch et al., 1977; Müller , 2003; Müller and Frerichs, 2013). As for methanogens, they can be conceptually divided into three classes based on their phenotypic and phylogenetic similarities (Anderson et al., 200 9): Class I methanogens include orders Methanococcales , Methanobacteriales and Methanpyrales ; Class II only includes order Methanomicrobiales and Class III only includes Methanosarcinales . Even though hydrogenotrophic Class I and II methanogens are important in maintaining low hydrogen partial pressure in the digester by using 6 hydrogen/carbon dioxide to produce biogas, the majority of methane is produced from acetate by Cl ass III, the acetoclastic methanogens (Mackie and Bryant, 1981). Although AD process heavily relies on the microbial activity, c ertain criteria can be controlled or regulated by scientists and engineers in order to maintain a healthy and productive system. The o perating temperature directly influences the microbial community and organic decomposition : while thermophilic process (49 - 57 ° C) generally has higher biogas productivity and is able to eliminate pathogens , mesophilic process (30 - 38 ° C) is more resistant to environmental variation s and requires less energy and maintenance cost (Chynoweth and Pullammanappallil , 1996; Monnet, 2003). The pH of the digestate also affects the performance of digestion: high acidity can inhibit the activity of both ac idogens and methanogens, but high alkalinity due to the excess accumulation of unionized ammonia is also toxic to the microbial community ( Mata - Alvarez et al., 2000 ). Consequently, a balanced carbon - to - nitrogen ratio is required to maintain a robust pH buf fering capacity and good gas production , which can be achieved by co - digesti ng more than one type of feedstock ( such as animal manure, food waste, and sewage ) (Monnet, 2003) . Hydraulic retention time (HRT) indicates the average time that liquid digestate remains in the AD reactor and is determined by the reactor design and feedstock, but in general a lower HRT (faster digestion rate) is preferred when evaluating an AD reactor (Mata - Alvarez et al., 2000) . 1.1.2. Configuration and current technologies of AD There is a wide variety of designs for AD reactor based on cost, feeding plan, so lid content and digestion stage . The simplest AD reactor can be constructed in - ground with a heavy - duty synthetic liner and a gas collecting furrow around. It requires minimum 7 investment ; however, this simplified design does not provide steady and controlled operating temperature , which means it will have low efficiency when implemented in regions with long and cold seasons. Most modern designs of AD reactors incorporate element to recycle the residual heat from biogas combustion to maintain its operating temperature. A plug - flow reactor utilizes the force from a pump during feeding and the motion of biogas when escaping from the digestate to achieve mixing, it is effective and ine xpensive when the feedstock has solid content of 10 - 14% . More r igorous mixing provides a more homogenized condition in relatively low solid feedstock (3 - 10% TS ) , which can be performed in a continuously stirred tank reactor, or CSTR: it uses recirculating pumps, propellers, or draft tubes to continuously mix the digestate in order to prevent solid buildup (Demirer and Chen, 2005; Wilkie, 2005) . A microbial activity specific design multi - stage anaerobi c digestion system has emerged and divides the AD process into several sequential stages. Each stage operates in a relatively small reactor with the environment tailored to its dominant microorganisms . It provides a much better refined configuration and an improve d volatile solids reduction comparing with previously mentioned designs, but it also requires higher capital and operational costs (Chenowyth, 1987; Kelleher, 2007) . Feedstock for c onventional AD process used to be one form of organic waste due to specific waste source from different applications ; for instance, AD on a cattle farm uses m anure as the feedstock, while AD i n a wastewater treatment plant deals with sewage sludge. However, recent studies have shown that co - digestion of two or more substrates could significantly enhance the biogas and methane production (Mata - Alvarez et al., 2014; 8 Astals et al. 2014 , 2015 ) : a nimal manure and sewage sludge are commonly good sources for concentrated anaerobic microorganisms and provide relatively robus t pH buffering capacity, although they contain either more recalcitrant organic molecules or higher moisture content; other feedstock such as food waste and pulp/paper waste might come with less methanogenic microorganisms, but they provide more high solub le organic matters (i.e. starch, grease, amorphous cellulose) which can be rapidly converted to volatile fatty acids (Cho et al., 1995 ; Zhai et al., 2015 ) . The combination of these substrates could provide higher organic loading, wider range of biodiversit y, balanced pH and nutrients, and synergistic relationship between substrates, all of which contribute to the improvement of AD performance. 1.1.3. Molecular genetic analyse s of microbial community in AD system Most molecular methods for identifying and classifying bacterial and archaeal use the 16S ribosomal RNA (16S rRNA) genes (sometimes referred as 16S r DNA) as a bio - marker, because ( a ) this gene is only about 1542 base pairs (bp) long, it can be quickly and cheaply copied and sequenced, an d ( b ) this gene is universally present in all bacteria and archaea . In addition, it evolves slowly and has conserved and essential function for each distinct species, which means the slight changes that have occurred provide clues as to how various organisms are clos ely related. Many analy tic approaches have been established and applied to the microbial studies of AD systems since methods for extracting and analyzing DNA from highly diverse microbial communities have improved recently. A variety of methods have been applied in the present study, including terminal restriction fragment length polymorphism (T - RFLP), clone library, and 454 pyrosequencing. Each of these analyses has its own advantage and disadvantages, which will be briefly discussed. 9 1.1.3.1. Terminal restrict ion fragment length polymorphism (T - RFLP) T - RFLP is a community profiling method which measures the size p olymorphism of terminal restriction fragments from a polymerase chain reaction (PCR) amplified marker. In detail, near full length of 16S rRNA gene se quences are first amplified from a mixed community DNA sample using either one or two primers that are fluorescently tagged. This mixed PCR product is then digested using one or several restriction enzymes, the length and quantity of the fragments with flu orescent label are determined by capillary gel electrophoresis. The resulting profiles of all fragments can be compared within and between samples to determine relative abundance and diversity; these profiles can also be compared to a database of fragments generated from known spe cies to tentatively identify these unknown fragment s (Marsh, 1999; Abdo et al., 2006). The primary limitation of T - RFLP method is that it cannot accurately identify a fragment peak of interest to a known species because the fragment itself cannot be directly sequenced. It is common that several species can be assigned to one fragment since the comparison is solely based on the presence and location of a restriction site. Moreover, noise or false peaks can also make T - RFLP pro file interpretation difficult. That is why i t is common to construct a clone library in parallel to the T - RFLP analysis in order to assess and interpret the T - RFLP profile. By comparing the T - RFLP profile to a clone library, it is possible to validate each of the peaks as genuine as well as to assess the relative abundance of each variant in the library (Liu et al., 1997; Moeseneder et al, 2001). For instance, a study using T - RFLP in complement with small clone libraries examined changes in the microbial di versity over a three and a half year period of an AD system operated at psychrophilic condition (4 - 15 °C) (McKeown et al., 2009) . 10 1.1.3.2. Clone library A clone library is a gene bank containing all DNA sequences (clones) from a single organism with accompanying information. With the help of these banks, unknown DNA sequences can be identified. Construction of a clone library involves creating many recombinant DNA molecules. A mixed community DNA extractant is first amplified using a 16S rRNA gene specific primer, and then the amplicons of the 16S rRNA gene fragments are ligated/inserted into cloning vector s (i.e. plasmid s); thirdly, these recombined vectors are transformed/introduced into competent cells (i.e. E. coli ) which will create a DNA library during growth and reproduction; and lastly, a screening process of these host cells using a culture plate with selective reagent is followed in order to select the library of interest . After a library is created, the cloned 16S co nstruct can be sequenced using the Sanger di - deoxy chain termination method to identify the composition of microorganisms in the original mixed community. Although clone libraries of 16S rRNA genes is currently one of the most widely used methods to simul taneously evaluate the composition and diversity of a microbial community, some rare and/or unculturable species can be often left out; in addition, the high biodiversity in a given AD system usually requires a very large number of clone sequences in order to adequately describe the entire community, which is always labor - intensive and expensive. For instance, Rivière et al. (2009) examined nearly ten thousand sequences using clone libraries generated from the AD systems of seven wastewater treatment plants around the world; although , they only discovered six universally dominant operational taxonomic units (OTUs) in all samples. To the contrary, the next - generation high - throughput sequencing technologies such as 454 pyrosequencing can parall el ize the 11 sequen cing process and offer the ability to achieve massive levels of sequence coverage compared to the traditional cloning and sequencing methods. 1.1.3.3. Next Generation Sequencing (NSG) method: 454 pyrosequencing Different from classic Sanger sequencing method, pyrosequencing relies on the detection of pyrophosphate released on nucleotide incorporation instead of chain termination with dideoxy - nucleotides (King and Scott - Horton, 2008; Wheeler et al., 2008) . This method is capable of sequencing about 400 to 600 Me ga b ases of DNA in only 10 hour running time (Voelkerding et al., 2009) ; and because of that, studies have been able to evaluate microbial diversity with a finer resolution in a wider range of environments compared to its predecessors. simply derived from the name of the and was awarded for this technology. The name 454 was the code name which the project used to be referred and has no special meaning (Pollack, 2003) . To prepare for the sequencing, amplified 16S rRNA gene is first fractionated into fragments of 300 - 800 bp with blunt ends where short adaptors are ligated onto. The fragments are then individually attached to a DNA - capture bead and amplif ied by emulsion PCR in an oil - water solution. Finally the beads with amplicons are captured in a PicoTiterPlate on a fabricated substrate and sequenced using GS FLX System (Fakruddin et al, 2012) . But a s with any other molecular sequencing method, pyrosequencing also has its limitations. For instance, this method detects long repeated nucleotides ( also known as homopolymers) which can result in accumulation of sequencing errors; and since most of pyrosequences contain none or only a few errors, accu mulated errors may be interpreted as a rare or pseudo OTU, which changes the estimation of community diversity and richness. 12 Another concern of pyrosequencing is the short length of reads generated, especially when identifying bacterial species. Clarridge (2004) reported that short sequence reads provide less phylogenetic information than those near f ull length of the 16S rRNA gene; although it has been commonly accepted that the information from the first 500 - bp region of 16S rRNA gene is sufficient to dis tinguish various bacteria that are associated with human ( Tang et al., 1998; Patel et al., 2000; Kattar et al., 2001; Hall et al., 2003) . In comparison, another commonly used modern benchtop high - throughput sequencing system - MiSeq from Illumina ® has much higher throughput of data per run ; that is 1.5 - 1.6 Giga bases of data at the speed of 60 Mb per hour. Additionally, MiSeq is considered a more accurate sequencing approach; it has a lower substitution error rate of only 1 substitution per 1000 b ase, and its indel (insertion or deletion of bases) production is very infrequently at the rate of less than 1 per 100,000 bases. However , MiSeq falls short when it comes to the length of reading ( Metzker, 2010; Glenn, 2011; Loman et al., 2012). 1.2. Fundament als of algae 1.2.1. Algae: a n up - and - coming star in chemical and environmental engineering The word algae refers to a diverse group of simple and mostly photoautotrophic organisms typically found in aquatic or moist environment. They vary from single - cell forms (i.e. Chlorella is used as dietary supplement , Microcystis causes harmful algal bloom) to complex multicellular forms (i.e. kelp for culinary use, decorative seawe ed for aquarium). Like many terrestrial plants , algae are able to convert light energy to chemical energy (i.e. sugar, starch) that is useful to all life forms as food and energy sources. On the other hand, many distinct advantages of algae make them stand out from the rest of 13 photosynthetic organisms: the simple cellular structure and lack of developed reproductive organ ensure the majority of carbon, nutrient and energy they utilize contribute to biomass growth and storage. In addition, since algal biomas s is not a substantial food source in many cultures, applying it as feedstock in biofuel and /or bio - based chemical industr ies does not raise concerns to the fuel debate, which is commonly mentioned in biofuel development using crop - based biomass such as soybean and corn (Pimentel and Patzek, 2005; Chisti, 2007) . From environmental point of view, it has been proved that nearly 50% of global carbon dioxide is removed by algae annually, and in return they produce more than 45% of oxygen into the atmosphere (Field, 1998). Moreover, due to the biodiversity and relati vely better robustness to changes in the environment, technologies using algae in the treatment of domestic and industrial wastewater has developed rapidly in the past few years. During the process, algae consume nutrients from the wastewater and carbon di oxide from the aerobic bacterial respiration; meanwhile, aerobic bacteria utilize oxygen produced by algae to decompose the organic materials. Economically speaking, this alternative method can reduce more than 50% of the aeration cost in conventional aero bic wastewater treatment (Pacheco et al., 2015). Last but not least, by varying its culture condition, algal biomass can adapt to accumulate high concentration of carbohydrate, lipid , protein, vitamins and /or minerals , which makes it a versatile feedstock in production of renewable fuels (i.e. ethanol, biodiesel, hydrogen, methane), food and cosmetic additives (i.e. agar, carrageenan , dietary fibers , food coloring, alginate), medicines (i.e. natural iodine, niacin, CoQ10, calcium and magnesium su lfate) , and fertilizers ( Shifrin and Chisholm, 1980; Chelf, 1990; Kay and Barton, 1991 ; Li, 2012 ). 14 1.2.2. Algal cultivation in pretreated AD effluent As briefly introduced in the previous sections , AD process cannot reduce or recover nutrients such as ammonia - nitrogen and phosphate, which are both considered as major pollutants to freshwater bodies. Therefore, AD effluent must be treated before discharged to the environment. Among various processes for nutrient management, alga l cultivation represents one of the best biological treatments with the advantages of faster nutrients uptake, year - round production, and higher photosynthetic efficiency (Kebede - Westhead et ies Program recommended that an integrated approach that combines wastewater treatmen t with algal biofuel production should be researched and developed (Sheehan et al., 1998 ; Mulbry et al., 2008 ). Unfortunately, due to the high turbidity and viscosity of t he AD effluent, it is not an optimal culture medium for photosynthetic algae (Hamdani et al., 2004) ; a pretreatment step to reduce the total solids (TS) content is highly recommended prior to applying to algal cultivation (Barnet et al., 1994 ). In wastewater treatment industry, processes such as coagulation aggregate s suspended solids into larger bodies to facilitate physical separat ion of liquid and solid (Global Health and Education Foundation, 2007). Various types of coagulation are being used to condition water befor e sedimentation and filtration. Conventionally used chemical coagulants like alum, lime, ferric chloride and ferrous sulfate (Sivaramakrishnan, 2008) might increase the capital cost of the entire process and introduce additional waste s to the environment. Instead, by applying electrical current to a waste stream, the suspended particles will change their surface charge and form an agglomeration which can be easily separated . This process is called electrocoagulation (EC) . While EC meth od is good in 15 removing suspended particles , heavy metals and compounds that cause biological/chemical oxygen demand (BOD/COD) in a highly conductive wastewater stream, it cannot recover some soluble organic and ammoniacal compounds (i.e. VFAs, ammonia) (G lobal Advantech, 2011) . Fortunately , these residual compounds can be further removed or reduced in an open - pond algae cultivation system : a mmoniacal compounds provide essent ial nitrogen source for algae, while VFAs can be consumed and converted into CO 2 by some bacteria (dissolved CO 2 is also carbon source for photosynthetic algae). 1.3. Fundamentals of lignocellulose biorefin ing Lignocellulose refers to plant biomass that consists of c arbohydrate polymers (cellulose and hemicellulose) and a class of complex aromatic polymer (lignin). It is the most abundantly available raw material and a n appropriate resource for producing biofuel s and value - added chemicals (Carroll and Somerville, 2009) . However, its heterogeneity and r ecalcitrance in both structure and composition mak e it economically difficult to be directly convert ed to sustainable biofuel (mainly bioethanol) . Therefore, numerous research efforts are dedicated to understand the changes of lignocellulose in the process of chemical/physical/biological conversion and the effect of different treatments on these changes. The process of converting lignocellulosic biomass into ethanol consists of three major steps: pretreatment, enzymatic hydrolysis, and fermentation. Due to the recalcitrant nature of lignocellulose, pretreatment that involves additional acid or alkali under elevated temperature and pressure can facilitate to break the long crysta l l ine structure of cellulose into shorter amorphous fragments and expose more surface area to enzymes. In contrast, hemicellulose is a branched polymer which consists of shorter chains of sugar units; during 16 the pretreatment of lignocellulosic biomass , most hemicellulose can be hydro lyzed into fermentable pentose and hexose sugar monomers such as glucose, xylose, mannose and galactose. However, as the severity of pretreatment increase s , degradation of hemicellulose can lead to the formation of aliphatic acids (i.e. acetic acid, formic acid, levulinic acid) and furan aldehydes (i.e. HMF, furfural), while lignin can be degraded to phenolics and other aromatic compounds; all these aforementioned unfermentable pretreatment by - products can pose inhibition to ethanol producing yeasts . Theref ore, a conditioning step such as washing and dewatering pretreated biomass is recommended to alleviate the inhibition issue ( Jönsson et al., 2013; Chen, 2014 & 2015 ; Ruan et al., 2013 ) . Enzymatic hydrolysis of biomass is a critical stage in biorefin ing process , in which enzymes cleave bonds in pretreated complex cellulose and hemicellulose and release simple sugar molecules (i.e. glucose, xylose) for further fermentation . However, the efficiency of enzymatic hydrolysis of lignocellulosic biomass is usually hindered by the high lignin content (Tatsumoto et al., 1988), because lignin can irreversibly adsorb enzyme proteins ( i.e. cellulase, xylanase, and beta - glucosidase) and reduce their availab ility during the hydrolysis of cellulose . Therefore, high enzyme loading is one of the reasons that encumber the industrialization of lignocellulosic biofuel. R ecent studies demonstrated that dosing exogenous surfactants such as T ween 20 or bovine serum albumin (BSA) prior to cellulase addition could significantl y improve the efficiency of lignocellulose hydrolysis (Tengborg et a l., 2001; Yang and Wyman, 2006) ; it is believed that these surfactants are able to deactivate the enzyme - binding sites on lignin and consequently increase the enzyme availability to the hydrolysis. Although the chemical composition of algae varies with species and culture conditions (Becker, 1994); generally, soft algae (exclude diatoms) 17 tend to accumulate more protein and chlorophyll when the nitrogen and carbon source in the culture medium are sufficient (Piorreck et al., 1984; Becker, 1994; Fleurence, 1999), which makes them an ideal candidate as feedstock and protein - rich cat alyst in a lignocellulose biorefin ery. 2. Goal, scope and objectives The long term goal of th is study is to develop an integrated and closed - loop biorefin ing system that reduces global water and carbon footprint while converting organic wastes into renewable energy . Figure 1 . 2 : Flow chart of proposed closed - loop biorefin ing system The scope of the entire study is shown in Figure 1. 2 . O rganic wastes (livestock manure, food waste, etc.) are collected and diluted to desired organic loading rate (%) before feeding to the anaerobic digester (CSTR design ). Three types of output in different 18 phases are produced from the AD process: biogas, solid digestate (AD fiber) and liquid digestate (AD effluent). The biogas is purified and combusted in a combined heat and power (CHP) cogenerator, from which power is supplied to the following processes suc h as EC and biorefin ing; and the residual heat is utilized to maintain the AD system at a certain temperature (meso - or thermo - philic). AD fiber has similar composition and structure as many secondary energy crops and residues, therefore can be used as fee dstock for biofuel (bio - ethanol) production (Teater et al., 2011; Yue et al., 2010 & 2011; Chen et al., 2012b). Liquid AD effluent is f irst treated using EC process to remove large organic particles which affect the turbidity and viscosity; treated effluen t is then applied to an algae cultivation system as medium. The water coming out of the algae pond contains very low concentratio n of nitrogen and phosphorus, and it can be either directly used for irrigation or further treated for potable water using simp le techniques such as reverse osmosis. And the algal biomass harvested can perform as a co - feedstock as well as a catalyst in the following biorefin ing process. Moreover, the processing effluent from the biorefin ery containing diluted alkali and small orga nic molecules (such as amino acids and VFAs) can be recycled as a nutritious pH buffer to dilute the original organic wastes for the AD . The specific objectives of this study are: (1) developing an optimized condition for anaerobic digestion of dairy manur e - food waste mixture, and assessing the microbial communities among various treatments; (2) culturing freshwater algae on EC treated AD effluent for nutrient removal and biomass production; and (3) using algae (algal hydrolysate) to improve enzymatic hydrolys is of lignocellulosic materials . 