,-_'. ”W IPIPIIf: J I..IM) P’I'IP, ’IPP Pin-é 'PI’III' IZ'II I'IINIII . "'r' IIW‘QI [- I \M' II I PIPIII II ' " ‘I .IP‘I‘II‘PPI I IIPlu I $3“ .101 -. '11' , . {IMP F-J-< 'I- ' -\ - _ . .V‘ 5‘Rflga-v." . . ‘ ;, ,~ - '{\_ . QPP . L‘\ \‘II\ ”P I \I P I'II I I'l‘::’:fi II: ’ 1 1. 131’ \ ' . “Q I’SI'I1 In..1}j\\:reié :- "1'1- 9 .‘ 5 P; *II'PFIEZ‘} SLI ht fiéifu: H . .‘ I’l‘f‘I‘I}; I,~ 5J1_«".I:C .7 ,- 1) ' -s ‘ 1:, , My: :rP‘u ‘I I}: {II-I {xiii II'Lfi‘x :,\;I;Tfih " “3. ‘5 I $1.521)...“ ,L' _ . 1' or, ‘-‘ 1D JI v \IIIII"P.IP III‘ |I "I'W x...» v 1‘ 1 1.1“ .J'.:- ““IvIHI :.I~ ‘ ‘ 1 I {1.1 .. ”1.1.11, 11:31:11,34111111.‘ at»: 193;. ‘ ‘H- }\ .5 3 JJ:;1.1'1.I[;LIV1P:HP I\:. I1 . '\ “.1111 «my; "1_:'\:|.,.‘1‘t{‘ .1 1L, IIfi'ZIITI‘II , l .495: 32' . . ' 1:11-- . 1'IIr_ -[P‘n.I)iI..' . .1. . I ’II:..I‘IP_II:Ir‘ ”PP; 'I‘PI.I.III n U0 j. .1 0’. ' PIIIP H: . ' IP11 1 . r . }- ”‘21:. )PPII VP" I.II;iI“.I A‘.rl."’1/.?. f‘I:1I:I‘IIkP::R’IILI‘.III ."I . 1‘": III..." I I i: Ill-1 {In IIIIIPIIII. ‘nfl's :ILIII. .. Ki?“ P.” ['QzII Ion,“ . “I I IIII."';'I ’,, 11,} ”#5..“ " EPA; 1:!‘1 II'PIIJPIY] R31 {Pu-jayfnfgd; “)3 \u n... III‘PPIII IIII-I\.I8 (1‘ 'v?‘ I$‘<|- NI 1 .P,"‘; III‘I' . NH PIP'I} I W" 11: ..; ’ .‘ - .11’..' ."'-1."""I"'.'.1;11" Us. -- “I -- 1 .. . My I‘IWP I I .‘I1 ..;. I"' IitIaIIIfIliuI‘ItPI-PI? I-IIIiI‘IiIP YPPT'PII‘I:”.~ ‘PJIPPIIII’OP‘U: III'IB 'IIJI‘tI‘II' I “P ' PII 1 . 'L. '0‘": MI ‘. " - II» 1.." I' .r “'14- P.1v '.. r 1 .l’XP (Pfing'fl l‘{.' .r: ,1”: IP'PI I'I“'.III P? 1%?” IIIIPIII IIP,I&*. If, IPH. "'II“ . ' . P W . I“ 11‘ .. '- HIP. 1.111.111.1- 1. .1.» 11.», .h‘. ‘ uu"{ {P.Ké" III.“ 2",: ” I'Ih: I 1" I “I ,L.'I»PIIh1W\I-'P'-"'j . .. a :L1II1IE:';1. »_._' .. ~ 111'“. 1 1111.); 1 . 11 .‘ .me 9» *1 .11111 IIu 1PI 'I‘ -. 1:1] (41", l PIZ‘P" I. '35. Pg‘ ~ U” ;III. 1‘1 "- 'II‘II :Pfi .I I :‘Pwa .1» 5%» %.»1 "1% IIIIII I IIIIIIIIIIII IIIIII b 31293 01066 2629 " “Ema! . Michigan em. University 3 v”-a’-.-o , u—rW“ .‘ This is to certify that the dissertation entitled Biogas Management by Controlled Feeding and Heating of a Dairy Manure Digester presented by Sarawoot Chayovan has been accepted towards fulfillment of the requirements for Civil and Sanitary Ph . D . degree in Engineering /fl%& m Major professor Date July 91 1984 MSU is an Affirmunw Action/Equal Opportunity Institution 0-12771 MSU RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from m your record. FINES will be charged if book is returned after the date stamped below. 3:110 1-6 ()5 OCTO 2008 usul BIOGAS IANAGEIENT'BY CONTROLLED FEEDING AND HEATING OF A DAIRY IANORE DIGESTER By Sarawoot Chayovan A DISSERTATION Snbnitted to Nichigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Civil and Sanitary Engineering 1984 I/ / was "I I ,0 . ‘, ABSTRACT BIOGAS NANAGENENT'BY CONTROLLED FEEDING AND HEATING OF A DAIRY NANURE DIGESTER By Sarawoot Chayovan Gas production dynamics were investigated using laboratory scale digesters fed daily with dairy manure and operated both at constant temperature and with imposed temperature fluctuations of t 3.3'C about a mean of 35.8.0. understanding digester dynamics would allow managing gas production to coincide more closely with its utilization, thereby reducing storage requirements. At constant temperature, a 14-liter control digester with a detention time of 19 days. fed with manure diluted to 25% and blended. behaved similarly to two 3—liter digesters fed whole manure at a detention time of 15 days. A second 14-liter digester fed with the diluted manure was operated with three phase relations between the 24 hour temperature cycle and the pulse feeding time. The higher the temperature at the time of feeding, the higher the peak gas production, up to 1.8 times the control. Gradually increasing the temperature after feeding results in sustained high gas production until the most rapidly degradable material is consumed. In all cases digester Operation was stable as indicated by pH. alkalinity and total daily gas production. A mathematical model based on three substrate fractions having each first order kinetics and the Arrhenius temperature relationship successfully predicted gas production dynamics as long as hydrolysis remained the rate limiting step and the volatile acid pool did not change rapidly. For whole manure digested at 36.4'C. the influent contained 19* fast fraction (K - 1.15 d“). 35% moderate fraction (x - 0.34 4"), and 46s slow fraction (x = 0.0085 0“). For diluted and blended manure digested at 35.8°c, the influent contained 35$ fast fraction (1 s 2.19 d"), 25% moderate fraction (I = 0.17 d"). and 41$ slow fraction (I ' 0.0075 d"). The temperature coefficient was found to be 1.25 corresponding to an Arrhenius activation energy of 42.5 heal/deg Kelvin. Results show that gas storage can be reduced as much as 52% using managed heating and feeding for a situation in which gas is productively utilized for only eight hours of the day. DEDI CATED TONYPAREQTS ii ACKNOILEDGENENTS I am grateful to Dr. John A. Eastman. my major professor. for his advice and support throughout my graduate studies. Bis intimate guidance and aid throughout this research program is gratefully appre- ciated. I am also grateful to Dr. John B. Gerrish for his interest. advice and support for this research. His electronics expertise has made the laboratory work more enjoyable. Thanks are extended to Dr. lekenxie L. Davis and Dr. Harold L. Sadoff of my advisory committee for their advice and suggestions. I would like to thank Ir. Gary Connor for his assistance in building the electronic equipment. Special thanks are also due Anne Prxybyla for her excellent help with the computer analysis, Bill Pres- son for his assistance in general laboratory work. and Maria DeRyke for her help in preparation of the manuscript. Finally, I wish to express my deepest gratitude to Napaporn. my wife. for her understanding, patience and sacrifices throughout the years of our graduate studies. iii II. III. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODUCTION BIOCHEMICAL AND MICROBIOLOGICAL BACKGROUND A. MICROBIAL ENERGETICS 1._ Oxidation-Reduction Reactions and Potentials 2. Free Energy Change 3. Adenosine Triphosphate (ATP) METABOLIC GROUPS INVOLVED IN ANAEROBIC FERMENTAIION DAIRY CATTLE IANURE 1. Chemical Composition 2. Substrate Biodegradability BIOCHEMISTRY 0F DAIRY MANURE DECOMPOSITION l. The Hydrolysis and Fermentation of Carbohydrates 2. Hydrolysis and Fermentation of Proteins 3. Hydrolysis and Fermentation of Lipids 4. Methane Formation LITERATURE REVIE' 0N PULSE FEEDING AND TEMPERATURE ' VARIATION EFFECTS A. B. C. EFFECT OF PULSE FEEDING EFFECT OF TEMPERATURE VARIATION PROCESS STABILITY 1. Process Instability 2. Biochemistry MATERIALS AND METHODS A. DESCRIPTION OF THE APPARATUS 1. The 3-Liter Digester System 2. The 14-Liter Digester System iv Page vii ix “a.“ C» q 10 10 15 18 18 24 25 26 28 28 30 31 31 32 34 34 34 38 Page B. EXPERIMENTAL PROCEDURES 38 1. Substrate 40 2. Experimental Program 41 C. ANALYTICAL TECHNIQUES 45 1. pH 46 2. Total Alkalinity 46 3. Total Solids 46 4. Total Volatile Solids 47 5. Chemical Oxygen Demand (COD) 47 6. Individual Volatile Fatty Acid 48 7. Gas Composition 52 8. Bubble Tube Calibration 53 V. EXPERIMENTAL RESULTS 56 A. EXPERIMENTAL GROUP I 56 1. Stable Period 56 2. Gas Production Dynamics 58 3. Substrate Degradation and COD Mass Balance 60 4. Volatile Fatty Acids - 61 5. Gas Composition 63 6. pH and Total Alkalinity 65 7. Gas Production during Extended Digester Operation 65 B. EXPERIMENTAL GROUP II 68 1. Stabilization and Replication of the The Digesters 68 2. Gas Production Dynamics 71 3. Comparison of the Bubble Tube and Wet Test Meter Results 77 4. substrate Degradation and COD Mass Balance 77 5. Volatile Fatty Acid Dynamics 79 6. Gas Composition Dynamics 85 7. pH and Total Alkalinity 85 8. Gas Production during Extended Digester Operation 88 VI. MAJHEIATTCAL MODEL OF GAS PRODUCTION DYNAMICS 90 A. MODEL DEVELOPMENT 90 1. Model for a Daily Pulse Feed Digester at Constant Temperature 91 2. The Model with Temperature Variations 98 B.- COMPARISON OF VARIABLE TEMPERATURE MODEL TO EXPERIMENTAL DATA 102 VII. VIII. IX. X. Page DISCUSSION OF THE RESULTS 106 A. DIGESTER STABILITY 106 1. Constancy of Daily Gas Production 106 2. Volatile Acids as an Indicator of Stability 107 3. Stability of pH and Alkalinity 108 4. Summary 109 B. GAS PRODUCTION DYNAMICS 109 1. Daily Pulse Feeding Effect 110 2. Thmperature Variation Effect 112 3. Combined Effect of Feeding and Temperature 113 C. THE RATE LIMITING STEP 116 D. COMPARISON OF TOTAL GAS PRODUCTION BETVEEN THE CONSTANT AND FLUCTUATING TEMPERATURE DIGESTERS 118 a. sum: " 120 MANAGEMENT IMPLICATIONS 122 A. HEATING COMPETITIVE VITH PRODUCTIVE GAS USE 122 B. HEATING COINCIDENT VITH PRODUCTIVE GAS USE 125 CONCLUSIONS 126 SUGGESTIONS FOR FUTURE IORK 130 APPENDICES: A. DATA FOR CHAPTER IV 131 B. DATA FOR CHAPTER V 133 C. DATA FOR CHAPTER VI 156 D. THEORETICAL GAIN IN GAS PRODUCTION 162 BIBLIOGRAPHY 165 vi Bl. B2. B3. B4. BS. LIST OF TABLES The Standard Redox Potentials (E ) and Standard Free Energy Change of Some Redox Couples of Interest in Anaerobic System. Comparison of Dairy Cow Manure and Domestic Primary Sludge. Fermentation of Amino Acids by One or More Species of Clostridium. - Chemical Characteristics of the Influent Manure. Experimental Program. Substrate Degradation and COO Mass Balance. Substrate Degradation and COO Mass Balance for Experimental Group II. Summary of Estimated Parameters for Mathematical Model. Estimated Kinetic Parameters for the Three Substrate Fractions. Calculated Feed Concentrations of Substrate Fraction. Evaluation of Gain in Total Gas Production Due to Temperature Fluctuations. Area Counts for Volatile Fatty Acids Standard Solution. Area Counts for Volatile Fatty Acids Standard Solution. Daily Gas Production recorded from let Test Meter Readings. Daily Gas Production Recorded by Vet Test Meters. Mean Gas Production Date for Experiment I. Mean Gas Production Data for Experiment II, Control. Mean Gas Production Data for Experiment IIA. vii Page 12 25 42 44 60 79 98 110 111 119 131 132 133 134 135 137 139 Table B6. B7. B9. B10. B11. B12. B13. B14. B15. B16. B17. B18. C1. C2. C3. C5. D1. Mean Gas Production Data for Experiment IIB. Mean Gas Production Data for Experiment IIC. Total Volatile Solids Data During Stable Period of Experiment I. Total Volatile Solids Data During Stable Period of Experiment II. Total COD Data During Stable Period of Experiment I. Total COD Data During Stable Period of Experiment II. Individual the Stable Individual the Stable Individual the Stable Individual the Stable Individual the Stable Individual the Stable Volatile Fatty Acid Concentrations During Period for Digester 1. Experiment I. Volatile Fatty Acid Concentrations During Period for Digester 2. Experiment I. Volatile Fatty Acid Concentrations During Period of Experiment II Control. Volatile Fatty Acid Concentrations During Period of Experiment IIA. Volatile Fatty Acid Concentrations During Period of Experiment IIB. Volatile Fatty Acid Concentrations During Period of Experiment IIC. Daily Gas Production for Extended Digester Operation without Feeding. Data for Estimation of Rate Constants and Initial Gas Potentials Experiment Data for Estimation of Rate Constants and Initial Gas Potentials Data for Estimation of Rate Constants and Initial Gas Potentials for the Slow and Moderate Fractions for I. for the Fast Fraction for Experiment I. for the Slow and Moderate Fractions for Experiment II. Data for Estimation of Rate Constants and Initial Gas Potentials Fortran Program for Comparison of Mathematical Model to for the Fast Fraction for Experiment II. Experimental Data. Theoretical Gain in Gas Production Due to Fluctuating Temperature for a Slowly Degradable Substrate. viii Page 141 143 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 164 LIST OF FIGURES Effect of Hydrogen Partial Pressure on the Free Energy of Conversion of Ethanol. Propionate. Acetate and HydrOgen during Methane Fermentation. Summary of Three-Stage Scheme Consisting of Four Metabolic Groups. Graphical Determination of the Refractory Fraction by the Long Term Batch Fermentation Method. Graphical Determination of the Biodegradable Fraction from Continuous Feed Anaerobic Digestion Using a Modified Kinetic Model. Pathways Involved in the anen Fermentation of the Major Insoluble Carbohydrates Present in Plants. Methane Production and Pool Size of Acetate versus Time after Feeding Large Digester. Effect of Temperature on um. Schematic of a 3-Liter Digester System. Schematic of a 14-Liter Digester System. Volatile Fatty Acid Standard Curves for the Higher Concentration Range. Volatile Fatty Acid Standard Curves for the Lower Concentration Range. Calibration Curves for Methane and Carbon Dioxide. Example Bubble Tube Calibration Curve. Daily Gas Production from Vet Test Meter Readings for Experiment I. Mean Gas Production During the Stable Period of Experiment I. ix Page 11 17 17 21 29 29 35 39 50 51 54 55 57 59 Figure Page 5-3. Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment I. 62 5-4. Methane Content in the Digester Head Space During the Stable Period of Experiment I. 64 5-5. Total Alkalinity and pH Data for Experiment I. 66 5-6. Gas Production During Extended Digester Operation without Feeding Following Experiment I. 67 5-7. Daily Gas Production using the Vet Test Meter for Experiment IIA and the Control Digester. 69 5-8. Daily Gas Production using the Vet Test Meter for Experiments IIB and IIC. 70 5-9. Mean Gas Production and Temperature During the Stable Period of Experiment II. Control. 72 5-10. Mean Gas Production and Temperature During the Stable Period of Ekperiment IIA. 73 5-11. Mean Gas Production and Temperature During the Stable Period of Experiment IIB. 74 5-12. Mean Gas Production and Temperature During the Stable Period of Experiment IIC. 75 5-13. Comparison of Daily Gas Production During the Stable Period of Experiment II by Bubble Counts and by Vet Test Meter Readings. 78 5-14. I Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment 11. Control. 80 5-15. Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment IIA. 81 5-16. Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment IIB. 82 5-17. Individual Volatile Fatty Acid Concentrations During the Controller Malfunction where the Temperature Remained between 37 and 38°C. 83 5-18. Methane Content in the Digester Head Space During the Stable Period for Experiment II. 86 5-19. Total Alkalinity and pH for Experiment II. 87 5-20. Gas Production During Extended Digester Operation without Feeding Following Experiment IIC. 89 Figure Page 6-1. Graphical Estimation of the First Order Rate Constant and the Initial Gas Potential for the Slow Fraction for Experiment I. 93 6-2. Graphical Estimation of the First Order Rate Constant and the Intial Gas Potential of the Slow Fraction for Experiment II. . 94 6-3. Graphical Estimation of the First Order Rate Constants and Intial Gas Potentials of (a) the Moderate and (b) the Fast Fractions for Experiment I. 96 6-4. Graphical Estimation of the First Order Rate Constants and Initial Gas Potential of (a) the Moderate and (b) the Fast Fraction for Experiment 11. 97 6-5. Comparison Between Model Results and Observed Data for Experiment I. (a) Digester 1 and (b) Digester 2. 99 6-6. Comparison Between Model Resutls and Observed Data for Experiment II. Control. 1' 100 6-7. Comparison Between Model Results and Observed Data for Experiment IIA. 103 6-8. Comparison Between Model Results and Observed Data for Experiment IIB. 104 6-9. Comparison Between Model Results and Observed Data for Experiment IIC. 105 8-1. Gas Storage Requirements for (a) Digesters with Uniform Feeding and Heating. and (b) Managed Digesters Using Conditions of Experiment IIA. 124 xi I. INTRODUCTION This investigation provides an understanding of the fluctuations in rate of gas production as a result of imposing daily pulse feeding and temperature fluctuations on a digester fed with dairy manure. Vith this information. digester feeding and heating programs can be developed to more closely co-ordinate biogas production with its subsequent utiliza- tion. resulting in reduciton of gas storage without wasting gas. thereby improving the economics of the process. In an ideal situation. all digester conditions such as temperature and feeding remain constant, resulting in constant methane generation rates. Also the uses of this methane would ideally remain constant throughout the day and week. Unfortunately. normal farm practices make such constant biogas usage impractical. Thus. in most cases. a high degree of gas utilization can only be obtained if large gas storage is provided or the production of methane and its utilization coincide. Because of high cost, storage for more than a fraction of one day's average gas production may not be economically justifiable. Gas storage of one day represents approximately one-third to one-half of the system cost for a 100-cow dairy (Heisler, 1981). For larger systems, the gas storage can represent an even greater fraction of the cost. In order to reduce gas storage requirements it may be desirable to control the rate of gas production by scheduling feeding and heating cycles such that a maximum rate of methane is produced during hours when energy demand on the farm is high. Part of the methane produced (up to 40* in Michigan winters) must be used to maintain the digester Operating l temperature and heating the influent manure. Depending on the detention time and amount of insulation, approximately 25 to 50 percent of the heat requirement is used to raise the incoming manure to the operating temperature. If this heating can be provided when the gas is not being heavily used for other productive purposes. reduction of gas storage needs would also result. For such schemes to work. they must not jeopardize digester opera- tion. Moreover, the effect of such temperature fluctuation and daily pulse feeding on the magnitude and timing of gas production. must be known. The specific objectives are: 1. To determine the ability of digesters to acclimate to fluctuating temperatures without loss'in total gas production; 2. To determine the amplitude and lag time of the 24-hour gas pro- duction cycle for a daily pulse feed digester: 3. To determine the amplitude and lag time of changes in the 24-hour gas production cycle caused by imposing temperature fluctuations on the daily pulse feeding; and 4. Tb develop a model from the experimental results such that some management strategies can be determined. II. BIOCHEMICAL AND MICROBIOLOGICAL BACKGROUND In this study. methane is produced through the anaerobic fermenta- tion of dairy cattle manure. In order to understand the processes occuring in anaerobic digesters and evaluate the experimental results effectively. it is necessary to review some basic knowledge of Biochem- istry and Microbiology involved in the anaerobic fermentation of dairy manure. The background material presented in this chapter covers A ) microbial energetics. B) metabolic groups involved in anaerobic fer- mentation. C) properties of dairy cattle manure. and D) the anaerobic fermentation of dairy manure. A. MICROBIAL ENERGETICS The diversity of chemical activities found among the microbes is ascribed to the method the microbes have of obtaining energy to drive their metabolisms. In anaerobic fermentation. microbes obtain energy by the oxidation of organic material using electron acceptors other than molecular oxygen. The chemical energy released by the oxidation-reduction reaction is transferred through the electron tran- sport system which is intimately linked with the interconversion of reducing equivalents and ATP. Some basic mechanisms of microbial ener- getics in anaerobic fermentation will be described in this section. For more detailed views of this subject. a number of text books such as Brock (1979). Gaudy and Guady (1980). Lynch and Poole (1979) should be consulted. 