This item is only available as the following downloads:
1 ANAEROBIC DIGESTION: BIOENERGY AND BIOFERTI LI ZER PRODUCTION FOR HAITI By REGINALD TOUSSAINT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013
2 2013 Reginald Toussaint
3 T his work is dedicated to those who believe in a better sustainable world.
4 ACKNOWLEDGMENTS This work was completed with the h elp and kindness of many individuals to whom I wish t o express my deepest gratitude. I especially thank my advisor, Dr. Ann C. Wilkie in the Soil and Water Science Department, for giving me the opportunity to gain valuable experience in the field of bioene rgy and sustainable technology. The tireless guidance, support and encouragement that she provided to me was crucial for a successful completion. My thanks are also extended to the members of my thesis committee, Dr. Kimberly Moore and Dr. John Erickson, f or their helpful advice. I am very appreciative for the support of my colleagues, Scott Edmundson, Ryan Graunke and Camilo Cornejo, at the Bioenergy and Sustainable Technology Laboratory who have helped me in many different ways to complete this research. I also thank my fellow Haitian students, especially Dakson, Joseph, Pascale and Lidiune, who have been indispensable in offering motivation and support over the course of this journey. I am grateful to the USAID/Watershed Initiative for National Natural En vironmental Resources Haiti project for pro viding my graduate scholarship. This research was supported through funding provided by the Florida Department of Agriculture & Consumer Services, Farm to Fuel Bioenergy Grants, Green Gas from Green Grass project. Finally, my deepest gratitude goes to my parents Francine and Lucois, and my brother and sisters for their love and unconditional support
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Environmental Challenges in Developing Countries Including Haiti ........................ 14 Integration of Anaerobic Digestion into Farming Systems: A Win Win Solution ...... 15 Application of Biogas Production in Developing Countries ................................ ..... 17 Microbiology of Anaerobic Digestion ................................ ................................ ....... 20 Anaerobic Digesters and Operating Conditions ................................ ...................... 22 Alkalinity and pH ................................ ................................ ............................... 23 Temperature ................................ ................................ ................................ ..... 24 Hydraulic Retention Time ................................ ................................ ................. 24 Organic Loading Rate ................................ ................................ ...................... 25 Carbon/Nitrogen Ratio ................................ ................................ ...................... 26 Feedstocks for Anaerobic Digestion ................................ ................................ ....... 26 Proximate Analysis and Feedstock Composition ................................ .................... 30 Ove rview of Anaerobic Digester Effluent ................................ ................................ 32 Sweet Sorghum as a Multipurpose Crop ................................ ................................ 34 The Relevance of Anaerobic Digestion Technology to Haiti ................................ ... 35 Thesis Rationale ................................ ................................ ................................ ..... 37 Hypotheses ................................ ................................ ................................ ............. 39 Objectives ................................ ................................ ................................ ............... 39 2 MATERIALS AND METHODS ................................ ................................ ................ 46 Characterization of Anaerobic Digested Effluent ................................ .................... 46 Nutrient Analysis of Anaerobic Digester Effluent ................................ .............. 46 Total nitrogen ................................ ................................ ............................. 46 Total ammonia nitrogen ................................ ................................ ............. 47 Total phosphorus ................................ ................................ ....................... 48 Elemental potassium ................................ ................................ .................. 48 Economic analysis ................................ ................................ ..................... 49 Field Experiment ................................ ................................ .............................. 49 Soil samplings ................................ ................................ ............................ 50 Fertilization ................................ ................................ ................................ 50 Plant sampling ................................ ................................ ................................ .. 51