19 Chapter 2 : Responses of Anaerobic Microorganisms to Different Culture Conditions and C orresponding Effects on Biogas Production and Solid Digestate Quality Abstract Microbial communities of anaerobic digestion have been intensively investigated in thepast decades. Majority of these studies focused on correlating microbial diversity with biogas production. The relationship between microbial communities and compositiona l changes of the solid digestate (AD fiber) has not been comprehensively studied to date. Therefore, the objective of this study was to understand the responses of microbial communities to different operational conditions of anaerobic co - digestion and thei r influences on biogas production and solid digestate quality. Two temperatures and three manure - to - food waste ratios were investigated by a completely randomized design. Molecular analyses demonstrate that both temperature and manure - to - food waste ratio g reatly influenced the bacterial communities, while archaeal communities were mainly influenced by temperature. The digestion performance showed that biogas productivity increased with the increase of supplemental food wastes, and there were no significant differences on carbohydrate contents among different digestions. The statistical analyses conclude that microbes changed their community configuration under various conditions to enhance digestion performance for biogas and homogenized solid digestate prod uction. 20 1. Introduction Anaerobic digestion (AD) is one of the oldest biotechnologies that mankind has practiced to treat organic wastes for several centuries. A complex anaerobic microbial consortium converts organic matter in the wastes into methane biogas - a carbon neutral and renewable energy source, and correspondingly alleviates the odor and pathogen problems. The classic AD systems often used animal manure or sewage sludge as feedstock to function as both source of nutrients and inoculum of anaerobic microorganisms (Humenik, et al., 2004) . However, due to the structural and nutritional limitation of manure and sludge, single - (Loehr, 1974) . Co - digestion of more than one type of feedstock was hence introduced to enhance AD performance of biogas production and total solids (TS) r eduction (Gou, et al., 2014, Liu, et al., 2009, Mata - Alvarez, et al., 2014) . On the other hand, the overall performanc e of an AD system depends on not only the composition of feedstock, but also operational parameters such as temperature (Safferman, et al., 2012) . The most conventional operational temperature levels are mesophilic (30 - 38 °C) and thermophilic (49 - 57 °C), and it has been proven that operational temperature is o ne of the most important determinants of the microbial community structure in an AD system (Safferman, et al., 2012, Song, et al., 2004) . Numerous studies have been conducted on the microbiology of anaerobic co - digestion system to correlate biogas production with microbial diversity (Dearman, et al., 2006, Lee, et al., 2009, Martin - Gonzalez, et al., 2011, Yu, et al., 2014, Zhang, et al., 2011) . However, the relationship between microbial communities and compositional changes of the solid digestate (AD fiber) have yet been widely reported (Yue, et al., 2013) . Several 21 recent studies have discovered that solid digestate has a similar cellulo se conversion potential with other energy crops and residues such as switchgrass and corn stover, and it can be used as a potential cellulosic feedstock for biorefining of fuel and chemical production (Chen, et al., 2014, 2012, Teater, et al., 2011, Yue, et al., 2010, 2011) . Therefore, a clear understanding on the relationship between mixed feedstock, microbial communities, biogas production, and solid digestate quality should be achieved in order to advance AD technology into a pretreatment unit operation for the next - gener ation fuel and chemical biorefining. The objective of this study was to delineate the responses of microbial communities to changes in substrate composition and reaction temperature of anaerobic co - digestion. Dairy manure was mixed with food waste as the substrates to feed anaerobic digesters. The 16S rRNA gene - based 454 pyrosequencing, Terminal Restriction Fragment Length Polymorphism (T - RFLP) and clone library were used to investigate the communities. Microbial assembly was also correlated with performan ce parameters such as daily biogas accumulation, TS reduction, biogas production, and solid digestate quality (cellulose, xylan, and lignin). 2. M aterials and methods 2.1. Feedstock Fresh dairy manure was collected from the Michigan State University dairy farm ( - 20 °C prior to use. Dairy cows were fed on an alfalfa and corn silage blend diet formulated according to the standard Total Mixed Rations (TMRs) (Nutrition, et al., 2001) . Food waste collected from cafeterias on campus 22 was homogenized using a commercial immersion blender (Waring WSB70, Waring, S tamford, CT) and stored at 4 °C prior to use. 2.2. Anaerobic digestion systems A cont in uously stirred tank reactor (CSTR) was used as the anaerobic digester in this study. Three different weight ratios of dairy manure to food waste were used as feeds for the a naerobic digesters: 100:0, 90:10 and 80:20 (based on dry weight). Each digester contained 5% TS. Two culture temperatures of 35 and 50 °C were tested. The hydraulic retention time (HRT) was 20 days. A completely randomized design (CRD) was applied on both factors of manure - to - food waste ratio and temperature. Six treatments with replicates were cultured on New Brunswick shakers (Eppendorf, Enfield, CT) at 150 rpm for 4 full HRTs (80 days). All digesters had a working volume of 0.50 L with 0.25 L headspace. The digesters were first purged with nitrogen gas for 30 second and then sealed with rubber septum caps. Daily biogas accumulation was measured using a water displacement system. Biogas sample from the digesters was collected for gas composition analysis. All digesters were fed every other day with 50 mL of aforementioned feed. Fresh feed was prepared every 14 days and stored at 4 °C. Before feeding, an equal volume (50 mL) of digestate was removed from the digesters as the digestate samples. 40 mL of the digestate samples were stored at - 20 °C for TS, cellulose, xylan, and lignin analyses. 10 mL of the digestate samples were stored at - 80 °C for microbial community analysis. The pH of all digesters was controlled above 6.70 by dosing 20% sodium hydroxide (NaOH). The operations of sampling, feeding, and pH adjustment were carried on using a Simplicity 888 automatic anaerobic chamber (PLAS Lab, Lansing, MI) purged with a medical grade specialty gas (85% nitrogen, 10% hydrogen and 5% carbon dioxide). 23 2.3. Analyt ical methods Methane and carbon dioxide content were quantified using a SRI 8610c gas chromatograph (Torrance, CA). The system was equipped with a thermal conductivity detector. The detector was kept at 150 °C during the analysis. Hydrogen and helium were carrier gases , and maintained at 21 psi. The biogas sample volume was 100 µL, and the syringe was purged three times before sample injection. Fiber composition of the digestate was analyzed according to the National Renewable Energy Laboratory (NREL) Analytical Procedu re (LAP) (Sluiter, et al., 2008) (A PPENDIX A) . The free sugars and starch was analyzed using a commercial starch assay kit (Catalog No. SA20. Sigma - A ldrich Co. LLC, St. Louis, MO). 2.4. Bacterial community analysis A Power - Soil DNA isolation kit (MO BIO Laboratories, Carlsbad, CA) was utilized to extract community genomic DNA from digestate samples (APPENDIX B) , and a NanoDrop TM Lite spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA) was applied to quantify the DNA extractions. Polymerase Chain Reactions (PCR) were conducted to ampli - CCTACGGGAGGCAGCAG - - CCGTCAATTCMTTTRAGT - as the forward and reverse primers, respectively (Brinkman, et al., 2011, Haas, et al., 2011, P reidis, et al., 2012) which targeted on the hypervariable V3 - V5 region of rRNA genes (APPENDIX C) 75 µL reaction solution contained 0.2 µM of each primer, 0.6 µL of high fidelity Taq polymerase (5 U µL - 1 ) (Life Tec hnologies TM , Grand Island, NY), 7.5 µL 10× AccuPrime PCR Buffer 24 II, and 20 - 25 ng DNA template. The reaction solution was mixed with DNase and RNase free water for PCR reaction. The amplification included an initial denaturing step at 95 °C for 5 minutes, f ollowed by 30 cycles of 3 temperature steps (denaturing at 95 °C for 45 seconds, annealing at 50 °C for 45 seconds, and elongation at 72 °C for 90 seconds), and a final extension at 72 °C for 5 minutes. The PCR products were purified using QiaQuick PCR Pro duct Purification kit (Qiagen, Valencia, CA). Purified amplicons were diluted to 0.5 ng dsDNA µL - 1 and sequenced using a Roche 454 GSFLX Titanium Sequencer at the Research Technology Support Facility of Michigan State University. All bacterial 16S rRNA amp licon sequences were trimmed, screened and analyzed using Ribosomal Database Project (RDP) Pyrosequencing Pipeline Initial Process tools (Cole, et al., 2014) with a minimum sequence length of 300 bp and no ambiguous bases. Chimeras were identified using USEARCH implemented UCHIME algorithm in reference mode with Silva Gold Alignment database (Edgar, et al., 2011) . Sequences were assigned with genus names at 80% confidence level by RDP Multi - Classifier and clustered at 97% similarity by Complete Linkage Clustering (Yue, et al., 2013) . 2.5. Archaeal community analysis DNA extra ctions from the previous step were also used for archaeal community analysis. The archaeal communities were examined using 16S rRNA gene - based Terminal Restriction Fragment Length Polymorphism (T - RFLP). The 16S rRNA gene was amplified (APPENDIX D) with arc haeal domain - specific primers 344aF - FAM (FAM - 5' - CGGGGYGCASCAGGCGCGAA - - GGYRSGGGTCTCGCTCGTT - (Koch, et al., 2006, Yue, et al., 2013) . A 100 µL reaction solution containin g 20 to 40 ng DNA template, 90 µL Platinum® PCR SuperMix (Invitrogen, Life Technologies 25 Corpora tion, Carlsbad, CA), 0.25 µM forward primer, 0.2 µM reverse primer, and 0.1 mg mL - 1 bovine serum albumin (BSA) was prepared for PCR reaction. The amplification included an initial denaturing step at 94 °C for 5 minutes, followed by 25 cycles of 3 temperatu re steps (denaturing at 94 °C for 1 minute, annealing at 50 °C for 45 seconds, and elongation at 71 °C for 100 seconds), and a final extension at 72 °C for 5 minutes. The PCR products were then purified using QiaQuick PCR Product Purification kit (Qiagen, Valencia, CA). The purified PCR products were subjected to restriction enzyme digestion with MspI (New England Biolabs Inc., Ipswich, MA). A 15 mL digestion mixture contained 300 - 400 ng of purified PCR product, 1.5 µL 10X enzyme buffer, 0.5 µL enzyme (20 U µL - 1 ), and 0.1 mg mL - 1 BSA. The digestion mixture was incubated at 37 °C for 3 hours and deactivated at 65 °C for 10 minutes. The digested DNA samples (7 µL) were sequenced at the Research Technology Support Facility at Michigan State University. In order to construct archaeal clone libraries, the 16S rRNA genes of four representative samples were amplified with archaea domain - specific primers of 344aF and 1119aR. Unlike the forward primer 344aF - FAM used in previous T - RFLP experiment, 344aF does not contai n any fluorescent label (FAM). TOPO TA cloning kit (Invitrogen, Life Technologies Corporation, Carlsbad, CA) with One Shot ® TOP10 Chemically Competent Escherichia coli was used for cloning (APPENDIX E) . A total of 192 clones were picked and screened for ea ch sample, and 96 clones with correct inserts were sequenced at the Research Technology Support Facility of Michigan State University. Archaeal 16S rRNA gene sequences obtained from clone libraries were processed using RDP. Phylogenetic affiliations were a nalyzed using the RDP Classifier at an 80% confid ence threshold at genus level. 26 2.6. Statistical analysis Two - way analysis of variance (ANOVA) and pair - wise comparison using Statistical Analysis System program 9.0 (SAS Institute Inc., Cary, NC) were conducted on biogas production, TS reduction, and AD fiber composition to compare the AD characteristics and performance among six experimental treatments (APPENDIX F) . Statistical software R with package Vegan was used to perform the operational taxonomic unit (OT U) - and phylotype - based analyses of both bacterial and archaeal communities. Specifically, non - metric multidimensional scaling analysis (NMDS) was used to correlate the dissimilarities between samples and the variations in phylotype abundance. The R packag e Vegan was also applied to estimate the diversity index H J ) and rarefaction curve for each sample based on clustered sequences. The sampling coverage ( C ) of each bacterial sample was C =1 - n 1 / N , where n 1 represents number of OTUs, and N is the total number of sequences in the sample (Esty, 1986) . Peak Scanner TM Software v1.0 (Applied Biosystems®, Life Technologies Corporation, Carlsbad, CA ) was applied to perform DNA fragment analysis; peaks below 50 fluorescence units were filtered out to eliminate the background noise. Comparisons of T - RFLP results among samples were conducted by T - Align (Smith, et al., 2005) . 27 3. Results and discussion 3.1. Characteristics of different feedstock Table 2.1 presents the characteristics of dairy manure and food waste, which includes total carbon (C), total nitrogen (N), free sugars and starch, structural carbohydrates (cellulose and xylan), as well as lignin. It is noticeable that even though food waste cont ained a similar amount of total carbon (47.8%) with dairy manure (43.7%), it contained significantly more available carbon ( p = 0 .001) but less cellulose ( p = 0 .05) and xylan ( p = 0 .007). This is due to the difference in diets between ruminant and human: d airy cows were fed on alfalfa and corn silage, both of which are fibrous lignocelluloses; while food waste from the university cafeterias contained less fiber but more free sugars and starch (34.6%, Table 2.1 ). Table 2 . 1 : Characteristics of feedstock * Dairy manure Food waste Comparison Total carbon (wt%) 43.7 47.8 p = 0.061 Total available carbon (wt%) 24.8 39.9 p = 0.001 Free sugars and starch (wt%) Not detectable 3 4 .6 p < 0.001 Cellulose (wt%) 22.7 16.5 p = 0.005 Xylan (wt%) 13.9 4.9 p = 0.007 Lignin (wt%) 28.4 11.9 p = 0.009 Total nitrogen (wt%) 2.1 5.3 p = 0.005 C:N ratio , total C 20.6 9.0 - C:N ratio, available C ** 11.8 7.5 - *: Data listed represent the average of two biological replicates. **: Available carbon excludes organic matters (i.e. lignin) that do not participate in AD . 28 high protein diet (meat and dairy). C/N ratio in the organic material plays a crucial role in the anaerobic digestion. A high C/N ratio indicates a rapid consumption of nitr ogen by the microbes, which could result in a reduction of biogas productivity (Verma, 2002) . While, a lower C/N ratio leads to carbon deficiency, ammonia accumulation and pH increase, all of which inhibit reproduction and metabolism of the methanogens (Chen, et al., 2008, Monnet, 2003) . Therefore, mixing food waste with dairy manure could provide a balanced C/N for a healthy and efficient AD process (Khalid, et al., 2011, Sievers and Brune, 1978, Weiland, 2006) . 3.2. Effects of feedstock composition and culture temperature on AD performance Figure 2. 1 and Table 2 .2 illustrate the performance of digesters under six different conditions. In general, methane content in biogas ranged from 58.3 to 67.6% without significant difference among all experimental runs ( p = 0 .117). Daily biogas production and carbon removed from the feed were significantly correlated to supplemental food waste ( p = 0 .001 and 0.004, respectively) and reaction temperature ( p = 0 .001 and 0.005, respectively). The interactions between these two factors also had significant impacts ( p = 0 .001 and 0.04 1, respectively) on the daily biogas production and carbon removal from the feed. With the increase of the food waste percentage, both mean biogas production and carbon removal were significantly improved. On the contrary, fiber analysis demonstrated that neither reaction temperature nor supplemental food waste had impacts on cellulose ( p = 0 .632 and 0.522, respectively), xylan ( p = 0 .478 and 0.253, respectively), and lignin ( p = 0 .998 and 0.165, respectively) contents in the solid digestate. In other words , this study demonstrated that within a certain range of variation in feedstock composition (mainly 29 controlled by available C/N ratio), AD system could adjust itself in order to maximize carbon utilization for biogas production and generate homogenized sol id digestate with similar carbohydrate content. Considering the potential application of solid digestate as a feedstock for biorefining of biofuel and chemical production (Chen, et al., 2014, 2012, Teater, et al., 2011, Yue, et al., 2010, 2011) , the AD function of homogenizing carbohydrate components in the solid dige state from different feeds might provide a solution to address the compositional diversity the issue of different lignocellulosic feedstock that lignocellulosic biorefining processes encounter. Tang et al. (2011) found similar functions among microorganism s belonging to different families and even different phyla. Their co - existence and functional redundancy could ensure a more effective and economic digestion to maximize carbon utilization and methane production . O Sullivan et al. (2005) also reported such adjustment of microbial community structure to enhance methane production and generate solid digestate with minimum composition changes. In order to further evaluate the AD performance between different combination of temperature and feed ratio, daily bio gas productivity was introduced to conduct the comparison. The daily biogas productivity [mL g - 1 TS reduction L - 1 digestion] was calculated by dividing the daily biogas accumulation [mL L - 1 digestion day - 1 ] by daily TS reduction [g TS reduction day - 1 ]. The biogas productivity data (Figure 2 .2 ) present that within the boundary of the experimental conditions temperature had no significant influence (p = 0.206) on biogas production. While, increasing the percentage of food waste in the feed significantly incre ased the biogas productivity (p = 0.002). It is apparent that easy - hydrolyzing carbon sources of free sugars and starch in food waste enhanced the AD performance to produce biogas, homogenized fiber quality of the solid digestate. 30 Figure 2 . 1 : Effects of feedstock composition and culture temperature on AD performance Figure 2 . 2 : Daily biogas productivity (daily biogas accumulation vs TS reduct ion) 0 20 40 60 80 100 0 300 600 900 1200 1500 100:0 90:10 80:20 100:0 90:10 80:20 35 °C - 50 °C TS reduction (%) Biogas production (mL L - 1 digestion day - 1 ) 0 200 400 600 800 1000 100:0 90:10 80:20 Biogas productivity (mL g - 1 TS L - 1 digestion) 35 °C 50 °C 31 Table 2 . 2 : Performance of anaerobic digestion * Daily b iogas (mL L - 1 AD day - 1 ) CH 4 (%) Carbon removal from the feed ** (%) Cellulose in residue (%) Xylan in residue (%) Lignin in residue (%) 35 °C 100/0 558.9 5 8.3 27.4 23.1 14.3 36.4 90/10 499.1 65.2 24.2 22.3 13.5 36.4 80/20 626.5 60.1 30.2 23.9 13.6 34.4 50 °C 100/0 554.4 59.0 27.2 25.3 14.8 38.0 90/10 642.1 5 8 .6 31.2 22.4 13.8 35.0 80/20 848.8 6 7 .6 40.9 23.4 13.7 34.3 *: Data listed represent the average of two biological replicates. **: Carbon removed from the feed means that the percentage of carbon in the feed has been consumed for biogas production. 3.3. Effects of feedstock composition and culture temperature on anaerobic microbes The aforementioned performance results demonstrate that AD can efficiently adjust itself to adapt into different nutrient conditions to maintain the performance of digestion. Considering that anaerobic microbes are the powerhouse of AD, the relationship between microbial communities, digestion conditions, biogas production, and digestate composition should be delineated in order to better understand microbial responses to digestion conditions, and enable engineering of anaerobic microbial communities to fulfill both biogas production and lignocellulose pretreatment. Metagenomic analysis was carried on to elucidate such relationship. The bacterial and archaeal communities were separately discussed in this section. 32 3.3.1. Anaerobic bacterial commun ity The pyrosequencing results demonstrate that the total bacterial 16S rRNA gene sequences in a sample ranged from 1,594 to 30,295 among 12 digestate samples (Table 2. 3). Although rarefaction curves ( Figure 2.3 ) demonstrate a great unsampled diversity ( C ) ranged from 89.0% to 98.8% with an average of 94.5%. Allers et al. (Allers, et al., 2008) community and they found that most of the gammaproteobacteria were covered with the bers from all 12 AD samples indicated a high sampling coverage of bacterial community. Additionally, a recent study by McMurdie and Holmes (McMurdie and Holmes, 2014) argued that if the total numbers o f sequences are vastly different among samples, commonly used rarefaction (or individual - based taxon resampling) technique is inadequate when comparing the relative proportions of each microbial strain across the entire community, rarefied counts might ove rlook the over - dispersion among biological replicates and suffer from a loss of power. Given the numbers of sequences in this study varied immensely among treatments and biological replicates, the normalization/rarefaction of the sequence numbers was not a pplied. 33 Figure 2 . 3 : Rarefaction curves for 12 samples obtained by 454 pyrosequencing. The curves demonstrate a great unsampled diversity across all 12 digesters. 34 Table 2 . 3 : The diversity and evenness of bacterial and archaeal communities calculated based on their 16S rRNA gene targeted sequencing. Temp R atio ID Bacteria Archaea N bact a OTU obs b C (%) c H bact d J bact e H arc d J arch e 35°C 100 : 0 MI1 30295 368 98.79 2.67 0.55 1.67 0.80 MI2 13987 227 98.38 2.78 0.59 1.85 0.77 90 : 10 MI3 13589 224 98.35 3.08 0.64 2.06 0.83 MI4 3492 282 91.92 3.07 0.71 1.98 0.77 80 : 20 MI5 4126 277 93.29 3.25 0.73 1.84 0.77 MI6 4116 301 92.69 3.46 0.75 1.74 0.76 50°C 100 : 0 MI7 3770 328 91.30 2.52 0.58 1.80 0.72 MI8 10416 220 97.89 2.73 0.58 1.52 0.85 90 : 10 MI9 10849 213 98.04 2.23 0.43 1.59 0.82 MI10 1594 176 88.96 2.81 0.66 1.66 0.80 80 : 20 MI11 2988 247 91.73 2.65 0.62 1.63 0.84 MI12 2689 204 92.41 2.62 0.59 1.60 0.82 Note: a. N bact is the total bacterial 16S rRNA gene sequences in the sample. b. OTU obs is the number of observed OTUs for an OTU definition. c. C d. H bact represents bacteria and arch represents archaea. e. J repres ents bacteria and arch represents archaea. Based on bacterial 16S rRNA gene targeted sequencing, bacterial diversities ( H bact ) of mesophilic digesters were much higher than thermophilic ( p = 0 .006). Several previous studies on anaerobic digestive microbial community also revealed similar trend (Sekiguchi, et al., 1998, Tiago, et al., 2004, vanLier, 1996) , which could be the reason why mesophilic AD is more robust to environmental changes than thermophilic process. However, the J bact ) indices did not show any significant difference among treatments 35 between OTUs. In addition, a combined dendrogram and heat map was generated to demonstrate the similarity of bacterial communities across all samples ( Figure 2.4 - a). Starting f rom the top of the dendrogram (left - side of the figure) , the first separation of clade s shows a community shift caused by reaction temperature. The cluster with samples MI7 - MI12 is thermophilic digesters (50 °C) and the one with samples MI1 - MI6 is mesophilic digesters (35 °C). At both temperatures, the digesters fed on dairy manure only (MI1 & 2 at 35 °C, and MI7 & 8 at 50 °C) were significantly differentiated from the ones fed on the mixture feed with 80:20 ratio (MI5 & 6 at 35 °C, and MI11 & 12 at 50 °C). Meanwhile, the bacterial communities in 90:10 digesters (MI3 & 4 at 35 °C, and MI9 & 1 0 at 50 °C) behaved like an intermediate state between the other two feed ratios, and their replicates illustrated closeness to either 100% dairy manure or 80:20 ratio digesters. This result indicated that bacterial community of an AD system gradually shif ted its structure with the change of the feedstock. Yue et al. (2013) also observed a bacterial community shift by supplementing corn stover into a dairy manure AD system. A heat map of the most abundant bacterial genera in 12 samples ( Figure 2.4 - a) demonstrate s a higher microbial density and diversity appeared in mesophilic digesters (MI1 - 6); moreover, digesters had higher manure content in the feedstock generally had higher microbial density. 