1. Ogidation-Redugtion Reactions and Potentials The breakdown of organic matter is generally oxidative and exergon- ic. In biological reactions. oxidation involves the removal of hydrogen or electrons. these being passed on to an acceptor. which is thereby reduced. In this way we can refer to the compound being oxidized as a hydrogen and/or electron donor. and the reaction sequence can be represented as: “DC A BH3 whore AH, and B are respectively the hydrogen donor and the hydrogen acceptor. Such a representation stresses two important features of oxidation reduction (or redox) reactions. Each oxidation is accompanied by a reduction. and secondly. the two are coupled through the transfer of reducing equivelents in the form of hydrogen or electrons. Bach redox couple such as AH,/A has a finite tendency to either donate its reducing equivalents and be oxidized (AH,<9 A) or accept them and be reduced (A-9 AH,). Vhen the two couples-are combined in a com- plete redox reaction. the net flow of the reaction is determined by the relative tendency of each couple to donate or accept reducing equi- valents. This tendency. or potential. can be measured and quantified by comparison with a standard redox couple. The standard redox couple is that present at the hydrogen electrode where hydrogen gas is in contact with hydrogen ions (protons) in solution in the presence of platinum as a catalyst. The reaction is a; 3 2H..- + 20- and the tendency to donate reducing equivalents. in this case as elec- trons. is measured as the voltage or potential of the electrical current generated when the electrode is coupled in series with another redox couple electrode. At 25'C, 1 atmosphere of hydrogen and pH 7. the potential 0‘ thO I'dOZ couple n3/2H.+ is -420 mV. Table 2-1 presents the standard redox potentials (at pH 7.0) of a number of redox couples of interest in anaerobic sytems. A couple of lower redox potential will always donate reducing equivalents to a couple of higher potential. The 003919 cog/CH. has E; of -240 mV so that in combination with the redox couple H3/2H+ the complete redox reaction is given by: 4H. + co. - cu4 + 23,0 with the hydrogen donating electrons and being itself oxidized. while the carbon dioxide accepts the electrons and is therefore reduced. In anaerobic ‘°t‘b°11" thi' CO, reduction reaction is mediated by methano- genie bacteria and is called methanogenesis. TABLE 2-1. Standard Redox Potentials and Standard Free Energy Change of Some Redox Couples of Interest in Anaerobic Systems. Redox Couple E;, .y AGo'. Kcal/mol e' 2n+ln . -420 -9.7 NADP+7NADPH —324 -7.5 NAD+INADH ~320 -7.4 ACETATE/C0, -290 -6.7 c0,/cn4 -240 -s.s so‘ln,s -220 -5.1 N07N0;' -360 -s.3 Hog/N0: -430 —9.9 2. Free Energy Change During the oxidation of a subsstrate. reducing equivalents are transferred in the direction of increasing redox potential. This transfer is accompanied by the release of energy. The magnitude of standard free energy change is given by the relationship: AG" =- -nFAB; (2-1) where AG" is the standard free energy change. n is the number of elec- trons transferred. F is the Faraday constant (96.649KJV"1 mol") and AB; is standard redox potential expressed in V. Standard free energy changes are provided in Table 2-1. Free energy changes are useful for determining if reactions or com- binations of reactions are thermodynamically possible. A chemical reaction can proceed only if the free energy change is negative or if it is coupled to another reaction such that the overalI reaction has a negative free energy change. The existence of such a negative free energy change does not. however. in itself. mean that the reaction will occur since. in many cases there is an activation energy which must be overcome. One role of enzymes is to mediate a reaction by reducing the activation energy and providing favorable kinetics. 3. on s be h ATP As in all living organisms. energy transformation in anaerobes is mediated by the ATP system. Generally. the reactions of catabolic path- ways are both oxidative and exergonic. The various specific dehydrogenases remove hydrogen from their substrates and donate them to one of a number of possible acceptors. Most often the acceptor is one of the pyridine nucleotides. NAD+ or NADP+. The reduced forms of these primary hydrogen acceptors are the carriers by which reducing equi- valents are transferred among the various metabolic reactions. NAD(P)H may be reoxidized by two general mechanisms. A coenzyme can donate its reducing equivalents -to the reduction of organic substrates. Examples of this include fermentations and the biosynthetic sequences of anabol- ism. Alternatively. the reducing equivalents can be donated to the next carrier in the respiratory chain with consequent transduction of the redox energy into ATP. This redox energy is captured by the reaction of adenosine diphosphate (ADP) and inorganic phosphate (Pi) to for. ATP. The energy conserved in the pyrophosphate bond is used for work when ATP is hydrolyzed either to ADP and Pi or to adenosine monOphosphate (AMP) and perphosphate (PPi). B. METABOLIC GROUPS INVOLVED IN ANAEROBIC EERMENTATION Effective bioconversion of organic matter to methane is a result of the combined and coordinated metabolic activity of a diverse. yet stable microbial papulation. This section describes the different metabolic groups and their syntrophic association in the anaerobic fermentation process. A general scheme of methanOgenesis which incorporates present knowledge of the microbiology and biochemistry of anaerobic fermentation will be presented. Until recently. methanogenesis was viewed as a two-stage process consisting of acid-formation and methane-formation (McCarty 1964. Kirsch and Sykes. 1971). In the first stage. the fermentative non-methanogenic bacteria. as a group. hydrolyze organic polymers and ferment the pro- ducts to organic acids. alcohol. CO3 and H3. NH, and sulfide. In the second stage. the end products of the metabolism of acid-forming bacter- ia in the first stage are converted to CH‘ and c0,. No methanogenic bacteria have been found. however. that utilize alcohols other than methanol or organic acids other than acetate and formate (Bryant et a1.. 1967. 1977). This finding indicates that the two-stage scheme is unsatisfactory. Bryant (1976) proposed a three-stage scheme by the addition of a new hypothetical group. the "H,-producing acetogenic bacteria". This metabolic group degrades pro- pionate and longer-chain fatty acids. alcohols and other organic acids with the production of acetate and H3. The "S organism" from Methanobacillug gmelianski . for example. represents this group and is a part of a syntrOphic association of two bacterial species. The "S organism" catabolizes ethanol to acetate and H3. The formation of 11a and acetate from ethanol is not energetically favorable unless Hz is used by methanogenic bacteria to reduce CO3 to CH4. Therefore efficeint removal 0f H, by the methanogenic bacteria is essential for the non-methanogenic bacteria to catabolyze acids and alcohol for growth. Figure 2-1 illustrates the relationship that exists between hydro- gen partial pressure and free energy available to the hydrogen-producing and hydrogen-consuming groups. At standard conditions (25°C. pH 7 and all reactants and products at unit activity). ethanol. butyrate and pro- pionate degradation are thermodynAmically unfavorable (Bryant et a1.. 1967: McInerney et a1.. 1979: Boone and Bryant. 1980). In order for energy to be available to the organism oxidizing propionate to acetate and hydrogen. for example. the partial pressure of H3 cannot exceed about 10'4 atmosphere (Thauer et a1.. 1977). Zeikus (1980) and Volfe (1979) added another metabolic group which .Aumanv huueuox noun .uouemum.uo noueuo>uoo no hmuemm noun one no cumuaeum «amen-m douche»: uo uoouum o . I NI. ”.1. .momaaunenuem enemaex undone nomouvhn cue euauoe< .euenoqnoum .Hla museum . 3:00.. en at m. 7 ml _ uzaa> .\R:\ announcers I.v' k WXMHmm o>am> noouooom once uuom owocoaom modeEomwomom nommoHouon uuom uoaaouucoo mCHHQEom poo annuauomsoa a new no naumuocoo amaze . :. cacao HMO ..mEsm :wmmeumm Hancom chose Houoz Houoom sawmuoEEH onwalgfi Suez How>uomou “muck umOB #03 40 experimental program. 1. Srhrtrate The dairy manure used as the substrate was obtained from a dairy farm near Michigan State University. This farm. owned by Mr. Ben Arend. has about one hundred milking cows and was considered a typical Michigan dairy farm. The animals were on a diet of approximately 70 percent corn silage. 20 percent mixed grain (corn and soy bean 50:50) and 10 percent hay. Small amounts of vitamins. salts and trace minerals were added. No antibiotics were incorporated in the animal feed. Straw was used as bedding in the barn. The manure was scraped from the barn floor with a front end loader. A one-inch mesh wire net was used to screen the manure to remove straw and other large particles that might have caused clogging problems in the laboratory digester Operation. The screened manure was placed in one-quart plastic bags each containing about 400 ml. These bags were wrapped with rubber bands and stored in a freezer within twelve hours. Vhen needed. the manure was removed from the freezer and thawed. Throughout the investigation. no extra organic or inorganic nutrients were added to the digesters. Once established. the pH of all digesters remained constant within one half of a pH unit without any acid/base addition. Suhrrrrtg for the 3-Lirer Digesters Full strength manure was fed to the 3-liter digesters for four months. A single 15 gallon batch of manure was collected. dispensed into bags and stored in the freezer so that the influent would be con- stant over the entire period of this phase of experimentation. Each 41 day. one bag of frozen manure was thawed at room temperature for about 1 to 1.5 hour. An appropriate volume of thawed manure was then fed to each digester. The chemical characteristics Of the influent manure for the 3-liter digesters are shown in Table 4-1. The COD and volatile fatty acid sam- ples were taken only during the stable period (after the digesters were Operated for about two detention times). The samples for other parame- ters were taken over the whole period of operation. The total volatile solids were relatively constant over the entire period of this experi- ment with a standard deviation of 2.5 percent. Sphrtrate for rhe 14-Liter Digesrers Another single batch of 20 gallons dairy manure was collected and frozen for feeding to the 14-liter digesters. Prior to feeding. thawed manure was diluted to 25 percent by adding three volumes of tap water to one volume of whole manure. The mixture was blended in a one gallon high-speed blender (Varing) for one minute. An appropriate volume was then fed to each digester. The chemical characteristics Of the influent for the l4-liter digesters are also listed in Table 4-1. LEW The experimental program was designed primarily to evaluate the dynamic rate Of gas production as a result Of daily pulse feeding at a constant temperature alone and combined with an imposed temperature fluctuation. Two sets of different size. completely mixed. manually fed digesters as described in the last section were Operated over a period of one year. Because of the pulse feeding. the substrate concentration in the digesters can never be constant so true steady state cannot be 42 a: use "he.m an van." mme.eH . n I. ll men I- it was .oo.o .. "\ea ..a~e eeeeeeeeeem e he" nee.” 4 gen ewe.ea .ooeo .. “\ea .Aeeeeaeaae neeea an can." naa.am a aeu.- eee.ee~ ~\ua .eeeaee eeuane Heeeaeao an efl.o ee.m an mm.o ae.m~ e ..ee~e. e_eeeue> Heeea NH ee.e ea.» on he.e oe.a we 2 ePOG eUHW eo>< z ebofl ovum e0>< “HOHOIdem «Heuu0u«m «caddies «Houseman Houmnlm eonflfldx HQOHHHEH oflu H0 nouunMHOHOIHIAQ mOOMIOAU eHIV Mdmflfi 43 achieved. The measured daily gas production. however. was found to be relatively constant after an acclimation period of two or three deten- tion times. From here on. this period of constant gas production will be referred to as a stable period. The demonstration of such a stable period for each experiment will be presented in the results. The exper- imental program is summarized in Table 4-2. It consists Of two major groups of experiments. descriptions of which follow. Experimental Group One was designed to evaluate the effect of daily pulse feeding alone. Two 3-liter daily-pulse-fed digesters were Operat- ed identically for duplication purposes. The digesters were fed with full strength manure at a constant temperature Of 36.40 t 0.5'C with a hydraulic retention time of 15 days. For the digester start up. active digester effluent from Baum's Dairy Farm (Springport. Michigan). was used for seeding. The effluent were collected in a five-gallon carboy and 2.5 liters was placed into each starting digester on the same day. Initially the digesters were fed 50 ml a day. This is gradually increased to the full amount of 150 ml a day. The digesters were Operated for about five detention times before an intensive program Of data collection for the stable period started. Data collection included the continuous measurement of gas production rate. volatile fatty acid samples five times a day. gas composition ana- lysis every two to four hours. and a daily sample of the effluent for determination of total volatile solids and chemical Oxygen demand. Experimental Group Two was conducted to study the combined effect Of daily pulse feeding with an imposed temperature fluctuation. THO 14-liter daily-pulse-fed digesters were Operated at a 19 day hydraulic retention time. Diluted dairy manure was used as the substrate. The Table 4-2. 44 Experimental Program. Experimental Group Operating Conditions II Daily pulse feeding at constant temperature Substrate: full strength dairy manure Hydraulic retention time: 15 days Operating temperature : 36.4 r 0.5'C Digesters: two 3-liter digesters with 2.25-liters operating volume each. Operated identically for duplication. Operation period: 4 months with daily feeding. 2 months without feeding Daily pulse feeding with fluctuating temperature Substrate: diluted dairy manure. 1:3 ratio (manure to tap water by volume) Hydraulic retention time: 19 days Operating temperature: 35.8 r 3.3.C. increased linearly for 12 hours. then decreased linearly for 12 hours Phase relation between feeding time and temperature cycle: A. feeding at the mid point of ascending temperature ramp _ B. feeding at the peak of the temperature cycle. C. Feeding at the bottom Of temperature cycle. Digesters: two 14-liter digesters with 9.5-liter Operating volume each. one Operated as described above. another as a control digester operated at a constant temperature Of 35.8 t 0.5'C Operating period: 6 months for A. B and C above: 2 months extended from C without feeding. 45 seeding procedure was the same as that for Emperimental Group One. except 9 liters of active effluent were blended before being placed into each starting digester. During the acclimation period. both digesters were Operated identically at constant temperature.v The digesters were considered to be stable after about three detention times. One digester was then imposed with a fluctuating temperature of t 3.3°C about the mean of 35.8.C. The temperature increased linearly for twelve hours. then decreased linearly for twelve hours. The other digester was Operated as a control unit with a constant temperature of 35.8 r 0.2'C. Using the first digester. three different phase relationships between the feeding time and the temperature cycle were studied and will be referred to as Experiments IIA. IIB and IIC from here on. For Exper- iment IIA. the digester was fed at the midpoint of the ascending temperature ramp. For Experiment IIB. feeding occured at the bottom. and for Experiment IIC the digester was fed at the peak of the tempera- ture cycle. After each change to a new phase relationship. the digester was Operated for about one more detention time before a stable period vas assumed. The stability of the operation will be presented in the next chapter. The data and sample collection programs were the same as for Experimental Group I. C. ANALYTICAL TECHNIQUES The parameters of interest in this study include pH. alkalinity. total sOlids. total volatile solids. con, individual volatile acids. gas composition and gas volume. The measurement procedures were based on Standrrd Merhogr. 14§h Edition (1975) unless otherwise described here. 46 1. 2! Measurements of pH were made shortly after samples were withdrawn. using a pH meter (Corning. Model 12) with combination electrodes. The reference half-cell had Ag/AgCl internal elements with a ceramic junc- tion. A commercial standard buffer solution with a pH of 7.00 at 25°C was used for calibrating the electrode and meter. Measurements were made to t 0.05 pH unit. 2. Tbtrl Alkalinity Total alkalinity was measured by titration to pH 4.5 using 0.02 N H3804. Results were reported in mg/l as CaCO,. The total alkalinity in the digester is composed of bicarbonate alkalinity and fatty acid alka- linity. The bicarbonate alkalinity which is the measure Of buffer capacity can be estimated using Equation 4-1. BA = TA - (0.76 x 0.833)(TFA) (4-1) where BA - bicarbonate alkalinity. mg/l as CaCO, TA - total alkalinity. mg/l as CaCO, TFA - total fatty acid concentration. mg/l as acetic acid The factor Of 0.76 is the estimated fraction of unionized volatile fatty acids at pH 4.5. and 0.833 is the conversion factor of mg/l as acetic acid to mg/l as CaCO,. 3. TO al i s TOtal solids is a measure of all material (other than water) present in sludge. both in suspension and in solution. Prior to sam- pling. the evaporating dish was heated at 550°C for 20 minutes and weighed after complete cooling in the desiccator. A freshly drawn sam- 47 ple of 20 to 30 ml was poured into the dish and weighed rapidly. The sample was dried at 103'C overnight. cooled in a desiccator and weighed. The total solids data were expressed as percent by weight which can be calculated by Equation 4-2. a total solids = ( V. - 'i )(100) (4-2) (Hz-'1) where V1 - weight of evaporating dish V3 = weight of wet sample and evaporating dish V, - weight Of dry solids and evaporating dish 4. TOta Volat e Solids The total volatile solids were measured from the dried solids of the above analysis by burning them completely in a muffle furnace at 550°C for 30 to 40 minutes depending on the size and concentration of the sample. The dishes were then air-cooled slightly and put in a desiccator for complete cooling before weighing. The percent tOtal volatile solids can be determined by Equation 4-3. S total volatile solids = l V, ' '4 )(100) (4-3) ('3";) where V. a weight of ash and evaporating dish V1. V, and V, are the same as in Equation 4-2. 5. Chemi 1 en emand COD The chemical oxygen demand was determined by the dichromate reflux method. The procedure was based on Standard MethgdsI 14th Edition (1215). Freshly drawn samples were diluted between 1:400 and 1:800 48 depending On the estimated COD concentrations of the samples. A 20 ml aliquot Of diluted sample was placed in a (COD) flask with 10 ml 0.25 N standard dichromate. 30 ml concentrated H3804 with A3380. and 0.4 g HgSO‘ and was refluxed for two hours. After diluting and cooling it was titrated with 0.25 N standard ferrous ammonium sulfate. 