6 Onion ................................ ................................ ................................ ......... 51 Corn and bean ................................ ................................ ........................... 52 Tissue analysis ................................ ................................ .......................... 52 Characterization of Sweet Sorghum as a Multipurpose Crop ................................ 53 Juice, Biomass, and Sugar Yield Dete rmination ................................ ..................... 54 Physicochemical Parameters ................................ ................................ .................. 55 Total Solids and Volatile Solids ................................ ................................ ........ 55 Fiber Analysis ................................ ................................ ................................ ... 55 Neutral Detergent Fiber ................................ ................................ .................... 56 Acid Detergent Fiber ................................ ................................ ........................ 56 In Vitro Organic Matter Digestibility ................................ ................................ .. 57 Nitrogen ................................ ................................ ................................ ............ 57 Total Phosphorus ................................ ................................ ............................. 57 Characterization of Sweet Sorghum Juice ................................ .............................. 58 pH, Electrical Conductivity and Brix ................................ ................................ .. 58 Alkalinity ................................ ................................ ................................ ........... 58 Potassium ................................ ................................ ................................ ......... 59 Chemical Oxygen Demand ................................ ................................ ............... 59 Methane Index Test ................................ ................................ ................................ 60 Methane Production Measurement ................................ ................................ .. 61 Statistical Analysis ................................ ................................ ............................ 61 3 CHARACTER IZATION OF ANAEROBIC DIGESTER EFFLUENT AS A FERTILIZER SOURCE FOR HAITI ................................ ................................ ........ 66 Characterization and Nutrient Composition of ADE ................................ ................ 69 E ffects of ADE on Vegetable Production ................................ ................................ 70 Effects on Onion Yield ................................ ................................ ...................... 70 Effect on Bean And Maize Yield ................................ ................................ ....... 71 Postharvest Soil Properties ................................ ................................ .............. 73 Discussion ................................ ................................ ................................ .............. 73 4 CHARACTERIZATION OF SWEET SORGHUM FOR BIOFUE LS AND FOOD PRODUCTION ................................ ................................ ................................ ........ 89 Crop Yield and Biomass Partitioning ................................ ................................ ...... 90 Feedstock Composition ................................ ................................ .......................... 91 Organic Matter, Fiber Content and Chemical Oxygen Content ........................ 91 In Vitro Organic Matter Digestibility ................................ ................................ .. 92 Nutrient Content ................................ ................................ ............................... 93 Methane Yield and Methane Production from Non Juice Components .................. 93 Methane Production Rate for Fibrous Compon ents Of Sweet Sorghum ................. 94 Proximate Analysis for Methane Yield and Kinetic Rate Prediction ........................ 95 Characterization of Sweet Sorghum J uice for Methane Production ........................ 96 Juice Yield and Chemical Characteristics ................................ ......................... 96 Anaerobic Digestion of Sweet Sorghum Juice ................................ .................. 97 Methane Yield of Sweet Sorghum Biomass ................................ ............................ 98
7 Application of Anaerobic Digestion of Sweet Sorghum ................................ ........... 99 Ensilage: A Potential Storage Technique of Sweet Sorghum Component for both Feed and Biogas Production ................................ ................................ ..... 101 Sweet Sorghum: A Crop for Fuel and Food Production in Hait i ............................ 102 5 CONCLUSION ................................ ................................ ................................ ...... 121 Fertilizer Value of Anaerobic Digester Effluent ................................ ..................... 122 Characterization of Sweet Sorghum for Food and Fuel Production ...................... 123 Practical Implication and Recommendations ................................ ........................ 124 LIST OF REFERENCES ................................ ................................ ............................. 126 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 135