36 a. Dendrogram and heat map of bacterial community (based on the most common genera from 454 pyrosequencing) in 12 samples Figure 2 . 4 : Dendrograms and heat map of the microbial community. 37 Figure 2.4 (cont d) b. Dendrogram and heat map of archaeal community (based on T - R FLP results) in 12 samples 38 - Classifier with a minimal bootstrap value of 80 was used to determine the bacterial taxa. A total of 23 phyla were assigned and overall 8.6% of total sequence was categorized as unclassified bacteria. At genus level, a total of 363 bacterial groups (275 classified and 88 unclassified) were identified. Bacteroidetes (46 - 69% at 35 °C, 16 - 28% at 50 °C), Firmicutes (20 - 45% at 35 °C, 45 - 62% at 50 °C), Proteobacteria (2 - 5% at 35 °C, 4 - 7% at 50 °C) and Spirochaetes (1 - 8% at 35 °C) were the most abundant phyla in all 6 treatments (12 digesters) ( Figure 2.5 - a). In addition, Thermotogae (18%) was only observed in thermophilic digesters with the 80:20 ratio. Synergistetes ( 1 - 2%) in mesophilic digesters and Chloroflexi (8 - 14%) in thermophilic digesters were also major components of their microbial communities ( Figure 2.5 - a, wide columns). Within these phyla, Clostridia (19 - 41% at 35 °C, 44 - 61% at 50 °C), unclassified Bacteroi detes (30 - 38% at 35 °C, 2 - 4% at 50 °C), Petrimonas (4 - 7% at 35 °C, 6 - 8% at 50 °C) and Bacteroides (1% at 35 °C, 1 - 2% at 50 °C) were highly abundant ( Figure 2.5 - a, thin columns). Thermophilic digesters tended to accumulate more Firmicutes while mesophilic o nes had significantly more Bacteroidetes . Class Clostridia comprised 91 - 98% of the phylum Firmicutes across 6 runs (12 digesters). Within phylum Bacteroidetes , unclassified Bacteroidetes was a significant component (p < 0.001) in mesophilic digesters. In a ddition, the fractions of Petrimonas in all 6 runs were similar (5 - 8%), but total amount of Bacteroides in the AD treatments was significantly lower than that in original dairy manure. 39 a. Assembly of dominant bacteria, where wide columns indicate dominant bacterial phyla and thin columns indicate dominant bacterial classes. b. Assembly of dominant archaea, based on T - RFLP results Figure 2 . 5 : Abundance of bacterial and archaeal communities. 0% 20% 40% 60% 80% 100% Original 100 : 0 90 : 10 80 : 20 100 : 0 90 : 10 80 : 20 manure - 35 C - 50 C Abundance of dominant bacteria Firmicutes (phylum) Bacteroidetes (phylum) Chloroflexi (phylum) Proteobacteria (phylum) Synergistetes (phylum) Thermotogae (phylum) Spirochaetes (phylum) Actinobacteria (phylum) Clostridia 0% 20% 40% 60% 80% 100% Original 100 : 0 90 : 10 80 : 20 100 : 0 90 : 10 80 : 20 manure - 35 C - 50 C Abundance of dominant archaea OTU1 OTU2 OTU3 OTU4 OTU5 OTU6 OTU7 OTU8 OTU9 OTU10 OTU11 OTU12 OTU13 OTU14 OTU15 OTU16 OTU17 OTU18 OTU19 OTU20 40 Figure 2.5 ( cont d) c. Clone library of archaeal communities at different temperature settings Phylum Bacteroidetes as one of the major bacterial groups in AD include several strains such as Flavobacterium johnsoniae , Sporocytophaga myxococcoides , and Cytophaga sp. that have been repeatedly reported as degraders of structural carbohydrates of plants (Coughlan and Mayer, 1992, L ednicka, et al., 2000, Mullings and Parish, 1984) . A recent study on bacterial community in anaerobic digesters (Yang, et al., 2014 ) also showed that unclassified Bacteroidetes was one of dominant taxa in lignocellulose - rich co - digestion systems. Besides cellulose/hemicellulose degradation, it has also been reported that many members of Bacteroidetes are proteolytic bacteria which can degrade protein and convert amino acids to acetate (Riviere, et al., 2009, Zehnder, 1988) . Class Clostridia was another major bacterial group in anaerobic digestion. As saprophytic bacteria, they commonly show high cellulolytic activity as well as capability of degrading volatile fatty acids such as butyrate and its analog compounds , which indicates these bacteria play an important role in cellulose degradation during the AD 10% 70% 20% 35 ° C 9% 90% 1% 50 ° C Methanobacterium Methanosarcina Methanobrevibacter Methanoculleus 41 (Riviere, et al., 2009 ; Wirth, et al., 2012 ; Tang et al. , 2004; Chouari et al. , 2005; Goberna et al. , 2009; Sasaki et al. , 2007; Weiss et al. , 2009; Sasaki et al. , 2011 ; Wiegel et al. , 2005; Goberna et al. , 2009; Sasaki et al., 2011; Tang et al., 2011 ) . Moreover, some strains of Clostridia can also utilize cellobiose and glucose generated from carbohydrate degradation to produce proton and hydrogen gas (Yang, et al., 2014) . Chlorflexi, Synergistetes, Spirochaetes and Thermotogae are other phyla that have been detected in the digesters. It has been reported that Chloroflexi have potential to treat wastes in anaerobic environment, such as thriving in naturally anaerobic dechlorination (Chandler, et al., 1998) , wastewater treatment processes (Bjornsson, et al., 2002) , and degrading carbohydrates (Riviere, et al., 2009, Sekiguchi, et al., 2001) . Synergistetes are able to consume amino acids and produce short chain fatty acids as well as sulphate for methanogenic archaea and sulphate - reducing bacteria (Vartoukian, et al., 2007) . They prefer mesophilic environment (Ganesan, et al., 2008) as was shown in this study. Spirochaetes can break down cellulose and other plant polysaccharides, and their optimum living temperature is also mesophilic (Lee, et al., 2013) . Noticeably, Thermotogae only appeared in thermophilic digesters with 80:20 ratio that had the highest biogas productivity among all treatments, which may be related to their capab ility of degrading different complex - carbohydrates and producing acetic acid, carbon dioxide and hydrogen gas (Conners, et al., 2006) . Non - metric multidimentional scaling (NMDS) analysis was performed based on the complete linkage clustering of 16S rRNA gene sequences of all 12 digesters (6 treatments with duplicates) ( Figure 2.6 - a). The differences of bacterial communities between two reaction temperatures were significant ( p = 0 .001), though, the supplemental food waste 42 did not have significant impact on the community s hift ( p = 0 .148). The biogas productivities were significantly different among treatments ( p = 0 .012), and the direction of its arrow indicates that digesters with 80:20 feed ratio had the highest biogas productivity ( Figure 2.6 - a). Similarly, the arrow of TS reduction shows an improved performance trend with elevated temperature and no - supplemental food wastes, even though the difference was not significant ( p = 0 .453). Fitting the dominant bacetrial taxa to the community distances reveals that phyla Bacte roidetes (p < 0.001), Synergistetes ( p = 0 .038) and Spirochaetes ( p = 0 .028) preferred mesophilic AD condition, while phyla Chloroflexi (p < 0.001), Thermotogae ( p = 0 .019) and Firmicutes ( p = 0 .004) tended to acccumulate more at thermophilic condition (50 °C). In addition, Firmicutes ( p = 0 .035) prefered the increased amount of supplemental food waste, while Bacteroidetes ( p = 0 .003) had higher abundance in 100% dairy manure digesters. Although both phyla Firmicutes (especially class Clostridia ) and Bacteroidetes were reported to be able to degrade crystalline fiber into organic acids (Flint, et al., 2008, Wan, et al., 2013, Yue, et al., 2013) , Sundberg et al. (2013) repo rted that Bacteroidetes were more susceptible to the environmental change caused by additional food waste, such as ammonia accumulation and pH fluctuation. The correlation between AD performance and bacterial community change also becomes obvious on this N MDS diagram. Bacterial communities tended to adapt themselves into different culture conditions and maximize their capability to convert all available carbon sources (free sugar, starch, protein, fat, hemicellulose, and easy - degradable cellulose) into biog as (Figure 2. 1 and Table 2 .2 ). As a result, differences in compositional cellulose ( p = 0 .626), xylan ( p = 0 .128) and lignin ( p = 0 .113) of the solid digestates among all six treatments were not significant, which means the bacterial 43 metabolism reached an equilibrium for each treatment and relatively homogenized carbohydrate composition in the solid digestate. a. NMDS diagram of bacteria Figure 2 . 6 : Nonmetric Multidimensi onal Scaling (NMDS) of bacteria and archaea . T he blue solid arrows demonstrate dominant phyla; the blue dashed arrows demonstrate dominant classes or genera; the ellipses demonstrate the dispersion of each factor using standard error of the weighted average scores. 44 Figure 2.6 (cont d) b. NMDS diagram of archaea 3.3.2. Anaerobic archaeal community H arch ) calculated from the aligned and clustered (0.03% cutoff) sequences (Table 2. 3) were relatively low, which indicated a relatively low diversity within ar evenness indices ( J arch ) of archaea also showed that archaeal communities had less variation. In Figure 2.4 - b, the archaeal dendrogram demonstrates community similarity across all treatments, and the heat map shows that several archaeal OTUs had higher density within mesophilic digesters. When temperature increased, archaea in the co - 45 digestion systems also shifted accordingly, though, the ones in 100% manure digesters were relatively consistent regardless o f temperature change. Statistically, reaction temperature ( p = 0 .001) had significant impacts on the change of archaeal community while the amount of supplemental food waste did not ( p = 0 .441). The community abundance of archaea based on T - RFLP test ( Figu re 2.5 - b) showed a relatively uniform assembly across all treatments. However, they were all significantly different from the archaea community in the original dairy manure. Moreover, similar to bacteria, archaeal communities in mesophilic digesters were m ore diverse than thermophilic digesters. Further phylogenetic affiliations based on clone library illustrated four genera of methanogenic archaea were detected in the digesters ( Figure 2.5 - c). In details, the abundance of Methanosarcina increased from 70% to 90% when reaction temperature was raised from 35 °C to 50 °C, while Methanobrevibacter reduced from 20% to non - detectable. Results also demonstrate a higher hydrogentrophic methanogen assembly (i.e. Methanobrevibacter , Methanobacterium and Methanoculleu s ) in mesophilic digeseters. The abundance change of Methanobrevibacter due to temperature was expected since the optimum temperature for both genera was 37 - 38 °C (Miller and Lin, 2002, Zellner, et al., 1987) . Methanocarsina is a genus that uses aceticlastic pathway to generate methane (Yue, et al., 2013) ; therefore, its dominance in the clone libraries illustrates that aceticlastic reactions of methanogensis were the dominant route to methane in all digesters. The NMDS analysis of archaeal community ( Figure 2.6 - b) shows the methane content in biogas was similar among all treatments ( p = 0 .117). The direction of the arrow illustrates that digesters at 35 °C had relatively higher methane content. Similar observation was reported previously (Gallert and Winter, 1997, Hashimoto, et al., 1981, 46 Mackie and Bryant, 1995) . The biogas productivities and TS reduction were discussed in previous bacterial NMDS section. Fitting the dominant archaeal genera to the community distances demonstrated that increasing the reaction temperature had significant impact on Methanosarcin a ( p = 0 .001) positively, but negatively on Methanobrevibacter ( p = 0 .031). 4. Conclusion A variety of molecular and statistical approaches were applied to examine the responses of microbial communities to the changes of digestion conditions and their impacts on biogas production and solid digestate quality. The biogas productivity increased si gnificantly with the increase of supplemental food waste. Reaction temperature did not show any significant effect within the experimental conditions. There were no significant differences on carbohydrate contents of solid digestate among six treatments. B oth Firmicutes and Bacteriodetes were dominant phyla found in all treatments; however, more Firmicutes were observed at higher digestion temperature and higher food waste content of the feedstock, while Bacteroidetes were prevailing in the mesophilic diges ters with higher manure content. The similarity of methane content among all six treatments and the analysis of archaea community b oth proved that methanogen community was lack of variation and it was only affected by reaction temperature. The co - existence of functional ly similar /redundant microorganisms (both bacteria and archaea) guaranteed rapid and effective utilization of organic matters for biogas production, which could also explain the relatively homogenize d composition of the solid fiber after AD regardless the deviation of community assembly . In - depth studies on the AD function of homogenizing solid digestate are urgently needed in order to develop an AD - based pretreatment method for lignocellulosic biorefining of biofuel and chemical production. 47 Chapter 3 : Using an environment - friendly system combining electrocoagulation process and algal cultivation to treat high strength wastewater Abstract This study investigated an alternative treatment approach using electrocoagulation (EC) and algae culture to reduce excess nutrients and turbidity in the liquid anaerobic digestion effluent as well as to accumulate algal biomass for potential chemicals and /or biofuels production. Batch culture demonstrates similar maximum growth rate (0.201 - 0.207 g TS L - 1 day - 1 ) from two dilutions (2 × and 5 × ) of the EC solution. E xcess ammonia was one possible growth inhibition factor for the culture in 1× EC medium, and hi gh nitrogen - to - phosphorus ratio might also have limited the growth of algal biomass. In addition, community assemblage of the fresh water alga e changed significantly in different dilutions of EC medium after 9 - day cultivation. Semi - continuous culture estab lished steady biomass productivities and nitrogen removal in 2 × and 5 × EC media . H owever, both conditions exhibited an increase of phosphorus removal rate which could be explained by the luxury uptake theory. Biomass composition analysis proved that algae cultured in medium with higher nitrogen concentration accumulated more proteins but less carbohydrates and lipids. 48 1. Introduction Technologically sound and cost effective waste management is crucial to both municipal and agricultural development and corr esponding well - being. While many conventional waste treatment plants are still using either landfill or relatively expensive and harsh chemicals to handle organic wastes, more and more modern operations have adopted inexpensive and environment - conscious ap proaches to reach the same goal. Anaerobic digestion (AD) is a sustainable technique which has been served for waste treatment and biofuel generation for centuries (Mata - Alvarez et al., 2000) : the process uses microorganisms to convert biodegradable materials to combustible biogas in the absence of oxygen (Chen et al., 2015; Yue et al., 2010) . Although AD proc ess significantly confines organic wastes and reduces the number of pathogens (Teater et al., 2011; Yue et al., 2013; Yue et al., 2010; Yue et al., 2011) , nutrients such as ammonium, nitrate and phosphate are remained and concentrated in the liquid digestate (Chen et al., 2012a; Chen et al., 2012b; Liu and Vyverman, 2015; Liu et al., 2015) . The most common application of this nutritious digestate is to directly apply in the field as liquid fertilize r (Chen et al., 2015) ; however, since arable soil tends to retain less nutrients from liquid ferti lizer than its solid form, such practice must be well regulated and excess liquid digestate needs to be further processed to reduce its eutrophication potential (Chen et al., 2012a; Chen et al., 2015; Macias - Corral et al., 2008; Smith et al., 2007) . In order to realize an eco - friendly manner to treat liquid AD digestate , studies and practices such as systematic algal culture in wastewater have emerged (Chen et al., 2012a; Karns et al., 1998; Mulbry et al., 2008; Mulbry and Wilkie, 2001; Pizarro et al., 2002) . These bioremediation processes combine nutrient uptake, dissolved oxygen enrichment, 49 and pH buffering to provide safe and effective ways to treat liquid AD digestate an d other wastewater. Meanwhile , algal biomass collected from these processes also has potential to produce alternative fuel and bio - derived chemicals. Nevertheless, due to the relatively high turbidity in the effluent , most of these commercial scale algae p onds have to have shallow bed to ensure sufficient access to light (Adey et al., 2011; Chen et al., 2012a; Mulbry et al., 2008; Mulbry and Wilk ie, 2001; Pacheco et al., 2015) , which could raise issues such as excess land use and water evaporation. Therefore, an additional step which can further reduce turbidity in the effluent was recommended prior to algal cultivation. Among various modern wastewater treatment techniques, electrocoagulation (EC) stands out as an electron driven coagulation method that eliminates chemical additives, reduces pathogens, and is able to han dle high strength wastewater (Liu et al., 2013; Liu et al., 2015; Mollah et al., 2004) . Industries such as paper, metal and mining all have stable and successful experience with EC treating their waste streams (Bellebia et al., 2012; Mollah et al., 2001; Parga et al., 2009) ; several recent studies on AD liquid digestate also reported that EC could significantly reduce chemical oxygen demand (COD), phosphate , and solution turbidity (Liu and Liu, 2015; Liu et al., 2013; Liu et al., 2015) . This study focused on integrating electrochemical technology and algal cultivation to develop a sustainable high - strength wastewater (i.e. liquid AD digestate) treatment system. The objectives of this study were to: (1) demonstrate the impact of EC treated liquid AD digestate (EC medium) on the fresh water algal assemblage; (2) determine an appropriate dilution of EC medium to achieve maximum nutrient removal and algal growth; and (3) analyze chemical composition of algal biomass cultured i n the EC medium. 50 2. Material and methods 2.1. EC treatment of liquid AD digestate Liquid digestate was collected from the commercial anaerobic digester at Michigan continuously stirred tank reactor (CSTR) and has an effective volume of 1800 m 3 . Feedstock of the digester consisted of roughly 60% dairy manure and 40% food waste (wet mass): dairy cows were fed on an alfalfa and corn silage blend diet according to the standard Total Mixed Rations (TMRs) for dairy cattle by Natural Research Council (NRC, 2001) , and food waste was mainly collected from campus cafeterias. The digester was operated at 35 °C with a hydraulic retention time of 25 days, and the digestate was sepa rated into liquid and solid portions using a screw press with 2 mm screen. The liquid digestate containing 4.8% total solids (TS, w/w), 3.1 g L - 1 total nitrogen (TN), 1.5 g L - 1 total phosphorus (TP), 21.5 g L - 1 chemical oxygen demand (COD), and a pH of 8.0, was used as the solution for the EC treatment. The EC treatment of the liquid digestate was carried out according to previous stud ies (Liu and Liu, 2015; Liu et al., 2015) with minor modifications. Briefly, the original liquid digestate was diluted 4 times using tap water and treated in a 50 L column EC reactor with anode surface area/volume ra tio as 0.124 cm - 1 (Figure 3. 1). A DC power supply (XPOWER TM 30 V, 5 A) was used to power the reaction, the current was maintained at 5 A, and the retention time was 4 hrs. After EC treatment, the effluent was centrifuged at 3500 rpm for 10 min and the supe rnatant (EC medium) was collected for algal culture. The original EC medium contained 0.03% TS (w/w), 350 mg L - 1 TN, 25.4 mg L - 1 TP, 907 mg L - 1 COD, 5.41 mg L - 1 total iron (Fe), and pH of 8.5. 51 Figure 3 . 1 : Sketch of column EC reactor (Liu and Liu, 2015) . 2.2. Preparation of algal inoculum A freshwater algal sample was collected from a pond locate d near the dairy farm at poured through a sheet of one - layer cheesecloth four times to screen out stones, debris, invertebrate larva (i.e. mayfly larva), and aquatic plants (i .e. duckweeds). The screened pond water was then centrifuged at 3750 rpm for 10 min to concentrate alga l biomass . The final algae concentrate was stored at 4 °C briefly before being applied as the inoculum for 52 following kinetics study and semi - continuous cultures ; the inoculum contained 0.34 g L - 1 algal biomass (dry weight) and the community composition of the original algal assemblage is shown in Figure 3. 4. 2.3. Kinetics study of algal culture in EC medium The effects of three EC medium concentrations: origi nal (1×), twice dilution (2×) and five - time dilution (5×), were investigated in the algal kinetics study with a set of two biological replicates. A 3 mL algal inoculum and 50 mL EC medium of each concentration were added to a 125 mL Erlenmeyer flask; a tot al of 12 flasks of each EC medium concentration were prepared and two were randomly sampled on day 1, 2, 3, 5, 7, and 9. The culture was conducted on orbital shakers (2.33 Hz, 150 rpm) at 22 ± 2 °C under continuous illumination using fluorescent lamps (100 µE m - 2 s - 1 ), and all culture media had a slightly alkaline pH (8.1 ± 0.4) due to the nature of AD effluent and EC process. A culture using only deionized water and the same algal inoculum was applied as blank for comparison, and a set of EC media in three concentrations without algal inoculation was also prepared as controls to eliminate nutrient loss due to the non - biological processes (i.e. volatilization) during culture ( APPENDIX G , Figure AP.1 AP.4 ) . An aliquot of 1 mL algal culture solution was collected to determine the optical density (OD 750nm ) of biomass, cell count, standardized biovolume, and algal community. An aliquot of 50 mL was centrifuged at 8000 rpm for 15 min to separate algal biomass from medium; biomass was dried overnight at 78 ± 3 °C, and the liquid medium was collected for measurements of TN, TP, Fe, and turbidity (OD 600nm ) (Sloof et al., 1995; Chen et al., 2012a) . 53 In order to determine growth rate, polynomial curve fitting based on the coefficient of determination (R 2 ) was appli ed to TS data. The concavity of fitted polynomial equation was measured by its second derivative, and the slope at inflection point of the polynomial was used to estimate the maximum growth rate. 2.4. Semi - continuous algal culture in EC medium The optimal EC m edium dilution based on the kinetics study was chosen for semi - continuous culture. Two 2 - L Erlenmeyer flasks were used to prepare biological replicates; each flask contained 60 mL algal inoculum and 1 L EC medium at the beginning of the culture. The cultu re was conducted on orbital shakers (2.33 Hz, 150 rpm) at 22 ± 2 °C under continuous illumination using fluorescent lamps (100 µE m - 2 s - 1 ). The lag and log growth phases were also determined by kinetics result; when algal growth reached to its maximum rate , an aliquot of 100 mL algal culture from each flask was sampled daily, and same amount of fresh diluted EC medium was added back into the flasks. Biomass optical density (OD 750nm ), cell count, community assembly, dry biomass weight, TN, TP, Fe and turbidi ty (OD 600nm ) were tested on the 100 mL daily culture sample using the same methods stated in previous section. Dry biomass collected at the end of the semi - continuous culture was ground using mortar and pestle for chemical composition analysis. 