6. In ividual Volatile Fatt i The individual volatile fatty acids in the influent and effluent of the digesters were analyzed on a gas-chromatograph using a flame ioniza- iton detector. The volatile fatty acids analyzed include acetic. prOpionic. butyric. iso-butyric. valeric and iso-valeric acids. A Varian 3700 gas chromatograph with a Varian CDS-lll data system and Model 9716 recorder were used in this analysis. A 6 ft 2.0 mm ID coiled glass column ‘(Supelco Cat. NO. 2-1721) packed with 10% SP-l200/PI H,PO‘ on 80/100 mesh acid washed Chromosorb V (Supelco Cat.NO. 1-1965. Supelco Inc.. Bellefonte. PA) was connected to a flame ionization detector. The gas chromatograph was Operated isothermally at 115°C. The detector and injection port temperatures were 250°C and 160°C respec- tively. For acetate concentrations of about 150 mg/l as COD and lower. these Operating conditions did not give a well-resolved acetate peak. After several trials. a column temperature Of 90°C. with the detector at 150°C and the injection port at 190'C yielded a much better resolved acetate peak. Thus. this Operating condition was used when acetate con- centrations in the samples were 150 mg/l as COO or lower. Before use. the column was conditioned overnight at a temperature of about 50°C above the Operating temperature with a nitrogen carrier gas flow rate of 49 30 ml/min. HydrOgen and air flow rates were adjusted to Obtain a maxi- mum sensitivity at 30 ml/min and 300 ml/min. respectively. The injeciton septum was replaced after 15-20 injections were made. The glass liner in the injection port was frequently checked for an excess accumulation of nonvolatile material which might cause a tailing peak or loss of sample. Vhenever appropriate the glass liner was replaced with a clean one. The data Obtained were analyzed either automatically by the exter- nal standard method or manually calculated from a set of calibration curves. Parts of the data were verified by injecting the same samples intOf another gas chromatograph (Perkin-Elmer 900) located in the Soil-Science laboratory at Michigan State University. The results from both machines compared within t 10 percent. Fresh standards were prepared for each set of analyses from a stock mixture containing a known amount of each pure volatile fatty acid of interest. The standard solutions were prepared by diluting this stock solution to an appropriate volume. Vhen data were analyzed using the external standard method. calibration samples were run for about every 3 or 4 samples analyzed. Three injections of one standard solution_were normally made and the areas averaged. Results were reported in mg/l as COO. Vhen the data were manually calculated. standard calibration curves were constructed for each set of samples. The standard calibra- tion curves are shown in Figures 4-3 and 4-4. In all cases. the response of each component was linear over the entire concentration range of the standard solutions. For acetate at low concentrations tailing of the solvent peak caused a non-zero intercept in Figure 4-4 (dotted line) which does not affect the calculated concentrations. The . 3 50 HB 700-— .iHB 600 500 U) 4.1 c 8 U 400 O H a? o u “1 300 .2 o o a. 200 lOCL 0'. I l l l O 1000 2000 3000 4000 FIGURE 4-3. VOlatile Fatty Concentration Range. mg/l as COD Acid Standard Curves for the Higher 51 .omnau acqu-uumeomou noted on» new nobuno mauve-um mqe< huuem o-ua~o> 000 me H\oE omm owm ovm com owH ONH om ov o _ _ _ _ _ _ _ _ com .vlw mmbenm o OCH d 8 P. “6 com N 3 .2 TL com 0: 3 0 an H 1. 3 00¢ 52‘ response of acetate in the lower range. however. remained linear with concentration. Samples for volatile acid analysis were prepared by removal of suspended solids and acidification. A 20 ml sample was placed in a plastic centrifuge tube. capped and centrifuged for 10 min. at 14.400 rpm. The supernatant was filtered through a glass fiber filter followed by .‘0.45 micrometer membrane filter (Millipore Type HA). The filtrate was acidified to a pH of 2 or below. capped and stored at 4'C. A one microliter sample was injected into the gas chromatograph using a 10 pl syringe ( 701N. Hamilton CO. Reno. Nev.). In most cases. peaks were well-resolved and baseline separation of all components was achieved. LW The gas composition was analyzed using the same gas chromatOgraph (Varian 3700) with a thermal conductivity detector. A 12 foot. 1/8 inch O.D. copper column was packed with 80/100 mesh Porapak Q (Vater Associ- ates. Inc.. Milford. Mass.). The column was operated isothermally at 50'C with detector and injection temperatures of 150°C and 190°C respec- tively. Helium at a flow rate of 30 ml/min. was used as a carrier gas. One milliliter samples were drawn from the digester head space with a one milliliter gas-tight syringe ( 1001N. Hamilton CO.) and injected immediately into the gas chromatOgraph. The major gases detected in the gas samples were nitrogen. methane. and carbon dioxide. The amount of nitrogen was small and primarily derived from the air that entered the digester during feeding. Therefore only methane and carbon dioxide were components of interest in the analysis. Standardization was accom- plished by injecting various volumes of pure methane and carbon dioxide (Scotty II Mix 109 and Mix 105. Supelco Inc.. Bellefonte. PA). 53 Figure 4-5 shows the peak area response to standard methane and carbon dioxide. The results of the analyses was normalized tO include only methane and carbon dioxide and reported in percent by volume. 8. Bungle Tube Calibration The bubble tubes. bubble counter and data acquisition device were described earlier. The calibration procedure will be detailed here. After several trials of sizes and shapes. the bubble tubes were specially made to match the expected range of gas production rates. Paraffin Oil (Vhite Saybolt. Viscosity 125/135) was added into the tube and the amount of oil was adjusted so that the differential head was about 1-1/4 to 2-1/2 inches depending on the range Of gas production rates to be measured. The higher the rate of gas production. the lower the differential head required to obtain a constant size and smooth ris- ing Of the bubbles. Vhen the rate Of gas production was relatively high. many bubbles tended to rise at one time (bursting). TO prevent this. the differential head was readjusted for each set of experiments. A wet test meter (Precision Scientific Co.. Chicago. IL) was connected to the outlet of the bubble tubes. During calibration. the wet test meter readings were made at 1/2 to 1 hour intervals over two or three days (feeding cycles). These readings were converted into rate Of gas production in mllhr and plotted against the corresponding bubble count data (Figure 4-6). showing a linear relationship over the entire range of interest. The correlation coefficients (R3) of the linear regression analyses for all calibrations were 0.95 or higher. The bubble count data were then translated into gas production rates in mllhr using these calibration curves. 6 S4 22_ 20_ 18_ :1) l6__ 4.) C 23 8 14... O "* 12__ o‘ 8’. a: 10.. .14 t0 3”. 3_ 6- 4... 2- o I l l l_ l o 02 0.4 0.6 0.8 1.0 FIGURE 4-5. Gas Injected. ml Calibration Curves for Methane and Carbon Dioxide. 55 ma .o>umu memueunmnau each Omanmm ommaeum .cfiE\moHnnsn .coHuoocomm mow mH qH NH OH m .wlv mmacnm _ L _ _ — _ vam.o u .HHOU mh.ma l xmh.m~ u w «HH unoEwuomxm ooa com com oov oom 'uorqonpOJd sea -:q/lm V. EXPERIMENTAL RESULTS The experimental data presented in this chapter are divided into two sections in accordance with the two. experimental groups; 1) 3-liter constant temperature. whole manure digesters: and 2) 14-liter . variable temperature. diluted manure digesters. A complete summary of the results. including statistical information. can be found in the Appendix. For a comparison Of the experimental results with the chemi- cal characteristics Of the influent manure. see Table 4-1. A. ELEERIMENTAL GROUP I The results presented in this section were Obtained from the two 3-liter. identically Operated digesters fed with whole manure. The data collected during the stable period include the gas production dynamics. the overall extent of substrate degradation in terms of total volatile solids and chemical oxygen demand. and the individual volatile acids and gas composition at different times during the 24 hour feeding cycle. In addition. the data on daily gas production during the extended period of digester Operation without feeding will be included. 1. St 1 Perio As defined in the previous chapter. the stable period is consi- dered to occur when the measured daily gas production is relatively constant for at least one detention time. The daily gas production for each digester measured using the wet test meter. is plotted in Figure 5-1. These data were recorded after the digesters had been Operated for about four detention times. The mean daily gas production 56 S7 .H «ceadueenm «on emcee-em none: ueeh eon noun eoauoeeoum use mafia: .Hln mmaoum mama .mmoo condom OHm mom com mam can mmm _ _ _ _ _ _ e L N N noumomfla d L m H noumomflo o l. v I. m m h p/I ’uoraanpoxa sea Ktreq 58 for the whole time period was 6.43 liters for Digester 1 and 6.48 liters for Digester 2 with standard deviations of 0.36 and 0.25 liter respectively. This variability was small and may have been caused. in part. by fluctuations in atmospheric pressure for which no correction was made. The variability was much less during the last 5 Idays when the continuous gas production data were taken. The mean daily gas pro- duction from wet test meter readings for this five day period was 6.14 liters for Digester l and 6.50 liters for Digester 2 with standard deviations of 0.16 and 0.12 liters respectively. 2. Gas Production Qynamigs Continuous readings of gas production were Obtained using bubble counts during the last five days of the stable period (Julian days 309 to 314). The means of these data are plotted in Figure 5-2 along with the 99% confidence intervals. The gas production curves for both digesters followed the same pattern. rapidly increasing in the first two hours after feeding. peaking at about 2 to 4 hours. then gradually decreasing almost linearly to the end Of the feeding cycle. The mean daily gas production calculated from the bubble counts is 6.7 lid for Digester 1 and 7.7 l/d for Digester 2. 11% and 18% higher than the wet test meter readings of the corresponding digesters. This discrepancy was significantly reduced for all later experiments as a result of a better adjustment of the oil level in the bubble tubes which led to a more consistent bubble size. Despite the difference in total daily gas production. the overall 24 hour patterns Of gas produc- tion for Digesters 1 and 2 were almost identical. 59 SCEL-—1 900- —" Digester l 702L-—d 602L-—- 502L-—- 402L-—— \ "/’\‘ SGQL I” GAS PROD” ML/HR 200. loo. — I TIME. HOURS 900.-fi 800. -— Digester 2 700. '— BDD. — 500. — 400. GAS PROD, ML/HR 300. 200. .1021'- TIME. HOURS FIGURE 5-2. Mean Gas Production During the Stable Period of Experiment I. Dashed lines are the 99$ confidence intervals on the mean (t test with n - 5). 60 TABLE 5-1. Substrate Degradation and COO Mass Balance for Experimental Group I. Effluent Parameters Influent DIG l DIG 2 TOtal Volatile Solids. g/l Mean 137.8 88.9 88.0 S.D. 3.5 3.8 4.4 S Reduction -- 35.5 36.1 Tbtal COD. g/l Mean 170.0 103.4 105.7 S.D. 11.5 10.4 9.3 S Reduction - 39.2 37.8 COD/TVS 1.23 1.16 1.20 Gas Produciton. l/d Mean -- 6.14 6.50 S.D. -- 0.16 0.12 Gas/COD. —- 0.97 1.06 9 Gas/COD 8 ratio Of gas measured by wet test meter to gas equivalent Of COD reduction (0.382 liters of CH. at 25°C and 1 atm is equivalent to 1 gram COD assuming digester gas contains 60S CH‘). 3. Substrate e rad tion and COO Mass Balance TOtal volatile solids and total COD reduction were determined in order to evaluate the efficiency of the Operating system. Because no oxidizing agent was added to the digesters. all the COD removed must be converted to methane or. in rare cases. hydrogen. Therefore the biogas equivalent of the measured COD reduction should balance the measured gas production. Table 5-1 summarizes the substrate degradation in terms of total volatile solids and COO. The reductions of total volatile solids were 35.5 and 36.1 percent and the COD removals were 39.2 and 37.8 percent for Digester l and Digester 2 respectively. This suggested that the ratio of COD removal 61 to total volatile solids removal is approximately 1.08. A mass balance Of the measured gas production ind COD reduction was calculated using a conversion factor of 0.382 liters CH. mt zs'c, 1 atm per gram COD. and assuming digester gas contains 60S CH‘. Comparing the measured gas production with the gas equivalent of the measured COD reduction gives ratios Of 0.97 and 1.06 for Digesters 1 and 2 respectively. Thus the discrepancy in the COD mass balance is less than 6 S. 4. Volatile Frtty Acids The volatile fatty acid pool size is important in the study of anaerobic fermentation because it indicates how well the acid produc- tion balances with its removal. Acetic acid is a particularly important intermediate because it has been suggested as a rate limiting step for the soluble part of the substrate. Samples for volatile acid analysis were taken at five times during the feeding cycle over two days during the stable period. The average values obtained in these analyses are plotted in Figure 5-3 for each digester. Because the concentrations Of individual C‘ to C‘ acids yore small. they are reported as a single group. The daily fluctuation of volatile acids in each digester over the 24 hour feeding cycle follows a similar pattern. After feeding. acetate increased about 4-5 fold due to the high level of acetate in the feed (filled circles) and remained relatively constant for about 4-5 hours. then slowly declined until the end of the feeding cycle. This follows the same pattern as the fluctu- ation in gas production rates. The high rate Of gas production with a relatively constant volatile acid concentration during the several 62 4000 ~— 0 Digester 1 <3 (J U) '5 3000 b Total VFA :1 CF- I ——43 E Cr—— . {.1 HP 0 .r o H H u 1000 m r-l o > 0 D C U m 3000 (O H \ U‘ E ~ 2000 'U "-1 U ‘1: ,3 W4 1000 — u 3 HAC g Cb“C6 0 l l l l l l 0 4 8 12 16 20 24 Time After Feeding, hr. FIGURE 5-3. Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment I. Solid symbols represent calculated concentrations immediately following feeding. 63 hours after feeding suggests that some components other than volatile acids in the influent manure are rapidly degradable. The C‘ to c‘ volatile acids varied in about the same manner as acetate. The pro- pionate concentrations in both digesters were very high and remained constant over the feeding cycle. 5. Gas Co osition Methane. carbon dioxide and nitrOgen were the major components found in the digester gas. The nitrogen content was small. increasing sharply to about 3 to 5 percent following feeding. then decreasing to 1 to 2 percent a few hours later. This nitrogen seems to be derived from air which entered the digesters during the feeding process. Therefore. the gas composition results have been normalized to include only methane and carbon dioxide. The methane content of the head space varied over the feeding cycle as shown in Figure 5-4. Both digesters showed a very similar variation of methane content. The methane percentage started declining Ifollowing the feeding and reached a minimum of 58S after about 8 to 10 hours. then started rising till the end Of the feeding cycle when the maximum was about 62S. The average methane content was 60S. Because the gas sample taken from the head space is the mixture Of newly produced gas and that remaining from earlier. the variation of methane content is also a function of the head space volume. For this experiment the head space volume is 0.75 liter which is about one fourth Of the digester volume. In a typical farm digester. the head space proportion is normally higher. Thus a smaller variation of methane content can be expected. 64 62 % Methane, ox O 58 Digester 1 62 % Methane, Ch 0 58 Digester 2 12 16 20 24 Time after Feeding, hr FIGURE 5-4. Methane Content in the Digester Head Space During the Stable Period of Experiment I. 65 6. pH and Total Alkalinity Results for pH and alkalinity are plotted in Figure 5-5. The effluent samples for both parameters were withdrawn just before feed- ing. Several samples for pH measurement were also taken at different times during the feeding cycle but showed no variation. 30.p The pH over the entire experiment was almost constant at about 7.65 to 7.70 for both digesters. The pH of the effluent was slightly higher than for the influent manure which had pH 7.40. The alkalinity data were also constant with the effluent value about 50S higher than that of the influent manure. Since the volatile acid concentration of the influent manure was much higher than that of the effluent. the increase in alka- linity of the digesters was primarily due to an increase in bicarbonate alkalinity. The increase in pH is due both to the removal of volatile acids in the influent and to the hydrolysis and fermentations of pro- teins which release ammonia as shown by Jewell (1980) and Eastman and Ferguson (1981). 7. Gas Production during Ertended Digester Qperation Virhout Feeding At the end of Experiment I. the Operation of the two digesters were extended without feeding until the gas production stopped. The daily gas production values recorded from the wet test meters are shown in Figure 5-6. The rates of gas production for the two digesters recorded over 37 days were almost the same. Gas production dropped sharply after the first day without feeding and continued dropping moderately for 10 days before remaining relatively constant for another three weeks. .H umelmnomum now due: an an. huumuueund amuou. .nln mun—Gnu mama .mwoo condom 66 com can con . com com com cam com 1 _ _ _ _ _ _ .o ,m v O? I Lb OH Um. Com ~63 o % IO IIO 1 cm um. dam 29.5 0 mm £832: a 1 cm .mo. _ _ m _ _ _ _ I or. uoooflam ~63 o I mm d ueooflem SEE o H ucosamcH o 67 .H uneaduoeum aquacuuom neaeoem neona«e cage-aeoo ecu-one: venueunm neuuem eouuoeeoum use .wln museum when .oommoum mcflooom moum< oEflB mm mm an «N om ma NH w v o _ _ _ _ _ d W _ ILWII IIId Jm O _ m Sandman C H moumomwo O p/I ’UOIuonpoza see 68 B. ELEERIMENTAL GROUP II The experimental data presented in this section describe the dynamics of gas production as a result of the combined effects of pulse feeding and temperature fluctuations. Other measured parameters include overall volatile solids and COO reduction. individual volatile acids and gas composition over the 24 hour feeding cycle. Also includ- ed are gas production data resulting from extended digester Operation without feeding. These results were obtained from two 14-liter. daily pulse fed digesters. Three different phase relationships between the feeding and temperature cycles were investigated using one digester. The other was Operated at a constant temperature as a control. For convenience the three phase relation experiments will be referred to as Experiments IIA. IIB and IIC in accordance with the phase relationships described in the previous chapter. The control unit will be referred to as Control. 1. Stabilization and Repdidation of the Two Digesters Data collection began on Julian Day 85 after the full amount Of influent manure had been fed for at least one detention time to each digester. This gave a detention time of 19 days. Time series of these data are shown in Figures 5-7 and 5-8. 0n Julian Day 111. the tempera- ture of the control digester was increased from 35.4'C to 35.8'C to match the average temperature of the other digester. For several days after the increase of temperature. gas production increased by about ten percent. then drOpped to about the same level as before. On sever- al Occasions. there were some problems with the temperature controller causing the temperature to remain at a high level (37 to 38°C) for some 69 .uoueomqn monumou emu mum m0 cmfiazh o2 ms 02 m: o: m3 2: mm cm 3 m _ _ d _ _ _ _ _ _ éHH mxm l o Ir oom.m.