Biochemically enhanced methane production from coal

For many years, biogas was connected mostly with the organic matter decomposition in shallow sediments (e.g., wetlands, landfill gas, etc.). Recently, it has been realized that biogenic methane production is ongoing in many hydrocarbon reservoirs. This research examined microbial methane and carbon dioxide generation from coal. As original contributions methane production from various coal materials was examined in classical and electro-biochemical bench-scale reactors using unique, developed facultative microbial consortia that generate methane under anaerobic conditions. Facultative methanogenic populations are important as all known methanogens are strict anaerobes and their application outside laboratory would be problematic. Additional testing examined the influence of environmental conditions, such as pH, salinity, and nutrient amendments on methane and carbon dioxide generation. In 44-day ex-situ bench-scale batch bioreactor tests, up to 300,000 and 250,000 ppm methane was generated from bituminous coal and bituminous coal waste respectively, a significant improvement over 20-40 ppm methane generated from control samples. Chemical degradation of complex hydrocarbons using environmentally benign reagents, prior to microbial biodegradation and methanogenesis, resulted in dissolution of up to 5% bituminous coal and bituminous coal waste and up to 25% lignite in samples tested.

BIOCHEMICALLY ENHANCED METHANE PRODUCTION FROM COAL by Aleksandra Opara A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of P hilosophy Department of The University of Utah August 2012 Civil and Environmental Engineering Copyright © Aleksandra Opara 2012 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of has been approved by the following supervisory committee members: , Chair Date Approved , Member Date Approved , Member Date Approved , Member Date Approved , Member Date Approved and by , Chair of the Department of and by Charles A. Wight, Dean of The Graduate School. Aleksandra Opara Michael L. Free 05/21/2012 D. Jack Adams 05/18/2012 John McLennan 05/11/2012 Jack Hamilton 05/01/2012 Brian McPherson 05/21/2012 Paul Tikalsky Civil and Environmental Engineering ABSTRACT For many years, biogas was connected mostly with the organic matter decomposition in shallow sediments (e.g., wetlands, landfill gas, etc.). Recently, it has been realized that biogenic methane production is ongoing in many hydrocarbon reservoirs. This research examined microbial methane and carbon dioxide generation from coal. As original contributions methane production from various coal materials was examined in classical and electro-biochemical bench-scale reactors using unique, developed facultative microbial consortia that generate methane under anaerobic conditions. Facultative methanogenic populations are important as all known methanogens are strict anaerobes and their application outside laboratory would be problematic. Additional testing examined the influence of environmental conditions, such as pH, salinity, and nutrient amendments on methane and carbon dioxide generation. In 44-day ex-situ bench-scale batch bioreactor tests, up to 300,000 and 250,000 ppm methane was generated from bituminous coal and bituminous coal waste respectively, a significant improvement over 20-40 ppm methane generated from control samples. Chemical degradation of complex hydrocarbons using environmentally benign reagents, prior to microbial biodegradation and methanogenesis, resulted in dissolution of up to 5% bituminous coal and bituminous coal waste and up to 25% lignite in samples tested. Research results confirm that coal waste may be a significant underutilized resource that could be converted to useful fuel. Rapid acidification of lignite samples resulted in low pH (below 4.0), regardless of chemical pretreatment applied, and did not generate significant methane amounts. These results confirmed the importance of monitoring and adjusting in situ and ex situ environmental conditions during methane production. A patented Electro-Biochemical Reactor technology was used to supply electrons and electron acceptor environments, but appeared to influence methane generation in a negative manner. Provision of electron acceptor environment might have given an advantage to methanotrophs present in the consortium. Availability of electron acceptors is a limiting step in methanotrophy under anaerobic conditions. TABLE OF CONTENTS ABSTRACT .............................................................................................................. iii 1. LITERATURE REVIEW .................................................................................... 1 1.1. Global Carbon Cycle and Methanogenesis ................................................ 1 1.2. Methane as a Green Fuel ............................................................................ 1 1.3. Taxonomy of Methanogens ......................................................................... 4 1.4. Methanogenic Pathways ............................................................................. 5 1.5. Biogenic Natural Gas from Complex Hydrocarbons ............................... 10 2. RESEARCH OBJECTIVES ............................................................................. 17 3. ORIGINAL CONTRIBUTION OF THE RESEARCH .................................... 18 4. MATERIALS AND EXPERIMENTAL PROCEDURES ................................ 19 4.1. Collection and Preparation of Materials .................................................. 19 4.2. Microbial Characterization ....................................................................... 23 4.3. Pretreatments of Carbonaceous Materials .............................................. 29 4.4. Gas Generation Tests in Serum Bottles .................................................. 31 4.5. Bench Scale Bioreactor Tests ................................................................... 36 5. RESULTS AND DISCUSSION ....................................................................... 38 5.1. Microbial Characterization ....................................................................... 38 5.2. Pretreatments of Carbonaceous Materials .............................................. 50 5.3. Gas Generation Tests in Serum Bottles .................................................. 57 5.4. Bench Scale Bioreactor Tests ................................................................... 88 6. CONCLUSIONS ................................................................................................. 96 6.1. Future Research ........................................................................................ 99 APPENDICES A. MEDIA COMPOSITION ................................................................................. 100 B. STANDARD OPERATING PROCEDURES .................................................. 102 C. LIST OF TESTED CHEMICAL PRETREATMENTS ................................... 139 D. MORPHOLOGY CHARACTERIZATION RESULTS ................................... 141 E. ENVIRONMENTAL INFLUENCES RESULTS ........................................... 145 F. RAMAN SPECTROSCOPY RESULTS .......................................................... 159 G. MICROBES IN LIQUID MEDIA RESULTS ................................................. 166 H. MICROBES IN COAL SAMPLES RESULTS ............................................... 170 I. MICROBIAL CONSORTIA IN PRETREATED COAL SAMPLES ............... 180 J. BENCH SCALE RESULTS ............................................................................. 189 REFERENCES ..................................................................................................... 195 1. LITERATURE REVIEW 1.1. Global Carbon Cycle and Methanogenesis On Earth, carbon constantly changes form through a set of complex aerobic and anaerobic conversions known as the carbon cycle (Figure 1). The atmosphere holds about 600-720 Gt of carbon, mostly as carbon dioxide and methane, 15% of which is fixed by plants annually through the process of photosynthesis (Figure 1a) [Falkowski et al., 2000; Killops and Killops, 2005; Opara, 2007]. When exposed to air, dead biomass decomposes back to CO2 (Figure 1b). In an anaerobic environment, such as subsurface sediments, wetlands or the rumen of animals, carbon is converted into methane. The organic matter is first fermented into fatty acids (such as acetate or formate), CO2, and H2 (Figure 1c) [Bryant, 1979; McInerney et al., 1979; Mah, 1982]. Then methanogenic microorganisms from the Archaea domain metabolize the fermentation products into methane (Figure 1d). Only a few percent of the biogenic methane is buried in suitable formations to form natural gas deposits [Thauer, 1998]. Most of it leaks into aerobic zones where it is oxidized photochemically and converted by menthanotrophic bacteria into CO2 (Figure 1e) [Cicerone, 1998; Zehnder and Brock, 1979]. Figure 1. Global carbon cycle (adapted from Ferry 2010a, Ferry 2010b). 1.2. Methane as a Green Fuel Methane is the main component of natural gas, fuel that - among other things - is used for residential and commercial heating, electricity generation, transportation, and as an industrial feedstock. Over 86% of natural gas used in the United States comes from within the country and 90% of imports come from North America [EIA, 2010b]. Because of its variety of uses, natural gas could replace coal as a main electricity source or crude oil as a main transportation fuel. Increased use of this domestically abundant fuel could reduce the American dependence on foreign oil. Additionally, methane offers many environmental benefits over other fossil fuels. In comparison to coal, burning natural gas releases half the amount of CO2, 80% less CO and NOx, and virtually no SOx, particulates or mercury. Comparison of air pollutant emissions resulting from combustion of natural gas, crude oil, and coal is given in Table 1. A transition period from fossil fuels towards renewable and Table 1. Air pollutant emissions by source in lbs/billion BTU energy consumed [EIA, 1999]. Pol lutant Natural Gas Crude Oi l Coal CO2 117,000 164,000 208,000 CO 40 33 208 NOx 92 448 457 SOx 1 1,122 2,591 Particulates 7 84 2,744 Mercury 0.000 0.007 0.016 sustainable fuels is needed. With expansion of currently identified resources, natural gas can be used in conventional steam-turbine power plants or to power vehicles as an exceptional transition fuel. The proven reserves of conventional natural gas in the United States amount to about 244 TCF (trillion cubic feet) [EIA, 2008], while the annual consumption reaches 23 TCF [EIA, 2010a]. Assuming no imports, conventional reserves will last for roughly 10 years. Since the American economy depends nearly as much on natural gas as it does on crude oil, the search for unconventional resources of natural gas within the United States has a high priority. In recent decades, capturing and burning landfill gas or methane produced in wastewater treatment plants became a significant portion of energy source for these facilities. Nevertheless, many undeveloped unconventional natural gas sources exist. Estimates of global identified sources of methane are given in Table 2. Some of these methane sources, such as landfill gas or municipal wastewater treatment, have been utilized. Intensive research is being performed around the world to tap into some other methane sources, including methane hydrates or gas produced by termites and Table 2. Estimated annual global sources of methane in Tg/year [Kvenvolden and Rogers, 2005; Opara, 2007]. Source Range Accepted average Natural 110-210 170 Wetlands 55-150 115 Termites and other insects 10-50 20 Oceans 5-50 15 Freshwater 10-35 5 Methane hydrates 0-10 10 Anthropogenic 300-450 375 Fossil fuel related 70-120 110 Rice paddies 20-120 110 Enteric fermentation 65-120 115 Animal waste 20-30 25 Domestic sewage treatment 15-80 25 Landfills 20-70 40 Biomass burning 20-80 40 other insects. However, large volumes of methane are being generated from area sources, i.e., wetlands and rice paddies or multiple point sources, i.e., enteric fermentation, which would be difficult to capture. 1.3. Taxonomy of Methanogens The taxonomy and ecology of methanogenic microorganisms have been studies thoroughly, due to their importance in the global carbon cycle. They are relatively slow growing anaerobes that are neither prokaryotes nor eukaryotes. They belong to a newest domain of life, proposed in 1990 by Woese and coworkers, Archaea (Figure 2) [Woese et al., 1990]. The creation of the domain of Archaea was based on the relationships derived from the 16S rRNA sequencing, which showed that they are not closely related to either of existing domains [Gupta, 1998]. It consists of two kingdoms: Crenarchaeota (thermophiles) and Euryarchaeota (extreme halophiles, methanotrophs, and methanogens) [Bintrim et al., 1997; Brown and Doolittle, 1997]. Furthermore, methanogens can be divided into five orders: Methanobacteriales, Methanosarcinales, Methanopyrales, Methanomicrobiales, and Methanococcales [Bapteste et al., 2005]. Within these five orders, there are over 50 described species of methanogens, which do not form a monophyletic group, although they all belong to Archaea. Methanogens are anaerobic microorganisms and cannot function under aerobic conditions. They are very sensitive to the presence of oxygen, even at trace levels. Usually, they cannot survive oxygen stresses for a prolonged time. Specific taxonomy of methanogenic microorganisms is given in Table 3. 