2.5. Analytical methods Cell density (cell per mL culture) was determined using a compound microscope (Nikon Eclipse 50i, 40× objective, 400× total system magnification) and a microscopic hemocytometer. The average biovolume (mL biomass per cell) of algal cells were meas ured using imaging software NIS - Elements D 3.00 (Nikon Instruments Inc., Melville, 54 NY). Standardized biovolume (mL biomass per mL culture) was calculated as the product of cell density and average cell biovolume. TN, TP, and Fe were analyzed using HACH tes ting reagent sets (HACH, Loveland, CO. Product #: 2714100, 2767245, 2415915, and TNT858, respectively). Carbohydrates in algal biomass were determined based on the analytical procedure by National Renewable Energy Laboratory (Van Wychen and Laurens, 2013) ( APPENDIX H ) . Protein content was measured using a bicinchoninic acid assay kit (BCA1, Sigma - Aldrich, St. Louis, MO) ( APPENDIX I ) . Crude lipid was measured using chloroform - methanol extraction method (Bligh and D yer, 1959) ( APPENDIX J ) . 3. Results and Discussion 3.1. Kinetics study Algal growth in EC medium with different dilutions were recorded based on dry weight, biomass density, and standardized biovolume (Figure 3. 2 , APPENDIX K ). There was no significant growth for the culture on the 1× EC medium (Figure 3. 2 - a) . It has been reported that the optimal algal culture medium made by the chemical treated AD effluent had a OD 600nm of 0.92 (Chen et al., 2012a) , in comparison, OD 600nm of the 1× EC medium in this study was 0.48, which indicates light penetration might not be a s ignificant factor of the slow growth in the 1× EC medium. On the other hand, the ammonia (NH 3 ) concentration in the 1× EC medium was relatively high, which could have the inhibitory effect on the algal growth (Azov and Goldman, 1982; Ohmori et al., 1977; Syrett, 1962) . C ompared to the culture on the 1 × EC medium, the cell growth on both 5× EC and 2× EC media showed much better growth (Figure 3. 2 - a). The c ell growth in the 5× EC medium had the shorter lag phase (less than 1 day) than the 2× EC medium (2 - 3 days), which 55 suggests that the freshwater alga e preferred relatively balanced and mild nutrient concentration. Figure 3 . 2 : Algal growth kinetics in diluted EC media based on (a) biomass dry weight (g TS L - 1 ), (b) biomass density (OD 750nm ), and (c) standardized biovolume (mL biomass per mL culture ) . Specific growth rate was calculated based on the dry weight regression models*. *The biomass dry weight ( g TS L - 1 ) regression models and corresponding R 2 values are: 0 0.4 0.8 1.2 1.6 0 2 4 6 8 10 Biomass dry weight (g TS L - 1 ) Time (day) 1X 2X 5X ( a ) 56 0 0.4 0.8 1.2 1.6 2 0 2 4 6 8 10 Biomass density (OD 750nm) Time (day) 1X 2X 5X ( b ) 0 0.005 0.01 0.015 0.02 0.025 0.03 0 2 4 6 8 10 Std. Biovolume (mL biomass mL - 1 culture) Time (day) 1X 2X 5X ( c ) 57 Figure 3 . 3 : Nutrients including (a) TN, (b) TP, ( c ) Fe, and ( d ) turbidity removal by algal growth in various EC media throughout 9 - day culture. 0 100 200 300 400 0 2 4 6 8 10 Total nitrogen (mg L - 1 ) Time (day) 1X 2X 5X 0 10 20 30 0 2 4 6 8 10 Total phosphorus (mg L - 1 ) Time (day) 1X 2X 5X ( b ) ( a ) 58 Figure 3. 3 Polynomial regression models ( ) based on biomass dry weight (g TS L - 1 ) was used to estimate growth rate and to compare growth kinetics among different treatments. Setting the second derivative of a continuous regression model to zero ( ) was used to locate the inflection point, or in this case, the maximum growt h 0 2 4 6 0 2 4 6 8 10 Total iron (mg/L) Time (day) 1x 2x 0 0.1 0.2 0.3 0.4 0.5 0 2 4 6 8 10 Turbidity (OD 600nm) Time (day) 1X 2X 5X ( c ) ( d ) 59 rate ( ). The results show that algae in the 5× EC medium reached to their maximum growth rate ( ) of 0.207 g TS L - 1 day - 1 at 2. 51 day; in the 2× EC medium, it was = 0.201 g TS L - 1 day - 1 at 6. 01 day; and in the 1× EC medium, the growth rate was negative and not used for comparison. The relatively similar and values indicate that the algal assemblage could adjust themselves in medium with elevated nutrient concentrations. Same trend was also observed in biomass density (OD 750nm ) and standardized biovolume (Figure 3. 2 - b and c). TN diagram (Figure 3. 3 - a) demonstrates that algae in the 5× EC medium had the highest nitrogen reduction (78.3%); it was 66.8% for the 2× EC medium, and only 16.7% for the 1× EC medium. As for the absolut e reduction of TN, algae in the 2× EC medium consumed the highest amount of nitrogen (128.2 mg L - 1 TN), the 5× EC medium reduced 63.4 mg L - 1, and the 1× EC medium reduced 56.2 mg L - 1 . TP diagram (Figure 3. 3 - b) shows that most available phosphorus in the 5× EC medium was depleted in the first two days of culture (70.1%), whereas in the 2× EC medium it was on the fifth day of culture (66.5%), and only 18.7% was reduced in the 1× EC medium in 9 days. It is noticeable that algal biomass in the 2× EC medium star ted to outgrow the culture i n the 5× EC medium after Day 7, it was possibly due to the completion of phosphorus in the 5× EC medium (1.7 mg L - 1 on Day 9). Choi and Lee (Choi and Lee, 2015) conducted a study to delineate the relationship between biomass productivity and N/P ratio , and concluded that TP removal rate greatly depended on N/P ratio and algal growth, but no strong correlation was observed between TN remo val and biomass productivity. Most iron (Figure 3.3 - c ) in the EC medium was introduced by the anode during EC reaction. The iron removal during algal cultivation was caused by both biological (algal 60 uptake and adsorption) and non - biological (participation) activities. The highest biological iron removal was approximately 50% from algal culture in the 2× and 5× EC media. While, algae in 1× EC medium were only able to consume about 18% of iron, and mostly within the first 2 to 3 days of culture. A lgae in the 2 × EC medium were able to utilize the ferric ion throughout the 9 - day culture, but the consumption rate in the 5 × EC medium slowed down after the third day. This is another indication that growth of algae in the 5 × EC medium was limited, and the consumption of one element might also rely on the other available nutrients. Many classic studies demonstrated the positive correlation between iron and algal growth (Aparicio et al., 1971; Rueler and Ades, 1987; Roche et al ., 1996). Specifically, Hopkins and Wann (1927) as well as Walker (1954) discovered that Chlorella requires a high minimum iron concentration to grow, and only ionized iron can be biologically available. Therefore, various uptake rates of the nutrients in different dilution of the culture media could also explain the algal community changes in this study. Meanwhile, the non - biological activities also removed a significant amount of iron in the media (APPENDIX G). It has been suggested by both freshwater and seawater studies that most ferric ions (Fe (III)) exist in a form of soluble chelates or suspended colloid ; the physical property of ferric chelates and colloid (mostly due to the electrostatic interactions among the molecules ) makes them difficult to si nk by gravity or centrifugation (Gunnars et al., 2002) . However, constant agitation and oxidization could change the form of ferric ions in the medium and made them unavailable to microorganisms. For instance, ferric ions can easily react with carbon dioxide dissolved in the medium and form in soluble ferric carbonate (Fe 2 (CO 3 ) 3 ). Interestingly, several studies using electron microscopic examination showed that more iron colloids were observed when a water body also 61 contained relatively high phosphorus content (Bernard et al., 1989; Tipping and Ohnstad, 1984) ; moreover, removal of phosphorus directly led to precipitation of ferric ions (Buffle et al., 1989; van Leeuwen et al., 2012) . Therefore, the non - biological iron reduction in this study was possibly due to the ferric ion precipitation caused by the disturbance of the colloidal electric charge during culture , insoluble chemical formation, as well as rapid phosphorus uptake by algae. 3.2. Effects of EC media on algal assemblage F igure 3. 4 illustrates the algal communities cultured on different EC media at Day 9 in comparison with the inoculum. The most dominant algal strain in the inoculum was Scenedesmus sp., the total amount of other two strains, Chlorella sp. and Pseudophormidium sp., was less than 10% in the inoculum. After nine days of culture, Scenedesmus in the control of the water medium still remained dominant, but Chlorella increased to about 20% of the entire community, and Pseudophormidium was disappeared in the water medium; the algal biomass collected from the culture on the water medium was significantly lower than 2× and 5× cultures due to the lack of nutrients. The culture on the 1× EC medium had a similar community composition on day 9 as the culture on the water medium; its biomass amount was also significantly lower than the cultures on the 2× and 5× EC media. It is mainly because of less algal growth on the 1× EC medium, so that the algal community maintained as it was in the inoculum. The most sign ificant community turnover was appeared in the 2× EC medium, where Chlorella became dominant strain (90%) and Scenedesmus reduced to less than 10%. This result indicates that Chlorella could tolerate higher nutrient concentration (Chen et al., 2012a; Parravicini et al., 2008; Roelke et al., 1999; Syrett, 1962) , while Scenedesmus consumed nutrients 62 faster when the growth was not restrained. In the 5× EC medium, dominant Scenedesmus took up to 80% of the entire algae assembly, while filamentous cyanobacteria Pseudophormidium and unicellular green algae Chlorella each counted for about 10%. Azov et al. (Azov and Goldman, 1982) observed the inhibition of excess NH 3 to algal growth, and proposed that converting NH 3 to non - toxic ammonium (NH 4 + ) by adjusting the culture pH to neutral to slightly acidic could prevent NH 3 toxicity in algal wastewater treatment system, which could be a solution to further enhance the algal growth on the EC medium (the pH of EC media are on alkaline side) . Figure 3 . 4 : Algae community assembly on day - 0 and day - 9 within various culture media. 3.3. Semi - continuous algal culture Given the similar maximum growth rates in both 2× and 5× EC media, these two conditions were carried out using 1 - L semi - continuous algal culture units. According to the maximum growth rates concluded from the kinetics study, the feeding was scheduled to 0% 20% 40% 60% 80% 100% Day-0 Final water Final 1X Final 2X Final 5X Chlorella Scenedesmus Pseudophormidium 63 st art on day 3 for the culture on the 5× EC medium, and day 7 for the culture on the 2× EC medium. Figure 3.5 demonstrates the biomass productivity from day 7 to day 19 (total of 13 days culture). The productivity in both media kept at steady states: the cul ture on t he 2× EC medium managed to produce 0.077 ± 0.004 g dried algae per liter per day (g TS L - 1 day - 1 ), which was 47% higher than the productivity from the culture on the 5× EC medium (0.052 ± 0.005 g TS L - 1 day - 1 ). For nutrient uptake, TN concentrations (Table 3.1) in the 2× EC and 5× EC media maintained steady at 84 and 22 mg L - 1 , respectively ; the TN daily consumption rates were 19.2 and 8.0 mg L - 1 day - 1 , respectively. However, the algal TP consumption rate s in both cultures increased thr oughout the experiment : from 1.35 to 3.28 mg L - 1 day - 1 in 2 × EC medi um and from 0.58 to 2.01 mg L - 1 day - 1 in 5 × EC medi um . Previous studies has reported this phenomenon as luxury uptake of phosphorus in algae, it occurs when the algae consume more phosphor us than required for growth without going through a prior starvation stage (Eixler et al., 2006) . Powell et al. (Powell et al., 2009) discovered that the polyphosphate accumulation of algae in high nutrient concentrations was mainly caused by luxury uptake. It was also reported that this mechanism has no effect on the algal growth (Azad and Borchard t , 1970) , which explains the biomass productivity in this study was kept steady, while the phosphorus uptake rate gradually increased in the cultures on both 2× and 5× EC media. Figure 3. 3 - d and Figure 3.6 shows that algal culture facilitated clarify and de - color the EC treated AD effluents (both 2× and 5×), and algal biomass cultured in the 2× EC medium had higher biomass density than that in 5× EC medium. These results clearly demonstrate that using th e combined EC - algae treatment system can significantly reduce the environmental impact of the AD liquid digestate. 64 Figure 3 . 5 : Algal biomass productivity from two different EC media in semi - continuous cultu re during steady growth phase. In addition, chemical analysis of algal biomass derived from the cultures on both 2× and 5× EC media (Table 3.2) demonstrated that 53.4% VS of the biomass from the 2× EC medium was protein, which was 13% higher than that from the 5× EC medium. It was possibly because of the significantly higher TN concentration in the 2× EC medium. On the other hand, algae from the 5× EC medium tended to accumulate more carbohydrates (36.6% VS, versus 27.4% VS in the 2× EC medium) and lipid (1 0.6% VS, versus 7.5% VS in the 2× EC medium). This pattern has been previous described in several studies of algal cultivation on nitrogen - rich wastewater streams (Garnier et al., 2014; Song et al., 2013; Yang et al., 2013) , and the conclusions were similar: excess inorganic nitrogen in cultural media could lead freshwater algae to accumulating more protein but less carbohydrates and lipids accumulation. However, Liu et al. (Liu and Liu, 2015) conducted a study using pure alga l strains to culture in EC treated AD digestate , in which the algal biomass was able to accumulate a relatively large amount of lipids. Liu s culture was also with pH control (pH 0 0.02 0.04 0.06 0.08 0.1 0 2 4 6 8 10 12 14 Biomass productivity (g TS L - 1 day - 1 ) 2X 5X Day - 5 Day - 9 Day - 13 Day - 17 65 6 - 7) and CO 2 feeding (5% CO 2 in the mixed gas. In contrast, the culture in this study utilized a mixed wild algal inoculum w ithout pH control or CO 2 supplement. It has been reported that CO 2 feeding can significantly increase the lipid content during the culture in the pH range of 6 - 7 (Widjaja et al., 2009) , which might be the main reason that Liu s study had higher lipid content in the algal biomass. T his presented study minimized the chemicals supplements to the system , and consequently reduce d the operational cost for wa stewater treatment facility . In addition, using mixed wild algal strains as inoculum could also provide a more robust and adaptive biome to the environmental variations. Table 3 . 1 : Chemical analysis of culture media before and after semi - continuous culture . TN (mg L - 1 ) TP (mg L - 1 ) Turbidity (OD 600nm ) Iron (mg L - 1 ) Before After Before After Before After Before After 2× EC medium 84.0 83.5 10.7 2.6 0.040 0.038 1.675 1.670 5× EC medium 21.5 22.0 14.5 8.3 0.028 0.029 0.843 0.856 Table 3 . 2 : Algal biomass chemical composition . Protein Carbohydrate Lipid Ash % VS % TS 2× EC medium 53.4 ± 0.5 27.4 ± 1.4 7.5 ± 1.9 14.2 ± 1.5 5× EC medium 47.3 ± 0.9 36.6 ± 0.8 10.6 ± 1.8 8.4 ± 1.0 66 (a) 2× EC medium (b) 5× EC medium Figure 3 . 6 : Comparing color and clarity of various fluids referred in this study. Original AD effluent 1 EC medium 2 EC medium Algae (in 2 ) 2 EC after culture Original AD effluent 1 EC medium 5 EC medium Algae (in 5 ) 5 EC after culture 67 4. Conclusion A study on the effect of EC treated AD liquid digestate on growth, nutrient uptake, and community assembly of freshwater algal assemblage was conducted. Kinetics revealed that algae cultured in two different dilutions of the EC media (2× and 5×) shared s imilar maximum growth rate (0.2 01 - 0.2 07 g TS L - 1 day - 1 ), and the culture on the 2× EC medi um had longer lag phase possibly due to the inhibition of excess ammonia. After 9 days of culture, Scenedesmus sp. remained as the most dominant taxon in the culture on the 5× EC medium, but Cholorella sp. in the 2× EC medium outgrew other taxa and took over 90% of the algal community. A semi - continuous culture further demonstrated that the culture on the 2× EC medium had significantly higher productivity (0.077 g TS L - 1 day - 1 ) than that the 5× EC mediu m (0.052 g TS L - 1 day - 1 ), though, both conditions showed significant advantage in nitrogen/phosphorus reduction and water clarification. It is recommended that combining EC treatment and algal cultivation could be an effective approach to be incorporated i n wastewater treatment process to deal with high - strength wastewaters such as animal manure and food wastes. 68 Chapter 4 : Effects of algal hydrolysate as reaction medium on enzymatic hydrolysis of lignocelluloses Abstract Effects of an algal hydrolysate on the enzymatic hydrolysis of lignocelluloses were examined using four bioenergy substrates (poplar, corn stover, switchgrass, and anaerobically digested manure fiber). Substrates were pretreated using dilute acid or alkali prior to hydrolysis. Hydrolysis reacti ons were conducted using the neutralized algal hydrolysate, citrate buffer, or deioinized water as reaction media. Results demonstrated that algal hydrolysate significantly improved the efficiency of enzymatic hydrolysis of lignin - rich or structurally recalcitrant biomass such as poplar and anaerobically digested manure fiber . This study showed that algal biomass can be used as not only a biofuel feedstock for direct diesel and ethanol production , but also a supplemental feedstock to enhan ce the performance of lignocellulosic biorefining. * Note: t he original article was published as Chen R, Thomas BD, Liu Y, Mulbry W, Liao W. 2014. Effects of algal hydrolyzate as reaction medium on enzymatic hydrolysis of lignocelluloses . Biomass and Bioenergy , 67:72 - 78. Minor formatting changes were made to satisfy the requirement of the graduate school at Michigan State University. 69 1. Introduction As aquatic photosynthesizing organisms, algae are considered as one of the most promising alternative bio - resources that could gradually replace fossil - based transportation fuels (Greenwell, et al., 2010, Sheehan, et al., 1998) . Compared to terrestrial energy crops, algae have major advantages such as using marginal land, faster growth, higher photosynthetic efficiency, year - around production, efficient uptake of nutrients in waste streams, and alleviating the global aquatic eutrophication potential (Groom, et al., 2 008, Kothari, et al., 2012, Mulbry and Wilkie, 2001, Zhang, et al., 2012) . Tremendous efforts have been made in the past decades to develop economic algal biofuel production (McHenry, 2009) . However, due to the complex composition of algal biomass, conversion processes that focus only on a single component, such as algal lipid, jeopardize the economic viability of algal biorefining. Full utilization of all algal components is critical to realizing the potential of algal biofuel and chemical products at commercial scale. Lignocellulosic materials such as agricultural and forestry residues, as well as he rbaceous and short rotation woody crops, have also been extensively examined as feedstock for biofuel production. At present, the major challenge of biofuel production from lignocellulose is to increase the productivity while minimizing the use of expensiv e chemicals and enzymes (Harmsen, et al., 2010, Singh and Trivedi, 2013, Wyman, 1996) . A ltering the lignin/cellulosic structure (Taherzadeh and Karimi, 2008) and introducing lignin - binding surfactants (Er iksson, et al., 2002) are two feasible solutions to expose more cellulose and hemicellulose in lignocellulosic materials to enzymatic attack. A recent study on converting algal biomass to bioethanol showed that a hydrolysate from algae grown in dairy manure improves the efficiency of enzymatic hydrolysis of a recalcitrant 70 lignocellulosic material (anaerobically digested manure fiber) by 50 - 80% (Chen, et al., 2012) . This result suggests that, in addition to direct production of biofuels using algae, components such as proteins, carbohydrates, and lipids from algal biomass can also be utilized to enhance lignocellulose conversion of biofuel production. The objective of this study was to extend our understanding of the effect of algal hydrolysate on the enzymatic hydrolysis of other lignocellulosic materials. In this study, we determined the hydrolytic efficiencies of four different lignocellulosic subst rates : poplar, corn stover, switchgrass, and anaerobically digested manure fiber (AD fiber) . These results will be useful for developing efficient and integrated biorefining processes of using lignocellulosic materials and algal biomass. 2. Methods and Materi als 2.1. Lignocellulosic feedstock Corn stover was harvested and collected in 2009 from a private farm in Muir, MI. Switchgrass w as collected from the Michigan State University Crop and Soil Science Teaching and Research Field Facility in 2010. Poplar hybrid s were planted in 1998 using a uniform spacing of 8x8 feet and harvested in fall of 2009. All of above substrates were dried and ground using an electric mill (Willey Mill, Standard Model No. 3; Arthur H. Thomas, Philadelphia, PA) with a sieve size of 2 mm. AD fiber was collected from the alfalfa and corn silage blended according to standard total mixed rations by the Natural Research Council (2001) (CSTR) operated at temperature of 35°C with a hydraulic rete ntion time of 30 days. AD 71 fiber was separated from the liquid digestate using a 5.5 kW FAN screw press with 2 mm screen. A sample of AD fiber was dried overnight at 75°C prior to use in these experiments. Compositional analyses of the glucan, xylan and li gnin content of substrates were conducted according to the analytical procedures for determination of structural carbohydrates and lignin in biomass provided by National Renewable Energy Laboratory (NREL) (Sluiter, et al., 2008) (APPENDIX A) . 2.2. Algae Algal biomass was grown using dilute AD dairy manure liquid effluent (USDA Dairy Research Unit, Beltsville, MD) recirculated in pilot - scale algal turf scrubber (ATS) raceways (Mulbry, et al., 2008) . Dominant species of the filamentous green algae assembly included Rhizoclonium hieroglyphicum (C.A. Agardh), Microspora willean a Lagerh., Ulothrix ozonata (Weber and Mohr) Kütz, R. hieroglyphicum (C.A. Agardh) Kütz and Oedogonium sp. (Chen, et al., 2012) . Wet algal biomass was first dewatered using 2 mm mesh nylon netting and then air dried to approximately 90% of total solids. Dry algae was milled to pass a 3 mm sieve and s tored at room temperature (Mulbry, et al., 2006) . The carbohydrate profile of algal biomass was analyzed using a concentrated acid hydrolysis method described by Chen et al. (2012) . Briefly, dry algal biomass was mixed with 75% (wt) sulfuric acid to a 3:5 sample - to - acid ratio (wt). The mixture reacted at room temperature for 30 min and was then h eated at 130°C for 10 min in an autoclave. Total protein and total fatty acid contents were determined by the Experiment Station Chemical Laboratories at the University of Missouri. 72 2.3. Pretreatment of feedstock and algal biomass The four substrates (AD fib er, switchgrass, poplar, and corn stover) were pretreated using either dilute sulfuric acid (H 2 SO 4 ) or sodium hydroxide (NaOH) solutions under the optimal conditions determined in previous studies (Ruan, et al., 2013, Teater, et al., 2011, Ucar, 1990, Yue, et al., 2011) (Table 4.1). Pretreated sample was neutralized to pH 5 using 30% NaOH or 20% H 2 SO 4 solutions and washed five times each with 500 mL deionized water until the supernatant was transparent. Pretreated sample was dewatered using eight - layer cheese cloth and t hen oven - dried overnight at 75°C. Glucan, xylan and lignin contents of the substrates were determined before and after pre - treatment. Table 4 . 1 : Pretreatment conditions for each feedstock (all concentrations a re in wt%) Feedstock Acid pretreatment condition Alkali pretreatment condition AD fiber 3% H 2 SO 4 , 130 °C , 2 h (Yue et al., 2011) 2% NaOH, 130 °C , 3 h (Yue et al., 2011) Poplar 5% H 2 SO 4 , 130 °C , 2 h ( unpublished data) 2.5% NaOH, 100 °C , 2 h (Ucar, 1990) Corn stover 2% H 2 SO 4 , 130 °C , 1 h (Ruan et al., 2013) 1% NaOH, 130 °C , 2 h (Teater et al., 2011) Switchgrass 2% H 2 SO 4 , 130 °C , 2 h (Ruan et al., 2013) 1% NaOH, 130 °C , 2 h (Teater et al., 2011) Algal biomass was hydrolyzed using a 5% TS loading using 4% (wt) H 2 SO 4 at 116°C for 30 min (Chen, et al., 2012) . The hydrolyzed mixture was neutralized to pH 5 using calcium carbonate (CaCO 3 ). Residual solid material was completely removed by centrifugation ( 2846 × g , 10 min). The liquid hydrolysate was saved and its carbohydrate profile was determined using high performance liquid chromatography (HPLC) and its protein content was measured using a bicinchoninic acid protein assay kit (BCA1 and B9643, Sigma - Aldrich, St . Louis, MO),. 