+m.mm I_ bow mm I. m (\ 'uorqonpoxa sea P/I mama .mxoo confine one mad can man can mom com me om mam a _ _ _ 4 _ _ _ _ m nomezoo l. o d I O D. «a; m h 3 m. A. a . I» . u .om mm —l Uov mm l. m . . T. / D. 70 .oHH mes mum nunolmuemuu new Home: ouch «om emu mama: moduoemoum new human mama .mmoa common mam cam mow cow mam oma mam owe mod _ _ _ _ _ _ _ _ _ aom.m+~.omu.a2oe l \xO/a . Ox 10.. lllllllllllll C e .1 e e e .-e e .1 UHH mxm O 0 e.. e e mama .maoa common owe med cam mom com mma omm. med ova mma . _ _ q _ _ _ _ _ _ oom.m+m.mmu.ane l a m mHH oxm a ti .aln mmbwnm 'uoruanpond sea P/I 1‘ p/t ’uoruanpoxd sea 71 time before being noticed. This problem was worst at the end of Exper- iment IIC because the temperature was checked at the peak of its cycle which was. coincidently. the value at which the controller was locked. Data from this period have not been used in the following presentation. In all cases. stable periods were assumed after digesters were fed with a constant amount of manure for at least two detention times (38 days). Vhen the phase relationship was shifted the digester was Operated for about one detention time before taking stable period data. From Julian Day 86 to 103. both digesters were operated at con- stant temperature. The means and standard deviations of daily gas production for this period were 6.61 r 0.17 1/d for the control diges- ter. and 6.70 t 0.28 l/d for the other digester. This indicates that a high degree of replication can be obtained for two digesters Operated under the same conditions. 2. Gas Produrtion byramicr Continuous measurement of gas production was obtained from bubble counts using the apparatus and procedure described in the previous chapter. Vhen digesters reached the stable period for each experiment. the bubble tubes were calibrated and six days of data were obtained to determine the mean and standard deviation for each 20 minute period. The results are plotted in Figures 5-9 to 5-12 along with the cor- responsing temperature cycle. Solid lines represent the mean values while the dashed lines represent the 99S confidence limits for the mean. lOOO. SOGL BOQL 7OQL 6021 SOQL 40%. GAS PROD” ML/HR SOZL 2821 lOELR 72 TEMP" C 01 $9 40. SO. FIGURE 5-9. Mean Gas Production Period of Experiment II. Control. intervals on the mean (t test. n 12. TIME, HOURS and Thmperature During the Stable Dashed lines are the 99S confidence 3 6). 73 1000. *- SOO. — BOO. 700. GOO. 500. 400. BOO- GAS PROD. ML/HR 200. 180. C 50. —- TEMP.. TIME. HOURS FIGURE 5-10. Mean Gas Production and Temperature During the Stable Period of Experiment IIA. Dashed lines are the 99S confidence intervals on the mean (t test. n - 6). 74 1000.- 900.— Boa.— g 700.— E z 800. cf 50a. 8 d 400. 93 o 300. 2021 we. 9 l I l r I l I l I I I U _ : 50.— % LLJ '— .- TIME. HOURS FIGURE 5-11. Mean Gas Production and Temperature During the Stable Period Of Experiment IIB. Dashed lines are the 99S confidence intervals on the mean (t test. n - 6). 75 mm. -——« 9w. 800. 7m. 5m. 500. 400. GAS PROD” ML/HR BZZL 200. lGEL C S@.-‘ TEMP” 395- I l I I I TIME, HOURS FIGURE 5-12. lean Gas Production and Temperature During the Stable Period of Experiment IIC. Dashed lines are the 99% confidence intervals on the mean (t test. n - 6). 76 Control Qigester 'ithin one and a half hours after feeding, the gas production rate increased from about 180 mllhr to the peak of 420 mllhr. The rate then decreased almost linearly to the end of the feeding cycle. Egperiment IIA For this experiment. the influent manure was fed at the midpoint of the ascending temperature ramp. Two hours after feeding, the rate of gas production reached a peak of about 650 mllhr then remained rela- tively constant for three hours. The rate started to decline shortly before the temperature reached its peak. suggesting that the readily degradable substrate was being depleted. The rate of gas production continued to decline until the minimum temperature was reached at which time gas production was only 85 mllhr. As the temperature again increased, the gas production rate also increased in a parallel fashion. figperimegg IIB For this experiment. the digester was fed when the temperature was at a minimum. The gas production reached a peak of 580 mllhr about one and a half hours after feeding. then declined gradually to about 140 mllhr at the end of the feeding cycle. ggpegimegt IIC In Experiment IIC, the digester was fed when the temperature was at its maximum. The gas production rate reached a peak of 860 mllhr about one hour after feeding. The gas production soon began to drop sharply until the minimum temperature was reached at which time the 77 rate was about 150 mllhr. The rate then slowly increased with the increasing temperature. 3. gogpggigon 2f the Bubble Tube and Wet Tegt Meter Result; Tb check the accuracy of the bubble counting method of measuring gas, the total daily gas production computed from the bubble counts is plotted in Figure 5-13 together with the wet test meter results for the stable periods. The data from both methods are fairly close. except for the control unit where the wet test meter results were consistently higher than the bubble count values. indicating an error of about 9% in the calibration of the bubble tube for that digester. 4. r ad tio n lass B Is e The substrate degradation during the stable period in terms of total volatile solids and COD reduction is summarized in Table 5-2. The results show that the reduction of COD is about 4% greater than for volatile solids in all cases with the same pattern for both parameters among the four digesters. Particularly interesting is that the vari- able temperature digesters ,had consistently greater removal than the constant temperature control. The mean daily gas production data were included in the table to determine the mass balance for the system. The mass balance was done by comparing the measured gas production with the calculated gas equivalent of the COD reduction. The calculation was based on the assumption that the temperature of the gas was at 25.0 and 1 atm. during measurement and that the digester gas contained 60% methane. These mass balance calculations show a maximum discrepancy of 15% which could be due to inaccuracies in COD measurement or assumed conversions. 78 8 _ CONTROL 0 Wet test meter 0 Bdiflecxmmt 7 — W 5 _ M— + A 4c 4 l 1 I I L l 128 129 130 131 132 133 8 .— 7 _ 'U \ '* 5 _ EXP IIA § 1 1 I L l l ‘8 114 115 130 131 132 133 5 re 8 8 - o. 3 7 — L9 6 .— L J I L J J 162 163 164 167 168 169 8 '— M 7 _. 5 — EXP IIC l l l l I l 190 191 192 193 194 195 Julian Days, 1983 FIGDEEIS-13. Comparison of Daily Gas Production During the Stable Period of Experiment II by Bubble Counts and by Iet Test Meter Readings. 79 TIBLE 5-2. Substrate Degradation and COD lass Balance for Experimental Group II. Inf. Effluent. g/l Parmmeter g/l Control Exp IIA Exp IIB Exp IIC TVS. s/l lean 34.4 19.9 16.7 17.7 17.0 S.D. 1.7 0.7 0.4 0.9 0.4 Removal. $ 42.2 51.5 48.6 50.6 COD. [/1 lean 38.9 20.9 17.2 18.5 17.9 S.D. 2.2 1.8 1.3 1.4 2.2 Removal. $ 46.2 55.7 52.5 53.9 COD/TVS 1.13 1.05 1.03 1.04 1.06 Gas Production. l/d Iet Test Meter -- 6.57 7.28 7.10 7.67 S.D. -- 0.11 0.25 0.13 0.16 Bubble Count - 6.05 7.29 7.08 7.52 S.D. - 0.08 0.33 0.41 0.35 Ga8(wet test)/COD - 1.15 1.06 1.09 1.15 Ga8(bubble oount)/COD -- 1.06 1.06 1.09 1.13 Gas/COD - ratio of the measured gas to the gas equivalent of COD reduction (0.382 liters of CD4 gt 25'C and 1 atm is equivalent to 1 gram COD assuming the digester gas contains 60% CH‘) 5. o i a t namics Fluctuations in the concentration of volatile acids as the result of combined daily pulse feeding and temperature fluctuations are presented in Figures 5-14 to 5-17. In all digesters. acetic acid pre- dominated followed by prOpionic acid. Butyric and iso-butyric acids had small but measurable concentrations and have been combined in the figures. Higher carbon volatile acids were barely detectable. In all cases the concentrations of volatile acids in the influent manure were higher than those in the digesters so that the volatile acid concentra- 80 .meuveou ueqscunou huuamvoalu neuuauaeeccec vouaumouao aceueunou sacrum. u««om .ueuueeu .HH aeoa«uoaum «c acquem e—aaum emu mefiuea necduauueeeeco mme< haven eumusuc> «suvupmmuu .vnlm uuaunm .E 55me .892 85 ON ma NH m c o . _ o . Jauwflo m m i m: mmTTmmH ///JQ I. OOH / ¢m> Hmuofi vm 0mm IL com cm S? V61! '91:»: 311121011 IL com 81 .meamoou mam-ouncu huuamvollm semuauueeoeeo nous—eomso neeaoquu ancpahu canon . «seeu>mmem E 5588 .8»? 85 3 o 2: com 1 8m .nnln mmacnm mo 89 I/fim ’PTDV 911991011 82 .uemuoou mum-canon huueauoaia saauauuueomeo meanneousc nee-ounce enema». wagon .mHH macaquenum we vcmuom cup-um on» menus: necuuauuuooeoo mfl0< human enaus~e> uaemm>mmeu .wutn muaonm Ha .mficmmm you? 99H. x cm 3 2 a 6 o _ _ o I cm W n. m. 1 2: w I'm.“ 1 ofl w. 1 8m 83 .ue«eoeu «aqueducu mucusnvoaam necuuauueooeeo nouaaeomao ueoaouqou «genius canon .u an me. an aoosuoa coca-Io: oueuauoaaoh emu ouoas eomuoeeuuaa usuaeuueeu on» «queen nee«uauueooeeu euo< Muaam ouquaue> useem>aveu .hnIn muacum o o m: .6 Ion / / m. m. I Lon: a w m 2: 1.2m ¢m> Haves. a m loom 0mm 84 tion increased sharply as a result of feeding. In each figure. the filled symbols represent the concentrations of acids calculated from mixing the influent manure with the digester contents. Control Unit As a result of feeding. acetate increased in the control digester from 31 mg/l to 142 mg/l. Over the next two hours. acetate continued to increase. peaking at 203 mg/l. before declining steadily to the end of the day. After the initial increase due to feeding. prOpionate and butyrate declined slowly throughtout the day. W The pattern of volatile acids in Experiment IIA is very similar to that of the control unit. except that all the individual acids declined faster. At the end of twelve hours all acids were nearly depleted. totaling only 15 mg/l. Experiment IIB In Experiment IIB. all volatile acids dropped sharply in the first hour following feeding and all except acetic acid declined over the rest of the daily cycle. The acetic acid showed small increases at several times. Again. butyrate and iso-butyrate were at very low con- centrations throughout. The overall level of total volatile acids was generally lower than in the control digester. ggperiment IIC In Experiment IIC. all volatile acid samples were taken when the digester temperature stayed between 37°C and 38°C due to a faulty tem- perature controller which was not noticed until after the experiment 85 was terminated. Therefore. these data show the effect of operating at a constant temperature about 2 to 3 'C higher than normal rather than with a variable temperature. under these conditions the volatile acid concentrations declined very rapidly in the 8 hours following feeding and remained at low levels for the rest of the cycle. 6. Gas 0 o iti n namic The gas composition data of Experiment II have been normalized to include only methane and carbon dioxide for the same reason described in Experiment I. Figure 5-18 shows the methane content cf the digester gas for Control. Experiment IIA and Experiment IIB (Experiment IIC data are not presented due to the faulty temperature controller). The fluc- tuations of methane content for the three experiments are similar; all have a minimum methane content at about 7 to 8 hours after feeding. Exeriment IIA. however. has twice as much fluctuation as the Control and Experiment IIB. This is due to the fact that a larger amount of gas. 4.6 liters. was produced during the 8 hours after feeding for Experiment IIA than for the Control and Experiment IIB which produced 2.7 and 3.7 liters respectively. 7. E al A al i Results for pH and alkalinity are plotted in Figure 5-19. The mean pH values for both digesters were almost equal at about 7.45 with a standard deviation of less than 0.1 unit throughout Experiment II. The mean pH of the influent manure was 8.1 with a standard deviation of 0.1. This influent pH was 0.7 unit higher than the influent manure of Experiment I due to stripping C02 during blending since the pH of the thawed manure measured before being diluted and blended was 7.4. about 62 % 60 Methane, 58 62 % 60 Methane, 58 62 % 60 Methane, 58 FIGURE 5-18. Stable Period - CONTROL "' J I I 1 I I 0 4 8 12 16 20 24 Time After Feeding, hr EXP IIA "' I l I I I 1 0 4 8 12 16 20 ?4 Time After Feeding, hr ’- EXP IIB e . ° _ . . I- h- I I I l 11 l I 0 4 8 12 16 20 24 Time After Feeding, hr Iethane Content in the Digester Eead Space During the for Experiment II. 87 .2 2.2.225. 3“ an 2:. 5323.2 33. .21 an: mama 5.32.. cam o2” 2H 9.3 cm...” 0.: SH £3 23 o .mmm 35:00 O ungflfi o 1 _ r 1 ‘I .14 Ir. 03 ea 4. mg 88 AI 5H 88 fl' ‘0023 52 M115 'hmmrtv cg. . mK o.m 88 the same as the influent manure of Experiment I. Total average alkalinities of the effluents for both digesters were about 5.000 38/1 ‘8 CaCO, compared with the influent of about 2.900 mail. The total alklinity of Experiment II was about one fourth that of Experiment I which is the dilution ratio for the influent manure. 8. G s ro no on ri n e er er tion M22121 At the end of Experiment IIC. the operation of the digester was extended without feeding. The mixing conditions remained the same and the temperature controller was corrected to the preper Ekperiment IIC pattern. The data for daily gas production recorded from the wet test meter is shown in Figure 5-20. The pattern for the decline in rate of gas produciton is similar to that of Experiment I. 89 .uHu accumuenum mouse—mom unqueom «nonuas eo«ueueno nouooumn usueeunm undue: uoauouuoum new .cNIn munenm when .uommoum mcwvoom Houmm mafia om mN ON mH OH m o _\w . \. qI — _ _ _ o {410 D H m 9 e s m a 1 0 am. a U. mO .u w... // D. h I. m VI. MATHEMATICAL MODEL OF GAS PRODUCTION DYNAMICS The observed gas production dynamics have been presented in Chapters 5. In this chapter. a mathematical model is formulated to describe the effects of pulse feeding and temperature fluctuations on manure digestion. Values for the model parameters were obtained from the constant temperature experiments and the periods of extended opera- tion without feeding. The theoretical results calculated from the mathematical model are graphically related to the experimental data from the other operating conditions. A. MODEL QEEELQPMENT For a homOgeneous substrate, the rate of reaction depends on the composition of the substrate as well as the temperature and pressure of the system. The rate of reaction of component A may be written as: EA - f(state of the system) - f(temperature. pressure. composition) (6-1) In the digesters being modeled in this investigation the pressure is held constant by the experimental conditions. Thus the reaction becomes: [A 3 f(temperature. composition) (5-2) In this investigation we are concerned with the forms of this functional relationship. A general model with constant temperature will be developed first. The Arrhenius law will then be incorporated 90 91 into the model when temperature fluctutions are considered. One assumption for this model is that the digester is operated under stable conditions and an active bacterial culture exists. 1. Mo el or Dai e eed Di ester 11.992;saat_12222r112r2 Methane production is directly correlated with substrate reduction in terms of chemical oxygen demand (COD). Because the sulfate and nitrate content of the influent manure are insignificant, the only way COD reduction can occur is through the conversion of organic material to methane and carbon dioxide. The initial amount of substrate can therefore be measured in terms of its ultimate gas potential (6'), the total amount of gas which could be produced from an infinite digestion period. In this model the ultimate gas potential represents the diges- ter contents immediately after feeding rather than the amount of substrate in the feed. Therefore. knowing the ultimate gas potential (G') iulgdiatoly after feeding and the volume of gas produced. the remaining gas potenr tial (G) in the digester can be calculated by: G - G° - It 2 dt (5‘3) where 6' - ultimate gas potential in the digester. liters of gas at 1 atm and 25 'C: G - gas potential in the digester at time t. liters of gas at 1 atm and zs‘c; and R a rate of gas production. l/d of gas at 1 atm and 25°C. 92 Figures 6-1 and 6-2 show semi-log plots for the rate of gas production versus time for the experimental results obtained from extended operation, without feeding, of digsters from Experiment I and IIC respectively. Interestingly. both plots show three approximately linear relationships. suggesting that the substrate in each digester can be approximated by three components. each following first order kinetics as described in the following equation. Bi ' IiGi (6'4) where [1 I rate constant for component i. d": 31 8 rate of gas production for component i. l/d of gas at 1 atm and zs‘c; and 61 - gas potential for component i. liters of gas at 1 atm and 25°C. The three components can be combined in terms of both rate of gas production and remaining gas potential: Gt 7 G: + Gs + Gs (6-6) where R3, R,. R, are the rates of gas production from the slow. moderate and fast fractions respectively. l/d 3 Rt is the total rate of gas produciton. l/d: G‘, G, and G, are the gas potentials of the three substrate fractions. liters: and Gt is the total gas potential. liters. 93 3.. 2— 1.54 -1 l-I» K1 = 1n (O.64/0.455) = 0.0085 d 0.9— 40 3'34 63= 0.64 =7s.3 liters 0.6— 0.5—I 0.4.. Gas Production, l/d 0.3— 1 l I T I 0 10 20 30 40 50 Tine After Feeding Stopped, Days FIGURE 6-1. Graphical Estimation of the First Order Rate Constant and the Initial Gas Potential for the Slow Fraction for Experiment I. l/d Gas Production, 94 0.8-- 4 K1 =1n(0.655/0.S) = 0.00912 d 30 1 “30.0083 d G: = 0.655 = 71.6 liters the Intial Gas Potential of the Slow Fraction for Experiment II. I 5 I I I I *T 10 15 20 25 30 Time After Feeding Stopped, Days Graphical Estimation of the First Order Rate Constant and 9 K' - effective constant temperature rate constant that gives the same gas production as I gives with variable temperature (I'll - 1.094). 95 3¢b3titntinl 3, 8 - dGi/dt into Equation 6-4 and integrating gives the following equation for a constant temperature digester. G1 ' 61°.Iit (6-7) Combining Equations 6-4. 6-5 and 6-7 gives at - r,c;o":t + r,0;o"a‘ + r,0:o"st (6-8) For Experiment I. K, and G: were obtained from the lowest part of thO OEIVO in Figure 5‘1 'hOIO Rt . R, since R, and R, are approximately zero due to substrate depletion. The slope is -K, and the intercept is R; - K,G: from which G: can be determined. The parameters. I, and G; were obtained by plotting R, - Rt - R, (Figure 6-3a) where R1 - 1,6;a"1t. Then. the slope is -K, and the intercept 1‘ K,G:. In a similar manner. I, and G: were obtained from Figure 6-3b by calculating R, n It - R, - 2,. The data and calcula- tions involved are presented in Appendices Cl and CZ. The results are summarised in Table 6-1. Values for these kinetics parameters for Experiment II were simi— larly obtained from Figures 6-2 and 6-4. The data for the moderate and slow fractions came from the extended operation of Experiment IIC without feeding. The values for K, and I, obtained from Figurea 6-2 and 6-4a were divided by a correction factor of 1.094 to account for the effect of the temperature cycle as described in Appendix D and then normalized to a reference temperature of 35.8'C. The data for the fast fraction came from the mean value of the stable period for the control digester. The data and calculations involved are presented in Appen- dices C3 and C4. The results are included in Table 6-1. 