1.4. Methanogenic Pathways There are two main methanogenic pathways: conversion of CO2 and H2, formate or alcohols, and conversion of methylated compounds or acetate to methane [Worm et al., 2010]. In the first methanogenic pathway, the substrate (hydrogen, formate) is the electron donor (Equations 1 and 2, Figure 3), while CO2 is the carbon source and the electron acceptor (Equation 3, Figure 3). As seen from Table 3, most of the methanogens can use hydrogen as the electron donor. This pathway, however, accounts for only a third of methane generated from freshwaters and bioreactors, such as domestic wastewater treatment facilities or landfills. Substrates in the Figure 2. Consensus drawing of the phylogenetic tree showing three domains of life. Groups of organisms and their location is based on rRNA sequence comparison [Woese et al., 1990; Lecointre and Le Guyader; 2006]. BACTERIA EUCARYA ARCHAEA thermotoga flavobacteria cyanobacteria proteobacteria gram positive bacteria green non-sulfur bacteria thermoacidophiles crenarchaea methanogens methylotrophs halophiles ogens euryarchaea animals ciliates green plants fungi flagellates Table 3. Taxonomy of methanogens [Garcia, 1990; Zinder, 1993]. Organism (Order, Family, Genus ) Morphology Substrates Examples of species Optimum growth conditions Temp. [oC] pH NaCl [M] Methanobacteriales Methanobacteriaceae Methanobacterium Rod H2/CO2, formate*, alcohols* alcaliphilum, wolfei 35-70 5.6-9.1 <1.7 Methanosphaera Coccus H2/methanol cuniculi, stadtmanae 35-40 6.5-6.9 NA Methanobrevibacter Short rod H2/CO2, formate arboriphilus, smithii, 30-39 7.0-8.0 NA Methanothermaceae Methanothermus Rod H2/CO2 fervidus, sociabilis 80-88 6.5 NA Methanosarcinales Methanosarcinacae Methanosarcina Irregular coccus H2/CO2*, methanol, methylamines, acetate alcaliphilum, mazei, thermophila 30-50 6.5-7.5 <0.7 Methanolobus Irregular coccus methanol, methylamines siciliae, tindarius 37 6.5 0.5 Methanococcoides Irregular coccus methanol, methylamines euhalobius , methylutens 28-37 6.8-7.3 <1.0 Methanohalophi lus Irregular coccus methanol, methylamines mahii, zhilinae 35-45 7.4-9.2 <2.5 Methanolohalobium Irregular coccus methanol, methylamines evestigatus NA NA NA Methanoseataceae Methanosaeta Filament acetate thermoacetophila 35-60 6.5-7.5 NA Table 3. (continued) Organism (Order, Family, Genus ) Morphology Substrates Examples of species Optimum growth conditions Temp. [oC] pH NaCl [M] Methanopyrales Methanopyraceae Methanopyrus Rod H2/CO2 thermophila 98 6.5 0.25 Methanomicrobiales Methanoinicrobiaceae Methanomicrobium Short rod H2/CO2, formate*, alcohols* mobile 40 6.1-6.9 NA Methanocul leus Irregular coccus H2/CO2, formate, alcohols* marisnegri 20-25 6.8-7.3 0.1 Methanogenium Irregular coccus H2/CO2, formate, alcohols* anulus, friitonii 20-60 6.2-7.3 <0.3 Methanocorpusculaceae Methanocorpusculum Irregular coccus H2/CO2, formate, alcohols* aggregans, parvum 30-37 6.4-7.5 <0.8 Methanospirillaceae Methanospiri l lum Long spirals H2/CO2, formate, alcohols hungatei 30-37 NA NA Methanococcales Methanococcaceae Methanococcus Coccus H2/CO2, formate* maripludis, 26-85 5.0-8.0 <1.7 Figure 3. Simplified view of three methanogenesis pathways (acetoclastic, methylotrophic, and hydrogenotrophic) with electron and hydrogen ion flows indicated. second group (acetate, methylated compounds) act as both, the electron donor and the carbon source [Vandecasteele, 2008]. Even though there are only two genera identified as using the acetoclastic pathway, acetate decarboxylation is responsible for about two-thirds of methane production in freshwaters and bioreactors [Zinder, 1993]. Conversion of CO2 is the only methanogenic pathway having a net negative electron flow (Figure 3). Moreover, only a handful of electron donors, including hydrogen, formate, and alcohols, have been identified as suitable for this pathway. HCO3 - + 8e- + 9H+ CH4 + 3H2O (1) H2 2e- + 2H+ (2) HCOO- + H2O HCO3 - + 2e- + 2H+ (3) The lack of electrons and availability of the electron donors could be the reason why there is not more methane produced through this pathway. Methylated compounds, on the other hand, can be simultaneously oxidized to CO2, releasing six electrons, and reduced to methane through the reaction with coenzyme B, accepting two electrons. Lack of electron acceptors could be the limiting factor in this case. Finally, during the acetoclastic pathway two electrons are donated through the conversion of the carboxylic group into CO2, while a series of reactions between the methyl group with coenzymes B, M, and tetrahydrosarcinapterin accepts two electrons, resulting in net zero free electrons [Ferry, 2011]. Methanogenic reactions are shown in Table 4, indicating the importance of electron donors and acceptors in the process. From the thermodynamic standpoint, the most favorable conditions for methanogenesis are negative Eh and low pH values (Figure 4). At neutral pH, methanogens require ORP (oxidation-reduction potential) of -400mV or lower [Khanal, 2008], (Figure 4). By controlling pH and ORP of the environment, an optimal generation of methane might be obtained. 1.5. Biogenic Natural Gas from Complex Hydrocarbons Biogenic gas, produced from anoxic decomposition of organic matter by microorganisms, is considered an unconventional natural gas resource. For many years, biogenic gas (or biogas) was connected mostly with the decomposition of organic matter in shallow anoxic sediments (e.g., wetlands, marsh gas, methane associated with or produced from municipal wastewater treatment facilities and landfills, or from rice paddies). Recently, it has been realized that biogenic methane production is ongoing in many hydrocarbon reservoirs. Coalbed methane (CBM), for Table 4. Most important methanogenic reactions in ordered from the most to least thermodynamically favored, as defined by free energy change [Zinder, 1993; Thauer, 1998]. Electron donor Carbon source Reaction ΔG [kJ/mol CH4] Formate CO2 4HCO2 - + H2 + H2O CH4 + 3HCO3 - -145 Hydrogen CO2 4H2 + HCO3 - + H+ CH4 + 3H2O -135 Alcohol CO2 2CH3CH2OH + HCO3 - 2CH3COO- + H+ + CH4 + H2O -116 Methanol Methanol 4CH3OH 3CH4 + HCO3 - + H2O + H+ -105 Methylamine Methylamine 4(CH3)3NH+ + 9H2O 9CH4 + 3HCO3 - + 4NH4 - + 3H+ -76 Acetate Acetate CH3COO- + H2O CH4 + HCO3 - -31 Figure 4. Pourbaix diagram of C-N-S-H-O system at 25oC. Red lines indicate boundary under neutral pH. Note that C(H3)4(g) species is a tetratritiated methane. instance, has been believed for many years to have thermogenic as well as biogenic origins through decomposition of organic matter occurring during early stages of coalification [Thomas 2002]. Recent studies show, however, that coalbed methane may also be of a more recent biogenic origin, produced through microbial degradation and utilization of complex carbon compounds [Ulrich and Bower 2008]. The mechanism of anoxic coal biodegradation is not well understood [Strąpoć et al., 2011]. It is believed, however, that after initial fragmentation and activation of coal compounds, fermentation and oxidation of intermediate compounds leads to methane precursors (Figure 5). Similarly, large biogenic shale gas plays, such as the Antrim Shale of the Michigan Basin or the Colorado shale in Alberta, have been recently described [Curtis 2002; Jarvie et al. 2007]. Biogenic natural gas wells, however, tend to have low production rates and are poorly researched, which discourages many operators [Shurr and Ridgley, 2002]. From a different perspective, mining of coal and other hydrocarbon sources (e.g., tar sands and oil shales) results in mountainous waste heaps of mineral waste and lower grade coal materials, a potentially useful, but under-utilized, fuel source. In the United States alone, accumulated culm and gob (waste products of anthracite and bituminous coal mining, respectively) are estimated to be about two billion tons [Akers and Harrison 2000]. Annually, about 55 million tons of waste coal are generated [Tillman and Harding 2004] and pile up on mine sites as unprofitable mountains or valley fills that can potentially contribute to generation of metal contaminated waters, or are gravitationally Figure 5. Simplified diagram of coal biodegradation (adapted from Strąpoć et al., 2011). ! " !# $ !# !# % !# " # & ' ' unstable and a risk for slope failure. Acid mine drainage and leaching of various metals into neighboring watersheds is a problem. Moreover, waste coal heaps often catch fire and release toxic gases into the atmosphere [Stracher and Taylor, 2004]. Successful conversion of even a fraction of this waste material into useful fuel could prove advantageous to the mining industry and the environment. Coal heaps could be designed in a similar fashion to modern landfills allowing for active methanogenesis and collection and utilization of collected gases. Enhancing the recovery of low production gas wells and stimulating the methanogenesis in depleting gas wells for methane production have been the research topic of scientists in the last few years. Most of the published research has been focused on characterizing the microbial populations present in hydrocarbon reservoirs. Limited attention has been given to the microbiology of crude oil fields. Using molecular techniques, Grabowski and colleagues (2005), Higashioka and colleagues (2010), and Pham and colleagues (2009) described the microbial diversity of various petroleum reservoirs. Gieg and colleagues (2011) characterized the microbial population responsible for methanogenesis and crude oil biodegradation in an Alaskan oil field. Youssef with colleagues (2009) and Strąpoć with colleagues (2009) described various roles of microorganisms in oil fields, including crude oil degradation and active methanogenensis. Coalbed methane has been by far the focus point of most of the research in this area; nevertheless, only a small amount of work has been published. Scott and Kaiser discussed in 1995 the potential for microbially enhanced recovery of CBM. In recent years, numerous studies have been published describing microbial populations present in coal seams and utilizing coal as the sole substrate. The diversity of the cultured organisms was large. Proteobacteria have been the bacteria most commonly associated with coal seams, even though their role has not been determined [Li et al., 2008; Penner et al., 2008, Midgley et al., 2010]. Bacteria from the phyla Firmicutes, capable of demethylating aromatic compounds, and Bacteroidetes, responsible for anaerobic degradation of cellulose, proteins, polysaccharides, and polyaromatic hydrocarbons, have been found in coal beds in large quantities as well [Shimizu et al., 2007; Strąpoć et al., 2008a]. Archaea were found to be present in coal seams but their relative abundance and type depended strongly on the location. Methanogens belonging to the archaeal phylum Euryarchaeota were common in coal beds but representatives from the phylum Crenarchaeota (extremophiles) were often present as well [Green et al., 2008; Strąpoć et al., 2008a]. By characterizing microbial populations present, many of the aforementioned studies suggested that coal biodegradation and methanogenesis are active within some coal beds. Some studies analyzed biogeochemistry of coalbed plays, connecting present microbial populations and chemistry of formation waters with the biogenic origin of natural gas [McIntosh et al., 2008; Rice et al., 2008; Strąpoć et al., 2008b; Warwick et al., 2008; Flores et al., 2008; McIntosh et al., 2010; Aikuan et al., 2010; Schlegel et al., 2011]. Regardless of all the indirect evidence of biogenic gas production in coal seams, the mechanism of anaerobic coal biodegradation is not well understood. Under laboratory conditions, many intermediate products were discovered (e.g., acetate, alkanes, long-chain fatty acids, and low molecular weight aromatics) [Orem et al., 2010]. Moreover, Formolo and colleagues (2008) analyzed the biodegradation indices of coals associated with biogenic coalbed methane in the Powder River Basin and San Juan Basin and observed a removal of n-alkanes and isoprenoids from the coal matrix. In recent years, the research focus shifted towards stimulating the biogenic methane production from coal by addition of nutrients and/or microbial consortia. Common nutrient amendments included phosphate, yeast extract, ammonia, trace metals, and vitamins [Jones et al., 2010; Strąpoć et al., 2011]. Added microbial consortia contained methanogens collected from coal seams and other natural environments (e.g., wetlands) [Jones et al., 2008]. Successful results reported gas generation potential between 101 - 8x103 SCF per ton per year [Strąpoć et al., 2011]. Even though many authors discuss addition of nutrients to stimulate methanogens present in coal, no published paper discusses the potential of stepwise chemical degradation of coal followed by biodegradation resulting in methane production. Moreover, the literature focuses mostly on coalbed methane, ignoring other potential applications; utilization of waste hydrocarbon material is not considered. No published studies discuss an ex-situ methane generation from coal or waste hydrocarbon materials and no large scale bioreactor designs have been presented for these applications. The technology for large-scale generation of biofuels exists; both gaseous fuels (e.g., biogas) and liquid fuels (e.g., biodiesel) have been successfully obtained from simple hydrocarbons. Furthermore, the availability of electron donors and acceptors as well as the oxidation-reduction potential, though being important factors in methanogenesis, are not discussed in the published literature. The studies focus instead on the bioavailability of carbon sources, availability of hydrogen ions, presence of methanogenesis inhibitory compounds, etc. 2. RESEARCH OBJECTIVES The research objectives are to: 1. Examine various minimally characterized microbial consortia from different environments for their potential use in methane generation from coal and other solid carbonaceous materials. 2. Evaluate methane and carbon dioxide generation potential from coal and other solid carbonaceous materials pretreated with various chemical reagents and microbial consortia. 3. Evaluate bench-scale generation of methane and carbon dioxide from coal and/or other solid carbonaceous materials and examination of the influence of environmental conditions, such as pH, salinity, and nutrient amendment. 4. Examine directly supplying electrons and electron acceptors to stimulate methanogenesis through the use of electro-biochemical reactors. 3. ORIGINAL CONTRIBUTION OF THE RESEARCH As original contributions, the research will: Examine the potential of degradation of complex hydrocarbon materials by chemical reagents prior to microbial biodegradation and methanogenesis; Test various hydrocarbon materials, including coal, lignite, and waste coal, since the published literature focuses mostly on enhancement of coalbed methane recovery, while other carbonaceous materials, such as waste solid hydrocarbons, are under-utilized; Examine ex-situ use of complex hydrocarbons in bench scale batch bioreactors, since the published literature in the field of complex hydrocarbons, such as coal, focuses mostly on in-situ methane production; Perform initial characterizations for unique microbial consortia and their potential for methane generation; Examine the influence of environmental conditions, such as pH, salinity, nutrient amendments, on methane and carbon dioxide generation in bench scale bioreactors; Test an electrobiochemical reactor to examine the influence of electrons and electron acceptor environments on methanognesis. 4. MATERIALS AND EXPERIMENTAL PROCEDURES 4.1. Col lection and Preparation of Materials 4.1.1. Coal Samples Coal and coal waste samples used in this study were provided from the Deer Creek Mine in Utah. The samples came from the same mining operation and types of mined materials to permit a more direct comparison of the results obtained. The coal sample had a total moisture content of 4.28% and 6.15% ash content as received. With over 76% carbon (dry) and a caloric value of about 14,000 BTU/lb it is classified as a bituminous coal. The coal waste product, as received, contained over 50% ash, 28% carbon (dry), and had a caloric value of about 4,400 BTU/lb. More specific elemental composition, as well as ash analysis for these coal and coal waste samples, is given in Tables 5 and 6. A commercially available North Dakota lignite sample was purchased as a bulk pack from Ward's Natural Science (#47-2133). Chemical composition of characteristic North Dakota coals is presented in Table 7, as given by Tang and colleagues (1996), and Gale and colleagues (1996). 