73 2.4. Enzymatic hydrolysis In order to determine the effect of the reaction medium on the enzymatic hydrolysis of the four substrates , separate hydrolysis reactions were conducted using pretreated substrates and de - ionized water, sodium citrate buffer (50 mM, pH 4.8) and neutralized algal hydrolysate. Aliquots (2 g DW) of each pretreated feedstock were combined with 20 g of each reaction medium and 17 g of de - ionized water. The resulting mixtures were steril ized using an autoclave (15 min, 120 C) and cooled to room temperature prior to the addition of cellulase (Accelerase 1500 ® , Genencor, Rochester, NY) to a final concentration of 25 FPU g - 1 dry feedstock. Samples were mixed using a shaker table orbiting at 150 rpm at 50°C for 72 h. Samples (1 ml) of hydrolysates were taken at hour 0, 24, 48 and 72. Hydrolysate samples were boiled for 5 min and filtered using Millex - GS 0.22 µm syringe filters prior to analysis of monosaccharide content. The net sugar concentrations during the enzymatic hydrolysis were calculated by subtracting the concentration of detected sugars in the algal hydrolysate medium from the total concentration of detected sugars in the hydrolysate samples . Overall glucan/xylan conversion o f each raw feedstock was used to determine the effects of pretreatment and of the reaction medium on the entire saccharification process. The overall glucan conversion is defined as the percentage of net glucose production over the glucose equivalent amoun t in the raw feedstock. The overall xylan conversion is defined as the percentage of net xylose production over the xylose equivalent amount in the raw feedstock. Enzymatic glucan/xylan conversion of pretreated feedstock was used to demonstrate the effe cts of different treatments on the enzymatic hydrolysis. The enzymatic glucan 74 conversion is defined as the percentage of net glucose production over the glucose equivalent amount in the pretreated feedstock. The enzymatic xylose conversion is defined as th e percentage of net xylose production over the xylose equivalent amount in the pretreated feedstock. Improvement of glucan conversion was calculated using net enzymatic glucan conversions in algal hydrolysate or buffer to divide net enzymatic glucan conve rsion in water for individual substrate . The improvement of glucan conversion was used to compare the effects of different pretreatments, substrates and reaction media on hydrolysis. 2.5. Analytical methods A Shimadzu 2010 HPLC system equipped with Bio - rad A minex ® HPX - 87P analytical column (300×7.8 mm) and a refractive index detector was used for determination of monosaccharide profiles. The mobile phase was ultrapure water (Synergy Ultrapure Water Purification System, MiliporeTM, Billerica, MA), the flow rat e was 0.6 mL min - 1 , and the column temperature was 80°C. High purity standards including glucose, xylose, galactose, arabinose, and mannose were purchased from Sigma - Aldrich (St. Louis, MO). 2.6. Statistical analysis A mixed linear model using the Statisti cal Analysis System program 9.2 (SAS Institute Inc., NC) was used to perform one - pair - wise comparison and simple main effects analysis (slicing) ( APPENDIX L ) . 75 3. Results and Discussion 3.1. Characteristics of f iber s and a lgae Compositional analyses of four raw lignocellulosic feedstock show that glucan is the most abundant component in all substrates (Table 4.2). Poplar contained the highest glucan content (45%) which makes it a good potential feedstock for bio - conversion to glucose. However, poplar also had the highest lignin content (23%) among the four substrates , which suggests that it would require more energy, and chemical/enzymatic inputs to process the same amount of poplar glucan compared to other substrates. Cor n stover and switchgrass are both commonly used as energy biomass substrates and they contain similar levels of glucan (40% and 37%, respectively) and lignin (19% and 21%, respectively), but different levels of xylan (30% in corn stover compared to 22% in switchgrass). AD fiber contains relatively less glucan (30%) and xylan (12%) compared to the other three substrates , though it has been recently discovered that AD fiber also has overall glucose conversion similar to many energy crops and residuals , which also makes it a potential feedstock for biofuel production (Yue, et al., 2011) . The compositions of each feedstock after acid or alkali pretreatments are shown in Table 4.3. Acid pretreatment significantly reduced the xylan content in all four substrates . This is likely because the random and amorphous structure of hemicellulose is e asily hydrolyzed by dilute acid solutions. Wyman et al. (2005, 2009) concluded that dilute sulfuric acid pretreatment can signifi cantly recover most of the hemicellulose from lignocellulosic biomass. This pretreatment also exposes cellulose to enzymes and facilitates high sugar conversion from subsequent biorefining processes. Alkali pretreatment removed more lignin from the substra tes , which also increases the accessibility of cellulose to enzymatic attack 76 (Taherzadeh and Karimi, 2008) . Both fiber structure and chemical composition varied among different pretreated feedstock, and would influence the efficiency of following enzymatic hydrolysis. Table 4 . 2 : Structural carbohydrate and lignin content of the raw lignocellulosic biomass Biomass Glucan (wt%) Xylan (wt %) Lignin (wt %) AD fiber 30.3±1.9 11.6±1.1 14.2±0.8 Poplar 44.8±0.5 20.5±0.5 22.6±0.5 Corn stover 39.7±0.7 29.9±0.5 18.6±0.7 Switchgrass 37.4±0.9 22.1±0.2 20.5±0.2 Table 4 . 3 : Structural carbohydrate and lignin of pretreated lignocellulosic biomass Biomass Pretreatment Glucan (wt%) Xylan (wt %) Lignin (wt %) AD fiber A cid 35.7±0.2 5.2±0.2 59.5±0.9 Alkali 51.2±0.5 16.3±0.3 31.6±0.7 Poplar A cid 62.1±1.2 0.0±0.2 32.3±0.1 Alkali 53.4±0.7 11.9±0.2 26.5±0.7 Corn stover A cid 63.1±0.6 11.3±0.4 28.6±0.5 Alkali 63.6±0.3 29.5±0.1 8.1±0.3 Switchgrass A cid 63.4±0.6 8.3±0.2 31.0±1.3 Alkali 58.7±0.1 28.9±0.1 15.4±0.2 Table 4 . 4 : Composition of Algal Biomass Carbohydrate (wt %) Primary components of carbohydrate (wt%) Protein (wt %) Fatty acid (wt %) Glucose Xylose Galactose Arabinose Mannose 27.50±0.6 10.98±0.5 3.68±0.1 6.61±1.3 1.75±0.7 2.89±0.5 30.65±1.1 4.06±0.3 77 The algal biomass predominantly consisted of filamentous green algae. The algae were cultured in nutrient - rich AD effluent. Consequently, the algal biomass contained relatively more protein (30%) and less fatty acid (4%) (Table 4.4) than green algae grown in natural waters (Becker, 1994) . Total carbohydrates comprised 27% of the algal biomass, and its major components were: gluc ose (11% of the total biomass dry matter ), galactose (6.6%), xylose (3.7%), mannose (2.9%), and arabinose (1.8%). It is difficult to refine biodiesel from the algal biomass due to the low fatty acids content. However, its high protein and sugar composition makes it a superior feedstock or supplemental feedstock to produce bio - alcohol and other value - added products such as fertilizer and fuel additives. Table 4 . 5 : Characteristics of algal hydrolysate * Protein ( g L - 1 ) Glucose ( g L - 1 ) Xylose ( g L - 1 ) Galactose ( g L - 1 ) Arabinose ( g L - 1 ) Mannose ( g L - 1 ) 5.31±0.50 3.44±0.16 1.28±0.33 1.86±0.42 0.56±0.10 0.99±0.15 *: The data are the average of two replicates with standard errors. 3.2. Enzymatic hydrolysis of lignocellu los ic substrates using algal hydrolysate as reaction medium The algal hydrolysate contained 8.3 g L - 1 of sugar (Table 4.5), of which glucose (3.4 g L - 1 ), galactose (1.9 g L - 1 ), and xylose (1.3 g L - 1 ) were the most abundant components. The hydrolysate contained 5.3 g L - 1 of total protein. Chen et al. (2012) reported that an algal hydrolysate with similar protein content increased the enzymatic glucan conversion from alkali pretreated AD fiber by nearly 50% compared to reactions performed in wa ter or buffer. 78 Table 4 . 6 : Sugar concentrations and overall glucan and xylan conversion s of different lignocellulosic feedstock a, b, c , d Pretreat - ment Feedstock Reaction Medium Glucose (g L - 1 ) Xylose (g L - 1 ) Galactose (g L - 1 ) Arabinose (g L - 1 ) Mannose (g L - 1 ) Overall Glucan conversion (%) Overall Xylan conversion (%) Acid AD fiber Water 2.86±0.0 0.17±0. 1 0.05±0.0 ND 0.07±0.0 9.6±0.0 1.5±0. 6 Buffer 3.69±0.1 0.20±0.0 0.05±0.0 ND 0.07±0.0 12.3±0. 4 1.7±0. 1 Algae 4.44±0.1 0.53±0.2 0.47±0.0 0.22±0.1 0.23±0.2 14.8±0. 3 4.6± 1.0 Corn stover Water 14.96±0.4 1.17±1.0 ND 0.01±0.0 0.07±0.0 33.5± 1.2 3.5± 1.7 Buffer 18.23±0.5 1.40±0.3 0.07±0.0 ND 0.09±0.0 40.8± 1.2 4.2±0. 9 Algae 18.99±0.2 1.94±0.7 1.15±0.1 0.30±0.0 0.69±0.1 42.5±0. 5 5.8± 1.8 Poplar Water 13.72±0.4 0.16±0.1 0.03±0.0 ND ND 32.4± 1.0 0.8±0. 4 Buffer 16.60±0.2 0.15±0.0 0.14±0.0 0.09±0.0 0.07±0.0 39.2±0. 5 0.8±0. 1 Algae 22.99±0.8 0.25±0.1 ND ND 0.04±0.0 54.3± 1.8 1.3±0. 5 Switchgrass Water 20.02±0.2 0.42±0.1 0.02±0.0 ND 0.08±0.0 53.0± 0.5 1.3±0. 4 Buffer 22.82±0.4 0.56±0.0 0.06±0.0 0.06±0.0 0.10±0.1 60.4± 1.1 1.7±0. 0 Algae 25.75±1.0 1.45±0.1 1.73±0.5 0.45±0.1 1.10±0.6 68.2± 2.6 4.4±0. 3 Alkali AD fiber Water 12.30±0.0 4.79±0.8 0.32±0.1 0.44±0.3 0.44±0.1 42.9±0. 1 43.6± 6.7 Buffer 12.54±0.3 4.41±0.2 0.34±0.1 0.90±0.2 0.55±0.0 43.7± 1.1 40.2± 1.6 Algae 18.11±1.4 6.02±0.5 1.75±0.1 1.61±1.1 1.38±0.4 63.2± 4.1 54.8± 4.2 Corn stover Water 24.15±1.1 7.05±0.2 ND 0.52±0.1 0.07±0.0 55.3± 2.3 21.4±0. 6 Buffer 25.84±0.8 6.78±0.2 ND 0.38±0.0 0.08±0.0 59.2± 1.6 20.6±0. 6 Algae 26.28±2.2 7.83±0.6 0.59±0.3 0.95±0.4 0.19±0.0 60.2± 4.5 23.8± 1.4 Poplar Water 12.38±0.2 3.42±0.0 ND ND 0.38±0.0 29.1±0 .5 18.1±0. 1 Buffer 13.15±0.3 3.53±0.6 ND ND 0.42±0.2 33.9±0. 7 21.9± 3.0 Algae 12.63±0.3 3.48±0.7 0.37±0.1 0.14±0.1 0.34±0.2 32.4±0. 8 18.7± 3.3 Switchgrass Water 19.08±1.4 5.62±0.2 ND 0.48±0.1 0.06±0.0 53.7± 4.0 18.0±0. 7 Buffer 19.56±1.0 5.33±0.6 ND 0.34±0.0 0.06±0.0 55.1± 2.7 17.1± 1.7 Algae 20.11±0.5 6.47±1.4 0.55±0.3 0.93±0.5 0.15±0.0 56.6± 1.3 20.8± 3.8 a : The overall conversion was calculated based on the chemical composition of the original raw feedstock. Do not confuse with the enzymatic conversion, which was calculated based on the chemical composition of pretreated feedstock. b: The data are the average of two replicates with standard errors. c: The data are from a 72 - hour enzymatic hydrolysis. d : ND represents not detectable. 79 Figure 4 . 1 : Enzymatic glucan conversion of differently treated feedstock using de - ionized water, citrate buffer, and algal hydrolysates as reaction medium . 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) a. acid pretreated AD fiber Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) b. alkali pretreated AD fiber Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) c: acid pretreated poplar Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) d: alkali pretreated poplar Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) e. acid pretreated corn stover Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) f: alkali pretreated corn stover Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) g: acid pretreated switchgrass Water Buffer Algae 0% 20% 40% 60% 80% 100% 0 20 40 60 80 Enzymatic glucan conversion Time (hour) h: alkali pretreated switchgrass Water Buffer Algae 80 The overall glucan conversion using different pretreatments and reaction media are presented in Table 6 and the kinetics of enzymatic glucan conversion in Figure 4.1. It is apparent that dilute acid was not an efficient pretreatment for AD fiber (Figure 4.1 - a), which had only 18% enzymatic glucan conversion using citrate buffer and 22% conversion using algal hydrolysate as medi a (corresponding to increases of 29% and 57%, respectively compared to conversions using the control medium of water). Alkali - pretreated AD fiber (Figure 4.1 - b) had a greater glucose conversion , 44% in citrate buffer and 64% in algal hydrolysate (corresponding to increases of 16% and 70%, respectively, compared to water). The difference on hydrolysis between acid and alkali treated AD fibers was caused by the high alkalinity of raw AD fiber, which was derived from the anaerobic digestion process (Yue, et al., 2010) . The high alkalinity of the AD fiber neutralized sulfuric acid during the dilute acid pretreatment and led to a decreased efficiency of the pretreatment process. The differences in pretreatment efficiencies were also related to the recalcitrant nature of AD fiber and the fiber structure after different pretreatment. AD fiber contains less easily digestible cellulose and hemicellulose branches than other agricultural residuals, and alkali is better than acid in terms of facilitating lignin removal and fi ber deconstruction. Hydrolyses in algal hydrolysate media for both acid and alkali treated AD fiber had better enzymatic glucan conversion than water and citrate buffer media. As it has been reported in a previous study (Chen, et al., 2012) , the protein and other small molecules in algal hydrolysate played as surfactants to bind the protein - active sites on lignin before introducing the enzymes, increase available active enzyme in the solution, and consequently enhance the performance of enzymatic hydrolysis. According to Table 4.3, pretreated AD fiber contained more lignin than other pretreated feedstock, which explained 81 the improvement of enzymatic efficiency by using algal hydrolysate (compared to citrate buffer) was more significant (p < 0.05) on AD fiber than other feedstock. For poplar, pretreat ments did not have significant effects on the enzymatic glucan conversion using water and citrate buffer media (Figure 4.1 - c & d). However, acid - pretreated poplar in algal hydrolysate medium had a significantly higher (p < 0.0001) enzymatic glucan convers ion (66%) than the alkali - pretreated poplar (47%) using the same medium. This was possibly due to the aforementioned function of algal hydrolysate that the algal protein and other molecules played as a surfactant to bind lignin (acid pretreated poplar had much higher lignin content than alkali pretreated poplar), and improved the hydrolysis performance of the acid pretreated poplar. For corn stover (Figure 4.1 - e & f), alkali pretreated fiber released more glucose (68 - 74%) than acid - pretreated (44 - 56% of en zymatic glucan conversion) in all three media, but the difference between citrate buffer and algal hydrolysate media was not significant in corn stover ( p = 0 .0598). For switchgrass (Figure 4.1 - g & h), acid - pretreatment showed better performance on enzymat ic glucan conversion (71 - 83%) than alkali - pretreatment (66 - 69%), while the effect of different reaction media on conversion was not significant under alkali - pretreatment ( p = 0 .5247). Corn stover and switchgrass are both agricultural residuals that have gr eat potential for biofuels production because of their relatively amendable fiber structure compared to poplar and AD fiber (Wyman, et al., 2009) . It is possible the majority of glucan from these two fibers be came hydrolytically available after the chemical pretreatments so that additional organic and/or inorganic chemicals from citrate buffer and algal hydrolysate did not show any enhancement as they did on structurally recalcitrant poplar and AD fiber. 82 3.3. Combi ned effect of pretreatment and reaction medium on the improvement of enzymatic hydrolysis In order to demonstrate the combined effect of pretreatment and reaction medium on the improvement of enzymatic hydrolysis of different feedstock, the enzymatic glucan conversion at 72 - hour (Figure 4.1) were compared against the conversion of control group using water as the medium (Figure 4.2). Overall, acid pretreated feedstock in both algal hydrolysate and citrate buffer had significantly better improvement of glucan conversion than alkali pretreated feedstock (Figure 4.2 - a & b). The improvement results along with the actual conversion (Table 4.3) indicated that acid - treated substrates were more recalcitrant than alkali - treated feedstock, and required pH buffer and surfactants to further facilitate the enzymatic hydrolysis in order to obtain better conversion. As for the algal hydrolysate medium, all four substrates under acid pretreatment had significant improvement (p = 0.0004, 0.0008, 0.0001, 0.005 for AD fi ber, corn stover, poplar and switchgrass, respectively) than water and citrate buffer. However , under alkali pretreatment, only AD fiber in algal hydrolysate showed a significant improvement ( p < 0 .0001) than other media (Figure 4.2 - b). According to the chemical composition of pretreated feedstock (Table 4.3), all the pretreated feedstock with higher lignin content had significant improvement of glucan conversion . Previous studies have reported that lignin can reduce the enzyme accessibility of cellulose and hemicellulose during hydrolysis of lignocellulosic biomass, which negatively affects the conversion (Tatsumoto, et al ., 1988) . These results suggest that organic compounds such as proteins and/or other small molecules in the algal hydrolysate may block the enzyme - absorption sites of lignin, which led to increased enzyme accessibility for the hydrolysis. It is also pos sible that the algal 83 hydrolysate could also provide a more chemical - and pH - balanced environment than the other reaction media. a. Acid pretreatment b. Alkali pretreatment Figure 4 . 2 : Effects of pretreatment and reaction medium on the improvement of enzymatic glucan conversion from different feedstock* * : all compared to the control group using water as reaction medium. 20% 22% 21% 14% 45% 27% 68% 29% 0% 20% 40% 60% 80% 100% AD Corn stover Poplar Switchgrass Improvement of glucan conversion*, acid pretreatment Buffer Algae 2% 7% 17% 2% 47% 9% 14% 5% 0% 20% 40% 60% 80% 100% AD Corn stover Poplar Switchgrass Improvement of glucan conversion*, alkali pretreatment Buffer Algae 84 4. Conclusion Algal hydrolysate as a reaction medium significantly enhanced the enzymatic hydrolysis of alkali pretreated AD fiber as well as acid - pretreated poplar. In contrast, the algal hydrolysate did not show a significant effect on the enzymatic hydrolysis of pret reated corn stover or switchgrass. This difference is likely caused by the fiber structure and lignin content of the pretreated lignocelluloses. Applying algal hydrolysate as reaction medium not only enhances enzymatic hydrolysis of recalcitrant lignocellu loses, but also eliminates use of pH buffer solution, which could make a significant contribution to lignocellulosic biorefining. 85 Summary Booming human population and expanding industrialization in the last few decades have had drastic and irreversible damage to the environment. In order to protect our limited natural resources and redefine a sustainable society to future generation s , environment - awareness must be implemented in all human activities. This study took a deep look at current issues in waste management, water recycle, and renewable bioenergy; it analyzed pros and cons of each individual practice and developed an integrate d solution to maximize each practice s potential and minimize their shortcomings. This closed - loop system managed to reduce the overall environmental impacts from organic wastes and convert them into renewable energy and value - added by - products. Chapter 1 of this dissertation conducted a thorough literature review to describe t he fundamentals, significance and current technical status of the study . Chapter 2 applied molecular and statistical approaches to test the response of anaerobic microbial community to the change of anaerobic digestion conditions, and their consequent influence on biogas productivity and solid digestate quality. It was discovered that even though the performance of digestion was enhanced with the increase of supplemental fo od waste , t his variation in feedstock did not have significant impact on the chemical composition of final solid digestate (AD fiber). In addition, the anaerobic microbial community demonstrated that they were able to adopt their configuration in order to maximize th e conversion of available carbon to biogas. Chapter 3 utilized the liquid digestate (AD effluent) from previous step in an electrocoagulation - algae combined treatment system to reduce its turbidity and nutrients 86 (i.e. ammonia - N, phosphate) concentration. I n the meantime , accumulated algal biomass in this system became a potential substrate in biofuel and chemical production. Results showed algae cultured in both 2× and 5× dilution of EC medium had reached s imilar maximum growth rate (0.201 - 0.207 g TS L - 1 day - 1 ), even though they experienced different lag phase. Inhibition of algal growth in the original EC medium (1×) was possibly due to the high unionized ammonia concentration. Semi - continuous algal culture based on the kinetics study illustrated higher biomass productivity in 2× than in 5× EC medium because of its more sufficient nutrients (i.e. phosphate) supply. Both cultures maintained steady growth and nitrogen consumption rates throughout the period of study, but their phosphorus consumption rate s k phosphorus uptake mechanism. The combined EC - algae treatment proved to be able to significantly and sustainably reduce the turbidity and eutrophication - causing nutrients in AD effluent. Chapter 4 explored a new application of algal biomass cultured in nitrogen - rich media. Acid hydrolysate from algal biomass was neutralized and applied to the enzymatic hydrolysis (EH) process on lignocellulosic substrates as buffer solution. Results showed significant enhance ment in EH efficiency on substrates containing higher lignin and structurally more recalcitrant (i.e. poplar and AD fiber) using algal hydrolysate in comparison with controlled treatment using either water or sodium citrate buffer. It was proposed that pro tein remained in the algal hydrolysate facilitated the EC process by serving as a lignin - binding surfactant and increase the availability of enzymes. The integrated system and its corresponding preliminaries as described previously provide a general concep t of a new and green way for treating waste biomass. Further 87 research is required to explore more possible substrates and mixing ratios for anaerobic co - digestion, optimizing the EC condition to reduce energy consumption and to provide efficient nutrients for algal culture, exploring other algal strains to increase the nutrient uptake rate, and improving the biorefining conditions for both algal biomass and lignocellulosic feedstock. In addition, because of the fast development of molecular genetic technolo gies, more advanced sequencing methods are also recommended in the future studies of microbial community (i.e. anaerobic bacteria, anaerobic archaea, and interaction of aerobic bacteria and algae). 88 APPENDICES 89 APPENDIX A : Procedure for analyzing structural carbohydrates and lignin content of lignocellulos ic biomass 1. Weigh 300.0 ± 10.0 mg of the biomass into a tared autoclave - safe bottle. 2. Add 4.92 ± 0.01 g of 72% sulfuric acid. Mix sample and acid with glass rod for 1 min. 3. Place bottles i n 30 ± 3 °C incubator for 60 min, stir samples with glass rods every 5 - 10 min. 4. After 60 min incubation, add 84.00 ± 0.04 mL deionized water to dilute the acid to a 4% concentration. Mix sample in solution gently, and autoclave the solution at 121 °C for 1 hour. Allow the bottles to cool down before removing the caps. 5. Transfer approximately 50 mL autoclaved solution into a clean and dry crucible with vacuum filter; apply pressure with an air pump to facilitate filtration. Measure the filtrate using spectroph otometer at 320 nm for dissolved lignin. Neutralize the rest filtrate with calcium carbonate to pH 5 - 6. Filter the n eutralized solution using 0.22 µ m pore size syringe filter for HPLC analysis of sugar composition. 