8 LIST OF TABLES Table page 1 1 Methane yiel d of animal manures and other organic wastes .............................. 40 1 2 Methane yield of selected agricultural wastes ................................ .................... 41 1 3 Theoretical characteris tics of typical substrate components. ............................. 41 2 1 Fertilizer application volumes and total N application rate for onion and corn .... 63 2 2 Fertilizer application volumes and total N application rate for bean .................... 63 2 3 Amounts of feedstock and inoculum used in the MIT for Sweet Sorghum juice 63 3 1 Physico chemical characteristic of anaerobic digester effluent (ADE) ................ 77 3 2 Economic value of ADE for bean, maize and onions at application rate based on total nitrogen ................................ ................................ ................................ .. 78 3 3 Economic value of ADE for bean, maize and onion production at application rates based on total phosphorus ................................ ................................ ........ 79 3 4 Effect of fertilizer treatment on onion ( Allium cepa ) growth and yield ................. 80 3 5 Effect of fertilizer treatment on bean ( Phaseolus vulgaris ) growth and yield ...... 80 3 6 Effect of fertilizer treatment on maize growth and yield ................................ ...... 81 3 7 Soil characteristic at time of planting ................................ ................................ .. 81 3 8 Soil properties after harvesting as affected by diff erent fertilizer treatment ....... 82 4 1 Partitioning of sorg hum plant biomass in g/plant. ................................ ............ 105 4 2 Sweet sorghum biomass yield ................................ ................................ .......... 105 4 3 Morphological characteristics of SS plant ................................ ......................... 106 4 4 Proxi mate analysis of fibrous components of SS ................................ .............. 106 4 5 Chemical oxygen demand of Sweet Sorghum components ............................. 107 4 6 Estimates of methan e production kinetics of SS components .......................... 107 4 7 Prediction of methane yield of sweet sorghum components from different independent variables ................................ ................................ ...................... 108
9 4 8 Sweet sorghum juice characteristics ................................ ................................ 108 4 9 Methane yield and energy value of Sweet Sorghum on a per ha basis ............ 108 4 10 Methane yield and energy value of Sweet Sorghum on Mg basis .................... 109
10 LIST OF FIGURES Figure page 1 1 Environmental and social benefits from i mplementing anaerobic d igestion. ..... 42 1 2 Integration of anaerobic digestion into farming systems, providing both fertilizers and en ergy. ................................ ................................ ......................... 43 1 3 Sequential metabolic phases in anaerobic digestion ................................ .......... 44 1 4 Energy consumption by fuel type in H aiti. ................................ .......................... 45 2 1 Par titioning of fiber in feedsto ck. ................................ ................................ ........ 64 2 2 Apparatus for measuring methane production through hydraulic displacement. ................................ ................................ ................................ ..... 65 3 1 Incre ase in onion yield by fertilization treatment compared t o the control .......... 83 3 2 Nutrient concentration in dry onion bulbs by fertilization treatment. .................... 83 3 3 Nutrient uptake by onion bulbs by fertilization treatment. ................................ ... 84 3 4 C h a racteristics of onion juice by fertilization treatment ................................ ....... 84 3 5 Nutrient contents of onion juice by fertilization treatment in mg/L ....................... 85 3 6 Percentage increase in bean yield compared to the control (no fertiliza tion) ...... 85 3 7 Percentage increase in maize yield compared to the control (no fertilization) .... 86 3 8 Tissue analy sis of bean (grain) by fertilization treatment ................................ .... 86 3 9 Nutrient uptake of bean (grain) by fertilization treatment ................................ .... 87 3 10 Tissue analysis of maize (grain) by fertilization treatment ................................ .. 87 3 11 Nutrient uptake of maize (grain) by fertilization treatment ................................ .. 88 4 1 Sweet sorghum biomass partitioning. Values are fresh weight in kg/ha ........... 109 4 2 Organic matter (OM) and in vitro organic matter (IVOMD) digestibility of SS components ................................ ................................ ................................ ...... 110 4 3 Crude protein and ash content of SS component ................................ ............. 110 4 4 Phosphorus content of SS compon ent ................................ ............................. 111