4.1.2. Coal Grinding All coal samples were pulverized to -200 mesh particle size in a ball grinder to provide maximum surface area. Sample preparation station was swept and washed Table 5. Analysis of bituminous coal and coal waste samples. Parameter Coal Waste Coal Total Moisture (as received) 4.28% 6.89% Ash (dry) 6.43% 57.61% Volatile Matter (dry) 48.13% 29.30% Fixed Carbon (dry) 45.44% 13.09% Carbon (dry) 76.60% 28.24% Gross Calorific Value [BTU/lb] (dry) 13,949 4,370 Sulfur (dry) 0.38% 0.30% Organic Sulfur (dry) 0.37% 0.16% Oxygen (dry) 9.15% 11.40% Hydrogen (dry) 6.02% 1.99% Nitrogen (dry) 1.42% 0.46% Table 6. Ash analysis of bituminous coal and coal waste samples. Component Coal Waste Coal SiO2 52.72% 62.05% Al2O3 13.16% 8.72% Fe2O3 5.27% 2.30% CaO 12.10% 16.88% MgO 1.50% 6.34% K2O 0.18% 1.55% Na2O 4.19% 0.35% SO3 8.89% 1.11% P2O5 0.75% 0.18% TiO2 0.90% 0.42% Table 7. Analysis of North Dakota lignite samples [Tang et al. 1996 and Gale et al. 1996]. Parameter Tang et al . , 1996 Gale et al . , 1996 Total Moisture (as received) 3.90% 23.3% Ash (dry) 5.58% 6.0% Volatile Matter (dry) 44.83% NA Fixed Carbon (dry) 45.69% NA Carbon (dry) 61.20% 67.2% Sulfur (dry) 0.25% 1.06% Oxygen (dry) 27.98% 26.5% Hydrogen (dry) 3.97% 4.3% Nitrogen (dry) 1.01% 1.0% with water and detergent. Grinding vessels, grinding balls, and all the dishes and tools were thoroughly washed. Bituminous coal consisted mostly of large and hard chunks; therefore, it was first manually crushed in a ceramic mortar to a size less than 8 mesh. Waste bituminous coal contained a large quantity of moisture; thus, drying was necessary prior to grinding. Moreover, the waste bituminous coal was upgraded prior to grinding, i.e., large inorganic rocks were removed from the material. Lignite material was soft and did not require manual grinding. In order to ensure minimal contamination, a 1 kg aliquot of crushed coal was ground for 30 minutes and discarded before the regular grinding started. Coal samples were ground for 3-4 hours and the resulting dust was screened through an 8-mesh sieve into a collection bucket. After 30 kg of coal dust were collected, the workstation and all the tools were thoroughly washed. 4.1.3. Col lection of Microbial Populations Microbial populations were cultured from coal samples and natural environments believed to be suitable for methanogens. Summary of collected samples is given in Table 8. Two 55-gallon drums containing bituminous coal and coal waste rock were obtained (for coal characterization see Section 4.1.1, page 19). Two sediment cores from the Great Salt Lake's wetlands and one from the Jordan River were sampled (Figures 6A and 6B). These three samples had a higher salt content [Jones et al., 2009]. Gas collected over the Jordan River sample site produced a self-sustaining flame, when ignited (Figure 6C). A 500 mL anaerobic digester sludge sample was collected from the Central Valley Wastewater Treatment Plant. Additionally, samples from eight locations Figure 6. Photographs of core samples from (A) the Great Salt Lake wetlands and (B) the Jordan River, and (C) self-sustained flame produced from the gas collected from the Jordan River. Table 8. Environmental samples collected for isolation of methanogenic microorganisms. Sample # of samples Source Notes bituminous coal, 2 Deer Creek coal mine, Utah Microbes were collected from coarse and from pulverized samples bituminous waste coal 2 lake sediment 1 The Great Salt Lake - south Higher salinity wetland sediment 1 The Great Salt Lake - Farmington Higher salinity river sediment 1 The Jordan River - the Legacy Nature Preserve Higher salinity digester sludge 1 Central Valley Wastewater Treatment Plant Anaerobic sludge oil seep 8 The Great Salt Lake - Rozel Point Higher salinity; hydrocarbon associated gas well 6 Drunkard Wash, Price, UT Coal-bed methane wells around the Rozel Point oil seeps in the Great Salt Lake and six locations from coal-bed methane wells were collected by Michael Peoples (Department of Metallurgical Engineering, University of Utah). 4.1.4. Microbial Cultivation and Morphology Characterization Prior to sampling, both, the media preparation station and the microbial sample station were cleaned and disinfected with 10% Clorox solution. The samples were collected from the sediments' core interior to avoid contamination with foreign microorganisms potentially present on the core outside surfaces. Grab samples were collected along the core length to ensure that a variety of microbial communities are captured. The remaining cores were wrapped with aluminum foil, sealed in plastic sample bags, and stored at about 5oC to preserve the natural moisture content of the sediments. Sterile 50 mL centrifuge vials were filled with collected environmental samples. Seven different sterile media were added to the prepared samples (see Appendix A). Sample vials were stored at about 24oC to allow for microbial growth. 4.2. Microbial Characterization 4.2.1. Morphology Characterization Six tenfold microbial sample dilutions were prepared with normal saline solution (0.85% NaCl) and plated on TSA plates (30g/L trypticase soy broth and 18mg/L agar) under sterile conditions. After three days, the morphology of the grown colonies was characterized accordingly to their form, size, surface, color, elevation, and margin. Moreover, the most representative plate from each series (i.e., one having 30-300 colonies) was counted. For a more detailed procedure description, see Standard Operating Procedure SOP: Cultivation of Microorganisms for Gas Generation Tests, Appendix B. 4.2.2. Environmental Inf luences The best CO2 and CH4 generating microbial populations from coal sample screening tests were selected (see Section 5.3.2.) and combined into five consortia (Table 9). The consortia were selected to contain both methane and carbon dioxide generating microbes that were not sensitive to oxygen exposure during culture. Following selection, the consortia were allowed to establish individual microbial concentrations within the consortia based on the culture medium. The influence of pH, temperature, and salinity on the created microbial consortia was examined. Growth curves were plotted using an indirect spectrophotometer optical density (OD) measurement. Thermo Spectronic Genesis 8 spectrophotometer was used. Since the Trypticase Soy Broth medium is yellow, the wavelength was set at 600 nm to minimize the effect of color on sample measurement. Direct counts were performed on selected samples in order to correlate the sample absorbance with colony counts. Even though both methods are not extremely precise, they provide a relevant indication of growth (OD measures the turbidity of the sample, which may differ between various microbes; while colony count takes into account only the aerobic, aerotolerant and culturable organisms). Table 9. Five created consortia and their microbial sources. Consortium Microbial Sources in1 2x coal, 2x coal waste, river sediments in2 2x oil seep, natural gas well in3 2x oil seep, lake sediment, 4x coal, 3x coal waste, natural gas well, river sediment in4 3x coal waste, 3x natural gas well, 2x oil seep in5 2x oil seep, coal, natural gas well Moreover, growth kinetics parameters can be assessed based on the created growth curves. Growth rate constant, μ [hr-1], denotes the number of generations that occur per unit time and can be calculated as: (4) where Nt and N0 are the amount of cells per milliliter at the time t and t0 (the initial time), respectively. The time required for the population to double, or the doubling time, g [hr], can be calculated from Equation 5. (5) 4.2.3. Oxygen Requirement Thioglycollate broth medium was used in determination of microbial oxygen requirement. This medium contains: dextrose, yeast extract, digest of casein (a mixture of nutrients), L-cystein and sodium thioglycollate (compounds removing oxygen from the medium), agar (slowing down the return of oxygen to the sample and thus creating an oxygen gradient), and resazurin (an oxygen indicator). The growth pattern in the tubes indicates the type of oxygen requirement for a given microbe or consortium (Figure 7). For a detailed description of the procedure, please see SOP: Oxygen Requirement Test with Thioglycollate, Appendix B. 4.2.4. Community Level Metabol ic Prof i l ing Each microbial species and microbial consortium or population has a specific and usually unique set of carbon compounds they can utilize as an energy substrate. Describing the pattern of carbon utilization by a given microbial species or Figure 7. Growth patterns in thioglycollate broth indicating: A - obligate anaerobes, B - aerotolerant anaerobes, C - facultative anaerobes, D - microaerophiles, E - obligate aerobes. consortium is called Community Level Physiological Profiling (CLPP). The CLPP system was developed by BIOLOG in the late 1980s for a rapid identification of clinically important bacteria. Three most common CLPP systems include GN (Gram negative bacteria), GP (Gram positive bacteria), and ECO (environmental) microtiter plates. GN and GP plates contain 95 carbon sources, while ECO plates contain 31 carbon sources. An ECO plate consists of 96 wells, each well contains one of 31 carbon sources, which are present in triplicates for the reproducibility purpose. The remaining three wells are filled with sterile DI water as a control. Additionally, each plate contains tetrazolium dye, which is transparent initially, but under respiration-dependent reduction, it turns purple. This colorimetric reaction can be monitored with time, indicating which of the carbon sources can be utilized by the consortium. Figure 8 shows the distribution of carbon sources on the ECO plate. 1 2 3 4 A Water β-methyl-D-glucoside D-galactonic acid γ- lactone L-arginine B Pyruvic acid methyl ester D-xylose D-galacturonic acid L-asparagine C Tween 40 i-erythritol 2-hydroxy benzoic acid 1-phenylalanine D Tween 80 D-mannitol 4-hydroxy benzoic acid L-serine E α-cyclodextrin N-acetyl-D-glucosamine γ-hydroxy butyric acid L-threonine F Glycogen D-glucosamic acid Itaconic acid Glycyl-L-glutamic acid G D-cellobiose Glucose-1-phosphate α-ketobutyric acid Phenylethyl-amine H α-D-lactose D,1- α-glycerol phosphate D-malic acid Putrescine Figure 8. Distribution of carbon compounds on the BIOLOG ECO plate. Each well is repeated in triplicates on the actual plate. Colored matrix corresponds to the type of compound; pink - carboxylic acids, blue - complex carbon sources, yellow - carbohydrates, green - phosphate-carbon, orange - amino acids, gray - amines. Adapted from Chazarenc et al., 2010. The BIOLOG plates were used in this research as a tool for metabolic profiling of microbial consortia. Dilution series was performed on the five microbial consortia until the desired concentration of approximately 103 CFU/mL were obtained. Using a multichannel pipette, 100 μL of the consortia were transferred into each well of the BIOLOG plate. The plates were covered by a moist towel and incubated at 30oC. The elevated temperature was chosen based on the results of environmental influences on the microbial growth experiments (see Section 5.1.2.). Color development in the wells was noted after 24, 48, 72, and 96 hours. Community Metabolic Diversity (CMD) factor was calculated by summing the number of positive responses observed (violet wells), while a related Functional Diversity (FD) factor was calculated using Equation 6. Similarity between the consortia was calculated accordingly to Equation 7, while the Variation within the results was calculated based on Equation 8. (6) (7) where a is a number of carbon sources used (indicated by the color development in the well) by both consortium i and consortium k, b is a number of carbon sources used by consortium k but not by consortium i, c is a number of carbon sources used by consortium i but not by consortium k, and d is a number of carbon sources not used by either consortium. (8) where F is the number of false results (i.e., number of carbon sources, in which the three replicates were not all positive or all negative). For a detailed description of the procedure, please see SOP: Community Level Physiological Profiling, Appendix B. 4.3. Pretreatments of Carbonaceous Materials 4.3.1. Dissolution Tests A small quantity of coal material (less than 1 g) was placed in a clean, sterilized mortar under sterilized hood. Using a sterilized pestle, rock was crushed until completely pulverized. More material was added and crushing continued. Pulverized material was placed in sterile containers and its weight was calculated. Chemical reagent was prepared and filtered into a sterile flask. A list of tested reagents' composition is provided in Appendix C). Desired amounts of ground samples were weighed under the sterilized hood and placed in 50mL centrifuge tubes. A 20mL aliquot of the prepared chemical reagent was added into the sample. Tubes were stored at room temperature for 14, 30, 90, and 120 days, and shaken thoroughly once a week. After the desired reaction time, samples were filtered. The filtration method was based upon the standard method 2540D Total Suspended Solids Dried at 103-105oC. It was used to determine the amount of solids remaining after coal material chemical dissolution. Using a vacuum filtration apparatus, glass microfiber filters (Whatman GF/C, 1.2 μm) were washed with DI water and dried at 105oC for an hour (or until a stable weight was achieved). Dried filters were weighed, placed in the filtration apparatus and a small aliquot of DI water was applied onto the filter to create a seal. Sample was mixed thoroughly and poured onto the filter and the suction was applied. The sample container and its cap were washed with DI water and detergent, if necessary, until all the solids were transferred onto the filter. Suction was applied until all the water was evacuated. Filter papers with solids were dried at 105oC for 24 hours (or until a stable weight was achieved), cooled in a desiccator, and weighed. The amount of solids was calculated as a difference between the final and initial weight of the filter. For more detailed procedure descriptions, see SOP: Coal Crushing - Small Scale, SOP: Chemical Dissolution of Coal - Sample Preparation, and SOP: Chemical Dissolution of Coal - Filtration (Appendix B). 4.3.2. Raman Spectroscopy In order to assess the biochemical degradation of coal materials, pulverized bituminous coal and coal waste were immersed in five liquid media and DI water (for media composition, see Appendix A). The Raman spectroscopy analysis was performed on the liquid phase of these samples after 48 hours and six months. The resulting spectra were compared to the spectra obtained from the five liquid media and DI water. A 3 mL aliquot of each sample was filtered through a 0.2 μm syringe filter into a glass vial and dried at 45oC. Samples were analyzed with a Raman Systems R-3000 QE portable spectrometer. 4.4. Gas Generation Tests in Serum Bottles 4.4.1. Microbes in Liquid Media Carbon dioxide, methane, and heavier gaseous hydrocarbons (C2-C6) generated from environmental samples, immersed in various media, were measured using gas chromatography (for environmental sample preparation see Section 4.1.4.). Gas chromatography analysis begins with turning on the computer and flaming the GC unit. The column and gastight syringe were cleaned with fresh air. The tip of the gastight syringe was placed in the sample vial through an opening made in the septum. The gases were allowed to fill the syringe up to the 500μL mark, the plunger was pushed down to the 200μL mark, while the syringe valve was closed, and excess gases were evacuated into DI water to prevent contamination with the atmospheric gases. After preparing the GC unit, the sample was injected into the injection port. Using calibration data, integrated peak areas were recalculated into gas concentrations. For a more detailed procedure description, see SOP: Gas Chromatography (Appendix B). 