6. Transfer the rest autoclaved solution int o the crucible with vacuum filter. Wash the solid particles remained on the vacuum filter with deionized water until the pH of flow through is close to 5 - 6. Leave the crucible in the 105 °C oven until a constant weight is achieved. Record the dry weight of insoluble solids before placing the crucible in the muffle furnace at 575 °C for 24 hours for ash content. 90 APPENDIX B : Procedure for DNA extraction 1. If samples are in liquid - sludge form, transfer all the contents in PowerBead Tubes into a sterilized 2 mL Collection Tube provided, add 1.5 to 2.0 mL of liquid sample depending on the viscosity of the sample; centrifuge at 10,000 rpm for 10 min and carefully remove the supernatant; pour the original content from PowerBead Tubes back. If samples are in soli d form, add 0.25 g directly into PowerBead Tubes. Vortex to mix. 2. Add 60 µL dissolved Solution C1 (a cell lysis reagent) into each PowerBead Tube and invert several times. 3. Secure PowerBead Tubes on a beadbeater, and leave the beadbeater on for 2 min. 4. Centrifuge PowerBead Tubes at 10,000 rpm for 30 sec at room temperature. 5. Transfer supernatant to a clean 2 mL Collection Tube, add 250 µL Solution C2 (an inhibitor removal reagent) and vortex for 5 sec. Incubate at 4 °C for 5 min. 6. Centrifuge at 10,000 rpm for 1 min at room temperature, carefully transfer 600 µL supernatant to a clean 2 mL Collection Tube. Add 200 µL Solution C3 (another inhibitor/cell debris removal reagent) and vortex briefly. Incubate at 4 °C for 5 min. 7. Centrifuge at 10,000 rpm for 1 min at room temperature, carefully transfer 750 µL supernatant to a clean 2 mL Collection Tube. Add 1.2 mL Solution C4 (high concentration salt solution to facilitate DNA binding on silica filter) and vortex for 5 sec. 8. Load 675 µL onto a Spin Filter and centr ifuge at 10,000 rpm for 1 min at room temperature. Discard the flow through. Repeat until all solution is centrifuged. 91 9. Add 500 µL Solution C5 (an ethanol based solution to clean DNA bound on silica filter) and centrifuge at 10,000 rpm for 30 sec at room te mperature. Discard the flow through. 10. Centrifuge at 10,000 rpm for 1 min at room temperature to remove residual ethanol. 11. Carefully transfer the Spin Filter in a clean 2 mL Collection Tube, add 100 µL Solution C6 (DNA elusion buffer) and centrifuge at 10,000 rpm for 30 sec at room temperature. Discard the Spin Filter. 12. Measure DNA concentration using NanoDrop spectrophotometer. Frozen samples for future u se if they have more than 25 ng µL - 1 dsDNA and the A260/A280 ratio (DNA purity test) close to 1.8. 13. If the d sDNA concentration or A260/A280 ratio is low, wash and concentrate DNA using the following steps: a. Add 4 µL of 5 M NaCl into each 100 µL DNA solution, invert 3 - 5 times. b. Add 200 µL of 100% (200 proof) cold ethanol, invert 3 - 5 times. c. Centrifuge at 10,000 rpm for 10 min at room temperature. d. Carefully decant supernatant; the residual ethanol can be further removed in a ventilated clean hood on ice. Resuspend precipitated DNA in sterile 10 mM Tris. Note: This procedure was modified based on the procedure fr om Power Soil DNA Extraction Kit , MOBIO 92 APPENDIX C : PCR procedure for 454 pyrosequencing of bacteria l 16S rDNA 1. Pilot PCR mixture (per sample) : 15.7 µL RNAse/DNAse free water , 2 µL 10× AccuPrime PCR Buffer II , 0.16 µL Taq polymerase , 0.4 µL Forward Primer (10 µM) , 1.34 µL DNA Template (~ 5 ng µL - 1 ) , 0.4 µL Reverse Primer with barcode (10 µM) (one barcode corresponding to only one sample) . 2. Regular PCR mixture (per sample) : 58.9 µL RNAse/DNAse free water , 7.5 µL 10× AccuPrime PCR Buffer II , 0. 6 µL Taq polymerase , 1.5 µL Forward Primer (10 µM) , 5 µL DNA Template (~ 5 ng µL - 1 ) , 1.5 µL Reverse Primer with barcode (10 µM) (one barcode corresponding to only one sample) . 3. Vortex each PCR tube to homogenize PCR mixture, then spin briefly (5 - 6 sec) in a centrifug e. 4. Place PCR tubes in thermos cycler, and the cycle as follows: (1) Initi al denaturing: 95 °C for 5 min ; (2) 30 cycles of amplification: denaturing at 95 °C for 45 sec , annealing at 50 °C for 45 sec , elongation at 72 °C for 90 sec ; (3) Final extension: 72 °C for 5 min ; (4) Storage: 4 °C for i nfinity time. 5. Mix 5 µL of PCR product with 1 µL loading dye (6×), and load the dyed PCR product onto a sheet of 1% agarose 1× TAE gel for electrophoresis (100 V) for approximately 20 - 30 min. 6. Transfer the gel sheet from previous step into EtBr (caution: carcinogen), stain for 5 min. Then transfer stained gel into water, de - stain for 10 min. 93 7. Observe the existence, position, width, and brightness of bands under UV light. If the band looks ideal in pilot PCR, proceed to actual PCR and s ubmit the product for 454 pyrosequencing. 94 APPENDIX D : PCR procedure for T - RFLP of archaea l 16S rDNA 1. Pilot PCR mixture (per sample) : 13.5 µL Platinum PCR SuperMix , 0.3 µL Forward Primer (344aF, 10 µM) , 0.3 µL Reverse Primer (1119aR, 10 µM) , 0.15 µL BSA (100× or 10 mg mL - 1 ) , 0.75 µL DNA Template (~ 5 to 10 ng µL - 1 ) . 2. Regular PCR mixture (per sample) : 90 µL Platinum PCR SuperMix , 2.5 µL FAM labelled Forward Primer (344aF - FAM , 10 µM) , 2 µL Reverse Primer (1119aR, 10 µM) , 1 µL BSA (100× or 10 mg mL - 1 ) , 4.5 µL DNA Template (~ 5 to 10 ng µL - 1 ) . ( Note: FAM labelled forward primer (344aF - FAM) is light sensitive. Conduct the following steps in low light/dark room for actual PCR. ) 3. Vortex each PCR tube to homogenize PCR mixture, then spin briefly (5 - 6 sec) in a centr ifuge. 4. Place PCR tubes in thermos cycler, and the cycle as follows: (1) Initial denaturing: 9 4 °C for 5 min ; (2) 30 cycles of amplification: denaturing at 9 4 °C for 1 min , annealing at 50 °C for 45 sec , elongation at 7 1 °C for 100 sec ; (3) Final extension: 72 °C for 5 min ; (4) Storage: 4 °C for i nfinity time. 5. Mix 5 µL of PCR product with 1 µL loading dye (6×), and load the dyed PCR product onto a sheet of 1% agarose 1× TAE gel for electrophoresis (100 V) for approximately 20 - 30 min. 6. Transfer the gel sheet from previous st ep into EtBr (caution: carcinogen), stain for 5 min. Then transfer stained gel into water, de - stain for 10 min. 95 7. Observe the existence, position, width, and brightness of bands under UV light. If the band looks ideal in pilot PCR, proceed to actual PCR and submit the product for T - RFLP analysis. 96 APPENDIX E : Procedure for archaeal 16S rDNA clon ing 1. TOPO Cloning reaction: mix 1 µL salt solution, 1 µL TOPO vector and 2 µL of water with 2 µL of archaeal PCR product (using un - labelled forward primer) for each reaction. Vortex gently and incubate for 20 - 30 min at room temperature. 2. Add 2 µL of the TOPO Cloning reaction from previous step into a vial of One Shot Chemically Competent E. coli and m ix gently. (Do not mix using pipette.) 3. Incubate on ice for 10 min, and heat - shock the cell at 42 ° C for 30 sec without shaking. Then immediately transfer to ice. 4. Add 250 µL S.O.C. medium at room temperature, cap the tube tightly and shake horizontally on a n orbital shaker (200 rpm) at 37 ° C for 1 hour. 5. While waiting, spread 40 µL of 40 mg mL - 1 X - gal solution on each plate and warm up the plates at 37 ° C for at least 20 min. 6. Spread 10 µL of the solution from each transformation and 20 µL S.O.C. medium on the pre - warmed selective plate (kanamycin) and incubate upside - down at 37 ° C overnight. 7. Pick successfully inserted white colonies using toothpicks and add into growth medium on a 96 well plate. Cover each well tightly to prevent cross - con tamination and evaporation. Culture on an orbital shaker (80 - 100 rpm) at room temperature overnight. 8. A biological replicate of the 96 well plate can be generated on the second day. 9. Sequence the cultures of inserted competen t cells using Sanger s method. Note: This procedure was modified based on the procedure for TOPO Cloning Kit . 97 APPENDIX F : Statistical analysis for AD performance a nal ysis Table AP.F.1: Two - way ANOVA: Biogas (mL/L AD) versus Temp, Ratio Source DF SS MS F P Temp 1 32179 32179.5 114.81 0.00 1 Ratio 2 73895 36947.6 131.82 0.00 1 Interaction 2 27483 13741.7 49.03 0.00 1 Error 6 1682 280.3 Total 11 135240 S = 16.74 R - Sq = 98.76% R - Sq(adj) = 97.72% Ta ble AP.F.2: Two - way ANOVA: TS reduced (%) versus Temp, Ratio Source DF SS MS F P Temp 1 22.431 22.4307 4.94 0.068 Ratio 2 25.627 12.8137 2.82 0.137 Interaction 2 60.454 30.2271 6.65 0.030 Error 6 27.258 4.5430 Total 11 135.770 S = 2.131 R - Sq = 79.92% R - Sq(adj) = 63.19% Table AP.F.3: Two - way ANOVA: Productivity (TS) versus Temp, Ratio Source DF SS MS F P Temp 1 6679 6678.9 2.0 1 0.206 Ratio 2 127444 63721.9 19.17 0.002 Interaction 2 362 181.1 0.05 0.947 Error 6 19946 3324.4 Total 11 154431 S = 57.66 R - Sq = 87.08% R - Sq(adj) = 76.32% Table AP.F.4: Two - way ANOVA: Cellulose in residue versus Temp, Ratio Source DF SS MS F P Temp 1 0.0001259 0.0001259 0.25 0.632 Ratio 2 0.0007174 0.0003587 0.73 0.522 Interaction 2 0.0004194 0.0002097 0.42 0.672 Error 6 0. 0029659 0.0004943 Total 11 0.0042286 S = 0.02223 R - Sq = 29.86% R - Sq(adj) = 0.00% 98 Table AP.F.5: Two - way ANOVA: Xylan in residue versus Temp, Ratio Source DF SS MS F P Temp 1 0.0000337 0.0000337 0.5 7 0.478 Ratio 2 0.0002053 0.0001027 1.75 0.253 Interaction 2 0.0000096 0.0000048 0.08 0.922 Error 6 0.0003529 0.0000588 Total 11 0.0006015 S = 0.007669 R - Sq = 41.34% R - Sq(adj) = 0.00% Table AP.F.6: Two - way ANOVA: Lignin in residue versus Temp, Ratio Source DF SS MS F P Temp 1 0.0000000 0.0000000 0.00 0.998 Ratio 2 0.0015936 0.0007968 2.47 0.165 Interaction 2 0.0004505 0.0002253 0.70 0.53 4 Error 6 0.0019389 0.0003232 Total 11 0.0039831 S = 0.01798 R - Sq = 51.32% R - Sq(adj) = 10.75% Table AP.F.7: Pairwise comparison: Biogas (mL/L AD) 35C, 100/0 35C, 90/10 35C, 80/20 50C, 100/0 50C, 90/10 50C, 80/20 35C, 100/0 0.0135 0.0077 0.2411 0.0203 <0.0001 35C, 90/10 0.0002 0.0707 0.0007 <0.0001 35C, 80/20 0.0012 0.6778 <0.0001 50C, 100/0 0.0046 <0.0001 50C, 90/10 <0.0001 50C, 80/20 Table AP.F.8: Pairwise comparison: TS reduced (%) 35C, 100/0 35C, 90/10 35C, 80/20 50C, 100/0 50C, 90/10 50C, 80/20 35C, 100/0 0.0249 0.0332 0.372 0 0.3037 0.9262 35C, 90/10 0.286 0 0.1551 0.0002 <.0001 35C, 80/20 0.206 0 0.0001 <.0001 50C, 100/0 0.8325 0.254 0 50C, 90/10 0.0016 50C, 80/20 99 APPENDIX G : Control tests for alga l growth (in DI water) and nutrient reduction (TN, TP, Iron) in batch kinetics study Figure AP . 1 : Algal growth in DI water based on cell density (OD 750nm , unitless), standardized biovolume (100 × dilution, unit: mL cell per mL cultu re), and biomass total solids (unit: g L - 1 ). Figure AP . 2: Total nitrogen reduction in control culture without algae inoculation. 0 0.5 1 1.5 2 0 2 4 6 8 10 Algae culture in DI water as control Time (day) OD750 (unitless) Std. Biovolume (100X) (mL/mL) TS (g/L) 0 100 200 300 400 0 2 4 6 8 10 TN (mg/L), control Time (day) 1X 2X 5X 100 Figure AP . 3: Total phosphorus reduction in control culture without algae inoculation. Figure AP . 4: Total iron reduction in control culture without algae inoculation. 0 10 20 30 0 2 4 6 8 10 TP (mg/L), control Time (day) 1X 2X 5X 0 2 4 6 0 2 4 6 8 10 Iron (mg/L), control Time (day) 1X 2X 5X 101 APPENDIX H : Algal carbohydrate analysis procedure 1. Weigh 25.0 ± 2.5 mg of the algal biomass into a tared autoclave - safe bottle. 2. Add 250 µL of 72% sulfuric acid. Vortex gently for mixing . 3. Place bottles in 30 ± 3 °C incubator for 60 min, vortex gently in every 5 - 10 min. 4. After 60 min incubation, add 7 mL deionized water to dilute the acid to a 4% concentration. Mix sample in solution gently, and autoclave the solution at 121 °C for 1 hour. Al low the bottles to cool down before removing the caps. 5. Neutralize the autoclaved solution with calcium carbonate to pH 5 - 6. Filter the n eutralized solution using 0.22 µ m pore size syringe filter for HPLC analysis of sugar composition. 102 APPENDIX I : Algal p rotein analysis procedure ( bicinchoninic acid assay ) 1. Mixing 2 mL bicinchoninic acid solution (Reagent A) with 0.04 mL copper (II) sulfate pentahydrate 4% (w/v) solution (Reagent B) for each testing sample (including standards). 2. For standard curve of protein, prepare 200, 400, 600, 800, and 1000 µg mL - 1 BSA solutions. Add 0.1 mL of each standard solution into 2 mL working reagent from previous step and vortex thoroughly. 3. For unknown samples, add 0.1 mL of each sample solution into 2 mL working reagent and vortex thoroughly. 4. Incubate mixed standards and samples at 60 ° C for 15 min. 5. Transfer incubated standards and samples into cuvettes and measure at 562 nm using a UV spectrophotometer. 6. Create a standard curve using A 562 readings of the reacted standards . Then curve - fit the unknown samples to acquire protein concentration. 103 APPENDIX J : Algal crude lipid extraction procedure 1. Weigh 0.5 g dry algal biomass. Add 2.5 mL chloroform and 5 mL methanol to dry algal sample and mix with a handheld homogenizer for 2 min. 2. Add additional 2.5 mL chloroform and homogenize for 30 sec. 3. Add 2.5 mL deionized water to the previous mixture and homogenize for 30 sec. 4. Filter the mixture through Whatman No. 1 filter paper on a Coors No. 3 Buchner funnel with slight suction by a n air pump. 5. Transfer filtrate liquid to a glass vial and allow complete separation of water phase (top layer) and chloroform - lipid phase (bottom layer), approximately in 2 min. 6. Carefully remove bottom layer with pipette and place in an aluminum tray. Leave the tray in a semi - covered and ventilated hood until all solvents have evaporated. 7. Weigh the residual (lipid) and calculate crude lipid concentration (%) based on initial weight of algal biomass. 104 APPENDIX K : Statistical analysis for alga l growth and nut rient reduction in kinetics study Table AP.K.1: Two - way ANOVA: TS versus Dilution, Time Source DF SS MS F P Dilution 2 0.0017060 0.0008530 200.16 0.000 Time 6 0.0051195 0.0008533 200.22 0.000 Interaction 12 0.0039369 0.0003281 76.98 0.000 Error 21 0.0000895 0.0000043 Total 41 0.0108519 S = 0.002064 R - Sq = 99.18% R - Sq(adj) = 98.39% Table AP.K.2: Two - way ANOVA: OD750 versus Dilution, Time Source DF SS MS F P Dilution 2 1.73650 0.868249 173.20 0.000 Time 6 2.41273 0.402121 80.22 0.000 Interaction 12 2.59112 0.215927 43.07 0.000 Error 21 0.10527 0.005013 Total 41 6.84562 S = 0.07080 R - Sq = 9 8.46% R - Sq(adj) = 97.00% Table AP.K.3: Two - way ANOVA: Std. Biovolume versus Dilution, Time Source DF SS MS F P Dilution 2 221454 110727 1161.18 0.000 Time 6 439473 73245 768.12 0.000 Interaction 12 358790 29899 313.55 0.000 Error 21 2003 95 Total 41 1021719 S = 9.765 R - Sq = 99.80% R - Sq(adj) = 99.62% Table AP.K.4: Two - way ANOVA: TN versus Dilution, Time Source DF SS MS F P Dilution 2 608483 304241 5364.96 0.000 Time 6 32066 5344 94.24 0.000 Interaction 12 7275 606 10.69 0.000 Error 21 1191 57 Total 41 649015 S = 7.531 R - Sq = 99.82% R - Sq(adj) = 99.64% 105 Table AP.K.5: Tw o - way ANOVA: TP versus Dilution, Time Source DF SS MS F P Dilution 2 2778.21 1389.10 7937.73 0.000 Time 6 192.80 32.13 183.62 0.000 Interaction 12 64.40 5.37 30.67 0.000 Error 21 3.67 0.17 Total 41 3039.08 S = 0.4183 R - Sq = 99.88% R - Sq(adj) = 99.76% Table AP.K.6: Two - way ANOVA: Fe versus Dilution, Time Source DF SS MS F P Dilution 2 55.439 27.7193 2818.79 0.000 Time 6 39.450 6.5750 668.62 0.000 Interaction 12 11.650 0.9708 98.72 0.000 Error 21 0.207 0.0098 Total 41 106.745 S = 0.09917 R - Sq = 99.81% R - Sq(adj) = 99.62% Table AP.K.7: Two - way ANOVA: Turbidity versus Dilution, Time Source DF SS MS F P Dilution 2 0.068179 0.0340896 3261.42 0.000 Time 6 0.236341 0.0393902 3768.53 0.000 Interaction 12 0.118767 0.0098972 946.89 0.000 Error 21 0.000219 0.0000105 Total 41 0.423506 S = 0.003233 R - Sq = 99.95% R - Sq(adj) = 99.90% 106 APPENDIX L : Statistical analysis for algae enhanced enzymatic hydrolysis of lignocelluloses Table AP.L.1: Differences of Least Squares Means Standard Effect fiber pretreatment medium _fiber _pretreatment _medium Error fiber*pretrea*medium AD H Buffer AD H Algae 0.05931 fiber*pretrea*medium AD H Buffer AD OH Buffer 0.05931 fiber*pretrea*medium AD H Buffer AD OH Algae 0.05931 fiber*pretrea*medium AD H Buffer CS H Buffer 0.04284 fiber*pretrea*medium AD H Buffer CS H Algae 0.04284 fiber*pretrea*medium AD H Buffer CS OH Buffer 0.04284 fiber *pretrea*medium AD H Buffer CS OH Algae 0.04284 fiber*pretrea*medium AD H Buffer Poplar H Buffer 0.04523 fiber*pretrea*medium AD H Buffer Poplar H Algae 0.04523 fiber*pretrea*medium AD H Buffer Poplar OH Buffer 0.04523 fiber*pretrea*medium AD H Buffer Poplar OH Algae 0.04523 fiber*pretrea*medium AD H Buffer SG H Buffer 0.05136 fiber*pretrea*medium AD H Buffer SG H Algae 0.05136 fiber*pretrea*medium AD H Buffer SG OH Buffer 0.05136 fiber*pretrea *medium AD H Buffer SG OH Algae 0.05136 fiber*pretrea*medium AD H Algae AD OH Buffer 0.05931 fiber*pretrea*medium AD H Algae AD OH Algae 0.05931 fiber*pretrea*medium AD H Algae CS H Buffer 0.04284 fiber*pretrea*medium AD H Algae CS H Algae 0.04284 fiber*pretrea*medium AD H Al gae CS OH Buffer 0.04284 fiber*pretrea*medium AD H Algae CS OH Algae 0.04284 fiber*pretrea*medium AD H Algae Poplar H Buffer 0.04523 fiber*pretrea*med ium AD H Algae Poplar H Algae 0.04523 fiber*pretrea*medium AD H Algae Poplar OH Buffer 0.04523 fiber*pretrea*medium AD H Algae Poplar OH Algae 0.04523 fiber*pretrea*medium AD H Algae SG H Buffer 0.05136 fiber*pretrea*medium AD H Algae SG H Algae 0.05136 fiber*pretrea*medium AD H Algae SG OH Buffer 0.05136 fiber*pretrea*medium AD H Algae SG OH Algae 0.05136 fiber*pretrea*medium AD OH Buffer AD OH Algae 0.05931 fiber*pretrea*medium AD OH Buffer CS H Buffer 0.04284 fiber*pretrea*medium AD OH Buffer CS H Algae 0.04284 fiber*pretrea*medium AD OH Buffer CS OH Buffer 0.0428 4 fiber*pretrea*medium AD OH Buffer CS OH Algae 0.04284 fiber*pretrea*medium AD OH Buffer Poplar H Buffer 0.04523 fiber*pretrea*medium AD OH Buffer Poplar H Algae 0.04523 fiber*pretrea*medium AD OH Buffer Poplar OH Buffer 0.04523 fiber*pretrea*medium AD OH Buffer Poplar OH Algae 0.04523 fiber*pretrea*medium AD OH Buffer SG H Buffer 0.05136 fiber*pretrea*medium AD OH Buffer SG H Algae 0.05136 fiber*pretrea*medium AD OH Buffer SG OH Buffer 0.05136 fibe r*pretrea*medium AD OH Buffer SG OH Algae 0.05136 fiber*pretrea*medium AD OH Algae CS H Buffer 0.04284 fiber*pretrea*medium AD OH Algae CS H Algae 0.04284 fiber*pretrea*medium AD OH Algae CS OH Buffer 0.04284 fiber*pretrea*medium AD OH Algae CS OH Algae 0.04284 fiber*pretrea*medium AD OH Algae Poplar H Buffer 0.04523 fiber*pretrea*medium AD OH Algae Poplar H Algae 0.04523 fiber*pretrea*medium AD OH Algae Poplar OH Buffer 0.04523 fiber*pretre a*medium AD OH Algae Poplar OH Algae 0.04523 fiber*pretrea*medium AD OH Algae SG H Buffer 0.05136 fiber*pretrea*medium AD OH Algae SG H Alg ae 0.05136 fiber*pretrea*medium AD OH Algae SG OH Buffer 0.05136 fiber*pretrea*medium AD OH Algae SG OH Algae 0.05136 fiber*pretrea*medium CS H Buffer CS H Algae 0.01237 107 fiber*pretrea*medium CS H Buffer CS OH Buffer 0.01237 fiber*pretrea*medium CS H Buffer CS OH Algae 0.01237 fiber*pretrea*medium CS H Buffer Poplar H Buffer 0.01906 fiber*pretrea*medium CS H Buffer Poplar H Algae 0.01906 fiber*pretrea*medium CS H Buffer Poplar OH Buffer 0.01906 fiber*pretrea*medium CS H Buffer Poplar OH Algae 0.01906 fiber*pretrea*medium CS H Buffer SG H Buffer 0.03091 fiber*pretrea*medium CS H Buffer SG H Algae 0.03091 fiber*pretrea*medium CS H Buffer SG OH Buffer 0.03091 fiber*pretrea*medium CS H Buffer SG OH Algae 0.03091 fiber *pretrea*medium CS H Algae CS OH Buffer 0.01237 fiber*pretrea*medium CS H Algae CS OH Algae 0.01237 fiber*pretrea*medium CS H Algae Poplar H Buffer 0.01906 fiber*pretrea*medium CS H Algae Poplar H Algae 0.01906 fiber*pretrea*medium CS H Algae Poplar OH Buffer 0.01906 fiber*pretrea*medium CS H Algae Poplar OH Algae 0.01906 fiber*pretrea*medium CS H Algae SG H Buffer 0.03091 fiber*pretrea*medium CS H Algae SG H Algae 0.03091 fiber*pretrea *medium CS H Algae SG OH Buffer 0.03091 fiber*pretrea*medium CS H Algae SG OH Algae 0.03091 fiber*pretrea*medium CS OH Buffer CS OH Alga e 0.01237 fiber*pretrea*medium CS OH Buffer Poplar H Buffer 0.01906 fiber*pretrea*medium CS OH Buffer Poplar H Algae 0.01906 fiber*pretrea*medium CS OH Buffer Poplar OH Buffer 0.01906 fiber*pretrea*medium CS OH Buffer Poplar OH Algae 0.01906 fiber*pretrea*medium CS OH Buffer SG H Buffer 0.03091 fiber*pretrea*medium CS OH Buffer SG H Algae 0.03091 fiber*pretrea*medium CS OH Buffer SG OH Buffer 0.03091 fiber*pretrea*medium CS OH Buffer SG OH Algae 0. 03091 fiber*pretrea*medium CS OH Algae Poplar H Buffer 0.01906 fiber*pretrea*medium CS OH Algae Poplar H Algae 0.01906 fiber*pretrea*medium CS OH Algae Poplar OH Buffer 0.01906 fiber*pretrea*medium CS OH Algae Poplar OH Algae 0.01906 fiber*pretrea*medium CS OH Algae SG H Buffer 0.03091 fiber*pretrea*medium CS OH Algae SG H Algae 0.03091 fiber*pretrea*medium CS OH Algae SG OH Buffer 0.03091 fiber*pretrea*medium CS OH Algae SG OH Algae 0.03091 fiber*pretrea*medium Poplar H Buffer Poplar H Algae 0.02396 fiber*pretrea*medium Poplar H Buffer Poplar OH Buffer 0.02396 fiber*pretrea*medium Poplar H Buffer Poplar OH Algae 0.02396 fiber*pretrea*medium Poplar H Buffer SG H Buffer 0.03414 fiber*pretrea*medium Poplar H Buffer SG H Algae 0.03414 fiber*pretrea*medium Poplar H Buffer SG OH Buffer 0.03414 fiber*pretrea*medium Poplar H Buffer SG OH Algae 0.03414 fiber*pretrea*medium Poplar H Algae Poplar OH Buffer 0.02396 fiber*pr etrea*medium Poplar H Algae Poplar OH Algae 0.02396 fiber*pretrea*medium Poplar H Algae SG H Buffer 0.03414 fiber*pretrea*medium Poplar H Algae SG H Algae 0.03414 fiber*pretrea*medium Poplar H Algae SG OH Buffer 0.03414 fiber*pretrea*medium Poplar H Algae SG OH Algae 0.03414 fiber*pretrea*medium Poplar OH B uffer Poplar OH Algae 0.02396 fiber*pretrea*medium Poplar OH Buffer SG H Buffer 0.03414 fiber*pretrea*medium Poplar OH Buffer SG H Algae 0.03414 fiber*pretrea*me dium Poplar OH Buffer SG OH Buffer 0.03414 fiber*pretrea*medium Poplar OH Buffer SG OH Algae 0.03414 fiber*pretrea*medium Poplar OH Algae SG H Buffer 0.03414 fiber*pretrea*medium Poplar OH Algae SG H Algae 0.03414 fiber*pretrea*medium Poplar OH Algae SG OH Buffer 0.03414 fiber*pretrea*medium Poplar OH Algae SG OH Algae 0.03414 fiber*pretrea*medium SG H Buffer SG H Algae 0.04192 fiber*pretrea*medium SG H Buffer SG OH Buffer 0.04192 fiber*pretrea*medium S G H Buffer SG OH Algae 0.04192 fiber*pretrea*medium SG H Algae SG OH Buffer 0.04192 fiber*pretrea*medium SG H Algae SG OH Algae 0.04 192 fiber*pretrea*medium SG OH Buffer SG OH Algae 0.04192 108 Table AP.L.1: (cont d) Effect fiber pretreatment medium _fiber _pretreatment _medium Adj P fiber*pretrea*medium AD H Buffer AD H Algae 0.0234 fiber*pretrea*medium AD H Buffer AD OH Buffer 0.0192 fiber*pretrea*medium AD H Buffer AD OH Algae 0.2262 fiber*pr etrea*medium AD H Buffer CS H Buffer 0.9397 fiber*pretrea*medium AD H Buffer CS H Algae 1.0000 fiber*pretrea*medium AD H Buffer CS OH B uffer 0.0067 fiber*pretrea*medium AD H Buffer CS OH Algae 0.0148 fiber*pretrea*medium AD H Buffer Poplar H Buffer 0.9105 fiber*pretrea*medium AD H Buffer Poplar H Algae <.0001 fiber*pretrea*medium AD H Buffer Poplar OH Buffer 0.0080 fiber*pretrea*medium AD H Buffer Poplar OH Algae 0.0014 fiber*pretrea*medium AD H Buffer SG H Buffer 0.3141 fiber*pretrea*medium AD H Buffer SG H Algae 1.0000 fiber*pretrea*medium AD H Buffer SG OH Buffer 0.0063 fiber*pretrea*medium AD H Buffer SG OH Algae 0.0212 fiber*pretrea*medium AD H Algae AD OH Buffer <.0001 fiber*pretrea*medium AD H Algae AD OH Algae 0.9891 fiber*pretrea*medium AD H Algae CS H Buffer <.0001 fiber*pretrea*medium AD H Algae CS H Algae 0.0005 fiber*pretrea*medium AD H Algae CS OH Buffer <.0001 fiber*pretrea*medium AD H Algae CS OH Algae <.0001 fiber*pretrea*medium AD H Algae Poplar H Buffer <.0001 fiber*pretrea *medium AD H Algae Poplar H Algae 0.3666 fiber*pretrea*medium AD H Algae Poplar OH Buffer <.0001 fiber*pretrea*medium AD H Algae Poplar OH Algae <.0001 fiber*pretrea*medium AD H Algae SG H Buffer <.0001 fiber*pretrea*medium AD H Algae SG H Algae 0.0062 fiber*pretrea*medium AD H Algae SG OH Buffer <.0001 fiber*pretrea*medium AD H Algae SG OH Algae <.0001 fiber*pretrea*medium AD OH Buffer AD OH Algae <.0001 fiber*pretrea*medium AD OH Buffer CS H Buffer 0.0155 fiber*pretrea*medium AD OH Buffer CS H Algae 0.0017 fiber*pretrea*medium AD OH Buffer CS OH Buffer 0.9960 fib er*pretrea*medium AD OH Buffer CS OH Algae 0.9461 fiber*pretrea*medium AD OH Buffer Poplar H Buffer 0.0344 fiber*pretrea*medium AD OH Buffer Poplar H Algae <.0001 fiber*pretrea*medium AD OH Buffer Poplar OH Buffer 0.9996 fiber*pretrea*medium AD OH Buffer Poplar OH Algae 1.0000 fiber*pretrea*medium AD OH B uffer SG H Buffer 0.6049 fiber*pretrea*medium AD OH Buffer SG H Algae 0.0055 fiber*pretrea*medium AD OH Buffer SG OH Buffer 1.0000 fiber* pretrea*medium AD OH Buffer SG OH Algae 1.0000 fiber*pretrea*medium AD OH Algae CS H Buffer 0.0014 fiber*pretrea*medium AD OH Algae CS H Algae 0.0128 fiber*pretrea*medium AD OH Algae CS OH Buffer <.0001 fiber*pretrea*medium AD OH Algae CS OH Algae <.0001 fiber*pretrea*medium AD OH Alga e Poplar H Buffer 0.0018 fiber*pretrea*medium AD OH Algae Poplar H Algae 0.0206 fiber*pretrea*medium AD OH Algae Poplar OH Buffer <.0001 fiber*pretrea*medium AD OH Algae Poplar OH Algae <.0001 fiber*pretrea*medium AD OH Algae SG H Buffer 0.0005 fiber*pretrea*medium AD OH Algae SG H Algae 0.0983 fiber*pretrea*medium AD OH Algae SG OH Buffer <.0001 fiber*pretrea*medium AD OH Algae SG OH Algae <.0001 fiber*pretrea*medium CS H Buffer CS H Algae 0.0406 fiber*pretrea*medium CS H Buffer CS OH Buffer <.0001 fiber*pretrea*medium CS H Buffer CS OH Algae <.0001 fiber*pretrea*medium CS H Buffer Poplar H Buffer 1.0000 fiber*pretrea*medium CS H Buffer Poplar H Algae <.0001 fiber*pretrea*medium CS H Buffer Poplar OH Buffer <.0001 fiber*pretre a*medium CS H Buffer Poplar OH Algae <.0001 fiber*pretrea*medium CS H Buffer SG H Buffer 0.4885 fiber*pretrea*medium CS H Buffer SG H Algae 0.6833 109 Table AP.L.1: (cont d) fiber*pretrea*medium CS H Buffer SG OH Buffer 0.0008 fiber*pretrea*medium CS H Buffer SG OH Algae 0.0057 fiber*pretrea*medium CS H Algae CS OH Buffer <.0001 fiber*pretrea*medium CS H Algae CS OH Algae <.0001 fiber*pretrea*medium CS H Algae Poplar H Buffer 0.2248 f iber*pretrea*medium CS H Algae Poplar H Algae <.0001 fiber*pretrea*medium CS H Algae Poplar OH Buffer <.0001 fiber*pretrea*medium CS H Algae Poplar OH Algae <.0001 fiber*pretrea*medium CS H Algae SG H Buffer 0.0356 fiber*pretrea*medium CS H Algae SG H Algae 1.0000 fiber*pretrea*medium CS H Algae SG OH Buffer <.0001 fiber*pretrea*medium CS H Algae SG OH Algae 0.0003 fiber*pretrea*medium CS OH Buffer CS OH Algae 0.9710 fiber*pretrea*med ium CS OH Buffer Poplar H Buffer 0.0001 fiber*pretrea*medium CS OH Buffer Poplar H Algae <.0001 fiber*pretrea*medium CS OH Buffer Poplar OH Buffer 1 .0000 fiber*pretrea*medium CS OH Buffer Poplar OH Algae 0.4390 fiber*pretrea*medium CS OH Buffer SG H Buffer 0.6519 fiber*pretrea*medium CS OH Buffer SG H Algae 0.0002 fiber*pretrea*medium CS OH Buffer SG OH Buffer 0.9737 fiber*pretrea*medium CS OH Buffer SG OH Algae 1.0000 fiber*pretrea*medium CS OH Algae Poplar H Buffer 0.0006 fiber*pretrea*medium CS OH Algae Poplar H Algae <.0001 fiber*pretrea*medium CS OH Algae Poplar OH Buffer 0.9835 fiber*p retrea*medium CS OH Algae Poplar OH Algae 0.1041 fiber*pretrea*medium CS OH Algae SG H Buffer 0.9274 fiber*pretrea*medium CS OH Algae SG H Algae 0.0006 fiber*pretrea*medium CS OH Algae SG OH Buffer 0.7739 fiber*pretrea*medium CS OH Algae SG OH Algae 0.9995 fiber*pretrea*medium Poplar H Buffer Poplar H Algae <.0001 fiber*pretrea*medium Poplar H Buffer Poplar OH Buffer 0.0009 fiber*pretrea*medium Poplar H Buffer Poplar OH Algae <.0001 fiber*pretrea*medi um Poplar H Buffer SG H Buffer 0.7679 fiber*pretrea*medium Poplar H Buffer SG H Algae 0.6615 fiber*pretrea*medium Poplar H Buffer SG OH Buffer 0. 