96 1- 0.9‘— 0.8-- 0.7-— 006 ‘- Gas Production, 1/d 0.5 '— / (a) 1n(4.25/O.151) = 0.335 d" 8 10 4.25 = 12.7 liters 0.335 6) ll I I I I I 6 8 10 (b) -1 K3 =1n(3.84/l.22) = 1.15 d G:=3.84/1.15 =3.34 liters 1.5" (3 C) l I [ I I I 0 0.2 0.4 0.6 0.8 1.0 Time After Feeding Stopped, Days FIGURE 6-3. Graphical Estimation of the First Order Rate Constants and Intial Gas Potentials of (a) the Moderate and (b) the Fast Fractions for Experiment I. 97 Time After Feeding Stopped, l/d 0 0.2 0.4 0.6 0.8 1.0 10 I I I I I -1 K3 = 1n(9.0/1.01) =2.19 d 1.0 -_9.0 = 4.1 liters — 2.19 l/d 1— 0.9‘- 0.8- 0.7 — 0.6— Kz=ln(2.7/O.353) 0.5— =o.203 d’l K2=O.186 d" G:=13.3 liters Gas Production, 0* 0.4 _' 0.3‘— o I I I I. A Time After Feeding Stopped, Days FIGURE 6-4. Graphical Estimation of the First Order Rate Constants and Initial was Potential of (a) the Moderate and (b) the Fast Fraction for Experiment II. 98 TABLE 6-1. Summary of Estimated Parameters for Mathematic Model (Normalized to wet test meter and constant temperature basis). Parameters Experiment I Experiment II Tr: 'c 36.4 35.3 r,, 4" 0.0085 0.0075 1,, 4" 0.335 0.168 ,, 4" 1.15 2.19 6;. liters 75.3 71.6 6:. liters 12.7 13.3 6:. liters 3.3 4.1 6;. 1119:. 91.3 89.0 0, - 0, - 0, 1.25 1.25 Tb 'PPIY th° model. 3: is plotted as a function of time using Equation 6-8. The model is compared with the experimental data in Fig- ure 6-5 for Experiment I and Figure 6-6 for Experiment 11 Control. The solid lines represent the means of the observed data while the dashed lines represent the predicted gas production rates. The areas under the curves between each line represent the gas production accounted for by each fraction of the substrate. 2. M 1 Tem er tu ri t on Variations in reaction rate as a function of temperature can gen- erally be represented by the Arrhenius equation: x = A,.E/RT (6-9) 99 908. -- 800' '—4 Digester l 780. -— 600..— 580. -— 400. 380. 200. GAS PROD. ML/HR 100. TIME, HOURS 90%. ‘— BOO. — Digester 2 703. —‘ 500. —‘ 509. —‘ 400. 300. GAS PROD.. ML/HR ZOO. 180. TIME. HOURS FIGURE 6-5. Comparison Between Model Results and Observed Data for Experiment I. (a) Digester l and (b) Digester 2. GAS PROD" ML/HR TEMP” C 100 1000.—r SOO.- BOO.—r 700u—~ BOO.-‘ SOO.-— 400. 300. 200. 100. 58.- 40.-— 39' IIIIIIII'III 0 4 8 12 16. 28 24 TIMEHOURS FIGURE 6-6. Cauparison Between Model Resutls and Observed Data for Experiment II. Control. 101 or, in legarithmic form 1n! = 1M. — E/R’l‘ (6-10) where R = rate constant; Ae - Arrhenius frequency factor: E - Activation energy: R - universal gas constant: and T - the absolute temperature. Strictly, the Arrhenius equation is applicable only to either a single stage reaction or to multistage raction in which the first step is rate détermining ('eber, 1972). The energy of activation. E. determines the fraction of the total number of molecules which are suffficiently activated at a given tem- perature to undergo reaction. The magnitude of E is therefore a direct determinant of the rate of a particular chemical reaction. The larger the value of E. the more the reaction is affeted by temperature. Ihen Equation 6-10 is evaluated against a reference temperature (Tr: It). the resulting expression is In III! a E(T-T’r)/R1"rr (6-11) or K = K‘OT’Tr (6-12) Where 0 is eE/RITr. In Equation 6-12, T and Tr may be expressed ,, celsius temperature rather than absolute temperature because the difference is the same in each case. Equation 6-4 now becomes: R: = Ki 0“: G: (6-13) 102 lhere 6, must be evaluated by substituting Equation 6-13 into Equation 6-3. The result is t Gi .. a: - I xi 0'1"": G, dt “-1“ O in which T varies with time. For computational purposes this is writ- ten in finite difference form: t Gi,t+At - G; -— 2 xi 91-1, 31,: At (6-15) 0 The overall gas production rate is still given by Equation 6-5. The model depends heavily on the value of the temperature coefficient, 0. which was estimated from the extended period following the last feeding of Experiment IIC with the same temperature cycle con- tinued. For each day a value of O was estimated from the ratio for the maximum to minimum gas production rates. These values were then aver- aged to give a mean 0 of 1.25 with a standard deviation of 0.02 (n-7). B. COMPARISON OE YARIABLE TEMPERATURE MODEL TO ERIMENTAL ATA To compare the model results with the observed data the numerical integration procedure was incorporated into a FORTRAN program (Appendix C5) and executed on a DEC PDP-11/23 computer using a At of 5 minutes and the actual temperature data (Figures 5-10 to 5-12) observed for Experimental Group II. The results are shown in Figures 6-7 to 6-9 and discussed in the next chapter. 103 1000. —m 900. — BOO. —-‘ GI: 70%. \ :J 660. o’ 500. 8 (1 400. 5’3 L3 300. 200. 100. 0. U [f 50. — Z LU *— _I 40. — N 30- I I I I I I I I I I I I 0 4 8 12 18. 20 24 TIME, HOURS FIGURE 6-7. Comparison Between Model Results and Observed Data for Experiment IIA. 188(7). 900. 800. 708. BOO. 580. 480. GAS PROD, ML/HR 300. 280. 100. 104 50. TEMP., C 40. 30. FIGURE 6-8 . TIME. HOURS Comparison Between Model Results and Observed Data for Experiment IIB. GAS PROD, ML/HR TEMP., C 105 1000. -— I 900. Boo. 700. 600. 500. 400. 300. 2m. 1w SO. 40. 30' ' I I I I . I I I I I I I 0. 4s 8. 12. 18. 20. 24 TIME, HOURS FIGURE 6-9. Comparison Between Model Results and Observed Data for Experiment IIC. VII. DISCUSSION OF THE RESULTS The discussion of the experimental results is organized around the following topics 1) evaluation of effects of the daily pulse feeding and temperature fluctuations on digester stability: 2) determination of the amplitude and timing of the 24 hour gas production cycle as a result of daily pulse feeding alone and combined with the temperature fluctuation cycle: 3) determination of the rate limiting step of the overall methane production process: and 4) comparison of total gas pro- duction between the constant and fluctuating temperature digesters. A. DIGESTER STABIL TY Information obtained from this investigation indicates that a daily pulse feed digester, with or without small temperature fluctua- tions. can be operated with considerable stability. The stability can be evaluated by three different parameters: 1) constancy of daily gas production; 2) volatile acid pool size and its fluctuations: and 3) stability of pH and alkalinity. 1. on ta of ai a Prod tion For each of the experiments, the daily gas production. measured using the wet test meter. showed a high degree of constancy following the initial transition period during start up or following a phase shift. For Experiment I. using full strength manure at constant tem- perature. the data were recorded after the digesters had been operated for four. 15-day detention times. The 30 days of recorded data (Figure 5-1) show a standard deviation of less than 6 percent of the 106 107 mean for both digesters. The control digester for Experiment II (255 dilution. constant temperature) showed a standard deviation of only 2.6$ of the mean over the 45-day period (Figure 5-7). Vhen temperature fluctuations were imposed during Experiments IIA, IIB and IIC. the digesters responded quickly with gas production (Fig- ures 5-7 and 5-8) remaining generally stable except during periods when the temperature increased due to controller malfunction. At those times the gas production increased significantly but returned to normal when the temperature returned to the proper pattern. This indicates that there was no imbalance between the various groups of anaerobic bacteria. 2. Iglgtile Agids as an Indigator of §§abili§y For all experiments, the overall level of volatile acids was stable. The imposed temperature fluctuations did not cause any imbal- ance in the acid pool from day to day. In all cases, the volatile acid pool increased sharply following feeding due to high concentrations in the influent manure, then declined toward the end of the feeding cycle indicating that acid removal was faster than its formation. In Experiment I, the concentration of total volatile acids in the influent was about ‘16.700 mgll as COD. Following feeding. the total volatile acid pool was about 3.500 mgll. declining to 2,600 mgll at the end of the cycle (Figure 5-3). The data from both digesters during the two day sampling period were nearly identical. indicating the ability of the digester to remove the high concentrations of volatile acid in the influent manure without causing an imbalance. The propionic acid in Experiment I, however. remained constant at 108 the relatively high levels of 2.000 and 1.700 mgll as COD for Digesters 1 and 2 respectively. Vhile the persistency of the propionic acid in Experiment I has not been explained. it was found for all the experi- ments conducted later with a 25% diluted influent manure and a 19-day detention time that propionic acid was nearly depleted at the end of the cycle. Therefore. it can be suggested that the propionic acid may be reduced by Operating‘at a higher detention time and/or by diluting the influent manure. In spite of the high level of propionate in Experiment I. no sign of imbalance in volatile acids has been observed. For Experiments IIA. IIB and IIC, where the digesters were imposed with temeprature variations, the results of the volatile acid pool fluctuations were much the same as for the control (constant tempera- ture). In general, the overall levels for total acids were less than in the control digester. For all cases in Experiment II. the concen- tration of total volatile acids in the influent was about 3.700 mgll as COD. Following feeding. the acid pool sizes were about 200 to 240 mgll. declining to only 60 mgll or lower depending on the experiment.- These very low concentrations of total volatile acids at the end of the feeding cycle demonstrated that the overall daily acid removal was fas- ter than its formation. In no case did volatile acid pools increase over the daily cycle. 3. Sta it B and Alkalini In all digesters. the effluent pH and alkalinity remained constant over the entire experimental period. Furthermore the average effluent P3 for Experiment I '48 within t 0.3 pH unit of that for Experiment II. The total alkalinity in the effluent was approximately proportional to 109 the influent manure strength. having values of 18,000 mg/l as CaCO, for Experiment I and 5.000 mgll for Experiment II. This high buffer capa- city ensured that the pR did not change detectably during feeding. 4.1mm The stability of the daily pulse feed digesters with or without temperature fluctuation has been discussed. The data for daily gas production, volatile acid pool. pH and alkalinity throughout this investigation demonstrated that the proposed operating conditions are perfectly feasible in terms of digester stability. B. GAS PRODUCTION DYNAMICS The experimental results showed that the rate of gas production varied greatly as a result of either daily pulse feeding or fluctuating temperature. In addition. the pattern of gas production can be con- trolled to a large extent by phase relationship between the feeding and temperature cycles. This section will first discuss the effects of daily pulse feeding and temperature variation separately. The combined effect will then be examined. The influent manure contains a wide variety of substrates having different rates of degradation. As shown in the previous chapter. the manure used in this study can be approximately divided into three com- ponent groups on the basis of degradation rate. Data for these fractions. labeled slow, moderate and fast for covenience. are summar- ized in Table 7-1. The initial gas potential has been divided by the digester volume to normalize the data. 110 TABLE 7-1. Estimated Kinetic Parameters for the Three Substrate Fractions. Parameter Fast Moderate Slow Experiment I Rate Constant (I) at 36.4'0. d“ 1.15 0.335 0.0035 Initial Gas Potential (0'). l gas/l digester 1.5 5.6 33.5 Experiment II Rate Constant (I) at 35.3'0, 6" 2.19 0.168 0.0075 Initial Gas Potential (0'). l gas/l digester 0.43 1.40 7.54 1. Dail Pu e edin Ef e In a constant temperature digester. the decline in gas production throughout the day due to pulse feeding results from the removal of substrate since the rate constants are not affected. Thus. most of the decline in gas production is caused by the removal of the fast fraction followed, to a lesser extent, by removal of the moderate fraction. The rate of degradation of the slow fraction is so low that gas production is unaffected by its removal within one day. These effects are clearly demonstated in Figures 6-5 in which the ordinate between each dotted line represents the rate of gas production for each fraction as calcu- lated from the mathematical model. The percentage of the total gas production contributed by each fraction at any time is determined by the product of the amount of that fraction present and the rate constant. The initial concentration of substrate at the beginning of the day is determined by the prOportional 111 TABLE 7-2. Calculated Feed Concentrations of Substrate Fractions. Gas Potential. 1 gas/l digester Detention Initial Effluent Feed Removal . Time (0). d 0' (t=0) G (t-24hrs) Coac.‘ 5' Experiment I Fast Fraction 15 1.5 0.47 15.5. 97 Moderate Fraction 15 5.6 4.0 28.3 86 Slow Fraction 15 33.5 33.2 37.3 11 Total -- 40.6 37.7 81.1 54 Experiment II Fast Fraction l9_ 0.43 0.05 7.3 99 Moderate Fraction 19’ 1.40 1.2 5.2 77 Slow Fraction 19 7.54 7.5 8.6 13 Total -- 9.37 8.75 21.1 59 ‘ Feed Conc. = 0G. - (0-1)G mixing of the feed manure with the digester contents. Thus a component which is rapidly degraded will have a low concentration in the reactor although its concentration in the feed may be high. This is illustrat- ed in Table 7-2 in which the feed concentrations of each substrate fraction are calculated from the mass balance equation. The percentage removals of each fraction are also shown in the table. A comparison of Experiment I with the Control of Experiment II shows significant differences in the fast and moderate fractions but not the slow fraction. In both experiments the slow fraction was larg- est: the difference in absolute magnitude is due to the four—to-one dilution of the feed manure in Experiment II. The fast fraction. however. was proportionally higher in Experiment II while the moderate 112 fraction was lower. It is suggested that these differences were caused by blending the manure when it was diluted so that particle size decreased and cell tissue was broken up. The increase in the rate conr stant for the fast material and decrease in the constant for the moderate material (Table 7-1) is also believed to be the result of blending the manure. The effect of the changes caused by blending was to decrease the contribution of the moderate fraction and increase the contribution of the fast fraction to the overall gas production rate for Experiment II Control compared with Experiment I. Also. the higher rate constant of the fast fraction in Experiment II resulted in a more rapid decline in gas production during the daily cycle. 2. Tegperature Variation Effect Throughout the experimental program there were several indications that the rate of gas production responds rapidly to temperature changes. Whenever the temperature controller malfunctioned resulting in a sudden increase or decrease in temperature of a few degrees. the gas production rate also increased or decreased immediately and dramat- ically. Vhen the temperature returned to normal. the gas production did also. Another piece of evidence showing the effect of temperature variation on gas production came at the end of Experiment IIC when the digester continued to Operate with the same temperature fluctuation but without additional feeding. The gas production over the next 7 days closely matched the temperature cycle imposed on the digester. From this period of extended operation, the temperature coefficient (0) was estimated as 1.25 corresponding to an Arrhenius activation energy of 113 42.5 kcal/degree Kelvin. 3. Combined Effect of Feeding and Temperature- As shown in Figures 5-10 to 5-12. imposing a temperature fluctua- tion 0f only i 3.3 degrees celsius about the mean caused major changes in the magnitude and timing of the peak gas production resulting from daily pulse feeding. These changes can be largely explained by the mathematical model develOped in Chapter 6. The discussion in this sec- tion will focus on each of the three phase relationships between the temperature cycle and the pulse feeding. discribing the resulting pat- tern of gas production in relationship to the model and explaining some of the descrepencies which remain. This information can then be used to develop strategies to provide better utilization of biogas by match- ing the gas production pattern to the energy needs of the farm. Experigent IIA In Experiment IIA (Figure 6-7). the calculated results from the model show that. following feeding. the rate of gas production contin- ued to rise slightly for several hours until the temperature reached its peak. During this period. the increase in the overall rate is conr tributed largely by the moderate fraction (R1), in spite of its lower rate constant. This is because the increase in rate due to rising tem- perature outweighs the effect of substrate removal which is relatively small with respect to its pool size. The rate of gas production con- tributed by the fast fraction (R,) remained relatively constant during the period of increasing temperature since the increase in the constant was offset by depletion of substrate. When the temperature began to fall gas production from the fast fraction declined most rapidly fol- ll4 lowed by the moderate and the slow fractions respectively. This can be explained similarly by the relative effects of the temperature and the change in individual substrate pool sizes. The model results for this experiment match the experimental data fairly well. especially in the important trends. Deviations from the experimental data occurred only in the first and last four hours when the model predictions were slightly low. In the last four hours the higher slope of the experimental data indicates a stronger temperature dependency than used in the model. The gas production pattern of Experiment IIA demonstrates that it is possible to obtain high sustained gas produciton over an eight hour working day by heating the digester at a rate sufficient to balance the substrate removal effect. Allowing the digester to cool off for the remaining 16 hours would conserve energy during this time. Total gas storage requirements would be substantially reduced in this case. ri C The model results for Experiment IIC (Figure 6-9) match the exper- imental Idata very well. Following feeding. the gas production rate is maximum because both the temperature and the substrate concentrations are highest. The rate of gas production. however. stays at this peak for only a short time because both the temperature and the amount of the fast substrate fraction are decreasing simultaneously. Although the gas production pattern during the first twelve hours is dominated by the decline of the fast fraction (R,), the increggg in gas production during the last twelve hours is due to the moderate fraction (R,). The only significant deviation of the model predictions 115 from the experimental data occurred in the middle of the cycle when the predicted ‘rate dropped too low. This observation and the lower slape for the experimental data during the period of increasing temperature indicate a slightly lower temperature dependency than used in the model. opposite to the observation from Experiment IIA. The gas production pattern of Experiment IIC might be useful in cases when a large amount of gas is needed for a short period of time. In practice this pattern might be achieved by heating the feed material to a temperature higher than the digester prior to pulse feeding it. The digester could then be allowed to cool down gradually over the remainder of the cycle to keep gas production low when it isn't needed. reducing storage requirements. figpgrimgnt IIB In Experiment IIB the results predicted by the model do not fit the experimental data well as shown in Figure 6-8. As will be explained below. it is beleived that this is largely due to formulating the model entirely around the hydrolysis of particulates and ignoring the volatile acid pool. The model predicts slowly increasing gas production caused by the moderate fraction (R3) since the amount of the fast fraction is decreasing while the temperature is increasing as happened in Experi- ment IIA. The predicted gas production peaks at the same time as the temperature and then falls off rapidly as temperature decreases. The experimental results show a peak soon after feeding followed by decreasing gas production throughout the remaining period. It is suggested that the discrepancy between the predicted and 116 observed results is due to changes in the acetic acid pool size which were not incorporated into the model. As shown in Figure 5-16, the acetic acid concentration was high immediately following feeding but decreased rapidly over the first few hours. The gas equivalent of the volatile acids removed in the first six hours is 1.07 liters which is close to the 1.25 liter discrepancy between the actual and predicted gas production during this time. Since the model includes the volatile acids in the fast fraction but does not include a separate degradation term. removal of these acids during the first 6 hours means they are not available for removal later. Thus. the actual gas production rate is lower than predicted by 0.94 liter during the middle eight hours. The explanation of this phenomenon is based on the observations of Stafford et al. (1980) that methane production is sppoximately propor- tional to acetic acid concentrations up to about 2.000 mg/l. Thus. the high acid concentration following feeding caused high methane produc- tion rates and hence high acetate removal rates. These high acetate removal rates could not be balanced by hydrolysis due to the low tem- perature. This effect was offset in Experiment IIA _and IIC by the higher rate of hydrolysis at the higher initial temperatures and did not affect the results. C. THE RATE LIMITING STEP The experimental results obtained in this investigation combined with literature information indicate that hydrolysis of particulate substrates is the rate limiting step in the overall anaerobic digestion process. The principal evidence for this statement comes from the var- iation in volatile acid concentration over the feeding cycle (Figures 117 5-3 and 5-14 to 5-17). After an initial increase in acid concentration due to feeding. the volatile acid pools declined throughout the remainder of the day. Thus volatile acid removal by methane production was faster than volatile acid production by hydrolysis and fermenta- tion. . The conclusion that hydrolysis of particulate substrate is the rate limiting step in manure digestion is supported by other investiga- tions working with dairy manure (Jewell et a1.. 