4.4.2. Microbes in Coal Samples with and without Nutrient Amendment Gas generation results combined with microbial colony counts were used to design the experimental matrix. Four microbial population categories as well as their consortia were selected (methane producers, carbon dioxide producers, producers of carbon dioxide and methane, and producers of other gases). The most representative TSA plate was chosen from each series of plated environmental samples (see Section 4.2.1.) and 2-5 mL of a medium used in a given environmental sample was added to it. Harvested microbes and liquid were collected from the plate and placed in a sterile 15 mL graduated centrifuge tube containing 10 mL of the appropriate medium. Tubes were vortexed and stored in room temperature. After 2- 3 days microbes were washed with saline solution and used to inoculate coal samples. Glass serum bottles (20mL, Wheaton #223742) were used as bioreactors. Four-gram aliquots of pulverized hydrocarbon material (coal, waste coal, and lignite) were placed in serum bottles; five milliliters of liquid solution and one and a half milliliter of microbial consortia were added; each vial had a headspace of approximately 14 milliliters. Microbial samples were centrifuged and washed with saline solution three times, in order to remove remaining carbon sources that could have been introduced through the culture media. A Teflon silicone septum (Wheaton #224173) was placed on top of a bottle and was sealed with an aluminum seal (Wheaton #224178) using a crimper. Three levels of nutrient amendments were selected; 0%, 10%, and 50%. Thus, the liquid solution added to hydrocarbon samples contained either only normal saline solution (corresponding to 0% nutrient amendment) or normal saline solution with 10% or 50% of additional nutrients. The composition of nutrient amendments was identical to liquid media, in which the consortia were cultured. Over 650 samples, containing solid hydrocarbons, microbial consortia from various environments, and nutrient solution amendments, were created. The samples were left at room temperature for 30 days and were not disturbed beyond the normal handling conditions. Produced gases were analyzed with gas chromatography. For more detailed procedure descriptions, see SOP: Gas Chromatography and SOP: Gas Generation Tests (Appendix B). 4.4.3. Microbial Consortia in Pretreated Coal Samples Coal samples were pretreated with three selected chemical pretreatments, analyzed for gas generation with GC after 14 days, inoculated with four selected microbial consortia, and analyzed for gas generation again after an additional 14 and 30 days, for a total of 44 days (Figure 9). Glass serum bottles (20mL, Wheaton #223742) were used as bioreactors. Four-gram aliquots of pulverized hydrocarbon material (coal, waste coal, and lignite) were placed in serum bottles; eight milliliters of liquid solution and two milliliters of microbial consortia concentrated by a factor of three were added. Four microbial consortia (consortium 1, 3, 4, and 5) were selected. Consortia were washed and centrifuged three times prior to usage. A Teflon silicone septum (Wheaton #224173) was placed on top of a bottle and was sealed with an aluminum seal (Wheaton #224178) using a crimper. The composition of chemical pretreatments is given in Table 10. Based on coal dissolution tests, lactic acid was selected as a carbon based pretreatment and a potential direct methane precursor, while Nickel/Alumina/Silica was selected as a non-carbon based catalytic pretreatment. Hydrogen peroxide was used both as a pretreatment, since it achieved high coal dissolution rates, and as a control, since all other pretreatments contained it. Sodium dodecyl sulfate (SDS) was selected as a Figure 9. Simplified test matrix for the combined chemical and microbial pretreatment tests. Table 10. Chemical pretreatment composition. Abbreviation Ni LA HP DI Hydrogen peroxide 3% 3% 3% - Sodium dodecyl sulfate (SDS) 0.001M 0.001M 0.001M - Active reagents 30mg/L NiCl2 20mg/L Al2O3 100mg/L SiO2 0.1M lactic acid - 0.85% NaCl surfactant instead of Tween 20. Both of these are relatively non-toxic for microorganisms, however, SDS (an anionic surfactant) performed better in coal dissolution tests than Tween 20 (a non-ionic surfactant). Additional tests, using the same testing matrix and conditions, were performed on four solid hydrocarbon samples (anthracite, subanthracite, bituminous coal, and coaly shale) as well as corn samples. This was done for a comparison of applicability of this technology to various carbonaceous materials. Moreover, a separate set of tests was performed using enzyme extracts as a pretreatment step. Microbial consortia were grown to the top of the exponential log phase and washed twice with normal saline solution. These concentrated cells were immersed in cold NP40 lysis buffer (chemical cell lysis) and placed in the bead-beater chamber filled with cold glass beads for three minutes (mechanical cell lysis). That allowed for rupture of microbial cell walls and extraction of enzymes. Some of the enzyme preparations, as well as some microbial inocula, were immobilized in alginate beads. One of the advantages of immobilization of cell homogenate is that multiple enzymes can be introduced to the reaction, eliminating the need for separate immobilization of multiple enzymes. Immobilized enzymes or microbes are more convenient to use, usually provide higher stability, and offer protection from the environment. Immobilization material allows for diffusion between the microbial cells and the environment. It also provides nutrients for the microbial growth and will eventually be degraded by the immobilized microbes, releasing them into the environment. Biodegradation of the immobilization material can be designed to be a slower or a faster process. Moreover, chemical degradation of the immobilization material can be performed after the microbes are delivered into the coal seam (e.g., exposure to citrate dissolves alginate material). Microbial immobilization would be a preferred technique during field injections into natural gas wells. For more detailed procedure descriptions, see SOP: Extraction and Immobilization of Enzymes and SOP: Immobilization of Microbial Cells (Appendix B). The samples were incubated at 30oC and were not disturbed beyond the normal handling conditions. Slightly elevated temperature was selected as the most suitable one for the created consortia (see Section 5.1.2.). Produced gases were analyzed with gas chromatography. 4.5. Bench Scale Bioreactor Tests Based on the results of combined chemical pretreatment and microbial gas generation tests, lactic acid pretreatment and inoculum 3 were chosen for the bench tests. Plastic, 500mL gas-tight cylindrical containers, obtained from the Energy and Geoscience Institute, were used as batch bioreactors (Figure 10). One set of bioreactors (electro-biochemical reactors; EBR) using titanium electrodes examined the direct electron provision influence on methane production. Each reactor contained 100 gram aliquots of pulverized hydrocarbon material (coal, waste coal, and lignite); 200 milliliters of liquid solution (nutrient media or normal saline solution), and 50 milliliters of four times concentrated microbial consortia. Additionally, about 150 grams of gravel was added to each reactor for equal distribution of solid, liquid, and gas phases (⅓ of each reactor volume) and to provide routes for gases to escape from the solid/liquid phase. Microbial samples were centrifuged and washed with saline solution three times, in order to remove remaining carbon sources that could have been introduced through the culture Figure 10. Bench scale bioreactors and electro-biochemical reactors. media. Control samples included: coal immersed in normal saline solution (to examine gas desorption from coal matrix and gas generation by native coal microbial population); microbial inoculum incubated in saline solution and gravel (to examine gases generated from microbes consuming dead cells); and microbial inoculum incubated in lactic acid pretreatment and gravel (to examine gases generated by decomposition of the chemical pretreatment). Aside from normal handling conditions, the samples were left undisturbed at about 23oC. One set of especially prepared electro-biochemical reactors were connected to 3.0V potential supplied by a power supply (TekPower HY1803D). Produced gases were analyzed with gas chromatography after 14, 28, and 44 days. 5. RESULTS AND DISCUSSION 5.1. Microbial Characterization 5.1.1. Morphology Characterization Collected coal and environmental samples were immersed in five different media and in DI water (see Appendix A). All the samples were plated on TSA plates in the dilution range of 10-1 to 10-6. After three days from plating, colony morphologies and plate counts were performed on all samples. Results are presented in Appendix D. 5.1.2. Environmental Inf luences 5.1.2.1. pH Influence on Microbial Growth Selected results from only one of five consortia are presented below, but all consortia were tested; the results are presented in Appendix E. Figure 11 shows consortium 3 growth curves under various pH conditions. Moreover, normalized microbial distribution graphs were prepared for all the measurements (Figure 12 shows and example of consortium 3 growth under various pH values, the remaining graphs are given in Appendix E). Interestingly, all tested consortia adjusted the pH towards their optimum level either by producing acids or by reducing sulfates (Figure 13). By combining these results, it can be concluded that the favorable pH for every consortium was in the range of 7.0-9.5. Moreover, it was found that none of the consortia grew well in pH below 6 or above 11. Most of the known methanogenic Figure 11. Consortium 3 growth under starting pH conditions ranging between 5.0 and 11.5 at about 23oC. Figure 12. Normalized consortium 3 microbial distribution under starting pH conditions ranging between 5.0 and 11.5. 0.0E+00 5.0E+08 1.0E+09 1.5E+09 2.0E+09 2.5E+09 3.0E+09 3.5E+09 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 10 20 30 40 50 60 70 80 Approximated Colony Count Absorbance Time [hrs] 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 0% 20% 40% 60% 80% 100% 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 Normalized % Absorbance pH 8hr 24hr 54hr 77hr Figure 13. Change of pH in consortium 3 after 77 hours at about 23oC. microorganisms favor conditions pH close to neutral (Table 3). Moreover, this was expected, since the microorganisms used were collected from neutral pH environments and cultured at neutral pH, making them predisposed to such conditions. 5.1.2.2. Temperature Influence on Microbial Growth Temperature dependence tests showed that all of the consortia performed best at 30oC, with little decrease in growth at 20oC (Figure 14). Many methanogenic microorganisms are thermophilic, preferring temperatures higher than room temperature (Table 3). The fact that these consortia were collected from natural environments with temperatures of 10-25oC and then grown and stored at room temperature explains why they adapted to or were naturally selected for 20-30oC temperature optimum. 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 0 10 20 30 40 50 60 70 80 pH Time [hrs] 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 Figure 14. Consortium 3 growth under temperature ranging between 10-60oC. 5.1.2.3. Salinity Influence on Microbial Growth All consortia preferred a slight addition of salts, with a significant advantage at 2g/L NaCl (Figure 15). This was expected as well, since many environments selected for microbial collection were characterized by elevated salinity. 5.1.2.4. Growth Kinetics Parameters Finally, growth kinetics parameters were calculated for every consortium and every environmental condition, over the exponential growth phase. As an example, Figure 16 shows the growth kinetics parameters for consortium 3 under various temperature conditions. All the remaining kinetics data is available in Appendix E. Growth kinetics data somewhat differs from the information obtained from the growth curves. Only the exponential growth stage is taken into account during these calculations, regardless of the extent to which it was maintained. For example, consortium 3 experienced the fasted initial growth at 40oC (Figure 14), which is also 0.0E+00 1.0E+09 2.0E+09 3.0E+09 4.0E+09 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 20 40 60 80 Approximate Colony Count Absorbance Time [hrs] 10C 20C 30C 40C 50C 60C Figure 15. Growth of consortium 3 under salinity (added as NaCl) ranging between 0-80 g/L at about 23oC. Figure 16. Growth rate constant, μ, and doubling time, g, for consortium 3 under various temperature conditions. 0.0E+00 5.0E+08 1.0E+09 1.5E+09 2.0E+09 2.5E+09 3.0E+09 3.5E+09 4.0E+09 0.0 0.2 0.4 0.6 0.8 1.0 1.2 0 10 20 30 40 50 60 70 80 Approximate Colony Count Absorbance Time [hrs] 0g/L 2g/L 10g/L 20g/L 40g/L 80g/L 0 2 4 6 8 10 12 14 16 18 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 10 20 30 40 50 60 Doubling time, g [hrs] Growth rate, u [hr-1] Temperature [C] u g evident in the growth kinetics parameters (Figure 16). Nevertheless, the growth began to slow down at this temperature after 9 hours and the final microbial concentration after 77 hours was much lower here than in the samples incubated at 20oC and 30oC. 5.1.2.5. Implications By selecting the aforementioned environments for microbial collection, a certain assumption was made about the optimal growth conditions of the desired methanogenic consortia. All samples were collected at 10-25oC and around neutral pH. At least a half of the samples came from higher salinity environments. By doing so, it was realized that the obtained populations would be best adapted to these conditions and would not include, for example, thermophilic microorganisms. Furthermore, the microbes were collected and stored at room temperature and around neutral pH. Therefore, the laboratory testing would need to be performed under the same or similar conditions. To achieve the best methane and/or CO2 production in different environments than the ones examined, microbes and consortia need to be developed based upon prevailing environmental conditions and chemistry; this holds for use in coal-bed seams or in bioreactors. For instance, when enhancing methane recovery from a deep coal seam that reaches temperatures above 40oC, methanogenic populations from similar environments will have to be collected. Moreover, consortia created in this research can adapt, to a certain extent, to different conditions than those found optimal herein, through gradual exposure to such conditions. As an example, consortium 3 did not show optimal growth at the pH of 6.0 but did grow favorably at pH 7.0-8.0. Transferring consortium 3 to growth medium and gradually lowering the pH and letting it adapt prior to the next transfer would possibly allow the microbial population to adapt to new lower pH conditions. However, placing consortium 3 in an environment with pH of 6.0 without attempting to adapt it first would result in microbes experiencing a shock, not growing as fast, and therefore, not metabolizing the substrates into methane or CO2 at the best possible rates. 