0037 fiber*pretrea*medium Poplar H Buffer SG OH Algae 0.0225 fiber*pretrea*medium Poplar H Algae Poplar OH Buffer <.0001 fiber*pretrea*medium Poplar H Algae Poplar OH Algae <.0001 fiber*pretrea*medium Poplar H Algae SG H Buffer <.0001 fiber*pretrea*medium Poplar H Algae SG H Algae <.0001 fiber*pretrea*medium Poplar H Algae SG OH Buffer <.0001 fiber*pretrea*medium Poplar H Algae SG OH Algae <.0001 fiber*pretrea*medium Poplar OH Buffer Poplar OH Algae 0.8986 fiber*pr etrea*medium Poplar OH Buffer SG H Buffer 0.6406 fiber*pretrea*medium Poplar OH Buffer SG H Algae 0.0005 fiber*pretrea*medium Poplar OH Buffer SG OH B uffer 0.9982 fiber*pretrea*medium Poplar OH Buffer SG OH Algae 1.0000 fiber*pretrea*medium Poplar OH Algae SG H Buffer 0.1174 fiber*pretrea*medium Poplar OH Algae SG H Algae <.0001 fiber*pretrea*medium Poplar OH Algae SG OH Buffer 1.0000 fiber*pretrea*medium Poplar OH Algae SG OH Algae 0.9972 fiber*pretrea*medium SG H Buffer SG H Algae 0.1205 fiber*pretrea*medium SG H Buffer SG OH Buffer 0.3836 fiber*pretrea*medium SG H Buffer SG OH Algae 0.8237 fiber*pretrea*medium SG H Algae SG OH Buffer 0.0008 fiber*pretrea*medium SG H Algae SG OH Algae 0.0035 fiber*pretrea*medium SG OH Buffer SG OH Algae 0.9999 110 A general linear model using the Statistical Analysis System program 9.2 (SAS institute Inc., NC) was applied to perform an analysis of variance (ANOVA) and multiple comparisons of the normalized glucose yield from all eight enz pair - wise comparison and slicing were carried out to determine the simple main effect of each factor (fiber, pretreatment and medium) and their interactions. Table AP.L.2: Tests of Effect Slices Num Den Effect fiber pretreatment medium DF DF F Value Pr > F fiber*pretrea*medium AD H 1 16 19.53 0.0004 fiber*pretrea*mediu m AD OH 1 16 58.33 <.0001 fiber*pretrea*medium CS H 1 16 16.96 0.0008 fiber*pretrea*medium CS OH 1 16 2.17 0.1604 fiber*pretrea*medium Poplar H 1 16 377.77 <.0001 fiber*pretrea*medium Poplar OH 1 16 3.30 0.0472 fiber*pretrea*medium SG H 1 16 12.26 0.0030 fiber*pretrea*medium SG OH 1 16 0.63 0.4378 fiber*pretrea*medium H Buffer 3 16 3.23 0.0504 fiber*pretrea*medium H Algae 3 16 160.32 <.0001 fiber*pretrea*medium OH Buffer 3 16 1.11 0.3745 fiber*pretrea*medium OH Algae 3 16 33.72 <.0001 fiber*pretrea*medium AD Buffer 1 16 20.49 0.0003 fiber*pretrea*medium AD Algae 1 16 1.71 0.2092 fiber*pretrea*medium CS Buffer 1 16 144.00 <.0001 fiber*pretrea*medium CS Algae 1 16 214.52 <.0001 fiber*pretrea*medium Poplar Buffer 1 16 38.17 <.0001 fiber*pretrea*medium Poplar Algae 1 16 749.46 <.0001 fiber*pretrea*medium SG Buffer 1 16 7.50 0.0146 fiber*pretrea*medium SG Algae 1 16 29.64 <.0001 fiber* pretrea*medium AD 3 16 31.63 <.0001 fiber*pretrea*medium CS 3 16 124.72 <.0001 fiber*pretrea*medium Poplar 3 16 314.61 <.0001 fiber*pretrea*medium SG 3 16 15.46 <.0001 fiber*pretrea*medium H 7 16 96.63 <.0001 fiber*pretrea*medium OH 7 16 15.51 <.0001 fiber*pretrea*medium Buffer 7 16 31.21 <.0001 fiber*pretrea*medium Algae 7 16 179.29 <.0001 111 REFERENCES 112 REFERENCES Abdo, Z., Schüette, U.M.E., Bent, S.J., Williams, C.J., Forney, L.J., Joyce, P., 2006. Statistical methods for characterizing diversity of microbial communities by analysis of terminal restriction fragment length polymorph isms of 16S rRNA genes. Environ Microbiol , 8, 929 - 38. Adey, W. H., P. C. Kangas, and W. Mulbry, 2011, Algal Turf Scrubbing: Cleaning Surface Waters with Solar Energy while Producing a Biofuel: Bioscience, v. 61, p. 434 - 441. Allers, E., C. Niesner, C. Wild , and J. Pernthaler, 2008, Microbes enriched in seawater after addition of coral mucus: Applied and Environmental Microbiology, v. 74, p. 3274 - 3278. Anderson, I.J., Ulrich, L.E., Lupa, B., Susanti, D., Porat, I., Hooper, S.D., Lykidis, A., Sieprawska - Lupa, M., Dharmarajan, L., Goltsman, E., Lapidus, A., Saunders, E., Han, C., Land, M., Lucas, S., Mukhopadhyay, B., Whitman, W.B., Woese, C., Bristow, J., Kyrpides, N., 2009. Genomic characterization of methanomicrobiales reveals three classes of methanogens. P LoS ONE , 4, e5797. Aparicio PJ, Cardenas J, Zumft WG, Ma Vega J, Herrera J, Paneque A, Losada P. 1971. Molybdenum and iron as constituents of the enzymes of the nitrate reducing system from Chlorella. 10 (7): 1487 - 1495 Astals S, Batstone DJ, Mata - Alvarez J, Jensen PD. 2014. Identification of synergistic impacts during anaerobic co - digestion of organic wastes Bioresour. Technol., 169, 421 427 Astals S, Musenze RS, Bai X, Tannock S, Tait S, Pratt S, Jensen PD. 2015. Anaerobic co - digestion of pig manure and a lgae: Impact of intracellular algal products recovery on co - digestion performance, Bioresource Technology, Volume 181, 97 - 104 Azad HS, Borchardt JA. 1970. VARIATIONS IN PHOSPHORUS UPTAKE BY ALGAE: Environmental Science & Technology, v. 4, p. 737 - 743. Azov, Y., and J. C. Goldman, 1982, FREE AMMONIA INHIBITION OF ALGAL PHOTOSYNTHESIS IN INTENSIVE CULTURES: Applied and Environmental Microbiology, v. 43, p. 735 - 739. Balch WE, Schoberth S, Tanner RS, Wolfe RS, 1977. Acetobacterium, a new genus of hydrogen - oxidiz ing, carbon dioxide - reducing, anaerobic bacteria. International Journal of Systematic and Evolutionary Microbiology. 24:4, 355 - 361. Balk, M., Weijma, J., Stams, A.J.M., 2002. Thermotoga lettingae sp. nov., a novel thermophilic, methanol - degrading bacterium isolated from a thermophilic anaerobic reactor. Int J Syst Evol Microbiol , 52, 1361 - 8. 113 Barnet, J.W., Kerridge, G.J., Russell, J.M., 1994. Effluent treatment systems of the dairy industry. Australian Biotechnology. 4 (1): 26 - 30. Becker, E. W. 1994. Microa lgae Biotechnology and microbiology. Cambridge University Press, Cambridge. Bellebia, S., S. Kacha, A. Z. Bouyakoub, and Z. Derriche, 2012, Experimental investigation of chemical oxygen demand and turbidity removal from cardboard paper mill effluents using combined electrocoagulation and adsorption processes: Environmental Progress & Sustainable Energy, v. 31, p. 361 - 370. Bernard, P. C., R. E. Vangrieken, and L. Brugmann, 1989, GEOCHEMISTRY OF SUSPENDED MATTER FROM THE BALTIC SEA .1. RESULTS OF INDIVIDUAL P ARTICLE CHARACTERIZATION BY AUTOMATED ELECTRON - MICROPROBE: Marine Chemistry, v. 26, p. 155 - 177. Bjornsson, L., P. Hugenholtz, G. W. Tyson, and L. L. Blackall, 2002, Filamentous Chloroflexi (green non - sulfur bacteria) are abundant in wastewater treatment pr ocesses with biological nutrient removal: Microbiology - Sgm, v. 148, p. 2309 - 2318. Bligh, E. G., and W. J. Dyer, 1959, A RAPID METHOD OF TOTAL LIPID EXTRACTION AND PURIFICATION: Canadian Journal of Biochemistry and Physiology, v. 37, p. 911 - 917. Boone, D.R. , Bryant, M., 1980. Propionate - Degrading Bacterium, Syntrophobacter wolinii sp. nov. gen. nov., from Methanogenic Ecosystems. Appl Environ Microbiol , 40, 626 - 632.. Brinkman, B. M., F. Hildebrand, M. Kubica, D. Goosens, J. del Favero, W. Declercq, J. Raes , and P. Vandenabeele, 2011, Caspase deficiency alters the murine gut microbiome: Cell Death & Disease, v. 2. Bryant MP, Wolin E a, Wolin MJ, Wolfe RS. (1967). Methanobacillus omelianskii, a symbiotic association of two species of bacteria. Archiv für Mikr obiologie 59:20 31. http://www.ncbi.nlm.nih.gov/pubmed/5602458. Buffle, J., R. R. Devitre, D. Perret, and G. G. Leppard, 1989, PHYSICOCHEMICAL CHARACTERISTICS OF A COLLOIDAL IRON PHOSPHATE SPECIES FORMED AT THE OXIC - ANOXIC INTERFACE OF A EUTROPHIC LAKE: Ge ochimica Et Cosmochimica Acta, v. 53, p. 399 - 408. Carroll A and Somerville C. 2009. Cellulosic Biofuels. Annual Review of Plant Biolog. Vol. 60, pp. 165 - 182 Chandler, D. P., F. J. Brockman, T. J. Bailey, and J. K. Fredrickson, 1998, Phylogenetic diversity of archaea and bacteria in a deep subsurface paleosol: Microbial Ecology, v. 36, p. 37 - 50. 114 Chelf P. 1990. Envrironmental control of lipid and biomass production in two diatom species. Journal of Applied Phycology. 2: 121 - 129. Chen H. 2014. Biotechnology of Lignocellulose: Theory and Practice. Springer. ISBN 978 - 94 - 007 - 6897 - 0. Chen H. 2015. Lignocellulose Biorefinery Engineering: Principles and Applications. Woodhead Publishing. ISBN 978 - 0 - 08 - 10035 - 6. Chen, R., B. D. Thomas, Y. Liu, W. Mulbry, and W. Liao, 2 014, Effects of algal hydrolyzate as reaction medium on enzymatic hydrolysis of lignocelluloses: Biomass and Bioenergy, v. 67, p. 72 - 78. Chen, R., M. M. Rojas - Downing, Y. Zhong, C. M. Saffron, and W. Liao, 2015, Life Cycle and Economic Assessment of Anaero bic Co - digestion of Dairy Manure and Food Waste: Industrial Biotechnology, v. 11, p. 127 - 139. Chen, R., R. Li, L. Deitz, Y. Liu, R. J. Stevenson, and W. Liao, 2012a, Freshwater algal cultivation with animal waste for nutrient removal and biomass production : Biomass & Bioenergy, v. 39, p. 128 - 138. Chen, R., Z. B. Yue, L. Deitz, Y. Liu, W. Mulbry, and W. Liao, 2012b, Use of an algal hydrolysate to improve enzymatic hydrolysis of lignocellulose: Bioresource Technology, v. 108, p. 149 - 154. Chen, Y., J. J. Cheng , and K. S. Creamer, 2008, Inhibition of anaerobic digestion process: A review: Bioresource Technology, v. 99, p. 4044 - 4064. Chenowyth, D.P. 1987, Anaerobic Digestion of Biomass. Ed. P. Chenowyth, R. Isaacson. New York: Springer. Chisti Y. 2007. Biodiesel from Microalgae, Biotechnology Advances 25, 294 - 306. Cho JK, Park SC, Chang HN. 1995. Biochemical methane potential and solid state anaerobic digestion of Korean food wastes. Bioresour. Technol., 52 (3), 245 253. Choi, H. J., and S. M. Lee, 2015, Effect of the N/P ratio on biomass productivity and nutrient removal from municipal wastewater: Bioprocess and Biosystems Engineering, v. 38, p. 761 - 766. Chouari R, Paslier DL, Daegelen P, Ginestet P, Weissenbach J, Sghir A. 2005. Novel predominant archaeal and bacterial groups revealed by molecular analysis of an anaerobic sludge digester. Environ Microbiol 7:1104 1115 115 Chynoweth, D. P. and Pullammanappallil, P. 1996. Anaerobic Digestion of Municipal Solid Wastes. in Palmisano, A.C. and Barlaz, M.A. eds. Micro biology of Solid Waste. CrC Press, Inc., Boca Raton, Florida., pp. 71 - 113 Clarridge JE. 3rd. 2004. Impact of 16S rRNA gene sequence analysis for identification of bacteria on clinical microbiology and infectious diseases. Clin Microbiol Rev. 17:840 62 Col e, J. R., Q. Wang, J. A. Fish, B. L. Chai, D. M. McGarrell, Y. N. Sun, C. T. Brown, A. Porras - Alfaro, C. R. Kuske, and J. M. Tiedje, 2014, Ribosomal Database Project: data and tools for high throughput rRNA analysis: Nucleic Acids Research, v. 42, p. D633 - D642. Conners, S. B., E. F. Mongodin, M. R. Johnson, C. I. Montero, K. E. Nelson, and R. M. Kelly, 2006, Microbial biochemistry, physiology, and biotechnology of hyperthermophilic Thermotoga species: Fems Microbiology Reviews, v. 30, p. 872 - 905. Coughlan, M., and F. Mayer, 1992, The cellulose - decomposing bacteria and their enzyme systems, in A. Balows, H. Truper, M. Dworkin, W. Harder, and K. Schleifer, eds., Prokaryotes: a handbook on the biology of bacteria, 2nd ed.: New York, Springer, p. 460 - 516. Dean R B. 1991. PROCESSES FOR WATER RECLAMATION. Waste Management & Research 9(5):425 - 430. Dearman, B., P. Marschner, and R. H. Bentham, 2006, Methane production and microbial community structure in single - stage batch and sequential batch systems anaerobically co - digesting food waste and biosolids: Applied Microbiology and Biotechnology, v. 69, p. 589 - 596. ogy, 21(8 - 9), pp. 1509 - 1514. Dong, X., Xin, Y., Jian, W., Liu, X., Ling, D., 2000. Bifidobacterium thermacidophilum sp. nov., isolated from an anaerobic digester. Int J Syst Evol Microbiol , 50 Pt 1, 119 - 25. Edgar, R. C., B. J. Haas, J. C. Clemente, C. Qui nce, and R. Knight, 2011, UCHIME improves sensitivity and speed of chimera detection: Bioinformatics, v. 27, p. 2194 - 2200. Eixler, S., U. Karsten, and U. Selig, 2006, Phosphorus storage in Chlorella vulgaris (Trebouxiophyceae, Chlorophyta) cells and its de pendence on phosphate supply: Phycologia, v. 45, p. 53 - 60. Eriksson, T., J. Borjesson, and F. Tjerneld, 2002, Mechanism of surfactant effect in enzymatic hydrolysis of lignocellulose: Enzyme and Microbial Technology, v. 31, p. 353 - 364. Esty, W. W., 1986, T HE EFFICIENCY OF GOOD NONPARAMETRIC COVERAGE ESTIMATOR: Annals of Statistics, v. 14, p. 1257 - 1260. 116 Fakruddin AC, Nur Hossain KSBM, Mazumdar RM, 2012. Pyrosequencing - principles and applications. International Journals of Life Science & Phama Research, 2(2), L 65 - 76 Field CB, Behrenfeld MJ, Randerson JT. 1998. Primary production of the biosphere: Integrating terrestrial and oceanic components. Science, 281 (5374): 237 - 240 Fleurence, J. (1999). Seaweed proteins: biochemical, nutritional aspects and potential u ses. Trends in food science & technology, 10(1), 25 - 28. Flint, H. J., E. A. Bayer, M. T. Rincon, R. Lamed, and B. A. White, 2008, Polysaccharide utilization by gut bacteria: potential for new insights from genomic analysis: Nature Reviews Microbiology, v. 6, p. 121 - 131. Gallert, C., and J. Winter, 1997, Mesophilic and thermophilic anaerobic digestion of source - sorted organic wastes: effect of ammonia on glucose degradation and methane production: Applied Microbiology and Biotechnology, v. 48, p. 405 - 410. Ga nesan, A., S. Chaussonnerie, A. Tarrade, C. Dauga, T. Bouchez, E. Pelletier, D. Le Paslier, and A. Sghir, 2008, Cloacibacillus evryensis gen. nov., sp nov., a novel asaccharolytic, mesophilic, amino - acid - degrading bacterium within the phylum 'Synergistetes ', isolated from an anaerobic sludge digester: International Journal of Systematic and Evolutionary Microbiology, v. 58, p. 2003 - 2012. Garnier, M., G. Carrier, H. Rogniaux, E. Nicolau, G. Bougaran, B. Saint - Jean, and J. P. Cadoret, 2014, Comparative proteo mics reveals proteins impacted by nitrogen deprivation in wild - type and high lipid - accumulating mutant strains of Tisochrysis lutea: Journal of Proteomics, v. 105, p. 107 - 120. Glenn T. 2011. Field guide to next - generation DNA sequencers. Mol. Ecol. Resour. 11, 759 769 Global Advantech, 2011. Electrocoagulation and advanced electrochemical oxidation. Technology Data Sheet TDS801 EN09. Global Health and Education Foundation. 2007. Safe Drinking Water is Essential: Conventional Coagulation Flocculation Sedimen tation. National Academy of Sciences. Goberna M, Insam H, Franke - Whittle IH. 2009. Effect of biowaste sludge maturation on the diversity of thermophilic bacteria and archaea in an anaerobic reactor. Appl Environ Microbiol 75:2566 2572 Gou, C. L., Z. H. Ya ng, J. Huang, H. L. Wang, H. Y. Xu, and L. K. Wang, 2014, Effects of temperature and organic loading rate on the performance and microbial community of anaerobic co - digestion of waste activated sludge and food waste: Chemosphere, v. 105, p. 146 - 151. 117 Greenw ell, H. C., L. M. L. Laurens, R. J. Shields, R. W. Lovitt, and K. J. Flynn, 2010, Placing microalgae on the biofuels priority list: a review of the technological challenges: Journal of the Royal Society Interface, v. 7, p. 703 - 726. Groom, M. J., E. M. Gray , and P. A. Townsend, 2008, Biofuels and biodiversity: Principles for creating better policies for biofuel production: Conservation Biology, v. 22, p. 602 - 609. Gunnars, A., S. Blomqvist, P. Johansson, and C. Andersson, 2002, Formation of Fe(III) oxyhydroxide colloids in freshwater and brackish seawater, with incorporation of phosphate and calcium: Geochimica Et Cosmochimica Acta, v. 66, p. 745 - 758. Haas, B. J., D. Gevers, A. M. Earl, M. Feldgarden, D. V. Ward, G. Giannoukos, D. Ciulla, D. Tabbaa, S. K. Highlander, E. Sodergren, B. Methe, T. Z. DeSantis, J. F. Petrosino, R. Knight, B. W. Birren, and C. Human Microbiome, 2011, Chimeric 16S rRNA sequence formation and detection in Sanger and 454 - pyrosequenced PCR amplicons: Genome Research, v. 21, p. 494 - 504. Hall L, Doerr KA, Wohlfiel LS, Roberts GD. 2003. Evaluation of the MicroSeq system for Identification of mycobacteria by 16S ribosomal DNA sequencing and its integration into a routine clinical mycobacteriology laboratory. J. Clin. Microbiol. 41:1 447 - 1453 Hamdani, A., Chennaoui, M., Assobhei, O., Mountadar, M., 2004. Dairy effluent characterization and treatment by coagulation - decantation. Lait. 84: 317 - 328. Harmsen, P. F. H., W. J. J. Huijgen, L. M. B. Lopez, and R. R. C. Bakker, 2010, Literature review of physical and chemical pretreatment process for lignocellulosic biomass, Energy Research Centre of the Netherlands. Hashimoto, A. G., V. H. Varel, and Y. R. Chen, 1981, ULTIMATE METHANE YIELD FROM BEEF - CATTLE MANURE - EFFECT OF TEMPERATURE, RATION CONSTITUENTS, ANTIBIOTICS AND MANURE AGE: Agricultural Wastes, v. 3, p. 241 - 256. Hopkins EF, Wann FB. 1927. Iron requirement for Chlorella. 84 (4): 407 - 427 Humenik, F., J. Martin, M. Rice, and K. Roos, 2004, Anaerobic digestion of animal manure, www.epa.g ov. Jönsson LJ, Alriksson B, Nilvebrant N - O. 2013. Bioconversion of lignocellulose: inhibitors and detoxification. Biotechnology for Biofuels. 6:16 Karns, J. S., C. J. Hapeman, W. W. Mulbry, E. H. Ahrens, and D. R. Shelton, 1998, Biotechnology for the elim ination of agrochemical wastes: Hortscience, v. 33, p. 626 - 631. Kattar MM, Chavez JF, Limaye AP, Rassoulian - Barrett SL, Yarfitz SL, Carlson LC, Houze Y, Swanzy S, Wood BL, Cookson BT. 2001. Application of 16S rRNA gene sequencing to 118 identify Bordetella hin zii as the causative agent of fatal septicemia. J. Clin. Microbiol. 38:789 - 794 Kay RA and Barton LL. 1991. Microalgae as food and supplement. Food Science and Nutrition. 30: 555 - 573. Kebede - Westhead, E., Pizarro, C., Mulbry, W., 2004. Treatment of dairy ma nure effluent using freshwater algae: elemental composition of algal biomass at different manure loading rates. Journal of Agricultural and Food Chemistry. 52: 7293 7296. 51. Khalid, A., M. Arshad, M. Anjum, T. Mahmood, and L. Dawson, 2011, The anaerobic digestion of solid organic waste: Waste Management, v. 31, p. 1737 - 1744. King, C., Scott - Horton, T. Pyrosequencing: A Simple Method for Accurate Genotyping. J. Vis. Exp. ( 11), e630. Koch, M., C. Rudolph, C. Moissl, and R. Huber, 2006, A cold - loving crenarchaeon is a substantial part of a novel microbial community in cold sulphidic marsh water: Fems Microbiology Ecology, v. 57, p. 55 - 66. Kothari, R., V. V. Pathak, V. Kumar, and D. P. Singh, 2012, Experimental study for growth potential of unicellular alga Chlorella pyrenoidosa on dairy waste water: An integrated approach for treatment and biofuel production: Bioresource Technology, v. 116, p. 466 - 470. Lednicka, D., J. Mergaer t, M. C. Cnockaert, and J. Swings, 2000, Isolation and identification of cellulolytic bacteria involved in the degradation of natural cellulosic fibres: Systematic and Applied Microbiology, v. 23, p. 292 - 299. Lee, M., T. Hidaka, W. Hagiwara, and H. Tsuno, 2009, Comparative performance and microbial diversity of hyperthermophilic and thermophilic co - digestion of kitchen garbage and excess sludge: Bioresource Technology, v. 100, p. 578 - 585. Lee, S. H., J. H. Park, H. J. Kang, Y. H. Lee, T. J. Lee, and H. D. P ark, 2013, Distribution and abundance of Spirochaetes in full - scale anaerobic digesters: Bioresource Technology, v. 145, p. 25 - 32. Li Y. 2012. Cultivation of algae on highly concentrated municipal wastewater as an energy crop for biodiesel production. Doct oral dissertation. University of Minnesota. Liu ZG, Stromberg D, Liu XM, Liao W, Liu Y. 2015. A new multiple - stage electrocoagulation process on anaerobic digestion effluent to simultaneously reclaim water and clean up biogas. Journal of Hazardous Material s. 285:483 - 90. 119 Liu, J., and W. Vyverman, 2015, Differences in nutrient uptake capacity of the benthic filamentous algae Cladophora sp., Klebsormidium sp and Pseudanabaena sp under varying N/P conditions: Bioresource Technology, v. 179, p. 234 - 242. Liu, K., Y. Q. Tang, T. Matsui, S. Morimura, X. L. Wu, and K. Kida, 2009, Thermophilic anaerobic co - digestion of garbage, screened swine and dairy cattle manure: Journal of Bioscience and Bioengineering, v. 107, p. 54 - 60. Liu, W. T., Marsh, T. L., Cheng, H., & For ney, L. J. (1997). Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied and environmental microbiology, 63(11), 4516 - 4522. Liu, Z. G., D. Stromberg, X. M. Liu, W. Liao , and Y. Liu, 2015, A new multiple - stage electrocoagulation process on anaerobic digestion effluent to simultaneously reclaim water and clean up biogas: Journal of Hazardous Materials, v. 285, p. 483 - 490. Liu, Z. G., Z. H. Ruan, Y. Xiao, Y. Yi, Y. J. J. Ta ng, W. Liao, and Y. Liu, 2013, Integration of sewage sludge digestion with advanced biofuel synthesis: Bioresource Technology, v. 132, p. 166 - 170. Liu, Z., and Y. Liu, 2015, Synergistic Integration of Electrocoagulation and Algal cultivation to Treat Liquid Anaerobic Digestion Effluent and Accumulate Algal Biomass Process Biochemistry, v. (in review). Loehr, R. C., 1974, Agricultural Waste Management: Problems, Processes and Approaches: New York, Academic Press. Loman NJ, Misra RV, Dallman TJ, Constant inidou C, Gharbia SE, Wain J, Pallen MJ. 2012. Performance comparison of benchtop high - throughput sequencing platforms. Natural Biotechnology. 