11 4 5 C u m ulative methane production of SS component measured from the MIT .... 111 4 6 Fitted cumulative meth ane production curve of SS component ........................ 112 4 7 Cumulative methane production of whole plant versus chipped stalk ............... 112 4 8 Fitted cumulative methane production of whole plant versus chipped stalk ..... 113 4 9 Correlation between methane yield and fiber content. ................................ ...... 113 4 10 Correlation methane yield and crude protein ................................ .................... 114 4 11 Correlation methane yield and crude protein ................................ .................... 114 4 12 Correlation methane yield and crude protein ................................ .................... 115 4 13 Correlation between constant rate and crude protein ................................ ....... 115 4 14 Correlation between constant rate and cellulose/lignin ................................ ..... 1 16 4 1 5 Sweet sorghum juice characteristics. Alkalinity values are in g/L CaCO 3 10) 116 4 16 Cumulative methane production from SS juice ................................ ................. 117 4 17 Fitted cumulative methane production of SS juice ................................ ............ 117 4 18 Methane of sweet sorghum plant. Estimates are for 1 ha ............................... 118 4 19 Methane of 1 ha sweet sorghum plant. Only the stem is used for digestion. .... 118 4 20 Integration of sorghum sorghu m for fuel and feed production. ........................ 119 4 21 Ensiling of sweet sorghum for feed and fuel production ................................ ... 120
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Sci ence ANAEROBIC DIGESTION: BIOENERGY AND BIOFERTIZER PRODUCTION FOR HAITI By Reginald Toussaint May 2013 Chair: Ann C Wilkie Major: Interdisciplinary Ecology Integration of energy and food production offers potential for simultaneously addressing the n eed for both food and energy security. The yields of bean, corn and onion were compared in a field study under different f ertilization regimes: anaerobic digester effluent (ADE), urea, fish emulsion (FE), and a control group receiving no fertilizer. Yields of onion, bean, and corn grown with ADE fertilization increased relative to the control by 42.4, 40.5, and 65%, respectiv ely. Yields of onion and bean grown under ADE fertilization were not significantly different from yields using either urea or FE. Corn yields from plants treated with ADE were equivalent to plants receiving urea and reached a higher yield than FE and contr ol groups. Additionally, the suitability of sweet sorghum as a multipurpose energy and food crop was evaluated. If all sweet sorghum components were diverted to anaerobic digestion, the methane yield could reach 9,850m 3 /ha. In a more integrated scenario, 6 ,600 m 3 /ha of methane can be obtained from the stem coupled to the production of 6.5 Mg/ha of grain for animal/human food, and 3.4 Mg/ha of leaves that can be left in the field to return nutrients and increase soil organic matter. Adoption of
13 anaerobic dig estion in an integrated energy and food production system would promote access to energy in rural areas of developing countries including Haiti, support the growth of local economies and sustain long term improvement in food security.
14 CHAPTER 1 INTRODUCTI ON Environmental Challenges in Developing Countries Including Haiti Providing clean energy and food security particularly to the developing world is an ambitious step towards a more sustainab le world A pproximately population and up to 90% of rural households in developing countries currently rely on raw biomass as their primary source of energy (Bruce et al ., 2000 ; FAO, 2010). The indoor air pollution caused by the combustion of unprocessed biomass in open fires or improperly functioni ng stoves represents a primary detriment to the health of women and young children expos ed to the soot produced by biomass combustion (Mihelcic, 2009; Bruce et al 2000). Additionally, u sing biomass as an energy source has environmental consequences in te rm s of deforestation, which can cause severe soil erosio n and decrease agricultural productivity In the context of world population growth global food production will need to substantially increase to sustain food security around the world (Roberts, 200 9). T he deterioration of agricultural lands particularly in developing countries is a question of concern that needs to be addressed to achieve and sustain increased food production (FAO, 2010). Beside s the requirement of maintaining and restoring soil q uality, intensification o f agricultural inputs such as water, pesticides and fertilizers has been considered as the most plausible prospect to address the need for increased agricultural productivity ( Tilman et al 2002; Pretty et al 2010). As current f ertilizer production is dependent on fossil fuels, any fluctuation in the cost of fossil fuels will be particularly detrimental for small farmers in developing countries who are not able to afford high fertilizer cost (Wilkie, 2007) Therefore, energy and food security are