5.1.3. Oxygen Requirement A photograph of the thioglycollate tubes after two-day incubation period at 30oC in the dark is given in Figure 17. The analysis of results in provided in Table 11. The analysis of the results indicates that consortia 1, 3, 4, and 5 are comprised of obligate and aerotolerant anaerobes, while consortium 2 contains facultative anaerobes, microaerophiles, and obligate aerobes. Consortia 1, 3, 4, and 5 were created from microbes cultured from a variety of environments and in a variety of media. Consortium 2, was created only from microbes cultured in two environments: lake sediments and a coal-bed methane well; and was cultured in DI water without any additional carbon source. 5.1.3.1. Implications The fact that unique facultative or aerotolerant anaerobic consortia capable of generating large amounts of either methane or CO2 can be screened and developed from naturally-occurring sources is a significant contribution of this research. Figure 17. Thioglycollate tubes inoculated with five consortia. Table 11. Growth pattern of five consortia (numbered 1 through 5) in thioglycollate tubes. 1 2 3 4 5 Bottom Throughout Throughout with higher concentration at the top Just below the surface Top Utilization of characterized and controlled strict anaerobic consortia in laboratory settings is difficult; in the field it is even more so, and the perceived need to conduct methane production enhancement research and testing using only strict anaerobes or in strict anaerobic environments has at least somewhat impeded development of this technology. Aerotolerant anaerobes can be utilized in the coal seams, where they will be exposed to oxygen during the injection process, as well as in the above ground applications, such as coal waste heaps or bioreactors, where the oxygen exposure can occur more frequently. Moreover, the use of aerotolerant microbes is possible when the use of oxidizing chemical agents (e.g., hydrogen peroxide) to degrade coal may be deemed beneficial. 5.1.4. Community Level Metabol ic Prof i l ing An example of BIOLOG plates after 0, 24, and 72 hours of incubation is given in Figure 18. Violet well indicates that the given carbon source has been utilized by a given consortium. Summarized results of metabolic profiling of the five consortia are presented in Figure 19. Consortium 3 was the most versatile, utilizing 29 carbon sources. Consortia 1 and 5 were also diverse, using 28 and 24 carbon sources, respectively. Consortium 4 utilized only slightly more than a half of available carbon compound. Finally, consortium 2 was able to metabolize only seven carbon sources. These observations are quantifiable using Community Metabolic Diversity and Functional Diversity calculations (Figures 20 and 21). While both of these factors represent the total number of substrates effectively metabolized by the microbial community; CMD is an absolute value, while FD is represented as a percentage. They both measure the metabolic diversity but do not identify the metabolized and non-metabolized carbon sources. None of the consortia were able to metabolize 2-hydroxy benzoic acid, also known as salicylic acid. Salicylic acid functions as a plant hormone and is chemically similar to acetylsalicylic acid - the active component of aspirin. That is surprising as the consortia components were selected for biodegradation of more complex carbon Figure 18. An example of inoculated BIOLOG plates after 0, 24, and 72 hours. Figure 19. Utilization of 31 carbon sources by five consortia after 96 hours incubation time at 30oC. Carbon sources: 1 - β-Methyl-D-Glucoside, 2 - D-Galactonic Acid, 3 - L-Arginine, 4 - Pyruvic Acid Methyl Ester, 5 - D-Xylose, 6 - D-Galacturonic Acid, 7 - L-Asparagine, 8 - Tween 40, 9 - i-Erythritol, 10 - 2-Hydroxy Benzoic Acid, 11 - L-Phenylalanine, 12 - Tween 80, 13 - D-Mannitol, 14 - 4- Hydroxy Benzoic Acid, 15 - L-Serine, 16 - α-Cyclodextrin, 17 - N-Acetyl-D-Glucosamine, 18 - γ-Hydroxybutyric Acid, 19 - L-Threonine, 20 - Glycogen, 21 - D-Glucosaminic Acid, 22 - Itaconic Acid, 23 - Glycyl-L-Glutamic Acid, 24 - D-Cellobiose, 25 - Glucose-1-Phosphate, 26 - α-Ketobutyric Acid, 27 - Phenylethyl-amine, 28 - α-D-Lactose, 29 - D,L-α-Glycerol Phosphate, 30 - D-Malic Acid, 31 - Putrescine. Figure 20. Community Metabolic Diversity of five consortia numbered 1 through 5. Figure 21. Functional Diversity of five consortia numbered 1 through 5. compounds than plant matter. Moreover, α-D-Lactose, milk sugar, was only decomposed by consortium 5, indicating that only this consortium contains lactic acid bacteria. On the other hand, all consortia were able to utilize the following: pyruvic acid methyl ester, Tween 40, Tween 80, L-serine, N-acetyl-D-glucosamine, glycyl-L-glutamic acid, and α-ketobutyric acid. Data variability is an indicator of reproducibility; in these experiments it measures variation within samples tested in triplicate (Figure 22). During the incubation period, consortium 5 had the highest variation, exceeding 25%. This may 0 5 10 15 20 25 30 0 20 40 60 80 100 CMD Time [hours] 1 2 3 4 5 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0 20 40 60 80 100 Functional Diversity Time [hours] 1 2 3 4 5 Figure 22. Variation of results with sample of five consortia numbered 1 through 5. indicate that consortium 5 contains slow growing organisms, which metabolize carbon sources slowly, even though at the end of the 96-hour incubation consortium 5 showed similar metabolic diversity to consortium 3. At the end of the incubation period, all consortia showed variation below 15%, representing good reproducibility of results. Percent (%) similarity indicates how functionally similar given two consortia are (Figure 23). Consortia 1 and 3 were nearly 97% similar at the end of the incubation period. Populations present in consortia 1 and 3 were collected from similar environments (Table 9), cultured in the same media, and might have contained the same microbial species at a similar population density. Consortia 2 and 3, and consortia 1 and 2 had a functional similarity of only about 30%. Consortium 2 contained three populations cultured from samples incubated with DI water only, while all other consortia were created from a larger number of populations, incubated with a variety of carbon based growth media (TSB, acetate, etc.). 0% 5% 10% 15% 20% 25% 30% 0 20 40 60 80 100 % Variation Time [hours] 1 2 3 4 5 Figure 23. % Similarity between consortia numbered 1 through 5. Therefore, it was expected that consortium 2 would show the lowest similarity to all other created consortia; however, it was not certain that it would also exhibit the lowest metabolic diversity. Nevertheless, since consortium 2 showed the lowest metabolic diversity and the lowest enhancement of methane production, it was not used in further tests. 5.2. Pretreatments of Carbonaceous Materials 5.2.1. Dissolution Tests Literature review of possible chemical reagents capable of coal degradation was performed by Dr. Amar Sathyapalan under supervision of Dr. Michael Free (Department of Metallurgical Engineering, University of Utah). Dr. Sathyapalan identified numerous metal catalysts, organic and inorganic acids, and other compounds that could break down aliphatic bonds in the coal structure and performed initial screening tests using aggressive and environmentally benign 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0 20 40 60 80 100 % Similarity Time [hours] 1-2 1-3 1-4 1-5 2-3 2-4 2-5 3-4 3-5 reagents. Up to 40% of bituminous coal, 50% of bituminous coal waste, and 100% of lignite were dissolved using aggressive chemical pretreatments in the preliminary study performed by Dr. Sathyapalan. Aggressive reagents included, for example, 5M sulfuric acid, 30% hydrogen peroxide, and 17M acetic acid, dipyridyl. Aggressive chemical pretreatments could be performed in ex-situ setup and would require neutralization steps prior to establishment of microbial populations. Results presented here were performed using reagents selected from Dr. Sathyapalan's tests and improved methods described in Appendix B. Figure 24 shows results of long-term digestion tests with environmentally benign reagents. These tests resulted in statistically insignificant dissolution of bituminous coal waste (Figure 24) and coal (data not shown) but achieved up to 25% dissolution of lignite. Additionally, several other reagents were tested for their ability to dissolve coal and are shown in Figure 24, including citric acid, ascorbic acid, Fenton's reagent (at normal pH and adjusted to about 4), and ferric citrate. None of these reagents were found to increase digestion of lignite, bituminous coal or bituminous coal waste above that achieved by previously tested reagents. The influence of surfactant type used in coal degradation was assessed (Figure 25). It was found that an anionic surfactant (SDS) increased the coal dissolution in comparison to a non-ionic surfactant (Tween 20), indicating that ionic forces might be important during coal breakdown. Selected reagents were tested for their ability to dissolve coal over 14, 30, and 90 days (Figure 26). Results indicate rapid dissolution kinetics and that reaction times longer than 14 days do not provide significant dissolution increases. Figure 24. Chemical dissolution of bituminous coal waste (upper graph) and lignite (lower graph) over 30, 90, and 210 days. Error bars correspond to values obtained with control treatment (DI water). All reagents contain 3% H2O2 and 0.5% Tween20 unless noted otherwise. AA - 0.1M acetic acid; DI - deionized water (no Tween20 or peroxide added); Et - 1% ethanol (no Tween20 or peroxide added); FeCl3-Dipyr - 25mg/L dipyridyl + 100mg/L FeCl3; FePorph - 40mg/L iron porphine (iron (III) meso-tetraphenylporphine-mu-oxo-dimer); LA - 0.1M lactic acid; NaF - 25mg/L sodium fluoride; NiAcAc - 60mg/L nickel acetyl acetone; NiAlSi - 30mg/L NiCl2 + 20mg/L Al2O3 + 100mg/L SiO2; NiSB - 60mg/L nickel Shiff base N,N'Bis(salicylidene)ethanediamino nickel II); P - 3% H2O2; PA - 0.1M phosphoric acid; SA - 0.1M sulfuric acid (no Tween20 added); T - Tween20; TP - Tween20 + H2O2; U - 100mg/L urea. 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% % Dissolution Pretreatment 30d 90d 120d 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% % Dissolution Pretreatment 30d 90d 210d Figure 25. Influence of the type of the surfactant on the dissolution of bituminous coal (upper, left), bituminous coal waste (upper, right), and lignite (bottom). Every treatment included 3% H2O2 and 0.5% Tween20 or 0.001M SDS. CA-citric acid, LA-lactic acid, AA-acetic acid, AsA-ascorbic acid. All the organic acids were added at a concentration of 0.1M. Figure 26. Chemical dissolution of lignite (upper, right), bituminous coal waste (upper, right), and bituminous coal (bottom) over 14, 30, and 90. Error bars correspond to values obtained with control treatment (DI water). P - 3% hydrogen peroxide; LA-TP - 0.1M lactic acid, 0.5% tween20, 3% hydrogen peroxide; U-TP - 100mg/L urea, 0.5% tween20, 3% hydrogen peroxide; AA-TP - 0.1M acetic acid, 0.5% tween20, 3% hydrogen peroxide; T - 0.5% tween20; TP - 0.5% tween20, 3% hydrogen peroxide. 0% 2% 4% 6% CA LA AA AsA % Dissolution 0% 2% 4% 6% CA LA AA AsA % Dissolution 0% 10% 20% 30% 40% CA LA AA AsA % Dissolution Tween SDS 0% 5% 10% 15% 20% 25% 30% 0.0% 2.0% 4.0% 6.0% 8.0% 0.0% 2.0% 4.0% 6.0% 8.0% P LA-TP U-TP AA-TP T TP 14-days 30-days 90-days Most tested chemical pretreatments contained hydrogen peroxide. Hydrogen peroxide is a strong oxidizer and a highly reactive oxygen species, capable of oxidizing organic matter. Moreover, at low concentrations (below 3%), it is harmless to most microorganisms, since they have catalase peroxidases (enzymes decomposing hydrogen peroxide into water and oxygen). Figures 24 and 26 show that 3% hydrogen peroxide achieved similar coal dissolutions as other chemical pretreatments, suggesting that oxidation might be a primary mechanism of coal dissolution. This indicates the need for developing coal biodegrading and methanogenic microbial consortia that are capable of biological oxidation and/or survival under low oxygen concentrations. As in all experiments, the results obtained are potentially biased due to methods used. Coal breakdown into smaller molecules, not resulting in complete coal dissolution, would not be registered by this technique. Furthermore, the formation of chemical precipitates or extent of microbial growth may have influenced these results significantly. There is a high probability of producing inorganic cations (e.g., metals) and anions (e.g., carbonate, hydroxide, sulfide) during chemical digestion of complex structures as coal. Binding of this species could occur, leading to formation of chemical precipitates retained during filtration, which would negatively influence the results presented. Figure 27 shows a dramatic example of lignite sample pretreated with 1M sulfuric acid, where a chemical reaction occurred, leading to a formation of crystals. Since the crystals were not observed immediately after the lignite dissolution, formation of such a precipitate could be avoided to circumvent possible detrimental effects on reservoir permeability, by controlling water chemistry. Moreover, many of the tested chemical pretreatments are also microbial Figure 27. Chemical precipitation remaining at the bottom of the sample container after a complete dissolution of lignite with concentrated sulfuric acid. The solution was allowed 30 days of reaction, even though after 1 day all lignite was dissolved. nutrients. Reagents like acetic acid, lactic acid or ethanol provide an easy carbon source for indigenous coal degrading and metabolizing microorganisms, potentially boosting their growth and abundance. Most microbes are larger than 1.2μm and would also be retained by the filter, potentially influencing the results. The high uncertainty of results is also reflected by the large error bars in Figures 24 and 26. 