30, 434 - 439. Macias - Corral, M., Z. Samani, A. Hanson, G. Smith, P. Funk, H. Yu, and J. Longworth, 2008, Anaerobic digestion of municipal solid waste and agricultural waste and the effect of co - digestion with dairy cow manure: Bioresource Technology, v. 99, p. 8288 - 8293. Mackie, R. I., and M. P. Bryant, 1995, ANAEROBIC - DIGESTION OF CATTLE WASTE AT MESOPHILIC AND THERM OPHILIC TEMPERATURES: Applied Microbiology and Biotechnology, v. 43, p. 346 - 350. Mackie, R.I., Bryant, M.P., 1981. Metabolic activity of fatty acid - oxidizing bacteria and the contribution of acetate, propionate, butyrate, and CO(2) to methanogenesis in cat tle waste at 40 and 60 degrees C. Appl Environ Microbiol , 41, 1363 - 73. Marsh TL, 1999. Terminal restriction fragment length polymorphism (T - RFLP): an emerging method for characterizing diversity among homologous populations of amplification products. Curr ent opinion in microbiology, 2(3), 323 - 327. 120 Martin - Gonzalez, L., R. Castro, M. A. Pereira, M. M. Alves, X. Font, and T. Vicent, 2011, Thermophilic co - digestion of organic fraction of municipal solid wastes with FOG wastes from a sewage treatment plant: Rea ctor performance and microbial community monitoring: Bioresource Technology, v. 102, p. 4734 - 4741. Mata - Alvarez J, Dosta J, Romero - Güiza MS, Fonoll X, Peces M, Astals S. 2014. A critical review on anaerobic co - digestion achievements between 2010 and 2013 R enew. Sustain. Energy Rev., 36, 412 427 Mata - Alvarez, J., Mace, S., & Llabres, P. 2000. Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives.Bioresource technology, 74(1), 3 - 16. McHenry, M., 2009, The algae val ue chain landscape - a competitive technical challenge, Perth, Western Australia, Division of Research and Development and the School of Engineering and Energy, Murdoch University. McKeown RM, Scully C, Enright AM, Chinalia FA, Lee C, Mahony T, Collins G, O'Flaherty V, 2009. Psychrophilic methanogenic community development during long - term cultivation of anaerobic granular biofilms. The ISME journal, 3(11), 1231 - 1242. McMurdie, P. J., and S. Holmes, 2014, Waste Not, Want Not: Why Rarefying Microbiome Data I s Inadmissible: Plos Computational Biology, v. 10. Metzker M. 2010. Sequencing technologies the next generation. Nat. Rev. Genet. 11, 31 46. Miller, T. L., and C. Z. Lin, 2002, Description of Methanobrevibacter gottschalkii sp nov., Methanobrevibacter thau eri sp nov., Methanobrevibacter woesei sp nov and Methanobrevibacter wolinii sp nov: International Journal of Systematic and Evolutionary Microbiology, v. 52, p. 819 - 822. Moeseneder, M. M., Winter, C., Arrieta, J. M., & Herndl, G. J. (2001). Terminal - restr iction fragment length polymorphism (T - RFLP) screening of a marine archaeal clone library to determine the different phylotypes. Journal of Microbiological Methods, 44(2), 159 - 172. Mollah, M. Y. A., P. Morkovsky, J. A. G. Gomes, M. Kesmez, J. Parga, and D. L. Cocke, 2004, Fundamentals, present and future perspectives of electrocoagulation: Journal of Hazardous Materials, v. 114, p. 199 - 210. Mollah, M. Y. A., R. Schennach, J. R. Parga, and D. L. Cocke, 2001, Electrocoagulation (EC) - science and applications : Journal of Hazardous Materials, v. 84, p. 29 - 41. Monnet, F. 2003. An Introduction to Anaerobic Digestion of Organic Wastes. Executive summary, Remade Scotland. http://www.biogasmax.co.uk/media/introanaerobicdigestion__073323000_1011_24042007 .pdf 121 Monnet, F., 2003, An introduction to anaerobic digestion of organic wastes, final report., Remade Scotland. Mulbry, W. W., and A. C. Wilkie, 2001, Growth of benthic freshwater algae on dairy manures: Journal of Applied Phycology, v. 13, p. 301 - 306. Mulbry, W., S. Kondrad, and C. Pizarro, 2006, Biofertilizers from algal treatment of dairy and swine manure effluents: characterization of algal biomass as a slow release fertilizer: Journal of Vegetable Science, v. 12, p. 107 - 125. Mulbry, W., S. Kondrad, and J. Buyer, 2 008, Treatment of dairy and swine manure effluents using freshwater algae: fatty acid content and composition of algal biomass at different manure loading rates: Journal of Applied Phycology, v. 20, p. 1079 - 1085. Mulbry, W., S. Kondrad, C. Pizarro, and E. Kebede - Westhead, 2008, Treatment of dairy manure effluent using freshwater algae: Algal productivity and recovery of manure nutrients using pilot - scale algal turf scrubbers: Bioresource Technology, v. 99, p. 8137 - 8142. Müller V, 2003. Energy conservation i n acetogenic bacteria. Appl. Enviorn. Microbiol., 69:11, 6345 - 6353. Müller, V. and Frerichs, J. 2013. Acetogenic Bacteria. eLS. DOI: 10.1002/9780470015902.a0020086.pub2 Mullings, R., and J. H. Parish, 1984, MESOPHILIC AEROBIC GRAM - NEGATIVE CELLULOSE DEGRA DING BACTERIA FROM AQUATIC HABITATS AND SOILS: Journal of Applied Bacteriology, v. 57, p. 455 - 468. NRC, 2001, Nutrient requirements of dairy cattle, in N. R. Council, ed., Washington, D.C., The National Academies Press. Nutrition, S. o. D. C., C. o. A. Nutrition, and N. R. Council, 2001, Nutrient Requirements of Dairy Cattle: Seventh Revised Edition, 2001, The National Academies Press. bacterial community during anaerobic digestion. Biotechnology and Bioengineering. 92 (7): 871 - 878 Obama, B. 2011. State of the Union Address. http://www.whitehouse.gov/the - press - office/2011/01/25/remarks - president - state - union - address Obour A. 2015. Improving the efficiency of cattle manure utilization for forage production. Cattle Research & Education Center, University of Florida/IFAS. http://rcrec - ona.ifas.ufl.edu/in - focus/IF7 - 21 - 06.shtml 122 Ohmori, M., K. Ohmori, and H. Strotmann, 1977, INHIBITION OF NITRATE UPTAKE BY AMMONIA IN A BL UE - GREEN - ALGA, ANABAENA - CYLINDRICA: Archives of Microbiology, v. 114, p. 225 - 229. O'Sullivan, C.A., Burrell, P.C., Clarke, W.P., Blackall, L.L., 2005. Structure of a cellulose degrading bacterial community during anaerobic digestion. Biotechnol Bioeng, 92, 871 - 8. Pacheco, M. M., M. Hoeltz, M. S. A. Moraes, and R. C. S. Schneider, 2015, Microalgae: Cultivation techniques and wastewater phycoremediation: Journal of Environmental Science and Health Part a - Toxic/Hazardous Substances & Environmental Engineering, v. 50, p. 585 - 601. Parga, J. R., V. Vazquez, H. M. Casillas, and J. L. Valenzuela, 2009, Cyanide Detoxification of Mining Wastewaters with TiO2 Nanoparticles and Its Recovery by Electrocoagulation: Chemical Engineering & Technology, v. 32, p. 1901 - 1908. P arravicini, V., K. Svardal, and H. Kroiss, 2008, Post - aeration of anaerobically digested sewage sludge for advanced COD and nitrogen removal: results and cost - benefit analysis at large - scale: Water Science and Technology, v. 57, p. 1087 - 1094. Patel JB, Leo nard DG, Pan X, Musser JM, Berman RF, Nachamkin I. 2000. Sequence - based identification of Mycobacterium species using the Microseq 500 16S rDNA bacterial identification system. J. Clin. Microbiol. 38:246 - 251 Pimentel D and Patzek TW. 2005. Ethanol producti on using corn, switchgrass, and wood; biodiesel production using soybean and sunflower. Natural Resources Research. 14: 65 - 76. Piorreck, M., Baasch, K. H., & Pohl, P. (1984). Biomass production, total protein, chlorophylls, lipids and fatty acids of freshw ater green and blue - green algae under different nitrogen regimes. Phytochemistry, 23(2), 207 - 216. Pizarro, C., E. Kebede - Westhead, and W. Mulbry, 2002, Nitrogen and phosphorus removal rates using small algal turfs grown with dairy manure: Journal of Applie d Phycology, v. 14, p. 469 - 473. Pollack A (22 August 2003). "Company Says It Mapped Genes of Virus in One Day". The New York Times Powell, N., A. Shilton, Y. Chisti, and S. Pratt, 2009, Towards a luxury uptake process via microalgae - Defining the polyphos phate dynamics: Water Research, v. 43, p. 4207 - 4213. Preidis, G. A., D. M. Saulnier, S. E. Blutt, T. - A. Mistretta, K. P. Riehle, A. M. Major, S. F. Venable, M. J. Finegold, J. F. Petrosino, M. E. Conner, and J. Versalovic, 2012, Probiotics stimulate entero cyte migration and microbial diversity in the neonatal mouse intestine: Faseb Journal, v. 26, p. 1960 - 1969. 123 Riviere, D., V. Desvignes, E. Pelletier, S. Chaussonnerie, S. Guermazi, J. Weissenbach, T. Li, P. Camacho, and A. Sghir, 2009, Towards the definitio n of a core of microorganisms involved in anaerobic digestion of sludge: Isme Journal, v. 3, p. 700 - 714. Roche JL, Boyd PW, McKay RML, Geider RJ. 1996. Flavodoxin as an in situ marker for iron stress in phytoplankton. Nature. 382: 802 - 805 Roelke, D. L., P. M. Eldridge, and L. A. Cifuentes, 1999, A model of phytoplankton competition for limiting and nonlimiting nutrients: Implications for development of estuarine and nearshore management schemes: Estuaries, v. 22, p. 92 - 104. Ruan, Z. H., M. Zanotti, Y. Zhong , W. Liao, C. Ducey, and Y. Liu, 2013, Co - hydrolysis of lignocellulosic biomass for microbial lipid accumulation: Biotechnology and Bioengineering, v. 110, p. 1039 - 1049. Rueler JG, Ades DR. 1987. The role of iron nutrition in photosynthesis and nitrogen as similation in Scenedesmus quadricauda (Chlorophyceae). Journal of Phycology. 23 (3): 452 - 457 Ruiz - Martinez, A.; Martin Garcia, N.; Romero, I.; Seco, A.; Ferrer, J. 2012 Microalgae cultivation in wastewater: Nutrient removal from anaerobic membrane bioreact or effluent. Bioresource Technol. 126, 247 253. Safferman, S. I., D. M. Kirk, L. L. Faivor, and W. - H. Wei, 2012, Anaerobic digestion processes, in R. Mohee, and A. Mudhoo, eds., Bioremediation and Sustainability : Research and Applications, John Wiley & So ns, p. 103 - 136. Sasaki D, Hori T, Haruta S, Ueno Y, Ishii M, Igarashi Y. 2011. Methanogenic pathway and community structure in a thermophilic anaerobic digestion process of organic solid waste. J Biosci Bioeng 111:41 46 Sasaki K, Haruta S, Ueno Y, Ishii M, Igarashi Y. 2007. Microbial population in the biomass adhering to supporting material in a packed - bed reactor degrading organic solid waste. Appl Microbial Biotechnol 75:941 952 Sekiguchi, Y., H. Takahashi, Y. Kamagata, A. Ohashi, and H. Harada, 2001, In situ detection, isolation, and physiological properties of a thin filamentous microorganism abundant in methanogenic granular sludges: a novel isolate affiliated with a clone cluster, the green non - sulfur bacteria, subdivision I: Applied and Environmental Microbiology, v. 67, p. 5740 - 5749. Sekiguchi, Y., Y. Kamagata, K. Syutsubo, A. Ohashi, H. Harada, and K. Nakamura, 1998, Phylogenetic diversity of mesophilic and thermophilic granular sludges determined by 16S rRNA gene analysis: Microbiology - Uk, v. 144, p . 2655 - 2665. 124 Sheehan, J., T. Dunahay, J. Benemann, and P. Roessler, 1998, A look back at the U.S. Department of Energy's aquatic species program: biodiesel from algae., National Renewable Energy Laboratory. Shifrin NS, Chisholm SW. 1980. Phytoplankton lipids:environmental influences on production and possible commercial applications. In Algae Biomass. 627 - 645. Elsevier/North Holland Biochemical Press Amsterdam Sievers, D. M., and D. E. Brune, 1978, CARBON - NITROGEN RATIO AND ANAEROBIC DIGESTION OF SWINE WASTE: Transactions of the Asae, v. 21, p. 537 - &. Singh, D. P., and R. K. Trivedi, 2013, Acid and alkaline pretreatment of lignocellulosic biomass to produce ethanol as biofuel: International Journal of Chem Tech Research, v. 5, p. 727 - 734. Sivaramakrishna n, C.N., 2008. Effluent treatments: Coagulation. Colourage. 55: 44 - 47. Sloof, J.E., A. Viragh, B. Vanderveer. 1995. Kinetics of cadmium uptake by green - algae. Water Air Soil Poll. 83:105 - 122. Sluiter, A., B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Temp leton, and D. Crocker, 2008, Determination of structural carbohydrates and lignin in biomass: Laboratory analytical procedure (LAP), Golden, CO, USA, National Renewable Energy Laboratory. Smith, C. J., B. S. Danilowicz, A. K. Clear, F. J. Costello, B. Wils on, and W. G. Meijer, 2005, T - Align, a web - based tool for comparison of multiple terminal restriction fragment length polymorphism profiles: Fems Microbiology Ecology, v. 54, p. 375 - 380. Smith, P., D. Martino, Z. Cai, D. Gwary, H. Janzen, P. Kumar, B. McCa Rice, B. Scholes, and O. Sirotenko, 2007, Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, in B. D. Metz, O.R., P. R. Bosch, R. D ave, and L. A. Meyer, eds.: Cambridge, United Kingdom and New York, NY, USA, Cambridge University Press. Song, P. P., L. Li, and J. G. Liu, 2013, Proteomic Analysis in Nitrogen - Deprived Isochrysis galbana during Lipid Accumulation: Plos One, v. 8. Song, Y. C., S. J. Kwon, and J. H. Woo, 2004, Mesophilic and thermophilic temperature co - phase anaerobic digestion compared with single - stage mesophilic - and thermophilic digestion of sewage sludge: Water Research, v. 38, p. 1653 - 1662. Sun Y, Cheng J. 2002. Hydrol ysis of lignocellulosic materials for ethanol production: a review. Bioresource Technology. 83(1), 1 - 11 Sundberg, C., D. Yu, I. Franke - Whittle, S. Kauppi, S. Smars, H. Insam, M. Romantschuk, and H. Jonsson, 2013, Effects of pH and microbial composition on odour in food waste composting: Waste Management, v. 33, p. 204 - 211. 125 Syrett, P., 1962, Nitrogen assimilation, in R. Lewin, ed., Physiology and Biochemistry of Algae: New York, Academic Press, p. 171 - 188. Taherzadeh, M. J., and K. Karimi, 2008, Pretreatment of lignocellulosic wastes to improve ethanol and biogas production: A review: International Journal of Molecular Sciences, v. 9, p. 1621 - 1651. Tang YQ, Fujimura Y, Shigematsu T, Morimura S, Kida K. 2007. Anaerobic treatment performance and microbial popul ation of thermophilic upflow anaerobic filter reactor treating Awamori distillery wastewater. J Biosci Bioeng 104:281 287 Tang YQ, Ji P, Yayashi J, Koike Y, W XL, Kida K. 2011. Characteristic microbial community of a dry thermophilic methanogenic digester: its long - term stability and change with feeding. Appl Microbiol Biotechnol. 91: 1447 - 1461 Tang YW, Ellis NM, Hopkins MK, Smith DH, Dodge DE, Persing DH. 1998. Comparison of phenotypic and genotypic technique for identification of unusual aerobic pathogeni c gram - negative bacilli. J. Clin. Microbiol. 36:3674 - 3679 Tatsumoto, K., Baker, J.O., Tucker, M.P., Oh, K.K., Mohagheghi, A., Grohmann, K., Tatsumoto, K., J. O. Baker, M. P. Tucker, K. K. Oh, A. Mohagheghi, K. Grohmann, and M. E. Himmel, 1988, DIGESTION OF PRETREATED ASPEN SUBSTRATES - HYDROLYSIS RATES AND ADSORPTIVE LOSS OF CELLULASE ENZYMES: Applied Biochemistry and Biotechnology, v. 18, p. 159 - 174. Teater C, Yue Z, MacLellan J, Liu Y, Liao W, 2011. Assessing solid digestate from anaerobic digestion as fe edstock for ethanol production. Bioresour Technol;102:1856 - 1862. Teater, C., Z. B. Yue, J. MacLellan, Y. Liu, and W. Liao, 2011, Assessing solid digestate from anaerobic digestion as feedstock for ethanol production: Bioresource Technology, v. 102, p. 1856 - 1862. Tengborg, C., Galbe, M., Zacchi, G., 2001. Reduced inhibition of enzymatic hydrolysis of steam - pretreated softwood. Enzyme Microb. Technol. 28 (9 10), 835 844. Tiago, I., I. Teixeira, S. Silva, P. Chung, A. Verissimo, and C. M. Manaia, 2004, Metabol ic and genetic diversity of mesophilic and thermophilic bacteria isolated from composted municipal sludge on poly - epsilon - caprolactones: Current Microbiology, v. 49, p. 407 - 414. Tipping, E., and M. Ohnstad, 1984, COLLOID STABILITY OF IRON - OXIDE PARTICLES F ROM A FRESH - WATER LAKE: Nature, v. 308, p. 266 - 268. Tyagi VK, Khan AA, Kazmi AA, Chopra AK. 2010. Enhancement of Coagulation Flocculation Process Using Anionic Polymer for the Post Treatment of UASB Reactor Effluent. Separation Science and Technology 45(5) :626 - 634. 126 Ucar, G., 1990, PRETREATMENT OF POPLAR BY ACID AND ALKALI FOR ENZYMATIC - HYDROLYSIS: Wood Science and Technology, v. 24, p. 171 - 180. Ueki, A., Akasaka, H., Suzuki, D., Ueki, K., 2006. Paludibacter propionicigenes gen. nov., sp. nov., a novel stric tly anaerobic, Gram - negative, propionate - producing bacterium isolated from plant residue in irrigated rice - field soil in Japan. Int J Syst Evol Microbiol , 56, 39 - 44. US Environmental Protection Agency. 2008. Revised National Pollutant Discharge Eliminatio n System Permit Regulation and Effluent Limitations Guidelines for Concentrated Animal Feeding Operations in Response to the Waterkeeper Decision, Federal Register, 73(225), pp. 70418 - 70486 v an Leeuwen, Y. M., K. P. Velikov, and W. K. Kegel, 2012, Stabilization through precipitation in a system of colloidal iron(III) pyrophosphate salts: Journal of Colloid and Interface Science, v. 381, p. 43 - 47. v an Wychen, S., and L. Laurens, 2013, Determinati on of Total Carbohydrates in Algal Biomass. , in N. R. E. Laboratory, ed., Contract, p. 275 - 300. v an Lier, J. B., 1996, Limitations of thermophilic anaerobic wastewater treatment and the consequences for process design: Antonie Van Leeuwenhoek Internationa l Journal of General and Molecular Microbiology, v. 69, p. 1 - 14. Vartoukian, S. R., R. M. Palmer, and W. G. Wade, 2007, The division "Synergistes": Anaerobe, v. 13, p. 99 - 106. Verma, S., 2002, Anaerobic digestion of biodegradable organics in municipal soli d wastes, Columbia University. Vijayaraghavan K, Yun YS, 2008. Bacterial biosorbents and biosorption. Biotechnol. Adv. 26, 266 291. Voelkerding, K. V., Dames, S. A., & Durtschi, J. D. (2009). Next - generation sequencing: from basic research to diagnostics. Clinical chemistry, 55(4), 641 - 658. Walker JB. 1954. Inorganic micronutrient requirements of Chlorella. II. Quantitative requirements for iron, manganese, and zinc. Archives of Biochemistry and Biophysics. 53 (1): 1 - 8 Wan, S. G., L. Sun, J. Sun, and W. S. Luo, 2013, Biogas production and microbial community change during the Co - digestion of food waste with chinese silver grass in a single - stage anaerobic reactor: Biotechnology and Bioprocess Engineering, v. 18, p. 1022 - 1030. Weiland, P., 2006, State of the art of solid - state digestion - recent developments, in F. N. Rohstoffe, ed., Solid - State Digestion - State of the Art and Further R&D Requirements, v. 24, Gulzower Fachgespräche, p. 22 - 38. 127 Weiss A, Jérôme V, Burghardt D, Likke L, Peiffer S, Hofstetter EM, Gabl er R, Freitag R. 2009. Investigation of factors influencing biogas production in a large - scale thermophilic municipal biogas plant. Appl Microbiol Biotechnol 84:987 1001 Wheeler, D. A.; Srinivasan, M.; Egholm, M.; Shen, Y.; Chen, L.; McGuire, A.; He, W.; C hen, Y. J.; Makhijani, V.; Roth, G. T.; Gomes, X.; Tartaro, K.; Niazi, F.; Turcotte, C. L.; Irzyk, G. P.; Lupski, J. R.; Chinault, C.; Song, X. - Z.; Liu, Y.; Yuan, Y.; Nazareth, L.; Qin, X.; Muzny, D. M.; Margulies, M.; Weinstock, G. M.; Gibbs, R. A.; Rothb erg, J. M. (2008). "The complete genome of an individual by massively parallel DNA sequencing". Nature 452(7189): 872 876. Widjaja, A., C. C. Chien, and Y. H. Ju, 2009, Study of increasing lipid production from fresh water microalgae Chlorella vulgaris: Jo urnal of the Taiwan Institute of Chemical Engineers, v. 40, p. 13 - 20. Wiegel J, Tanner R, Rainey FA. 2005. An introduction to the family Clostridiaceae. In: Dworkin M (ed) The prokaryotes: an evolving electronic resource for the microbiological community, 3rd edn. Springer, New York, 654 678 Wilkie, A. 2005. Anaerobic Digestion of Dairy Manure: Design and Process Considerations, Dairy Manure Management: Treatment, Handling, and Community Relations, NRAES - 176, pp. 301 - 312. Wirth, R., E. Kovacs, G. Maroti, Z . Bagi, G. Rakhely, and K. L. Kovacs, 2012, Characterization of a biogas - producing microbial community by short - read next generation DNA sequencing: Biotechnology for Biofuels, v. 5. Wyman, C. E., 1996, Handbook on bioethanol: production and utilization: W ashington D.C., USA, Taylor & Francis. Wyman, C. E., B. E. Dale, R. T. Elander, M. Holtzapple, M. R. Ladisch, and Y. Y. Lee, 2005, Coordinated development of leading biomass pretreatment technologies: Bioresource Technology, v. 96, p. 1959 - 1966. Wyman, C. E., B. E. Dale, R. T. Elander, M. Holtzapple, M. R. Ladisch, Y. Y. Lee, C. Mitchinson, and J. N. Saddler, 2009, Comparative Sugar Recovery and Fermentation Data Following Pretreatment of Poplar Wood by Leading Technologies: Biotechnology Progress, v. 25, p . 333 - 339. Yamada, T., Sekiguchi, Y., Imachi, H., Kamagata, Y., Ohashi, A., Harada, H., 2005. Diversity, localization, and physiological properties of filamentous microbes belonging to Chloroflexi subphylum I in mesophilic and thermophilic methanogenic slu dge granules. Appl Environ Microbiol , 71, 7493 - 503. Yang, B., Wyman, C.E., 2006. BSA treatment to enhance enzymatic hydrolysis of cellulose in lignin containing substrates. Biotechnol. Bioeng. 94, 611 617. 128 Yang, F., R. Chen, Z. Yue, W. Liao, and T. Marsh, 2014, Phylogenetic Analysis of Anaerobic Co - digestion of Animal Manure and Corn Stover Reveals Linkages between Bacterial Communities and Digestion Performance: FEMS Microbiology Ecology, v. (in review) Yang, J.C., Chynoweth, D.P., Williams, D.S., Li, A. , 1990. Clostridium aldrichii sp. nov., a cellulolytic mesophile inhabiting a wood - fermenting anaerobic digester. Int J Syst Bacteriol , 40, 268 - 72. Yang, Z. K., Y. F. Niu, Y. H. Ma, J. Xue, M. H. Zhang, W. D. Yang, J. S. Liu, S. H. Lu, Y. F. Guan, and H. Y. Li, 2013, Molecular and cellular mechanisms of neutral lipid accumulation in diatom following nitrogen deprivation: Biotechnology for Biofuels, v. 6. Yu, D., J. M. Kurola, K. Lahde, M. Kymalainen, A. Sinkkonen, and M. Romantschuk, 2014, Biogas productio n and methanogenic archaeal community in mesophilic and thermophilic anaerobic co - digestion processes: Journal of Environmental Management, v. 143, p. 54 - 60. Yue Z, Teater C, MacLellan J, Liu Y, Liao W, 2010. A sustainable pathway of cellulosic ethanol pro duction integrating anaerobic digestion with biorefining. Biotechnol Bioeng; 105(6):1031 - 1039. Yue, Z. B., C. Teater, J. MacLellan, Y. Liu, and W. Liao, 2011, Development of a new bioethanol feedstock - Anaerobically digested fiber from confined dairy oper ations using different digestion configurations: Biomass & Bioenergy, v. 35, p. 1946 - 1953. Yue, Z. B., C. Teater, Y. Liu, J. MacLellan, and W. Liao, 2010, A Sustainable Pathway of Cellulosic Ethanol Production Integrating Anaerobic Digestion With Biorefini ng: Biotechnology and Bioengineering, v. 105, p. 1031 - 1039. Yue, Z. B., R. Chen, F. Yang, J. MacLellan, T. Marsh, Y. Liu, and W. Liao, 2013, Effects of dairy manure and corn stover co - digestion on anaerobic microbes and corresponding digestion performance: Bioresource Technology, v. 128, p. 65 - 71. Zehnder, A. J. B., 1988, Biology of anaerobic microorganisms: New York, Wiley. Zellner, G., C. Alten, E. Stackebrandt, E. C. Demacario, and J. Winter, 1987, ISOLATION AND CHARACTERIZATION OF METHANOCORPUSCULUM - PAR VUM, GEN - NOV, SPEC - NOV, A NEW TUNGSTEN REQUIRING, COCCOID METHANOGEN: Archives of Microbiology, v. 147, p. 13 - 20. Zhai N, Zhang T, Yin D, Yang G, Wang X, Ren G, Feng Y. 2015. Effect of initial pH on anaerobic co - digestion of kitchen waste and cow manure, W aste Management, Volume 38, 126 - 131 Zhang, C., Liu, X., Dong, X., 2004. Syntrophomonas curvata sp. nov., an anaerobe that degrades fatty acids in co - culture with methanogens. Int J Syst Evol Microbiol, 54, 969 - 73. 129 Zhang, Q. T., J. Ma, G. Y. Qiu, L. Li, S. Geng, E. Hasi, C. Li, G. Y. Wang, and X. Y. Li, 2012, Potential energy production from algae on marginal land in China: Bioresource Technology, v. 109, p. 252 - 260. Zhang, Y., E. M. Zamudio Canas, Z. W. Zhu, J. L. Linville, S. Chen, and Q. He, 2011, Robustn ess of archaeal populations in anaerobic co - digestion of dairy and poultry wastes: Bioresource Technology, v. 102, p. 779 - 785. Zverlov VV, Hiegl W, Kock DE, Kellermann J, Kollmeier T, Schwarz WH, 2010. Hydrolytic bacteria in mesophilic and thermophilic deg radation of plant biomass. Eng. Life Sci. 10:6, 528 - 536