1980) and municipal sludge (Eastman and Ferguson. 1977). Furthermore. Eastman and Ferguson showed that fermentation of soluble hydolysis products was much faster than the hydrolysis process itself. This observation has also been assumed to hold in this investigation. Although the basic pattern of volatile acid decline was true for all experiments. the rate and extent of the decline varied for each experiment. In general. both the rate and extent of volatile acid decline was faster in the variable temperature experiments than in the constant temperature experiments. In addition. the pattern of volatile acid decline, especially for acetic acid. roughly approximates the decline in gas production for each experiment. The similarity in pattern between the acetic acid pool size and the gas production makes sense when the role of acetic acid is exam- ined. Ihen the acetic acid pool size is constant. the rate of methane production must equal. and be controlled by. the rate of hydrolysis and fermentation. Also Stafford et al. (1980) showed that. up to about 2.000 mall. the rate of methane production is approximately proportionr al to the acetic acid concentration. This indicates that the acetic acid pool size in balanced digestion may be largely controlled by the 118 rate at which acid COD. produced by hydrolysis is being converted to methane. Knowledge of the rate limiting step provides improved understand- ing of digester kinetics. As long as hydrolysis of particulates ‘remains the rate limiting step. the balance between acid formation and acid removal should not be damaged by pulse feeding or by temperature fluctuations. The balance can. however. be upset by pulse feeding of soluble substrates or particulates such as starch which have a very high rate of hydrolysis. Furthermore. knowing that particulate hydrolysis is the rate lim- iting step results in considerable simplification in the formulation of the mathematical model because only the first step of the multistage reaction need be considered in most cases. D. TOT 8 CTION THE CONST UCTUATING TEMPERA IGES Both the experimental and theoretical results in this study indi- cate that a fluctuating temperature digester produces more gas than a constant temperature digester Operated at the same mean temperature. The data for daily gas production for Experimental Group II are summarized in Table 7-3. The data measured by the wet test meter show that all the experiments imposed with temperature fluctuations have a higher total daily gas production than the constant temperature control unit by about 8 to 10 percent. For Experiment IIC. the increase in gas production was 17 percent, about half of which is estimated to be caused by the average temperature being 0.4'C higher than for the other units. This estimate assumes a temperature coefficient of 1.25. 119 TABLE 7-3. Evaluation of Gain in Total Gas Production Due to Temperature Fluctuations. Parameter Control Exp IIA Exp IIB Exp IIC Average Temperature..C 35.30 35.32 35.77 ' 36.20 TVS Removal, S 42.2 51.5 48.6 50.6 Increase Over Control, 1 22 15 2O COD Removal. 8 46.2 55.7 52.5 53.9 Increase Over Control. 8 21 14 17 Daily Gas Production. l/d Vet Test Meter 6.57 7.28 7.10 7.67 Increase Over Control, i 11 8 17 Calculated (Model) 6.27 6.55 6.54 6.32 Increase Over Control. S 4 4 9 The increase in gas production rates are substantiated by the increase in substrate removal both in terms of volatile solids and COD (Table 7-3). In addition. the theoretical results calculated from the mathematical model developed in the previous chapter support the exper- imental observations in trend if not magnitude. These data suggest that the rate of degradation is non-linear with increasing temperature such that an increased removal at higher temperatures more than offsets decreased removal at lower temepratures resulting in a net gain of gas production for each temperature cycle as compared with the Control. The 9 percent increase calculated for Experiment IIC shows the effect of the higher average temperature as well as the temperature fluctua- tion in a manner paralleling the wet test meter results. 120 E. SUMMAR! This discussion can be summarized by relating the information presented above to the objectives stated in Chapter 1. The first objective was to determine the ability of anaerobic digesters to acclimate to fluctuating temperature without loss in total gas production. Not only was it found that there was no loss in gas production when temperature fluctuations were imposed on the digestion of dairy cow manure. but gas production actually increased about 9S. This result was also predicted by the mathematical model although with a lesser increase. The second and third objectives were to determine the amplitude and lag time of the 24-hour gas production cycle caused by daily pulse feeding alone and in combination with an imposed temperature fluctua- tion. This investigation has shown that the amplitude of the gas production cycle can be controlled to a large extent by the phase rela- tionship between the pulse feeding and the temperature ramp. The higher the digester temperature at the time of feeding. the higher is the peak gas production and increasing the temperature after feeding can sustain high gas production until the most readily degradable material is consumed. The phase relationship did not. however. sub- stantially change the timing of the initial large rise in gas production. The fourth and final objective was to develOp a model from the experimental results such that some management strategies can be deter- mined. Such a model has been successfully developed based only on constant temperature experiments and on the periods of extended Opera- tion without feeding. The kinetic parameters obtained from these 121 periods of operation were used to predict the effects of imposed tem- perature fluctuations. The predicted results corresponded closely to the observed results in two cases. The discrepencies in the third case can be explained by the fact that volatile acid ultilization was not expressly incorporated into the model. VIII. MANAGEMENT IMPLICATIONS To minimize the gas storage requirement. feeding and heating of a digester must be scheduled such that the net gas produced during the high demand hours matches that demand. Consequently. a minimum frac- tion of the daily gas production remains to be stored. In this chapter. various management strategies for the reduction of gas storage will be discussed. Experimental data will be used to demonstrate how a gas storage requirement can be reduced compared to a conventional digester operated with uniform feeding at constant temperature. To maintain digester temperature or impose a desired temperature fluctuation. energy is required for heating. Energy for heating may come from burning digester gas directly or from utilizing waste heat from productive processes such as electricity generation. In the form- er case. heating requirements are in competition.with productive uses while in the latter case. heating coincides with productive uses. Furthermore. some portion of the heating requirements can be met by heating the influent separately to a temperature higher than that in the digester. The discussion in this chapter will be organized around these considerations. A. TE PROD CTI E S US Many productive uses of digester gas consume the gas without the generation of waste heat that can be diverted to digester heating. Examples include boiler operation. space heating and crop drying. In this case. gas storage requirements can be reduced by heating the digester during times when gas is not being productively utilized 122 123 and/or by increasing gas production at times when demand is highest. To illustrate the potential for reduction in gas storage needs through digester management. a hypothetical situation will be examined. In this example. productive gas requirements are uniformly high for an eight hour working day and digester heating requires 25% of the daily gas production. For illustration purposes an idealized case will be considered in which 1005 of the gas is used. For a digester with uniform feeding and heating. the gas producr tion would also be uniform as shown in Figure 8-1a. In this case the - storage requirement would be 50% of the total daily gas production. To illustrate the case of a managed digester. the pattern of Experiment IIA will be used to overlay the gas requirements as shown in Figure 8-1b. The sustained high gas production in this case requires increasing the temperature to offset the reduction of rapidly degrad- able substrate so the gas requirement for heating is not uniform but is twice as large for the twelve hours beginning at 1 AM dropping to zero at 1 PM. In addition. the pulse feed will be made one hour early. at 7 AM to allow time for gas production to rise by 8 AM. As shown by the area between the curves in Figure 8-1b. the gas storage requirement has been reduced to 24% of the total daily gas production. As an added benefit. the total gas production in this case can be expected to be 5 to 10 percent higher. A similar analysis can be made for cases in which it is desired to have short periods of very high gas production. In these cases. the pattern of Experiment IIC would be appropriate. The digester should then be heated during the period when gas is not being productively used so that feeding would occur at. or shortly before. the point of 124 Storage Requirement 300 .— Gas Utilization 3 § 200 3 fl Gas ’5 9; Production .3 / / m o ------------ 3 e 0 I 1 Heating 25% A! I I M 4 8 N 4 8 M Storage Requirement 300 ._ / Gas ‘ :y/ggf 6511...... 1 J1 / ~ 8 3 200 — / ~Z // w-I to I H I - \ \ // . g g I 4 Production I \ n. 13 100 r- I ' \(///” 3 , \ w _ _ v/ \ V ’ fl \\ I, ------- fl \‘ “1’ Heating 25% E ‘ ‘ ‘- 0 1 L’ 40 - 30 E L l I I 1 J IM 4 8 N 4 8 M Time, h FIGURE 8-1. Gas Storage Requirements for (a) Digesters with Uniform Feeding and Heating, and (b) Managed Digesters Using Conditions of Experiment IIA. 125 maximum temperature. As an alternative. some of the heating could be accomplished by increasing the temperature of the influent material prior to adding it to the digester. This would keep gas production in the digester lower until it was needed. B. HEATING COINCIDENT 'ITR PRODUCTIVE GAS USE Vhen digester gas is used to generate electricity only about 20 to 25 percent of the energy is actually converted to electricity. The remaining 75 to 80 percent is converted to heat, about 75* of which can be recovered for heating the digester and/or the influent material. This recovered heat is more than sufficient to maintain digester tem- perature. Because digester heating would occur at the same time as productive uses. the effect would be to sustain the gas production at relatively high levels as long as the rapidly degradable fraction had not been depleted. On a dairy farm high electrical demand typically occurs twice a day during the milking operation. Vhile it may not be possible to exactly match the gas production cycle to this demand. storage require- ments would be reduced if the digester were fed, twice a day approximately one hour before milking with the waste heat being used to sustain the gas production for several hours. By also heating the influent manure. the digester temperature could be increased sharply at the time of feeding to more closely coordinate gas production with utilization. The mathematical model. with some refinements, can be used to make more accurate predictions of gas production patterns for management strategies such as this which were not experimentally exam- ined in this study. IX. CONCLUSIONS Based on the results Of this study, the following conclusions can be made for dairy manure digesters Operated with daily pulse feeding at constant temperature or with small temperature fluctuations. 1. Once established. a dairy manure digester can be operated in a stable manner in conjunction with pulse feeding and temperature fluctuations. Stable Operation was achieved for all the conditions tested as indicated by low volatile acids. and constant pE (£0.05 units). alkalinity (i105) and daily gas production (*6i). Dairy manure contains a wide variety of substrates having different rates of degradation. some extremely rapid. The initial rise in gas production immediately following feeding is primarily due to substrates other than volatile acids since the acid pool did not decline to nearly the extent that gas production increased. Hydrolysis Of particulate substrates is the rate limiting step in the overall anaerobic digestion of dairy manure. VOlatile acid pool size never increased and the literature indicates that hydro- lysis products are fermented to acids as rapidly as they are produced. The rate of gas production responds rapidly to temperature changes in either direction. This was true both for the gradual tempera- ture changes intentionally imposed and for sudden temperature changes which occurred accidentally. 126 5. 127 Iithin a daily cycle. the rate of gas production varies greatly as a result of pulse feeding and temperature fluctuations. The pat- tern Of gas production can be controlled to a large extent by proper timing of the phase relationship between the feeding and temperature cycles. The constant temperature control digester showed a peak gas production 1.7 times the average occurring 1.5 hours after feeding. Feeding the digester at the peak of the temr perature cycle caused. a sharp peak 1.8 times as great as for the control but Occurring at about the same time: gas production then decreased rapidly. Feeding the digester at the midpoint of the ascending temperature ramp caused a peak gas production about 1.4 times as great as the control: high gas production was maintained for 6 hours due to increasing temperature before falling rapidly with declining temperature. A fluctuating temperature digester produces about 10% more total gas in 24 hours than a constant temperature digester Operated at the same mean temperature. This reflects the non-linear nature of the Arrhenius temperature function. For moderately or slowly degradable substrates. increased removal rates at higher tempera- tures more than Offset decreased‘ removal rates at lower temperatures for a net gain in gas production. Degradation of rapidly degradable substrates is nearly complete in 24 hours in all cases so they do not contribute to the increased gas production due to fluctuating temperatures. 7. 128 A mathematical model based on first order kinetics and the Arrhen- ius temperature relationship successfully predicted gas production dynamics as long as hydrolysis remained the rate limiting step and volatile acid pool size did not change rapidly. The data showed that the substrate could be approximated as three fractions based on the relative rates of degradation. For the whole manure diges- ters the size of these fractions in the influent as a percentage of the total gas potential and the first order rate constants at 36-4°C were: Fast Fraction: 19% with K 8 1.15 d'“1 Moderate Fraction: 35% with K - 0.335 (1'1 Slow Fraction: 467. with r = 0.0035 6" For the blended and diluted manure these variables for the influent with the digester at 35.8'c 'orc: Fast Fraction: 35% with K = 2.19 (1"1 Moderate Fraction: 255 with K = 0.168 d"1 516' Fraction: 41s with r . 0.0075 6“ The temperature coefficient was estimated as 1.25 corresponding to an Arrhenius activation energy of 42.5 kcal/deg Kelvin. The precision of the model for predicting the timing Of gas produc- tion can be improved by directly incorporating changes in the volatile acid pool which can be significant for some phase rela- tions between feeding and heating cycles. Gas storage requirements can be substantially reduced by managing the feeding and temperature cycles. For a hypothetical situation in which gas is productively utilized eight hours a day. gas sto- 129 rage requirements can be reduced from 50* of daily gas production for a constant temperature. uniform feed digester to 24% by feeding at the midpoint of an ascending temperature ramp. X. SUGGESTIONS FOR FUTURE WORK Based on the results of this investigation. the following ideas are suggested for future research: 1. To verify that the results of the fluctuating temperature experi- ments are applicable to full strength manure. the conditions of Experiments IIA and IIC should be repeated with whole manure. The combination of pulse feeding and temperature fluctuations may not result in stable operation with other types Of substrate and the distribution of substrate fractions is likely to vary with type of substrate. Therefore. similar experiments should be conducted with a variety of waste materials. The dynamics of volatile acid utilization need to be incorporated directly into the model. TO do this. kinetic data for at least acetic acid must bbe Obtained under similar Operating conditions of pulse feeding and fluctuating temperature in a stable digester. A theoretical study should be made with the mathematical model to determine the range of gas production patterns which would be predicted under various management strategies. Then does which would be most useful and those which would most severely test the model could be studied experimentally for further verification. 