5.2.2. Raman Spectroscopy A total of 30 Raman spectra were obtained. The findings and shortcomings of the Raman analysis will be discussed on one example only; the remaining data is provided in Appendix F. Figure 28 shows the spectra of the coarse waste coal Figure 28. Spectra of bituminous coal waste sample immersed in yeast, urea and phosphate medium for 48 hours (CWR-6 Fresh) and for six months (CWR-6 Old), and of the yeast, urea phosphate medium (YUrPh). immersed in yeast, urea and phosphate medium. All spectra were normalized with respect to the DI water spectrum. The blue line, representing the yeast, urea and phosphate medium, has one strong peak at 747 cm-1, caused by the S-S and C-S region [Qian and Krimm, 1992]. Spectra obtained from coal waste immersed in the medium for 48 hours (red line; secondary y-axis) and for six months (green line; primary y-axis) do not contain this peak. Two new peaks are visible on the red plot: 1160 cm-1 and 1344 cm-1. The first one is caused by the inorganic carbonates [Socrates, 2004], while the latter one - by the NH3 bending [Centeno and Collery, 2000]. Spectra of coal waste immersed in the medium for six months shows only the 1344 cm-1 peak. Such results strongly indicate that the medium is being utilized. The presence of the inorganic carbonates can be a result of decomposition of the yeast extract, whereas the ammonia is primarily a product of utilization of urea. However, these results do not indicate the decomposition of coal nor do they disprove it. 1800 2000 2200 2400 2600 2800 3000 3200 3400 0 100 200 300 400 500 600 700 800 900 1000 400 500 600 700 800 900 1000 1100 1200 1300 1400 Counts Wavenumber [cm-1] YUrPh CWR-6 Old CWR-6 Fresh ~1160 cm -1 (ref.: "inorganic carbonates") ~1344 cm -1 (ref.: "NH 3 ~747 cm bending") -1 (ref.: "C-S stretching") LC-MS analysis of hydrocarbons present in the liquid phase is the only technique capable of providing a solution fingerprint and therefore proving that coal decomposition was occurring. The largest shortcoming of chromatography techniques are the columns, which are used for the detection of very specific groups of hydrocarbons. For instance, a detection of alkenes or PAHs could be performed; however, it would still not provide a complete picture of the liquid phase composition. 5.3. Gas Generation Tests in Serum Bottles 5.3.1. Microbes in Liquid Media 5.3.1.1. Jordan River Sediments Samples were incubated at about 23oC for a month and a half. Nick Dahdah from the Energy and Goescience Institute (EGI) at the University of Utah provided GC training. Figure 29 gives gas concentrations generated from Jordan River sediments in various media. Only methane and carbon dioxide were produced at significant levels (above 10,000 ppm). Carbon dioxide is an important indicator of carbon biodegradation as well as a direct methane precursor. The selected microbial consortia were the most active in 50% TSB medium (a solution prepared at 50% recommended strength), producing the largest amount of both methane (896,036 ppm) and carbon dioxide (126,006 ppm). Tryptic Soy Broth medium, having the largest quantity of carbon sources and providing a balanced mixture of other nutrients, was expected to generate the best results. On the other hand, deionized water contains no carbon sources or other nutrients and was expected to result in the smallest productivity, if the sediments did not contain significant amounts of Figure 29. Methane (black bars) and carbon dioxide (gray bars) generation from the Jordan River sediments in various media. organic matter. Only 10.6-ppm CH4 (almost five orders of magnitude less than in case of the 50% TSB medium) and 37,063-ppm CO2 were generated from Jordan River sediment samples incubated with DI water. Deionized water was used as a control medium and it was expected that little gas generation would be observed. All other media produced larger amounts of both CH4 and CO2. Since minute amounts of methane were generated from samples incubated in DI water and methane amounts increased significantly when additional carbon sources were provided, it was concluded that the collected Jordan River sediments contained low concentrations of easily biodegradable organic matter or lacked other essential nutrients for methanogenesis. Had the sediments contained all the necessary nutrients, microbial populations would have converted the carbon present into methane. High methane and carbon dioxide concentrations obtained from the Jordan River sediment samples incubated with media indicate that the methanogenic populations were present in these samples and that the produced gases were at least partially the result of metabolic conversion of carbon compounds present in the provided media. 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 0 200,000 400,000 600,000 800,000 1,000,000 TSB YUrPh Ac Lc AcPhY DI CO2 Concentration [ppm] CH4 Concentration [ppm] Medium 5.3.1.2. Digester Sludge The analysis of the gases generated by the digester sludge under various conditions show a completely different pattern from the Jordan River sediments (Figure 30). Only minute amounts of carbon dioxide (3,799 ppm) and methane (1,547 ppm) were produced and no other gases were detected after incubation in 50% TSB. The highest production of methane (509,678 ppm) was achieved from a mixture of yeast, urea and phosphate. Surprisingly, methanogenic bacteria were also very active in the DI water sample and produced 432,020 ppm of methane. This indicates that the digester sludge sample was rich in nutrients and organic matter that could be easily broken down to simple degradation products. That was to be expected from municipal wastewater sludge. Addition of various media with high carbon content to a sludge containing many simple hydrocarbons might have been growth inhibiting. Finally, regardless of the media type used, heavier hydrocarbons were only detected in insignificant amounts in GC analysis. Figure 30. Methane (black bars) and carbon dioxide (gray bars) generation from the digester sludge in various media. 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 0 100,000 200,000 300,000 400,000 500,000 600,000 YUrPh DI AcPhY Lc Ac TSB-50 CO2 Concentration [ppm] CH4 Concentration [ppm] Medium 5.3.1.3. Great Salt Lake Sediments The largest amounts of CH4 and CO2 obtained from the Great Salt Lake's wetland sediments (707,340 ppm and 84,270 ppm, respectively) were produced from the incubation in 50% TSB solution, while the lowest concentration of methane (13.5 ppm) was obtained from the control DI sample (Figure 31). This is not surprising, as tryptic soy broth provides a balanced mixture of nutrients as well as carbon sources. Deionized water on the other hand does not provide any nutrients or carbon sources. Therefore, gases produced from inoculation in DI water were most likely a result of decomposition of organic matter from wetlands remaining within microbial cells and the decomposition of dead microbial cells. The sediments from the Great Salt Lake produced lower concentration of gases than the samples from the wetlands (Figure 32). However, they followed similar trend and generated the largest amount of methane when tested in a 50% TSB solution. The Great Salt Lake wetlands are an environment that sustains a diverse ecosystem. Plants growing in the wetlands fix carbon, nitrogen, and other nutrients from the atmosphere and soil. After the plants die, these elements are returned into the environment. On the other hand, the Great Salt Lake sediments do not sustain any plant growth and consist of largely inorganic sands. Therefore, it was expected that a higher gas generation would be obtained from the wetland sediments than the lake sediments, due to the higher availability of the essential nutrients. Figures depicting gas generation by microbes cultured from the coal, oil seep, and natural gas well samples are given in Appendix G. Figure 31. Methane (black bars) and carbon dioxide (gray bars) generation from the Great Salt Lake wetland sediments in various media. Figure 32. Methane (black bars) and carbon dioxide (gray bars) generation from the Great Salt Lake sediments in various media. 5.3.2. Microbes in Coal Samples with and without Nutrient Amendment The main aim of this part of the research was to screen for microbial consortia capable of utilizing complex carbon compounds as a main food source and survive exposures to the atmospheric gases. The main expected gaseous byproducts of metabolism were carbon dioxide and methane. The samples were analyzed after a 30-day reaction period at about 23°C and were not agitated. Control samples were created by adding normal saline solution to pulverized samples without the addition of external microbial population, and are represented on the following graphs by solid black and dashed gray lines for methane and CO2, respectively. Any gases detected from control samples were a result of desorption from coal under atmospheric conditions and generations by native microbial populations present in coal. Additional control samples included the same concentration of microbial inocula suspended in normal saline solution (0.85% NaCl) without any coal material. Gases generated from these controls most likely came from degradation of the dead cells by the remaining populations. Depending on a microbial inoculum source, 3.89 ppm to 5.40 ppm methane and 1,210 ppm to 1,597 ppm carbon dioxide were generated. These amounts are insignificant in comparison to the results obtained from the samples being tested and are not included in the figures. Pulverized bituminous coal and bituminous coal waste materials, kept in closed containers, were tested for outgassing after six months. Approximately 10-ppm methane and 4,900-ppm CO2 were generated in the bituminous coal containers, while bituminous coal waste produced about 13-ppm methane and 3,000-ppm CO2 through outgassing. A matrix with over 650 samples consisting of coal materials, microbial consortia from various environments, and saline or nutrient solution amendments, was developed and tested. For nutrient composition, see Appendix A. This matrix represents a large microbial screening test in which many of the microbial samples did not generate higher methane and/or carbon dioxide concentrations than the control samples, and are therefore not included in the results and discussion presented. For a complete set of results, please refer to Appendix H. 5.3.2.1. No Nutrients Added Generation of methane or CO2 from coal samples with no nutrient amendment is significant in this study, since it indicates microbial breakdown of coal and subsequent utilization of the intermediate compounds. Up to 5.4-ppm methane and 1,600-ppm carbon dioxide were generated from "no coal" controls, i.e., most likely from microbes utilizing carbon present in the dead microbial cells. Small amounts of additional gases were desorbed from coal matrix and produced by native microbial populations, as represented by horizontal lines on Figures 33 - 41. The highest concentration of methane produced from bituminous coal samples was nearly 300 ppm, while carbon dioxide exceeded 6,000 ppm (Figure 33). Generation of both methane and carbon dioxide from lignite was considerably higher, reaching nearly 450 ppm and over 100,000 ppm, respectively (Figure 34). Bituminous coal waste samples generated less methane than either coal or lignite (over 250 ppm) but generated over 14,000-ppm CO2 (Figure 35). The highest concentrations of methane were obtained from the samples inoculated with consortia cultured from coal (C) or waste coal (WC) environments. Figure 33. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with no nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. Figure 34. Methane (black bars) and carbon dioxide (gray bars) generated from lignite samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with no nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 0 50 100 150 200 250 300 350 C1 WC1 WC2 C2 LS1 OS1 OS2 OS3 C3 C4 OS4 C5 OS5 WC3 OS6 GW1 DS1 OS7 WC4 Carbon Dioxide [ppm] Methane [ppm] Microbial Source 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 0 50 100 150 200 250 300 350 400 450 500 WC4 WC2 WC1 C1 LS1 C10 RS1 C2 OS3 OS1 C9 C4 C7 GW1 WC6 WC7 OS4 WC8 RS2 Carbon Dioxide [ppm] Methane [ppm] Microbial Source Figure 35. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal waste samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with no nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. This was expected, since coal populations contain native methanogens that are adapted to this environment. The highest concentrations of carbon dioxide were produced from samples inoculated with consortia cultured from other environments. These included oil seep (OS), natural gas wells (GW), lake sediments (LS), digester sludge (DG), and river sediments (RS). This is an important finding, indicating that introduction of non-native species could increase the rate of hydrocarbon biodegradation, with CO2 as an end product. It can be assumed that under proper environmental conditions (e.g., sufficient amount of hydrogen ions, appropriate temperature, etc.), a part of generated carbon dioxide would ultimately be converted to methane. Moreover, concentrations of both methane and carbon dioxide obtained from samples with no nutrient amendments, containing an inoculated consortia, were 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 0 50 100 150 200 250 300 WC1 C2 C1 WC2 LS1 OS2 OS1 OS3 C4 DS1 GW1 OS7 OS6 OS4 C5 WC5 OS5 C8 OS9 Carbon Dioxide [ppm] Methane [ppm] Microbial Source higher than in control samples containing only normal saline solution and lower concentrations of native microbial populations. This strongly indicates that introduction of non-native species and/or higher concentrations of native species into solid hydrocarbon materials could potentially enhance the rates of gas production. Production of methane directly from coal sources showed a two- to seven-fold increase over control samples. 5.3.2.2. 10% Nutrients Added Initial nutrient addition might be necessary to stimulate the methanogenic population, i.e., stimulate microbial growth to a point that a sufficient number of microbes would be present to produce readily measurable results. Amendment with 10% nutrient solution did not result in significant additional amounts of methane production from bituminous coal, with one exception generating over 550 ppm (Figure 36). However, carbon dioxide concentrations in all coal samples were significantly increased, ranging from 5,000 to 30,000 ppm. Similar results were observed with lignite and bituminous coal waste samples. Gases analyzed in the headspace of lignite samples reached about 300-ppm methane (with one exception of 2,000 ppm), while CO2 generation was stimulated in all samples to 20,000-90,000 ppm (Figure 37). Bituminous coal waste material produced up to 400-ppm CH4 and 10,000-30,000 ppm CO2 (Figure 38). Figure 36. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with 10% nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. Figure 37. Methane (black bars) and carbon dioxide (gray bars) generated from lignite samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with 10% nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 0 100 200 300 400 500 600 WC1 GW2 C1 C2 WC2 LS1 WC5 OS1 OS3 OS2 RS1 WC4 C4 C3 C6 GW3 OS8 GW4 LS2 Carbon Dioxide [ppm] Methane [ppm] Microbial Source 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000 0 500 1,000 1,500 2,000 2,500 WC4 WC1 WC2 C1 LS1 C2 WC9 RS1 OS3 OS1 C4 C10 C3 OS2 C7 WC6 WC5 WC7 WC10 Carbon Dioxide [ppm] Methane [ppm] Microbial Source Figure 38. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal waste samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with 10% nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 5.3.2.3. 