15 considerably intertwined and instability in the fossil fuel chain drives food instability and malnutrition Innovative agronomic practices to better utilize nutrients from sources other than synthetic fertilizers including recycling elem ents from organic materials s uch as crop residues, manures or household wastes, and biological nitrogen fixation, are crucial to maximize food production partic ularly in developing countries The situation of Haiti is a concrete example of poor natural res ource management and environmental deterioration which characteriz es many developing countries. As biomass provides more than 70% of the energy consumption of the country, the reliance on firewood has intensified deforestation and subsequent widespread so il erosion issues As a result, Haiti is ran k ed as the lowest in the region in terms of agricultural productivity (UNEP, 2010). Additionally, the absence of any functioning waste management system compromises the quality of drinking water to the population The occurrence of the 2010 earthquake and the subsequent cholera outbreak highlighted the precarious ness of the living conditions of the Haitian people. In order to address the complex nexus of public health, energy and the environment within Haiti and m ost other developing countries, a multifaceted solution that strengt hens agricultural productivity while providing sustainable services such as clean energy, potable water and adequate sanitation is needed. Anaerobic digestion of organic residues generat ed on household farm s is one such example capable of producing renewable energy and fertilizer while managing organic waste Integration of Anaerobic Digestion into Farming System s : A Win Win Solution Anaerobic digestion (AD) is the microbial degradation of organic matter under an oxygen starved environment by a consorti um of microorganisms that produc e biogas, a mixture of methane carbon dioxide, and minor amounts of other gases such as
16 ammonia, hydrogen sulfide, hydrogen and nitrogen (Wilkie et al 200 8). Anaerobic digestion has been used for centuries for energy production from organic wastes (Braun, 2007). Simultaneously, AD generates a rich plant nutrient material, which can serve as fertilizer or soil amendment (Angelidaki et al 2011). The biogas can be combusted to produce heat and electricity or can be directly used for cooking, lighting or for heating water. Integration of anaerobic digestion technology into existing farming systems valorizes organic wastes generated on the farm and offers a sus tainable tool for waste management. By providing a clean and sustainable energy source and recycling plant nutrients, anaerobic digestion links food and energy production and, therefore contributes to address ing possible issues between food and fuel produ ction (FAO, 2010). Potential impacts of biofuel production from food based biomass on global agricultural markets and on the environment in terms of resource use and land change are controvers ial and justify the adoption of technolog ies that promote valori zation of agricultural residues for energy production (FAO, 2008). Moreover, AD combines the production of renewable energy and biofertilizer with other environmental benefits such as greenhouse gas emission mitigation, reduction of odors, pollution contro l, and effective pathogen removal (Angelidaki et al 2011; Ward et al 2008 Wilkie 20 08 ). Most small farmers in Haiti generate abundant nonedible biomass from either crop production or livestock ; th ese can be used as feedstock s for anaerobic digestion. Even though some crop residues should be left in the field for increasing soil organic matter, surplus biomass can be diverted to anaerobic digestion rather than burned i n field, a current practice in many places of the country (FAO, 2010). Another scenar io of
17 integration of anaerobic digestion into existing farming system s consists of using multipurpose crops that are capable of simultaneously producing food, feed and fuels Application o f Biogas Production i n Developing Countries Most households in deve loping countries including Haiti suffer from a lack of clean and sustainable energy to meet their basic subsistence needs for a minimum level of human comfort. Energy is needed primarily for cooking, lighting, and sometimes for powering certain household appliances and devices. Energy needs for cooking account for more than 9 0% of the household energy demand in rural areas ( Rajendran et al 2012). Unprocessed biomass such as wood, crop residues, and anima l dungs are the main energy sources for cooking. T he reliance on biomass as an energy source has led to uncontrolled deforestation, a common environmental concern in most developing countries. In addition, burning firewood for c o oking produces smoke and soot particles contributing to indoor air pollution, which