130 APPENDICES APPENDIX A 131 TABLE A1. Area Counts for Volatile Fatty Acids Standard Solution (Data for Figure 4-3). Area Counts mgll as COD 1 2 3 Average Std. Dev. Acetic Acid 392.91 41.127 44.817 51.622 45.855 5.324 785.82 97.189 96.180 101.995 98.455 3.107 1571.64 197.132 217.029 202.007 205.389 10.371 2357.45 330.095 325.286 -- 327.690 3.400 4714.90 649.235 653.615 -- 651.425 3.097 9429.80 1.299.219 1.278.542 -- 1.288.880 14.621 Propionic Acid 527.14 89.521 93.774 90.606 91.300 2.210 1054.28 178.838 172.134 175.748 175.573 3.355 2108.56 334.581 324.080 349.493 336.051 12.770 3162.83 523.466 527.944 -- 525.705 3.166 6325.67 1.043.196 1.040.493 --- 1.041.844 1.911 12651.34 2.080.229 2.010.550 -- 2.045.390 49.270 iso-Butyric Acid 172.47 25.684 27.573 27.133 26.797 988 344.93 56.899 54.418 56.995 56.104 1.461 689.87 107.984 -- 116.462 112.223 5.995 1034.80 175.860 177.527 -- 176.694 1.179 2069.61 359.316 353.453 -- 356.384 4.145 4139.22 727.149 696.297 -- 711.723 21.816 Butyric Acid 174.21 31.893 33.410 32.261 32.521 791 348.42 62.893 60.379 62.844 62.039 1,438 696.84 117.845 123.253 121.657 120.918 2.779 1045.25 184.531 185.930 --- 185,230 989 2090.51 365.645 364.299 -- 364.972 952 4181.01 731.939 708.388 -- 720.164 16.653 iso-Valeric Acid 94.05 17.842 18.216 16.750 17,603 762 188.10 34.156 33.288 33.554 33.666 445 376.20 64.448 69.895 67.767 67.370 2.745 564.30 101.290 101.829 -- 101.560 381 1128.60 203.934 202.889 -- 203.412 739 2257.21 407.712 400.414 -- 404.063 5.160 Valeric Acid 93.11 17.406 18.082 17.146 17.545 483 186.21 34.160. 32.464 33.990 33,538 934 372.44 61.174 64.528 66.512 64.071 2,698 558.64 98.466 99.410 --- 98.938 668 1117.29 196.532 195.756 -- 196.144 549 2234.57 396.016 380.126 --- 388.071 11.236 132 TABLE A2. Area Counts for Volatile Fatty Acids Standard Solution (Data for Figure 4-4). Area Counts mgll as COD 1 2 3 Average Std. Dev. Acetic Acid . 37.72 115.124 92.311 104.618 104.018 11.418 75.45 266.264 149.491 166.268 160.641 9.656 150.90 278.060 267.539 259.972 268.524 9.084 301.80 463.633 481.052 484.642 476.442 11.237 Propionic Acid 50.59 85.431 85.957 84.791 85.393 584 101.17 167.372 171.468 174.897 171.246 3.767 202.34 345.867 351.846 347.280 348.337 3.135 404.68 673.094 690.746 671.634 678.491 10.638 iso-Butyric Acid 28.29 50.682 49.020 54.686 51.463 2.913 56.59 99.294 98.496 99.596 99.129 568 113.18 206.588 213.284 202.122 207.331 5.618 226.36 397.008 399.782 388.316 395.035 5.982 Butyric Acid ' 28.29 62.964 51.458 51.568 55.330 6.611 56.59 101.600 97.686 102.036 100.441 2.396 113.18 191.304 195.470 194.788 193.854 2.235 226.36 371.728 380.276 362.886 371.630 8.695 APPENDIX B 133 TABLE B1. Daily Gas Production recorded from Wet Test Meter Readings. l/d (Data for Figure 5-1). I.D. Daily Gas Production. l/d 1982 Digester 1 Digester 2 284 6.91 6.76 285 6.75 6.85 286 6.58 6.33 287 6.66 6.51 288 6.60 6.13 289 6.48 6.26 290 6.90 6.48 291 6.16 6.27 292 6.72 6.69 293 6.92 6.28 294 - - 295 7.20 7.14 296 6.74 6.86 297 - 6.40 298 - 6.48 299 6.07 6.44 300 - - 301 6.20 ~— 302 6.28 - 303 6.40 6.44 304 6.39 6.53 305 6.08 6.26 306 - - 307 6.12 6.21 308 5.85 6.06 309 6.02 6.42 310 6.10 6.61 311 6.09 6.48 312 6.06 6.36 313 6.41 6.63 Mean 6.43 6.48 Std. Dev. 0.36 0.25 n 25 25 134 TABLE B2. Daily Gas Production Recorded by Wet Test Meters. l/d (Data for Figures 5-7 and 5-8). CONTROL EXP IIA EXP IIB EXP IIC JD Gas Prod JD Gas Prod JD Gas Prod JD Gas Prod 85 6.59 85 6.71 134 - 170 - 86 6.45 86 6.92 135 6.52 171 -- 87 6.63 87 6.66 136 6.72 172 7.32 88 6.82 88 7.19 137 6.91 173 7.35 89 6.62 89 7.03 138 7.31 174 6.82 90 6.55 90 6.83 139 7.19 175 7.16 91 6.37 91 6.18 140 7.60 176 7.04 92 6.78 92 6.59 141 7.54 177 7.16 93 6.32 93 6.47 142 7.48 178 6.76 94 6.57 94 6.10 143 7.28 179 6.70 95 6.45 95 6.67 144 7.43 180 6.90 96 6.97 96 6.55 145 7.02 181 7.12 97 - 97 6.68 146 7.86 182 6.77 98 6.70 98 6.84 147 6.81 183 7.07 99 6.63 99 6.74 148 7.12 184 9.15 100 6.73 100 6.92 149 6.94 185 8.93 101 - 101 6.78 150 6.70 186 8.16 104 6.67 104 7.09 151 7.05 187 7.76 105 6.52 105 6.78 152 6.83 188 8.68 106 6.49 106 6.88 153 6.74 189 7.39 107 6.61 107 7.16 154 6.97 190 7.43 108 6.84 108 7.29 155 7.15 191 7.68 109 6.65 109 7.05 156 7.07 192 7.54 110 6.74 110 7.22 157 7.10 193 7.72 111 6.53 111 7.18 158 7.76 194 7.81 112 6.94 112 7.16 159 8.04 195 7.85 113 6.88 113 7.00 160 7.05 196 9.57 114 7.22 114 7.14 161 6.96 197 8.96 115 7.19 115 7.00 162 7.09 198 8.96 116 7.32 116 7.34 163 7.08 199 9.25 117 7.20 117 7.16 164 6.92 200 9.90 118 6.94 118 7.20 165 7.00 201 8.83 119 6.83 119 6.94 166 7.14 202 8.70 120 6.57 120 7.38 167 7.02 203 9.70 121 6.88 121 6.73 168 7.30 204 9.42 122 6.78 122 7.49 169 7.20 205 9.42 123 6.48 123 7.02 206 - 124 6.50 124 6.48 207 7.86 125 6.42 125 7.56 208 8.72 126 -— 126 7.81 209 7.42 127 6.67 127 7.08 210 8.96 128 6.70 128 7.45 211 7.93 129 6.47 129 7.22 212 5.35 130 6.50 130 7.77 213 8.64 131 6.68 131 7.34 214 I 8.31 132 6.61 132 7.73 215 8.31 133 6.46 133 -- 135 TABLE B3. Mean Gas Production Data for Experiment 1. Gas Production, mllhr Time. No. of DIGESTER 1 DIGESTER 2 Hours Points Mean Std.Dev Mean Std.Dev 0.33 4. 185. 21. 244. 58. 0.67 4. 199. 45. 283. 54. 1.00 4. 266. 43. 299. 67. 1.33 4. 305. 4. 378. 24. 1.67 4. 318. 16. 392. 37. 2.00 4. 333. 34. 386. 19. 2.33 4. 318. 20. 414. 33. 2.67 4. 333. 26. 413. 23. 3.00 4. 334. 25. 410. 30. 3.33 4. 342. 30. 411. 37. 3.67 4. 345. 43. 405. 36. 4.00 4. 352. 53. 406. 37. 4.33 4. 354. 49. 381. 30. 4.67 5. 332. 26. 372. 39. 5.00 5. 322. 26. 365. 44. 5.33 5. 323. 24. 388. 28. 5.67 5. 318. -25. 386. 30. 6.00 5. 310. 19. 384. 27. 6.33 5. 302. 29. 368. 10. 6.67 5. 296. 24. 357. 12. 7.00 5. 294. 23. 356. 20. 7.33 5. 322. 50. 356. 29. 7.67 5. 323. 46. 354. 21. 8.00 5. 319. 48. 358. 21. 8.33 5. 320. 47. 336. 13. 8.67 5. 310. 39. 343. 18. 9.00 5. 315. 41. .348. 28. 9.33 5. 304. 45. 337. 20. 9.67 5. 308. 39. 352. 14. 10.00 5. 314. 49. 340. 12. 10.33 5. 302. 35. 338. 33. 10.67 5. 297. 38. 334. 12. 11.00 5. 298. - 47. 328. 23. 11.33 5. 285. 34. 325. 20. 11.67 5. 287. 39. 323. 26. 12.00 5. 296. 51. 321. 13. 12.33 5. 291. 51. 322. 11. 12.67 5. 291. 43. 320. 22. 13.00 5. 288. 51. 315. 19. 13.33 5. 278. 40. 307. 12. 13.67 5. 287. 42. 306. 15. 14.00 5. 287. 55. 300. 26. 14.33 5. 270. 45. 289. 24. 14.67 5. 266. 34. 297. 33. 15.00 5. 271. 42. 289. 37. 15.33 5. 273. 38. 303. 31. 15.67 5. 266. 34. 292. 26. 16.00 5. 260. 42. 291. 32. 136 TABLE 33 Cont. Gas Production. mllhr Time. No. of DIGESTER 1 DIGESTER 2 Hours Points Mean Std.Dev Mean Std.Dev 16.33 5. 266. 38. 294. 24. 16.67 5. 257. 32. 296. 37. 17.00 5. 264. 44. 288. 31. 17.33 5. 263. 45. 291. 30. 17.67 5. 246. 42. 287. 20. 18.00 5. 244. 46. 272. 29. 18.33 5. 247. 37. 281. 21. 18.67 5. 245. 40. 263. 26. 19.00 5. 244. 40. 280. 42. 19.33 5. 249. 42. 270. 38. 19.67 5. 238. 36. 275. 25. 20.00 5. 242. 48. 267. 23. 20.33 5. 242. 48. 263. 21. 20.67 5. 237. 43. 275. 24. 21.00 5. 233. 38. 269. 28. 21.33 5. 232. 44. 274. 26. 21.67 5. 230. 40. 270. 33. 22.00 5. 227. 38. 257. 34. 22.33 5. 221. 31. 256. 25. 22.67 5. 219. 33. 245. 26. 23.00 5. 197. 13. 252. 27. 23.33 5. 193. 12. 259. 21. 23.67 5. 203. 21. 256. 24. 24.00 5. 192. 19. 251. 23. Daily Total = 6680. Daily Total = 7672. 137 TABLE B4. Mean Gas Production Data for Experiment II. Control. Time. No. of Gas Prod. mllhr Temperature. 'C Hours Points Mean Std.Dev Mean Std.Dev 0.33 6. 133. 57. 35.76 0.09 0.67 6. 215. 57. 35.80 0.06 1.00 6. 353. 18. 35.81 0.06 1.33 6. 423. 59. 35.81 0.07 1.67 6. 398. 21. 35.80 0.07 2.00 6. 400. 23. 35.81 0.08 2.33 6. 368. 23. 35.81 0.08 2.67 6. 356. 14. 35.81 0.08 3.00 6. 366. 11. 35.81 0.08 3.33 6. 362. 37. 35.81 0.08 3.67 6. 377. 20. 35.81 0.08 4.00 6. 362. 18. 35.81 0.07 4.33 6. 354. 13. 35.81 0.07 4.67 6. 356. 21. 35.81 0.07 5.00 6. 350. 7. 35.80 0.08 5.33 6. 366. 29. 35.80 0.08 5.67 6. 321. 31. 35.80 0.08 - 6.00 6. 321. 11. 35.80 0.08 6.33 6. 326. 17. 35.80 0.07 6.67 6. 332. 14. 35.80 0.07 7.00 6. 318. 17. 35.80 0.08 7.33 6. 307. 21. 35.80 0.08 7.67 6. 296. 13. 35.80 0.08 8.00 6. 302. 21. 35.79 0.08 8.33 6. 297. 25. 35.79 0.09 8.67 6. 307. 12. 35.79 0.08 9.00 6. 296. 16. 35.79 0.09 9.33 6. 320. 34. 35.79 0.09 9.67 6. 301. 14. 35.78 0.09 10.00 6. 293. 16. 35.78 0.08 10.33 6. 297. 20. 35.78 0.08 10.67 6. 289. 9. 35.79 0.08 11.00 6. 301. 15. 35.79 0.08 ~11.33 6. 281. 17. 35.78 0.08 11.67 6. 279. 13. 35.78 0.08 12.00 6. 283. 45. 35.78 0.08 12.33 6. 257. 36. 35.78 0.08 12.67 6. 260. 11. 35.78 0.07 13.00 6. 255. 9. 35.78 0.08 13.33 6. 265. 12. 35.78 0.08 13.67 6. 260. 17. 35.78 0.08 14.00 6. 245. 20. 35.78 0.08 14.33 6. 236. 14. 35.79 0.08 14.67 6. 231. 19. 35.79 0.08 15.00 6. 228. 15. 35.80 0.07 15.33 6. 195. 23. 35.81 0.06 15.67 6. 173. 29. 35.82 0.06 16.00 6. 182. 20. 35.83 0.07 138 TABLE B4 Cont. Time. NO. Of Gas Prod. mllhr TVmperature. 'C Bours Points Mean Std.Dev Mean Std.Dev 16.33 6. 179. 21. 35.83 0.06 16.67 6. 166. 18. 35.82 0.06 17.00 6. 172. 21. 35.82 0.06 17.33 6. 168. 26. 35.81 0.05 17.67 6. 154. 5. 35.80 0.05 18.00 6. 152. 21. 35.80 0.05 18.33 6. 178. 57. 35.80 0.05 18.67 6. 153. 44. 35.80 0.06 19.00 6. 149. 12. 35.80 0.05 19.33 6. 155. 20. 35.79 0.05 19.67 6. 161. 21. 35.80 0.06 20.00 6. 166. 22. 35.80 0.06 20.33 6. 161. 15. 35.80 0.05 20.67 6. 154. 13. 35.80 0.04 21.00 6. 152. 13. 35.80 0.04 21.33 6. 146. 18. 35.80 0.04 21.67 6. 146. 8. 35.80 0.04 22.00 6. 149. 11. 35.80 0.03 22.33 5. 142. 14. 35.79 0.03 22.67 5. 143. 14. 35.79 0.04 23.00 5. 142. 16. 35.79 0.05 23.33 5. 191. 67. 35.79 0.06 23.67 6. 119. 44. 35.79 0.04 24.00 6. 178. 45. 35.80 0.03 Daily TOtal - 6053. Ave. Temp. - 35.80 139 TABLE BS. Mean Gas Production Data for Experiment IIA. Time. No. of Gas Prod.. mllhr Tbmperature. 'C Rours Points Mean Std. Dev Mean Std. Dev 0.33 6. 355. 18. 35.30 0.30 0.67 6. 510. 41. 35.65 0.31 1.00 6. 581. 35. 36.13 0.33 1.33 6. 586. 44. 36.18 0.33 1.67 6. 610. 37. 36.21 0.31 2.00 6. 660. 50. 36.68 0.31 2.33 6. 657. 80. 36.74 0.28 2.67 6. 615. 44. 36.72 0.30 3.00 6. 654. 51. 37.07 0.29 3.33 6. 636. 44. 37.32 0.31 3.67 6. 619. 46. 37.35 0.32 4.00 6. 697. 85. 37.56 0.30 4.33 6. 641. 56. 37.95 0.31 4.67 6. 605. 46. 37.97 0.31 5.00 6. 626. 95. 38.04 0.30 5.33 6. 614. 81. 38.52 0.31 5.67 6. 550. 42. 38.60 0.30 6.00 6. 571. 53. 38.62 0.30 6.33 6. 587. 60. 39.01 0.37 6.67 6. 555. 33. 39.12 0.46 7.00 6. 523. 52. 39.12 0.48 7.33 6. 438. 46. 38.83 0.50 7.67 6. 447. 28. 38.52 0.47 8.00 6. 428. 36. 38.50 0.47 8.33 6. 362. 37. 38.34 0.48 8.67 6. 353. 41. 37.92 0.48 9.00 6. 390. 84. 37.87 0.48 9.33 6. 313. 52. 37.83 0.49 9.67 6. 248. 49. 37.36 0.49 10.00 6. 257. 34. 37.26 0.48 10.33 6. 255. 38. 37.25 0.48 10.67 6. 193. 50. 36.88 0.50 11.00 6. 206. 33. 36.66 0.49 11.33 6. 195. 27. 36.64 0.49 11.67 6. 145. 26. 36.41 0.49 12.00 6. 180. 46. 36.07 0.48 12.33 6. 215. 104. 36.04 0.47 12.67 6. 137. 13. 35.93 0.47 13.00 6. 137. 12. 35.50 0.47 13.33 6. 152. 15. 35.44 0.47 13.67 6. 131. 20. 35.42 0.47 14.00 6. 112. 14. 34.99 0.48 14.33 6. 133. 10. 34.87 0.47 14.67 6. 130. 22. 34.85 0.48 15.00 6. 86. 16. 34.56 0.49 15.33 6. 90. 16. 34.29 0.47 15.67 6. 104. 17. 34.28 0.48 16.00 6. 71. 17. 34.09 0.48 140 TABLE BS Cont. Time. No. Of Gas Prod.. mllhr Temperature. °C Hours Points Mean Std. Dev Mean Std. Dev 16.33 6. 91. 16. 33.71 0.49 16.67 6. 106. 9. 33.67 0.47 17.00 6. 85. 7. 33.61 0.48 17.33 6. 74. 12. 33.18 0.49 17.67 6. 94. 6. 33.10 0.49 18.00 6. 98. 18. 33.09 0.49 18.33 6. 80. 39. 32.81 0.36 18.67 6. 100. 47. 32.63 0.36 19.00 6. 107. 33. 32.61 0.36 19.33 6. 171. 50. 32.88 0.34 19.67 6. 134. 35. 33.15 0.36 20.00 6. 136. 38. 33.17 0.36 20.33 6. 172. 40. 33.32 0.35 20.67 6. 164. 33. 33.71 0.36 21.00 6. 154. 35. 33.76 0.36 21.33 6. 177. 37. 33.80 0.35 21.67 6. 194. 38. 34.28 0.34 22.00 6. 169. 43. 34.35 0.36 22.33 6. 167. 46. 34.35 0.34 22.67 6. 220. 39. 34.75 0.34 23.00 6. 188. 39. 34.93 0.35 23.33 6. 190. 47. 34.93 0.36 23.67 6. 207. 55. 35.20 0.35 24.00 6. 232. 50. 35.51 0.36 Daily Total I 7287. Ave. Temp.- 35.82 141 TABLE B6. Mean Gas Production Data for Experiment IIB. Time. No. of Gas Prod.. mllhr Temerature. 'C Eours Points Mean Std.Dev Mean Std.Dev 0.33 5. 377. 81. 32.50 0.26 0.67 5. 463. 27. 32.55 0.26 1.00 5. 556. 31. 32.60 0.23 1.33 6. 583. 36. 32.87 0.24 1.67 6. 553. ‘ 54. 33.15 0.24 2.00 6. 528. 48. 33.17 0.25 2.33 6. 555. 67. 33.31 0.26 2.67 6. ~521. 76. 33.72 0.27 3.00 6. 499. 62. 33.76 0.26 3.33 6. 500. 67. 33.79 0.24 3.67 6. 510. 62. 34.28 0.25 4.00 6. 476. 54. 34.36 0.24 4.33 6. 459. 45. 34.36 0.24 4.67 6. 512. 60. 34.74 0.27 5.00 6. 458. 38. 34.92 0.19 5.33 6. 441. 28. 34.95 0.23 5.67 6. 463. 35. 35.18 0.23 6.00 6. 431. 20. 35.50 0.24 6.33 6. 403. 15. 35.49 0.19 6.67 6. 411. 21. 35.57 0.16 7.00 6. 399. 21. 36.01 0.16 7.33 6. 366. 13. 36.05 0.15 7.67 6. 366. 25. 36.07 0.15 8.00 6. 374. 24. 36.56 0.18 8.33 6. 337. 21. 36.69 0.24 8.67 6. 308. 16. 36.70 0.24 9.00 6. 328. 24. 37.06 0.27 9.33 6. 306. 20. 37.32 0.27 9.67 6. 281. 22. 37.33 0.26 10.00 6. 289. 12. 37.52 0.27 10.33 6. 292. 12. 37.92 0.26 10.67 6. 283. 20. 37.95 0.27 11.00 6. 276. 21. 38.01 0.27 11.33 6. 294. 20. 38.52 0.28' 11.67 6. 277. 28. 38.57 0.27 12.00 6. 263. 21. 38.58 0.28 12.33 6. 291. 39. 38.89 0.10 12.677 6. 279. 13. 39.00 0.23 13.00 6. 276. 22. 39.01 0.24 13.33 6. 238. 10. 38.72 0.25 13.67 6. 235. 19. 38.40 0.24 14.00 6. 233. 15. 38.37 0.24 14.33 6. 212. 16. 38.24 0.25 14.67 6. 205. 16. 37.79 0.24 15.00 6. 198. 13. 37.76 0.25 15.33 6. 196. 10. 37.73 0.25 15.67 6. 177. 17. 37.24 0.24 16.00 6. 187. 18. 37.14 0.24 142 TABLE B6 Cont. Time. No. of Gas Prod.. mllhr Temerature, 'C Bours Points Mean Std.Dev Mean Std.Dev 16.33 6. 183. 7. 37.13 0.24 16.67 6. 159. 12. 36.75 0.23 17:00 6. 175. 10. 36.52 0.23 17.33 6. 185. 11. 36.51 0.22 17.67 6. 155. 15. 36.29 0.20 18.00 6. 162. 12. 36.02 0.02 18.33 6. 163. 16. 36.02 0.03 18.67 6. 153. 15. 35.93 0.02 19.00 6. 156. 10. 35.49 0.02 19.33 6. 159. 14. 35.43 0.03 19.67 6. 158. 10. 35.42 0.03 20.00 6. 152. 8. 34.97 0.02 20.33 6. 150. 8. 34.86 0.03 20.67 6. 155. 5. 34.85 0.02 21.00 6. 147. 9. 34.55 0.02 21.33 6. 152. 9. 34.28 0.02 21.67 6. 143. 7. 34.24 0.02 22.00 6. 143. 9. 34.07 0.00 22.33 6. 150. 9. 33.68 0.00 22.67 6. 147. 8. 33.65 0.02 23.00 6. 134. 4. 33.60 0.02 23.33 6. 136. 9. 33.12 0.05 23.67 6. 144. 11. 32.98 0.22 24.00 6. 201. 34. 32.96 0.24 Daily Total - 7075. Ave. Temp.= 35.77 143 TABLE B7. Mean Gas Production Data for Experiment IIC. Time. No. of Gas Prod. mllhr Temperature. “C Bours Points Mean Std.Dev Mean Std.Dev 0.33 6. 568. 123.’ 39.26 0.07 0.67 6. 824. 69. 39.54 0.03 1.00 6. 860. 30. 39.57 0.02 1.33 6. 772. 23. 39.25 0.03 1.67 6. 779. 22. 38.97 0.04 2.00 6. 770. 10. 38.95 0.04 2.33 6. 645. 23. 38.80 0.03 2.67 6. 586. 62. 38.37 0.04 3.00 6. 519. 23. 38.34 0.04 3.33 6. 511. 32. 38.30 0.04 3.67 6. 460. 43. 37.80 0.04 4.00 6. 456. 38. 37.73 0.04 4.33 6. 482. 60. 37.72 0.03 4.67 6. 420. 19. 37.30 0.04 5.00 6. 410. 27. 37.11 0.03 5.33 6. 417. 43. 37.09 0.03 5.67 6. 381. 35. 36.84 0.03 6.00 6. 379. 23. 36.50 0.02 6.33 6. 373. 45. 36.47 0.03 6.67 6. 360. 45. 36.36 0.02 7.00 6. 332. 52. 35.93 0.02 7.33 6. 307. 30. 35.88 0.03 7.67 6. 294. 43. 35.86 0.02 8.00 6. 279. 45. 35.40 0.02 8.33 6. 264. 38. 35.29 0.00 8.67 6. 254. 36. 35.28 0.02 9.00 6. 225. 56. 34.95 0.04 9.33 6. 209. 42. 34.71 0.03 9.67 6. 216. 36. 34.68 0.03 10.00 6. 174. 16. 34.49 0.03 10.33 6. 196. 39. 34.12 0.02 10.67 6. 193. 16. 34.09 0.02 11.00 6. 176. 21. 34.03 0.02 11.33 6. 159. 8. 33.59 0.04 11.67 6. 173. 12. 33.51 0.04 12.00 6. 177. 30. 33.50 0.04 12.33 6. 132. 5. 33.08 0.06 12.67 6. 159. 22. 32.93 ' 0.04 13.00 6. 180. 36. 32.92 0.04 13.33 6. 238. 30. 33.23 0.03 13.67 6. 178. 11. 33.49 0.04 14.00 6. 185. 30. 33.49 0.04 14.33 6. 224. 14. 33.66 0.03 14.67 6. 189. 22. 34.06 0.04 15.00 6. 183. 8. 34.07 0.04 15.33 6. 215. 22. 34.12 0.04 15.67 6. 212. 26. 34.62 0.04 16.00 6. 207. 22. 34.67 0.03 144 TABLE B7 Cont. Time. No. Of Gas Prod. ml/hr Temperature. 'C Hours Points Mean Std.Dev Mean Std.Dev 16.33 6. 178. 18. 34.68 0.03 16.67 6. 242. 28. 35.12 0.05 17.00 6. 194. 18. 35.27 0.05 17.33 6. 199. 19. 35.27 0.05 17.67 6. 249. 22. 35.55 0.04 18.00 6. 187. 7. 35.85 0.03 18.33 6. 207. 18. 35.86 0.03 18.67 6. 273. 40. 35.99 0.03 19.00 6. 222. 11. 36.43 0.03 19.33 6. 212. 22. 36.45 0.03 19.67 6. 239. 22. 36.47 0.03 20.00 6. 256. 21. 36.95 0.02 20.33 6. 215. 15. 37.04 0.02 20.67 6. 212. 11. 37.05 0.02 21.00 6. 276. 16. 37.44 0.02 21.33 6. 245. 11. 37.67 0.02 21.67 6. 232. 21. 37.67 0.03 22.00 6. 294. 19. 37.90 0.03 22.33 6. 250. 24. 38.26 0.03 22.67 6. 238. 21. 38.28 0.04 23.00 6. 268. 16. 38.38 0.03 23.33 6. 264. 21. 38.88 0.02 23.67 6. 242. 28. 38.92 0.02 24.00 5. 280. 60. 38.93 0.03 Daily Total 8 7523. Ave. Temp. - 36.20 145 TABLE BS. TOtal Volatile Solids During the Stable Period of Experiment I (in percent). Influent Effluent ID Sample 1 Sample 2 Digester 1 Digester 2 307 13.84 13.46 9.54 9.25 309 13.72 13.85 8.79 8.63 310 12.93 14.19 8.85 9.56 311 14.44 14.30 9.24 8.29 312 14.27 13.80 8.92 8.63 313 - 13.99 8.69 8.71 314 13.73 13.80 8.66 9.31 316 13.69 13.39 9.35 8.80 317 14.48 13.43 8.37 8.21 318 13.53 - 8.45 8.63 Mean 8.89 8.80 Std. Dev. 0.38 0.44 S TVS Reduction 35.5 36.1 146 TABLE B9. Tbtal Volatile Solids Data During Stable Period of Experiment II (in percent). JD Influent Effluent 1983 Control Exp IIA Exp IIB Exp IIC 128 3.42 1.95 1.69 - - 3.38 1.93 1.72 - -- 130 3.70 2.02 1.70 - - 3.78 1.97 1.65 - - 132 3.40 2.10 - 1.67 - -- 3.66 2.04 1.63 - - 134 3.49 2.03 1.69 - - 3.26 2.07 1.70 - - 135 3.16 1.90 1.65 -- - 3.13 1.88 1.60 -- -- 164 3.50 - - 1.75 ~- 3.39 -- - 1.91 - 166 3.64 - - 1.73 - 3.46 - - 1.69 -- 167 3.57 - -- 1.76 - 3.72 - - 1.64 -- 168 3.24 - -- 1.81 - 3.21 - -- 1.89 - 191 3.49 - - -- 1.66 ' 3.47 - -- - 1.75 193 3.37 - - - 1.72 3.53 - - - 1.68 211 3.60 - -- - - 3.35 - - -- - 214 3.57 - -- - - 3.32 - - - - 216 3.24 - -- - - 3.34 - - -- - Mean 3.44 1.99 1.67 1.77 1.70 Std. Dev. 0.17 0.07 0.04 0.09 0.04 Note: For each sample, two replicates were analyzed. 147 TABLE B10. Tbtal COD Data During Stable Period of Experiment I (in mgll). JD Influent Effluent 1983 Sample 1 Sample 2 Digester 1 Digester 2 316 182.528 180.544 98.208 107.136 317 162.032 -- 106.704 104.728 318 158.080 -- 98.800 104.728 319 185.368 -- 112.404 114.376 -- -- 116.348 114.376 320 169.936 162.032 88.130 88.999 Mean 169.996 103.432 105.724~ Std. Dev. 11.508 10.401 9.310 S COD Reductions -- 39.16 37.81 148 TABLE B11. TOtal COD Data During the Stable Period of Experiment II (in mgll). JD Influent Effluent 1983 Control Exp IIA Exp IIB Exp IIC 128 38.801 21.164 17.637 - 40.917 - 19.400 -- 130 38.720 20.064 16.896 -- -- 21.120 15.840 -- 131 - 22.880 17.600 -- -- 24.640 15.840 -- -- 19.972 - -- 132 38.456 20.102 19.228 -- 41.952 - 15.732 -- 134 37.393 19.131 16.522 -- -- 19.131 17.392 -- 166 37.374 - -- 17.892 - - -- 19.880 168 38.093 -- -- 17.062 46.029 - -- - 38.093 - - - 169 39,680 - - 17.856 37.299 - - 17.856 192 38.464 -- -- -- 17.308 - -- - -- 16.155 194 38.417 -- -- - 21.073 - -- - -- 17.244 212 38.417 -- - -- - 213 39.856 - - -- -- 215 37,600 - - - - 37.600 - - -- -- Mean 38.883 20.192 17.209 18.490 17.945 Std. Dev. 2.220 1,819 1,323 1,413 2.151 S Red. -' 46.22 55.74 52.45 53.85 149 TABLE B12. Individual Volatile Fatty Acid Concentrations During the Stable Period for Digester 1. Experiment I. in mgll as COD (Data for Figure 5-3). Time JD Tbtal 1982 EAc HP iEB EB iHV EV EC VFA 11 AM 323 _ 182 1970 331 369 21 0 O 324 234 2182 85 0 108 6 0 325 194 1747 32 0 72 148 0 Ave. 203 1966 149 123 67 51 0 2559 1 PM 323 1150 2077 200 247 197 0 8 324 642 1833 62 140 109 20 6 Ave. 896 1955 131 194 153 10 7 3346 4 PM 323 1096 2220 326 319 222 47 O 324 850 1960 72 76 92 54 6 Ave. 973 2090 199 198 157 50 3 3670 11 PM 323 616 2108 109 0 192 0 32 324 669 2082 2 19 94 25 0 Ave. 642 2095 56 10 143 12 16 2974 6 AM 324 255 2093 93 0 43 0 1 325 327 2092 8 10 0 4 0 Ave. 291 2092 50 5 22 2 0 2462 TABLE 813. 