50% Nutrients Added Nutrient amendments at 50% levels act as both stimulation for methanogenic consortia growth as well as a carbon source for methanogenesis. Increase of nutrients to 50% resulted in significant increase of generated gases from bituminous coal materials; where up to 200,000 ppm of methane and 50,000-ppm carbon dioxide were produced (Figure 39). One microbial consortium was stimulated enough, with 50% nutrient solution, to produce 110,000 ppm of CH4 from lignite, while most lignite samples did not generate above 500-ppm CH4 (Figure 40). CO2 was produced in the range of 40,000 to 120,000 ppm from lignite. Good microbial stimulation was achieved in bituminous coal waste samples, where 120,000-ppm methane and up to 60,000-ppm carbon dioxide were produced (Figure 41). 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 0 50 100 150 200 250 300 350 400 450 WC1 C1 C2 WC2 LS1 C4 OS2 OS1 OS3 RS1 GW5 C3 C7 OS9 WC5 GW6 GW3 GW4 GW7 Carbon Dioxide [ppm] Microbial Source Figure 39. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with 50% nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 0 10,000 20,000 30,000 40,000 50,000 60,000 0 50,000 100,000 150,000 200,000 250,000 WC4 WC1 C6 WC5 C2 C7 GW2 GW3 C1 WC2 LS1 GW5 OS2 OS1 RS1 OS3 Carbon Dioxide [ppm] Methane [ppm] Microbial Source 0 10,000 20,000 30,000 40,000 50,000 60,000 0 500 1,000 1,500 GW3 C1 WC2 LS1 GW5 OS2 OS1 RS1 OS3 Figure 40. Methane (black bars) and carbon dioxide (gray bars) generated from lignite samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with 50% nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 0 20,000 40,000 60,000 80,000 100,000 120,000 WC4 WC9 C10 C2 WC1 C6 LS1 WC2 C1 OS2 OS1 OS3 C4 RS1 C11 WC7 C3 WC5 WC6 GW3 Carbon Dioxide [ppm] Methane [ppm] Microbial Source 0 20,000 40,000 60,000 80,000 100,000 120,000 0 500 1,000 1,500 C2 WC1 C6 LS1 WC2 C1 OS2 OS1 OS3 C4 RS1 C11 WC7 C3 WC5 WC6 GW3 Figure 41. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal waste samples inoculated with various microbial consortia (represented by the x-axis) after 30 days incubation period with 50% nutrient amendment. Solid black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 5.3.2.4. Significance The results obtained represent a one-time measurement after a 30-day testing period. Consequently, it is not known whether the maximum conversion to methane or CO2 was achieved or whether partial gas pressures in the reaction vessels limited the amount produced. Furthermore, only limited conclusions on conversion kinetics can be drawn from the results obtained. 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 WC5 WC1 WC4 C6 OS10 C2 GW3 GW7 C7 C1 GW6 OS11 C9 GW2 OS9 C4 OS2 OS3 OS1 LS2 Carbon Dioxide [ppm] Methane [ppm] Microbial Source 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 0 500 1,000 1,500 2,000 2,500 GW2 OS9 C4 OS2 OS3 OS1 LS2 It is known that particle size will influence the extent of methanogenesis [Green et al., 2008]. The smaller the coal particle size, the larger the surface area accessible to the microbes; particle size and surface area is a cubic function relationship. Therefore, the results presented here represent a best-case scenario. High surface areas and high permeability can be obtained in the subsurface environments through hydraulic fracturing, but fractured materials would produce a large variation of particle sizes. Analysis of duplicate samples showed standard deviation in methane concentration of 0.02 ppm at low concentrations to 20.82 ppm at high concentrations, or a deviation of about 0.16% to 8.34%. Carbon dioxide measured from duplicate samples indicated a standard deviation between 1.2% to 20.4%. Using an equation from Buswell (1930), the maximum theoretical conversion of nutrient organic content to methane and carbon dioxide can be calculated (Equation 9). CcHhOoNnSs + yH2O xCH4 + (c - x)CO2 + nNH3 + sH2S 9) x = 0.125(4c + h - 2o - 3n + 2s) 9a) y = 0.25(4c - h - 2o + 3n + 2s) 9b) Addition of nutrient solution increased the degradation of coal materials and subsequent methanogenesis. Higher nutrient concentration should have also resulted in a larger microbial population, generating larger quantities of gases. Only small gas concentrations were generated from "no coal" saline controls (3.89 ppm to 5.40 ppm methane and 1,210 ppm to 1,597 ppm carbon dioxide), representing gas produced from decomposition of microbial cells. Larger gas amounts would be expected to come from cell degradation when samples were incubated in the supplied carbon-based nutrient media, where microbial populations were able to establish a higher cell density. Only a small fraction of generated methane (1-2%) and carbon dioxide (1-3%) was a direct result of nutrient conversion from the samples amended with 50% nutrients, based on a maximum theoretical nutrient conversion calculated from Equation 9 (Table 12). The highest carbon dioxide concentrations were generated from lignite samples at all nutrient amendment levels. Lignite is the lowest metamorphosed and the softest hydrocarbon rock used in this study. High concentrations of carbon dioxide indicate the faster biodegradation of lignite, compared to other coal samples. Methane concentrations detected at the highest levels from lignite samples amended with no and 10% nutrients could be a result of a direct CO2 conversion. Regardless of the nutrient amendment level, methane generated from bituminous coal waste material was at the same order of magnitude as that produced from bituminous coal. This indicates that a similar microbial population was established within each tested samples. With no additional nutrients, carbon dioxide concentrations detected from bituminous coal waste samples were over twice as high as those produced from bituminous coal. These results suggest that culm, gob, and potentially other waste hydrocarbon materials could be converted to useful fuel. Larger volumes of coal waste material would be necessary to generate the same methane concentration as the equivalent coal material, since the coal waste samples were upgraded in these tests, excluding larger inorganic rocks from the matrix. Where available space is not a limitation, a design similar to landfill heaps could be Table 12. The best CH4 and CO2 production [ppm] from test results for lignite, bituminous coal, and bituminous coal waste, compared to the maximum theoretical gas generation from nutrient amendments. Nutrient Bituminous Coal Lignite Coal Waste 0% nutrient CH4 0 290 440 270 CO2 0 6,100 130,000 14,000 10% nutrient CH4 428 580 2,000 400 CO2 286 29,000 95,000 33,000 50% nutrient CH4 2,143 200,000 110,000 120,000 CO2 1,429 50,000 120,000 60,000 used for methane generation from coal waste. The materials would be piled on an impermeable liner, with gas collection pipes placed throughout the heap, and a liquid distribution system on the surface (for the application of chemical pretreatment, microbial inoculum, and nutrient solution). Presence of inorganic material within such a heap would not be a problem and could actually increase the permeability of the system. In cases when space is not readily available, methane could efficiently be produced from upgraded coal waste materials using above ground bioreactors, in a similar manner that agricultural biogas is generated. A novel approach presented in this experiment is based on the assumption that methanogenic microorganisms are not strict anaerobes. None of the microbial cultivation or testing was performed under anaerobic conditions. Agar plates were exposed to the atmosphere and the air present in the headspace of serum bottles was not evacuated prior to experiments. All other studies examining the potential of microbially enhanced coal-bed methane recovery (MECBM), as well as the microbiology textbooks, decisively state that methanogens cannot survive in the presence of oxygen [Barker, 2010; Ulrich and Bower, 2008; Jones et al., 2008]. If that were true, designing a full scale MECBM operation and delivering the microorganisms to the subsurface would prove difficult if not impossible. The same would be true for organic carbon sources present on the surface, such as waste coal. By consciously designing the test in order to select methanogenic microorganisms capable to survive in the presence of oxygen, the approach presented in this study offers a simpler and refreshing look at the prospects of MECBM. 5.3.3. Microbial Consortia in Pretreated Coal Samples 5.3.3.1. Chemical Pretreatments Chemical pretreatments generated significant amounts of gases from all coal samples (Figure 42). The best gas generation was obtained with lactic acid chemical pretreatment. This was expected, as lactic acid is a carbon source, which stimulated native methanogens and acted as an electron donor in methanogenesis. Under lactic acid pretreatment, approximately 170-ppm, 240-ppm, and 500-ppm methane was generated from bituminous coal, bituminous coal waste, and lignite, respectively. About 35,000-ppm and 110,000-ppm CO2 was generated from the bituminous coal and bituminous coal waste pretreated with lactic acid pretreatment. High concentrations of carbon dioxide (above 200,000 ppm) obtained from lignite pretreated with catalytic reagent (Ni) and 3% hydrogen peroxide (HP) indicate that this coal matrix contains much higher amounts of carbonates than bituminous coals. Moreover, it is also the least metamorphosed coal sample used in this study and is therefore more readily oxidized and/or biodegraded. Figure 42. Methane (black bars) and carbon dioxide (gray bars) generated from chemically pretreated bituminous coal, bituminous coal waste, and lignite samples after 14 days incubation period. Solid black line and dashed gray line represent the DI control samples of methane and carbon dioxide, respectively. According to the Buswell equation (Equation 9), no microbial methane or carbon dioxide should have been generated from the pretreated samples, due to a high oxygen component from an addition of 3% hydrogen peroxide. Oxygen is the preferred electron acceptor, because it is the most electronegative species. Before it is depleted, no other species can become reduced. 0 10,000 20,000 30,000 40,000 0 50 100 150 200 Ni LA HP CO2 [ppm] CH4 [ppm] Pretreatment Bituminous Coal 0 50,000 100,000 150,000 0 100 200 300 Ni LA HP CO2 [ppm] CH4 [ppm] Pretreatment Coal Waste 0 50,000 100,000 150,000 200,000 250,000 0 200 400 600 Ni LA HP CO2 [ppm] CH4 [ppm] Pretreatment Lignite Figure 43. Methane (black bars) and carbon dioxide (gray bars) generated from chemically pretreated coaly shale, anthracite, sub-anthracite, bituminous coal, and corn samples after 14 days incubation period. Solid black line and dashed gray line represent the DI control samples of methane and carbon dioxide, respectively. 0 20,000 40,000 60,000 80,000 100,000 0 5 10 15 20 25 30 NiAlSi LA H2O2 CO2 [ppm] CH4 [ppm] Pretreatment Coaly Shale 0 50,000 100,000 150,000 200,000 250,000 0 20 40 60 80 NiAlSi LA H2O2 CO2 [ppm] CH4 [ppm] Pretreatment Anthracite 0 50,000 100,000 150,000 200,000 250,000 0 20 40 60 80 100 NiAlSi LA H2O2 CO2 [ppm] CH4 [ppm] Pretreatment Sub-Anthracite 0 50,000 100,000 150,000 200,000 250,000 0 100 200 300 400 NiAlSi LA H2O2 CO2 [ppm] CH4 [ppm] Pretreatment Bituminous Coal 0 10,000 20,000 30,000 40,000 50,000 0 5 10 15 20 􀀁􀅎NiAlSi 􀀁􀅌LA 􀀁􀅈H2O2 CO2 [ppm] CH4 [ppm] Pretreatment Chemical Pretreatment: Corn 5.3.3.2. Other Carbonaceous Materials Results obtained from four additional coal sources and corn pretreated with three chemical reagents, are shown in Figure 43. These samples followed the same general trends as the main coal materials tested, i.e., the highest concentration of both methane and carbon dioxide were obtained from the lactic acid pretreatment. Moreover, the less metamorphosed the coal, the more methane it generated when pretreated chemically. This is an interesting finding, since it is generally accepted that the higher the coal rank, the larger the amount of adsorbed natural gas content. This means that the main source of methane in these chemical pretreatment tests was not gas desorption from coal matrix but rather microbial methane production. Corn samples, on the other hand, did not generate a lot of methane or carbon dioxide when pretreated chemically, indicating that the plant material did not contain a large native methanogenic population. This was to be expected. Coal samples were extracted from underground where conditions appropriate for methanogenesis exist (e.g., low oxygen content) and where the native microbial populations were evolving for millions of generations and adapting to this environment. Corn plants, however, grow fast, are harvested and processed, and are constantly exposed to high oxygen atmospheric conditions, environments where methanogens are not normally present in significant numbers. 5.3.3.3. Microbial Pretreatments Approximately 30% and 40% more methane was generated from bituminous coal and bituminous coal waste samples inoculated with microbial consortia after a 14- day reaction time as compared to control samples. Up to 50% and 25% more carbon dioxide was generated from the same bituminous coal and coal waste samples when inoculated with the microbial consortia. Lignite samples tripled the amount of methane, when inoculated with microbial consortia. However, CO2 detected from microbially pretreated samples was lower than that of the control samples. This could indicate that the microbial reduction of carbon dioxide to methane was occurring, if the proper environmental conditions were present (e.g., availability of hydrogen ions and electrons). Figures are attached in Appendix I. 5.3.3.4. Combined Chemical and Microbial Pretreatments Results obtained with bituminous coal waste are shown as an example in Figure 44, the remaining data is given in Appendix I. No significant methane production was observed after the 14-day incubation period in any of the samples tested. However, increased CO2 concentrations were observed in bituminous coal and bituminous coal waste samples. The best gas production was achieved with consortia 1 and 3 for bituminous coal waste, consortia 3 and 4 for bituminous coal, and consortia 1, 4, and 5 for lignite. Moreover, lactic acid pretreatment generated the highest gas amounts. After additional 30 days, a significant increase of gas generation was observed. Up to 300 ppm more CH4 and 70,000 ppm more CO2 were generated from bituminous coal waste samples (Figure 44). Gases produced from saline control samples were 7.9-ppm CH4 and 5,040-ppm CO2. Methane concentration increase after an additional 30 days was expected; methanogenesis is a slow process and requires times longer than 30-50 days in most biogas production facilities (e.g., municipal wastewater treatment facilities, landfills, agricultural biogas, etc.). Figure 44. Methane and carbon dioxide generated from chemically pretreated bituminous coal waste samples, inoculated with four microbial consortia. Upper graph - 14-day incubation, bottom graph - 44-day incubation. Solid blue and red lines represent the DI control samples of methane and carbon dioxide, respectively. Violet and orange lines represent microbial control samples of methane and carbon dioxide, respectively. Moreover, the best results obtained from combined chemical and microbial pretreatments represent a 3.5-fold improvement in methane generation and a 11.5- fold improvement in CO2 generation over the gases produced from chemical pretreatments alone (marked by a blue and red line for methane and carbon dioxide on Figure 44). Results obtained with bituminous coal samples followed the same trend; however, lignite samples did not show any significant increase of gases after 44-days incubation. One possible explanation might be that lignite, being the least 0 10,000 20,000 30,000 40,000 50,000 0 50 100 150 200 Ni-1 Ni-3 Ni-4 Ni-5 LA-1 LA-3 LA-4 LA-5 HP-1 HP-3 HP-4 HP-5 CO2 [ppm] CH4 [ppm] Pretreatment + Consortium 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 0 50 100 150 200 250 300 350 Ni-1 Ni-3 Ni-4 Ni-5 LA-1 LA-3 LA-4 LA-5 HP-1 HP-3 HP-4 HP-5 CO2 [ppm] CH4 [ppm] Pretreatment + Cosortium CH4 CO2 CH4-Chem_ctrl CH4-Micro_ctrl CO2-Chem_ctrl CO2-Micro_ctrl metamorphosed of the tested coals, underwent a more complete dissolution under chemical pretreatment, yielding toxic compounds, such as metals, certain inorganics, or even high concentrations of organic compounds. Moreover, it is possible that lignite dissolution changed the physicochemical properties of the solution (e.g., pH, ORP, etc.) to less favorable ones for microbial activity. Results obtained here suggest that a large initial release of gases can be obtained by applying a chemical pretreatment containing an oxidant, a surfactant, and an organic acid to the coal. Following adjustment to a suitable environment, sustained gas generation can be achieved by inoculating pretreated coals with selected microbial consortia. Microbial gas generation is characterized by a slower kinetics than the chemical release of these gases, however, significant concentrations can be achieved in a time frame as short as 30 days. 5.3.3.5. Physicochemical Parameters Since a large discrepancy of results was obtained between the bituminous coal and lignite samples, the liquid phase from the tested samples was analyzed (Figure 45). It was discovered that while the pH of bituminous coal and coal waste samples was circum-neutral (6.8-7.8), regardless of the pretreatment, the pH of all lignite samples was below 4.0. Lignite materials generated high carbon dioxide concentrations (above 15%), even from samples incubated only with saline solution. This indicates that the lignite matrix either contains a high concentration of adsorbed CO2 or that carbonate species are present in high concentrations, acidifying the solution. High carbonate content is common in sub-bituminous coals, where humic acid salts do not metamorphose into coal but decompose into Figure 45. pH and ORP (Ag/AgCl reference electrode) of the liquid phase collected from chemically pretreated bituminous coal waste (black bars), bituminous coal (gray bars), and lignite (white bars) samples. carbonates instead. Methanogenic organisms prefer neutral pH range (Table 3), as do the consortia created in this research (Figure 11). Therefore, low pH generated in the lignite samples was not optimal for the microbial population used in these tests. Furthermore, lactic acid pretreatment acidified all the coal samples to the greatest extend. It was expected, since lactic acid has an acid dissociation constant, pKa, of 3.86. Moreover, cell metabolism might have been stimulated upon addition of carbon-based nutrient (lactic acid), resulting in microbial acid production. Moreover, the electrochemistry of methanogenesis requires a low redox potential, one that provides a suitable reduction environment. None of the samples developed negative ORP values (measured with a Ag/AgCl reference electrode) (Figure 45). The lignite samples showed the highest redox potential among all coal sources. Lignite contained 28% oxygen, while the bituminous coal and bituminous coal waste samples contained only 9% and 11% oxygen, respectively. With the highest chemical and microbial degradation of lignite material, large quantities of oxygen would have been released, causing high ORP values. Furthermore, it was expected that the 0 2 4 6 8 10 Ni LA HP DI pH Chemical Pretreatment 0 50 100 150 200 250 Ni LA HP DI ORP [mV] Chemical Pretreatment lactic acid pretreatment would generate the lowest oxidation-reduction potential, since it acts as an electron donor. However, the lowest ORP values from coal sources were obtained with the Nickel/Alumina/Silica (Ni) pretreatment (20 to 90 mV lower than lactic acid pretreatment). Both nickel and aluminum ions are low in the electrochemical series, meaning that they are reducing agents. Moreover, it was expected that all chemical pretreatments would generate environments with higher potential than that found in normal saline control samples, since all the pretreatments contained 3% v/v hydrogen peroxide, which is a strong oxidizing agent. However, this was not the case. For instance, coal waste immersed in normal saline solution resulted in the highest ORP (154 mV), lactic acid pretreatment resulted in ORP of 149 mV, while 3% hydrogen peroxide created an environment with the ORP of 120mV. The lowest ORP of coal waste sample (90 mV) was achieved with Nickel/Alumina/Silica pretreatment. High ORP values indicate the environment suitable for oxidation reactions, such as degradation of organic matter. Such an environment could be beneficial in the initial stages of the methane production, when the main goal is degradation of complex carbon sources. However, an additional step would be necessary to decrease the potential to levels suitable for methanogenesis (generally, negative ORP values with -400 mV being an optimum [Khanal, 2008]). 5.3.3.6. Solid vs. Liquid Phase Methane and CO2 generation from solid (coal re-suspended in a normal saline solution) and liquid phase (chemical pretreatment) was investigated separately (Appendix I). Generally, solids produced more methane and carbon dioxide than Figure 46. Methane and carbon dioxide generated from the solids and liquid phase of the bituminous coal waste sample after 58 days of combined chemical and microbial pretreatment. liquid phases did (results obtained with the bituminous coal waste sample are shown as an example in Figure 46). This can be explained by the fact that centrifuged coal particles contained higher carbon concentration than was dissolved in the liquid phase. Moreover, the native microbial population was concentrated with the solids during the coal washing procedure, involving centrifugation. Native populations contain microbes that are the best equipped for degradation of a given coal and subsequent production of gases. Moreover, combined concentrations of gases produced from solid and liquid phase alone were generally lower than those produced when the two phases were combined. There are many potential reasons for that phenomenon; e.g., inadvertently, a fraction of dissolved carbon in the solution as well as a fraction of the native microbes might have been lost with the discarded solution during the washing procedure. 0 50 100 150 200 250 300 Ni+3 LA+3 HP+3 CH4 Concentration [ppm] Pretreatment + Consortium 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 Ni+3 LA+3 HP+3 CO2 Concentration [ppm] Pretreatment + Consortium Liquids Solids Total 5.3.3.7. Immobilized Microbial and Enzyme Pretreatments No increase in gas production was observed in the samples pretreated with an enzyme extract. Samples pretreated with an immobilized enzyme extract generated more carbon dioxide, which could have been caused by a biodegradation of the alginate (Figure 47 shows results obtained with the bituminous coal waste as an example; the remaining results are given in Appendix I). Immobilized microbial consortia produced insignificant amounts of methane and less carbon dioxide than the non-immobilized consortia over a 28-day incubation period (Figure 48 shows results obtained with the bituminous coal waste as an example; the remaining results are given in Appendix I). This indicates that the diffusion through the immobilization material was not occurring. Some possible solutions to this problem include: use of a less dense alginate preparation, application of alginate degrading reagents (e.g. citrate), use of different immobilization materials (e.g., carrageenans, poly-vinyl alcohols, high-surface porous materials, such as activated carbon, etc.). Figure 47. Methane (black bars) and carbon dioxide (gray bars) generated from bituminous coal waste samples after 14 days incubation period with immobilized and non-immobilized enzyme preparation. Black and gray lines represent the DI control samples of methane and carbon dioxide, respectively. 0 500 1,000 1,500 2,000 2,500 3,000 0.0 1.0 2.0 3.0 4.0 5.0 CO2 Concentration [ppm] CH4 Concentration [ppm] Enzyme Prep Figure 48. Methane generated from bituminous coal waste samples after 28 days incubation period with non-immobilized (black) and immobilized (gray) microbial consortia. 5.3.3.8. Statistical Significance of the Results For all GC analyses a standard gas with a known concentration of methane and CO2 was run with each 40 samples tested. Table 13 shows a summary of results obtained using three methane standards and two CO2 standards over 18 measurements. Standard deviations depended on gas concentration and were generally higher at higher concentrations. Moreover, 14 samples were analyzed twice (Table 14). The second analysis showed generally a lower result, which was expected on such a small scale. Finally, a set of 12 duplicate samples was prepared and analyzed, for repeatability of results (Table 15). The standard deviation between the duplicates was generally lower than 10%. Table 13. Average standard deviation obtained from 18 measurements of methane and carbon dioxide standard gases. Concentration of a standard Average standard deviation Methane [ppm] 14.9 0.21 102 0.81 995 8.86 Carbon Dioxide [ppm] 1,010 13.4 10,000 100.1 Table 14. Results of methane and carbon dioxide concentrations obtained from duplicate measurements of 14 samples. CH4 [ppm] CO2 [ppm] 240.81 240.67 242,333 242,097 506.57 506.86 103,734 103,648 189.27 189.17 207,723 207,647 8.27 8.26 10,053 10,070 10.43 10.24 12,141 12,108 11.49 11.33 7,635 7,605 12.08 12.11 10,934 10,890 Table 15. Average results of methane and carbon dioxide concentrations obtained from 12 duplicate samples. CH4 [ppm] CO2 [ppm] 7.89 10,556 8.77 11,404 8.84 13,781 10.99 13,928 11.84 28,692 13.64 30,314 18.68 35,711 171.08 41,934 194.75 109,806 236.74 111,289 249.66 223,991 505.26 233,479 5.4. Bench Scale Bioreactor Tests Large quantities of gas were released after application of chemical pretreatments. The reactors became over-pressurized and the gasses were evacuated immediately after the initial GC measurement (four hours after preparation) for the safety reasons. A significant increase of methane generation was observed from all coal samples after 14 days, as compared to "no microbes" saline controls and "no coal" microbial controls (Figure 49 shows results obtained with bituminous coal; for a complete set of results, please refer to Appendix J). Significant improvement of carbon dioxide production was observed in comparison to "no microbes" saline controls. However, the CO2 results obtained were four times lower than the gas produced from "no coal" lactic acid microbial controls. This could be an indication that a part of produced carbon dioxide was converted to methane. Another possible explanation would be that the coal environment provided suitable conditions for microbial transformation of the nutrient to other compounds, instead of CO2. Finally, there is also a possibility Figure 49. Methane (blue bars) and carbon dioxide (red bars) generated from microbiochemically treated bituminous coal samples after 14, 28, and 44 days. Blue and red lines represent "no microbes" saline controls of methane and CO2, respectively. Light blue and orange lines represent "no coal" microbial controls of methane and CO2, respectively. that a large portion of the nutrient was absorbed on the pulverized coal and was less available to the microbes. Bituminous coal and bituminous coal waste materials produced the same order of magnitude of both methane and carbon dioxide. Bituminous coal waste generated more carbon dioxide than the bituminous coal sample, perhaps because the coal waste material was exposed to the atmosphere and was therefore more easily 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 0 200 400 600 800 1,000 1,200 CO2 [ppm] CH4 [ppm] Bituminous - 14 days 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 0 500 1,000 1,500 2,000 CO2 [ppm] CH4 [ppm] Bituminous - 28 days 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 400,000 450,000 0 50,000 100,000 150,000 200,000 250,000 300,000 350,000 NaCl +Bugs LA +Bugs LA +Bugs +Volt CO2 [ppm] CH4 [ppm] Bituminous - 44 days CH4 CO2 CH4_NaClctrl CH4_LActrl CO2_NaClctrl CO2_LActrl biodegradable. Lignite samples produced significantly different results than the bituminous coal and bituminous coal waste samples (Appendix J). Nearly 100% CO2 was produced from lignite amended with lactic acid treatment and microbial inoculum, which also resulted in methane concentration as low as 150 ppm. Lignite was also the only coal sample that produced more carbon dioxide when bio-chemically amended as compared to "no coal" lactic acid control. The gas generation from the bituminous coal, bituminous coal waste, and lignite followed the same trends as in the previous gas generation tests with smaller quantities of coal materials, indicating reproducibility and scalability of the results. After the initial increase of methane concentration in the samples amended with nutrients, the methane concentrations decreased (Figure 50). Surprisingly, after 44 days, coal samples containing only the microbial amendment produced over a 100 times more methane than the samples containing additional carbon nutrients. Based on the common sense and the results obtained from the smaller scale gas generation tests, it was expected that samples amended with carbon-based chemical pretreatment/nutrient and microbes would generate the highest methane concentrations. It is possible that environmental conditions created by the chemical reagents (i.e., pH and ORP) were not the most suitable for methanogenesis, as experienced with the lignite samples in the smaller scale tests (Section 5.3.3.5). Samples amended with a lactic acid treatment generated considerably higher CO2 concentrations, which might have caused acidification of the liquid phase below the optimal methanogenic conditions. The dissociation constant of lactic acid is 3.86, while that of carbonic acid is 6.35. Figure 50. Methane generated from microbiochemically pretreated bituminous coal and bituminous coal waste samples after 14, 28, and 44 days. Blue line shows samples immersed in normal saline solution, red lines shows samples immersed in lactic acid treatment, and green lines show samples immersed in lactic acid treatment and placed in electro-biochemical reactors. Bottom figures show the results on a smaller y-axis. Two-fold improvement in CO2 generation was observed in bituminous coal samples, and over a four-fold improvement was observed in bituminous coal waste samples. Lactic acid pretreatment contained 3% hydrogen peroxide, which might have resulted in fast oxidation of coal ma