150 in mg/l as COD (Data for Figure 5-3). Individual Volatile Fatty Acid Concentrations During the Stable Period for Digester 2. Experiment I, Time JD Tbtal 1984 RAc RP iEB EB iEV EV EC VFA 11 AM 323 256 2223 339 0 57 0 0 324 104 1504 16 276 289 134 0 325 165 1507 621 235 0 14 12 Average 175 1745 325 170 115 49 4 2583 1 PM 323 525 1894 233 516 158 95 20 324 727 1490 203 216 67 40 37 Average 626 1692 218 366 112 68 28 3110 4 PM 323 645 1774 234 435 163 55 3 324 828 1550 47 74 111 0 0 Average 736 1662 140 254 137 28 2 2959 11 PM 323 733 1810 76 0 200 0 5 324 573 1703 0 19 48 17 0 Average 653 1756 38 10 124 8 2 2591 6 AM 324 280 1688 53 137 38 0 2 325 312 1514 0 8 0 6 0 Average 296 1601 26 72 19 3 l 2018 151 TABLE 814. Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment II Control. in mg/l as COD (Data for Figure 5-14). Time Individual VFA Tbtal JD EAc RP iEB BB VFA 2:00 PM 129 50 3 0 0 57 130 28 4 0 2 34 131 32 5 0 8 45 132 22 5 0 2 29 133 25 6 0 4 35 Average 31.4 4.6 0 4 40 4:00 PM 129 206 49 3 4 262 131 186 44 4 20 254 133 172 32 5 27 236 Average 188 41.7 4 17 250.7 5:00 PM 130 203 44 5 6 258 6:00 PM 129 198 50 4 10 262 131 184 32 2 18 236 Average 191 41 3 14 249 8:00 PM '129 188 42 2 6 238 130 186 36 1 1 224 131 186 27 2 24 239 133 185 24 2 29 240 Average 186.2 32.2 1.75 15 235.2 12:00 PM 129 127 25 0 6 158 130 138 21 0 0 159 Average 132.5 23 0 3 158.5 1:30 PM 132 103 12 0 9 124 3:00 PM 132 80 12 0 4 96 9:00 AM 130 22 5 0 3 30 131 31 7 0 0 38 Average 26.5 6 0 1.5 34 TABLE 815. 152 Individual Volatile Fatty Acid Concentrations During the Stable Period of Experiment IIA, in mgll as COD (Data for Figure 5-15). Time Individual VFA Total JD EAc RP iEB EB VFA 2:00 PM 129 10 4 0 0 14 130 20 4 0 0 24 131 0 3 0 5 8 132 12 3 0 4 19 133 8 4 0 2 14 Average 10 3.6 0 2.2 15.8 4:00 PM 129 162 45 2 3 212 131 151 42 3 20 216 133 149 32 3 29 213 Average 154 39.7 2.7 17.3 213.7 5:00 PM 130 148 46 2 18 214 6:00 PM 129 135 42 3 3 183 131 131 35 2 22 190 Average 133 38.5 2.5 12.5 186.5 8:00 PM 129 103 35 0 2 140 130 117 32 1 5 155 131 123 22 0 16 161 133 104 16 0 22 142 Average 111.8 26.2 0.2 11.2 149.5 12:00 PM 129 15 4 0 4 23 130 23 4 0 3 30 Average 19 4 0 3.5 26.5 1:30 AM 132 10 3 0 3 16 3:00 AM 132 10 3 0 2 16 9:00 AM 130 0 2 1 6 9 131 0 4 0 O 4 Average 0 3 0.5 3 6.5 153 TABLE B16. Individual Volatile Fatty Acid Concentrations During the Stable Period Of Experiment IIB. in mg/l as COD (Data for Figure 5-16). Individual VFA Total Time ID EAc RP iEB+EB VFA 2:00 p.m. 165 54.0 3.0 4.0 61.0 166 51.0 4.0 4.0 59.0 Average 52.5 3.5 4.0 60.0 4:00 p.m. 165 75.0 24.0 7.0 106.0 166 76.0 25.0 4.0 105.0 Average 75.5 24.5 5.5 105.5 6:00 p.m. 165 89.0 25.0 6.0 120.0 166 75.0 26.0 6.0 107.0 Average 82.0 25.5 6.0 113.5 8:00 p.m. 165 54.0 21.0 4.0 79.0 166 50.0 20.0 4.0 74.0 Average 52.0 20.5 4.0 76.5 10:00 p.m. 165 44.0 12.0 3.0 59.0 166 40.0 12.0 3.0 55.0 Average 42.0 12.0 3.0 57.0 12:00 p.m. 165 55.0 3.0 4.0 62.0 166 58.0 _3.0 4.0 65.0 Average 56.5 3.0 4.0 63.5 9:00 a.m. 165 34.0 0.0 3.0 40.0 166 38.0 0.0 2.0 40.0 Average 36.0 0.0 2.5 40.0 154 TABLE B17. Individual Volatile Fatty Acid Concentrations During the Stable Period Of Experiment IIC, in mg/l as COD (Data for Figure 5-17). Individual VFA Total Time ID BAc HP iEB+BB VFA 2:30 p.m. 214 18.0 38.0 tr 56.0. 215 13.0 14.0 tr 27.0 Average 15.5 26.0 tr 41.5 4:00 p.m. 214 115.0 13.0 8.0 136.0 215 135.0 28.0 12.0 175.0 Average 125.0 20.5 10.0 155.5 7:00 p.m. 215 41.0 25.0 tr 66.0 216 56.0 21.0 tr 77.0 Average 48.5 23.0 tr ' 71.5 10:00 p.m. 214 12.0 9.0 tr 21.0 215 18.0 12.0 tr 30.0 Average 15.0 10.5 tr 25.5 155 TABLE B18. Daily Gas Production for Extended Digester Operation without Feeding (wet test meter results. data for Figures 5-6 and 5-20). Exp I Exp IIC Days Dig. 1 Dig. 2 1 6.44 6.70 7.52 2 3.30 3.12 3.86 3 2.35 2.22 2.31 4 2.08 1.96 1.98 5 1.60 1.52 1.69 6 1.43 1.47 1.51 7 1.10 1.09 1.31 8 0.92 0.98 - 9 0.71 0.80 0.97 10 - - 1.08 11 - r- 0.92 12 0.61 0.59 - 13 - -- 0.74 16 - - 0.66 19 0.55 -- - 19.5 - r- 0.55 20 0.52 0.48 - 24.5 - - 0.54 26 0.50 0.46 - 27 - - 0.50 30 r- -- 0.50 37 0.50 0.48 APPENDIX C 156 TABLE C1. Data for Estimation of Rate Constants and Initial Gas Potentials for the Slow and Moderate Fractions for Experiment I (wet test meter results. data for Figures 6-1 and 6-3a). Time, Overall Rate (Rt). l/d Calculated‘ R, - Rt-R1. l/d Days Dig. 1 Dig. 2 8,. 1/d Dig. 1 Dig. 2 0 - - 0.64 - -- 0.5 6.44 6.70 0.64 5.80 6.06 1.5 3.30 3.12 0.63 2.67 2.49 2.5 2.35 2.22 0.63 1.72 1.59 3.5 2.08 1.96 0.62 1.46 1.34 4.5 1.60 1.52 0.62 0.98 0.90 5.5 1.43 1.47 0.61 0.82 0.86 6.5 1.10 1.09 0.61 0.49 0.48 7.5 0.92 0.98 0.60 0.32 0.38 8.5 0.71 0.80 0.60 0.11 0.20 11.5 0.61 0.59 0.58 0.03 0.01 18.5 0.55 v- 0.55 - 'r 19.5 0.52 0.48 0.54 -- -- 25.5 0.50 0.46 0.52 -- -- 36.5 0.50 0.48 0.47 - '- 47.5 0.32 -- 0.43 - -- ‘ Calculated 8, - K,G:exp(-K,t) where K, and G: are Obtained from Figure 6-1. 157 TABLE C2. Data for Estimation of Rate Constants and Initial Gas Potentials for the Fast Fraction for Exp I (bubble count results normalized to wet test meter basis. data for Figure 6-2b). Time. NOrmalized R . l/d Calculated R, - Rt-R1-R, Days Dis. 1 DIs- 2 3.. lid 3,. 1/d Dig. 1 01;. 2 0 4.39 5.73 0.64 4.25 - -- 0.1 7.39 8.34 0.64 4.11 - 3.59 1.15 7.79 8.03 0.64 4.05 3.10 3.34 0.2 7.06 7.73 0.64 3.93 - 3.11 0.3 7.06 7.22 0.64 3.85 2.57 2.73 0.4 6.66 6.85 0.64 3.72 2.30 3.49 0.5 6.42 6.49 0.64 3.60 2.18 2.25 0.6 5.96 5.90 0.64 3.48 1.84 1.78 0.7 5.60 5.86 0.64 3.37 1.60 1.86 0.8 5.29 5.49 0.64 3.25 1.40 1.60 0.9 4.92 5.19 0.64 3.15 - 1.41 1.0 4.04 4.96 0.63 3.04 0.36 1.28 1.5 3.30 3.12 0.63 2.57 0.09 -- Notes: 1. Normalized Rate - (Bubble count. l/hr)x(24 hr/d)x(Factor) where Factor - wet test meter ave. rate/bubble count ave. rate - 6.14/6.68 for Dig. 1 and 6.50/7.67 for Dig. 2 2. Calculated Ri - KiGzexp(-Kit) where K, and G; are Obtained from Figures 6-1 and 6-3. 158 TABLE C3. Data for Estimation of Rate Constants and Initial Gas Potentials for the Slow and Moderate Fractions for Exp II (fluctuating temperature results: Ave. Tbmp. I 36.25'C; wet test meter basis; data for Figures 6-2 and 6-4a). Time. Overall Rate Calculated. Days (Rt), l/d R,, l/d R, s Rt-R,. 1/d 0.5 7.96 0.65 7.31 1.5 3.87 0.64 3.23 2.5 2.31 0.64 1.67 3.5 1.98 0.63 1.35 4.5 1.69 0.63 1.06 5.5 1.51 0.62 0.89 6.5 1.31 0.62 0.69 8.5 0.97 0.60 0.37 9.5 1.08 0.60 0.48 10.5 0.92 0.59 0.33 12.5 0.74 0.58 0.16 15.5 0.66 0.57 0.09 19.0 0.55 - - 24.0 0.54 -- -- 26.5 0.50 - -- 29.5 0.50 - -- ‘ Calculated 8, - K,G;exp(K,t) where K, and G: are Obtained from Figure 6-2. 159 TABLE C4. Data for Estimation of Rate Constants and Initial Gas Potentials for the Fast Fraction for Exp II (bubble count results from Control digester normal- ized to wet test meter basis: data for Figure 6-4b). Time. Normalized Rate Calculated D.y‘ Rte lld ' [15 lld R’s lld R3 3 Rt-k—RI 0 3.47 0.54 2.23 0.70 0.05 11.00 0.54 2.22 8.25 0.1 9.75 0.54 2.20 7.02 0.2 9.25 0.54 2.16 6.55 0.3 8.08 0.54 2.12 5.42 0.4 7.82 0.54 2.09 5.20 0.5 7.15 0.53 2.05 4.56 0.6 6.15 0.53 2.02 3.60 0.7 4.50 0.53 1.99 1.98 0.8 3.91 0.53 1.95 1.42 0.9 3.81 0.53 1.92 1.36 1.0 3.40 0.53 1.89 0.98 Notes: 1. Normalized Rate - (Bubble count. l/hr)x(24 hr/d)x(Factor) where Factor - wet test meter ave. rate/bubble count ave. rate - 6.57/6.05 2. Calculated R1 - KiGzoxp(Kit)9(35'8-36’25) where K, and G; are Obtained from Figures 6-2 and 6-4a; O 8 1.25. 160 TABLE C5. Fortran PrOgram for Comparison of Mathematical Model to 10 Experimental Data. PROGRAM‘VAEPLT REAL RP(72).TTME(72).T(72).GT(72) REAL K1,!2,K3 R130.0165 [230.236 K3=2.88 61347900. 02‘11900. 03‘2950. GlZERO‘47900. GZZERO‘II900. G32ERO'2950. TEETA1‘1.2 TEETH-1.2 DELT*0.0034722 TIME(1)'0.3333 TYPE 5 FORMAT(' TYPE PLOT OUTPUT FILE NAHE'a/I CALL ASSTGN(2.'TT:'.‘1.'NEI') TYPE 6 FORMAT(' TYPE TEMP. FILENAME'a/I CALL ASSIGN(1.'TT:',-1,'OLD') TYPE 7 FORMATI' TYPE OUTPUT FILE NAME'.’) CALL ASSIGN(3.'TT:',-1,'NE") SUM'0.33333 SUMA'0.0 SUMB'0.0 SUMC'0.0 READ(1.102) DO 10 K'l.72 READ(1.103)T(K) CONTINUE READ(1.100) EEAD(1.103)TR TYPE .,'TRP',TR IRITE(3 .1047TEETA1 .mETAZ 'MITE(3.105)K1.K2.K3 'RITE(3.106) OT‘J)‘0.0 DO 490 131.4 SUIA'SUMA*(K2.62‘(TEETA2“(T(J)'TE))‘DELTI OZ‘GZZERO'SUIA‘ - IF(GZ.LE.0.0)TYPE ’.'32 IS LESS TEEN ZERO' SUIB‘SUIB+(K1‘GI‘(TEETAI‘.(T(J)'TE))‘DELT’ Gl‘GlZERO'SUMB IF(GI.LE.0.0)TYPE .p'Gl IS LESS TEEN ZERO' SUMC'SUMC+(K3‘G3‘(THETA1"(T(J)-TR))‘DELT) G3‘GSZERO'SUMC 480 490 500 101 102 100 103 104 105 106 107 161 IF(G3.LE.0.0)TYPE ‘.'03 IS LESS TEEN ZERO' CONTINUE CONTINUE ercklacla(TRETA1"(T(J)-TR)I RPZsKZ‘GZ‘(THEIA2"(T(J)-TR)) RP3-K3‘G3‘(TBETA1“(T(J)-TR)) RP(J)-(RP1+RP2+RP3)/24. GT(J)=Gl+GZ+G3 IRITE(3.107)TIME(J).61.62.G3.GT(J).RP1.RP2.RP3.RP(J) SUM-SUI+0.3333 TIME(J+1)-SUM IRITE(2.101)TIME(J).RP(J) CONTINUE FORMAT('RD'.2615.7) FORMAT(/) FORMAT(' ') FORMAT(14X.F10.2) FORMAT!'TRETAl-‘.F6.4.2x.'THETA2-'.F6.4) FORMAT('k1-'.F7.5.2x.'K2-'.F7.5.zx.'K3-'.F7.5) FORMATU TIME ,. 61 02 G3 G‘m'r RPl +293 RTOT') FORMAT(F5.2.F8.0.F8.0.F8.0.F8.0.F8.0.F8.0.F8.0.F8.0) CALL CLOSE(1) CALL CLOSE(2) CALL CLOSE(3) STOP END APPENDIX D APPENDIX D THEORETICAL GAIN IN GAS PRODUCTION It was found in the analysis of the mathematical model that the higher gas production of the fluctuating temperature digester lies. in part. in the temperature dependence term based on the Arrhenius equa- tion. The higher the activation energy or the temperature fluctuation, the higher the gas production will be compared with a digester at con- stant tempersture. The demonstration. of this relationship will be presented as follows. Equation G-ll from Chapter 6 can be rewritten as I - Krerp [-1szr - run-r; (1H) s where TI: 8 TT,. Let I' - effective constant temperature rate that gives the ssme gas production 18 It gives with variable temperature. substitute Equa- tion D-l into Equation 6—4 and integrate with respect to time: t t Ix’c apt-Bu, - Dani] dt - I re dt (9'2) lhen 6 remains relatively constant over a feeding cycle, G can be removed from Equation D-2 giving: t It! “pl-Bu“, - Tutti.) dt - K't (1)-3) 162 163 Assume that the temperature fluctuation is a linear function of time as follows: T . T.‘x " It (”-4) where m - rate of temperature change. 'C/day. Substituting Equation D-4 into Equation D-3 and integrating with t - 0.5 day. half a cycle. gives: [v/xr ' RI:/B(Tmax-Tmin)[°39[’E(Tr-Tmax)’21:] - expl-ECI'r-Tun) any] (1)-5) Equation D-S gives the ratio of the gas production with a linearly fluctuating temperature to the gas production at constant temperature. Using the values of activation energy (E I 42.5 Kcal/degree Kelvin) and temperature range of 6.65’c (Ti‘: 3 39,51'C, 1min a 32.92'C) for the extended period of Experiment IIC, the calculated result is: x'lrr a 1,093 (D-6) Therefore, the estimated value of the gas production from the slow and moderate fractions in the fluctuating temperature digesters (EXP IIA. IIB. and IIC) are 9.3% higher than the gas production from these fractions in the control unit. However. since the fast fraction is almost completely degraded in 24 hours, gas production from this frac- tion is not much affected by temperature fluctuations. The estimation of gain in gas production for fluctuating temperature digesters com- pared to constant temperature digesters for various ranges of temperature fluctuation (AT) calculated by Equation D-6 are shown in Table D-l. 164 TABLE D1. Theoretical Gain in Gas Production Due to Fluctuating _Temperature for a Slowly Degradable Substrate (percent increase over constant temperature). E AI. '0 rca1l'r 0‘ 3.00 3.00 6.63 10.00 10 1.05 0.1 0.3 0.5 1.2 20 1.11 0.4 1.2 2.0 4.7 30 1.17 0.9 2.6 4.6 10.7 40 1.23 1.7 4.7 8.3 19.3 60 1.30 2.6 7.4 13.2 31.4 60 1.37 3.8 10.7 19.4 46.9 70 1.44 5.2 14.7 26.9 66.8 so 1.32 6.8 19.3 35.9 91.8 ‘ O : exp(E/ET:) BIBLIGSRAPHY BIBLIOGRAPHY Barker. B. A. 1961. "Fermentation of Nitrogenous Organic Compounds.” I!g_fl;g£g£_g. Vbl. II. Gunsalus. I. C.. and Stanier. R. Y.. Eds.. New York: Academic Press. Boone. D. R. and I. P. Bryant. 1980. "Propionate-Degrading Bacterium. Syntrophobacter volinii sp. nov. gen. nov.. from lethanogenic Ecosystem." MW 9. 626- Brock. T. D.. 1979. Biglggy_gj_!flggggggggig_s. 3rd Edition. California: Prentice Hall. Inc. Bryant. I. P. 1976. "The Microbiology of Anaerobic Degradation and lethanogenesis." in B.G. Schelegel (ed.) Symposium 23 Microbial flggggy_gggzgggigg. Gottingen: Germany. E. Goltze KG. p. 107. Bryant. I. P. E. A. 'blin. I. J. 'olin. and R. S. Iolfe 1967. "lethanobacillus omelianskii. a Symbiotic Association of Two Species of Bacteria.” A£g§;_!i§;gbig;‘ 59:20 Bryant. I. P. . L. L. Campbell. C. A. Reddy. and I. R. Crabill. 1977. "Growth of Desulfovibrio in Lactate or Ethanol ledia Low in Sulfate in Association with Ha-Utilizin‘ usthanogenic Bacteria." Appl. Environ. Iigrgbio. 33:1162 Chen. Y. R. and A. G. Hashimoto. 1978. "Kinetics of Methane Fermenta- tion." in C.D. Scott (ed.) Pr n iotec W. New York: John Iiley- p- 269. . Chen. T. R. et a1. 1980. "Effect of Temperature on lethane Fermentation Kinetics of Beef-Cattle Eanure." iotgghgol. and Bioegg, 10:325 Chung. K. T. 1972. "An Ecological Significance of Hydrogen Utilization in lethanogenesis." Abgtgl A, g, AmI Sgg, Iigrgbiol, p. 64. Chynoweth. D. P.. and R. A. Nah. 1971. "Volatile Acids Formation in Sludge Digestion." in Anaerobic Biological Treatment Processes. Adv. in §h15‘_§g£‘. 105:41 Doelle. B. I. 1975. W. New York: Academics Press. Eastman. J. A. 1977. Sglghilizgtign of Organic Cgrhgn During the Acid Phggg 21 Anaerobic Digestion. Ph.D. Thesis. Uhiv. of Iashington. 165 166 Eastman. J. A.. and J. F. Ferguson. 1981. "Solubilization of Particulate Organic Carbon During the Acid Phase of Anaerobic Digestion." J. 'aeer Poll, Control £ed.. 53:352 Garber. I. F. 1954. "Plant-Scale Studies of Themophilic Digestion at Los Angelos." Sewage and Ind, Wastes. 26:1202 Gaudy. A. F. and E. T. Gaudy. 1980. Microbiology for Environmental ‘ Seientiste an; Engineers New York: McGraw Hill. Inc. Eashimoto. A. G.. V. B. varel. and Y.R. Chen. 1979. "Factors Affecting Methane Yield and Production Rate." ASAE paper No.79-4583. ASAE. St. Joseph. MI Bawkes. D. R.. and B. V. Young. 1980. "Design and Operation of Laboratory-Scale Anaerobic Digesters: Operating Experience with Poultry Litter." eggicelEurel Iesee. 2:119 Beisler. M.. 1981. "Biogas Filtration and Storage." Paper Presented at the "Methane Technology for Agriculture Conference." Cornell University. Ithaca. New York. Heukelekian. E.. and P. Muller. 1958. "Transformation of Some Lipids in Anaerobic Sludge Digestion." Sewage, IndI 'estes. 30:1108 Robson. P. N.. S. Bousfield and R. Summers. 1974. "Anaerobic Digestion of Organic Matter." in Critieel Reviews ie EnviromenEel geeeeel. Cleaveland: Chemical Rubber Co.. p. 131. Robson. P. N.. et a1.. 1981. "The Microbiology and Biochemistry of Anaerobic Digestion." in Meeeene Production From Ageieeitural end 20.21119 gastee. ch. 3. p. 10 Bungate. R. E.. 1975. "The anen Microbial Ebosystem." Anneel Review of Eeelogz end Szetematiee. 6:39. Iannotti. et a1.. 1973. "Glucose Fermentation Products of aninococcus albus Grown in Continuous Culture with Vibrio succinogenes: Changes Caused by Interspecies Transfer of B3," II Bgct. 114:1231. Jeris. J. S.. and P. L. McCarty. 1965 "The Biochemistry of Methane Fermentation Using C34 Tracers." J, Eager Poll. genErol Eed.. 37: 2 Jewell. I. J. et a1.. 1976. "Bioconversion of Agricultural Wastes for Pollution Control and Energy Conservation." Final Report. ERDA-NSF-741222A01. Ithaca: Cornell University. Jewell. I. J. et a1.. 1980. "Anaerobic Fermentation of Agricultural Residue-Potential for Improvement and Implementation." Vol. II 0.8. Dept. of Energy Report. NO. DE-AC02-76ET20051. 167 Kirsch. E. J.. and R. M. Sykes. 1971. "Anaerobic Digestion in Biological laste Treatment." Progress in Indesteial Mieroeiology. 9:155 Kugelman. I. L. . and K. K. Chin. 1971. "Toxicity. Synergism and Antagonism in Anaerobic Waste Treatment Processes." AdvI Chem,I See, 105:55 Kugelman. I. J.. and P. L. McCarty. 1965. "Cation Toxicity and- Stimulation in Anaerobic Isste Treatment." J, 'aeer Poll, gonerol Beg. 37:97 Latham. M. J.. 1979. "The Animal as an Environment." in J. M. Lynch and N. J. Poole (eds.) Mi r i l olo ° Conce tual roa London: Lackwell Scientific. Latham. I. J. and M. J. lblin. 1977. "Fermentation of Cellulose by Ruminococcus flavefaciens in the Presence and Absence of Methanobacterium ruminatium." Appl. Eeviree, Microbiel,. 34 297 lmwrence. A. I. and P. L. McCarty. 1969. ”Kinetics of Methane Fermentation in Anaerobic Treatment. Part II." 1, later Poll, Congrol £ed.. 41: 21 Leng. R. A.. 1973. "Salient Featrures of the Digestion of Pastures by Ruminants and Other Herbivores." in G. I. Butler and R. I. Bailey (eds.) m t Bio hemi t Herb e. vol. 3. London: Academic Press. Lynch. J. M.. and.N. J. Poole. 1979. Mieroeiel Eeelogy; A Qoneepteel Appeeeeh. London: Blackwell Scientific. p.81. McCarty. P. L.. 1964. "Anaerobic Iaste Treatment Fundamentals. Part III Toxic Materials and Their Control." Peelie gorge. 95:91. McInerney. M. J.. et a1.. 1979. "Anaerobic Bacterium that Degrades Fatty Acids in Syntrophic Association with Methanogens." A££!1_ Mihggeiol. 122: 129. McCarty. P. L.. 1981 "One Hundred Years of Anaerobic Treatment." in D. A. Stafford. B. I. Iheatley and D. E. Hughes (eds.) Anee£212_ Di gee§ion London: Applied Science. Mountfort. D. O.. and R. A. Asher. 1978. "Changes in Proportions of AOOt‘tO and C0; Used as Rethane Precursors During the Anaerobic Digestion of Bovine Wastes." l a d v i r i lo 35:648 O'Rourke. J. R.. 1968. Kinetic; of Anaeroeie waete Treatmeet a; Redeeed Teeperaeeree. Ph.D. Thesis. Stanford University. 168 Pfeffer. J. T. and G. E. Quindry. 1978. "Biological Conversion of Biomass to Methane. Beef Lot Manure Studies." Report No. UILU-ENG-78-2011. May. Urbana: Univ. of Ill. Prin and Lankhorst. 1977. FEMS Microbiology Letters. 1:255 Robbins. J. E.. M. T. Arnold and S. L. Lacher. 1979. "Methane Production from Cattle Waste and Delignified Straw." Infeceien apfl Mitx. 38: 175 Smith. P. H.. and R. A. Mah. 1966. "Kinetics of Acetate Metabolism During Sludge Digestion." pplI Mieroeial 14:368 Speece. R. E.. and J. A. Kern. 1970. "The Effect of Short-Term Temperature Variations on Methane Production." J. Wage; Pel;I gentro; Beg. 42:1990 Stafford. D. A.. D. L. Hawkes. and R. Horton. 1980. Me h e Prod c feem [aete Organie Matter. Boca Raton. Florida. CRC Press Inc. t r M ho for he ination of Water and Wastewater. 14 1976. APHA.. AWWA. and WPCF. (Jointly Published) Washington. Thauer. R. I. et a1.. 1977. "Energy Conservation in Chemotrophic Anaerobic Bacteria." BecEeriel, Rev.. 41: 100 Van Velsen. A. F. M.. and G. Lettinga. 1980. "Effect of Feed Composition on Digester Performance." in D. A. Stafford. B. I. Wheatley and D. E. Hughes (eds.) Anaerebie Digeetion. London: Applied Sci. p.113 Van Velsen. A. F. M.. et a1.. 1979. "Anaerobic Digestion of Piggery Waste. 3. Influence of Temperature." e§hI J, Agrie, §eil 27: 255 Varel. V. E.. A. G. Hashimoto. and Y. R. Chen. 1980. "Effect of Temperature and Retention Time on Methane Production from Beef Cattle Waste." eel. eed Eny, Mieereie, 40:217 Weber. W. J.. 1972. i och m 1 e se for Wa er ali C ntrol New york: Wiley-Interscience. Wolfe. R. S.. 1979. "Microbial Biochemistry of Methane: a Study in Contrasts." in J. R. Quayle (ed.) Mieroeia; Biechemietey. V01. 21. Baltimore: University Park Press. p. 293. Wolin. M. J.. 1974. "Metabolic Interactions Among Intestinal Microorganisms." Agar. 1. glin, N'utrI 27:1320 Ziekus. J. G.. 1980. "Microbial Populations in Digesters." in D.A. Stafford. B.I. Wheatley and D. E. Hughs (eds.) Anaeroeic 211111122: London: Applied Science p. 61. “000000