can cause serious health issues such as respiratory diseases and lung cancer (Bruce et al 2000 ) Biogas produced from anaerobic digestion of wastes or dedicated crops can be a cleaner substitute energy source for daily cooking requirements Utiliz ation of biogas for cooking would contribute to reduc ing indoor air pollution and consequent related illnesses, while simultaneously avoid ing the detrimental consequences of deforestation (Cornejo and Wilkie, 2010) In rural areas of Haiti lighting energy needs are predomina ted by the use of kerosene. Because the household electricity supply in most villages is very limited, utilization of kerosene is in most cases the only option. Although energy needs for lighting in rural households represents only a sm all share of their total energy demand, it is of crucial importance. Fuels for lighting are often a significant energy expense and the proportion of this cost in the household budget can be substantial. Providing a clean
18 energy source for lighting w ould a llow rural families to have more productive and comfortable lives. Lighting allows c hildren to study and do their homework at night, which translate s to a subsequent increase in school performance and quality of education The combustion of biogas in a bio gas lamp produces high incandescence and induces brightness. Biogas can be a substantial energy source for lighting in rural households (Rajendran et al 201 2 ). Biogas can be directly burned in boilers or burners for cooking in rural households. Direct co mbustion of biogas in natural gas burners is also possible and has been applied in many countries. Another use of biogas combustion is the production of heat for either on or off site utilization. When used directly for cooking, lighting or for heat produ ction, biogas does not need to undergo any upgrading and can be used directly (Wilkie, 2007) A more efficient utilization of biogas is the cogeneration of heat and power (CHP). However, before CHP conversion, biogas needs to undergo condensation and parti culate removal, compression, cooling and drying. The content of hydrogen sul f ide and halogenated hydrocarbons have to be remove d from the biogas since most gas engines have maximum limits for these substances (Wilkie, 2008) Despite the efficiency of CHP, its application is limited by the need of biogas upgrading and expensive equipment, and therefore may not be the best option for rural Haiti. Implementation and utilization of biogas by rural households can provide environmental, economi c, and social benef its (Figure 1 1 ). For instance, improved health conditions and ch ange in lifestyle for women were observed after installation of household biodigesters in many developing countries (Rajendra et al 201 2 ). After the launch of biogas projects in villages in India firewood use dropped drastically from
19 1,048 to 410 kg /year and kerosene use from 120 to 46 L /year (Agoramoorthy and Hsu, 2008). Unlike fossil fuels, biogas produced from AD is renewable since it is obtained from biomass, which is a living storage o f solar energy through photosynthesis. Production of renewable energy from AD will increase security of local and national energy suppl ies and subsequently diminish reliance on imported fuels (Wilkie, 2006) In addition, anaerobic digestion of organic wast es can considerably improve sanitation in developing countries According to the WHO (2010), over 2.6 billion humans are living without access to improved sanitation. In developing countries, less than 50% of the solid waste generated in urban areas is col lected by the municipalities and private sector (Katukiza et al (2012) Anaerobic digestion has been used in different countries to treat human and organic wastes and therefore contribute s to address ing the major challenges of solid waste management and p ublic health which are present in most developing countries. In this regard, AD can also contribute to reducing the volume and disposal cost of a variety of waste s (Rajendra et al 201 2 ). In addition, the production of renewable energy from biomass and t he generation of bioferti li zer by recovering the nutrients in the anaerobic digester effluent ( ADE ) provide a closed carbon and nutrient cycle (Figure 1 2 ) (Wilkie, 2008 ) The biogas is burned to produc e energy and the CO 2 is released to the atmosphere for re uptake by vegetation through photosynthesis. Some carbon components and the inorganic fraction of the feedstock are retained in the ADE and can be applied to agricultural fields, improving soil organic matter and soil fertility. Implementation of AD ca n be adequately incorporated into both conventional and organic farming system s where the biogas can
20 be used to offset energy needs and the biofertilizer used to offset the need for purchasing synthetic fertilizers Microbiology o f Anaerobic Digestion Anae robic digestion of organic residues is a multistep metabolic process involving consorti a of microorganisms working synergistically through four distinct phases: hydrolysis, acidogenesis, acetogenesis, and methanogenesis (Figure 1 3) During the hydrolysis phase, complex organic compounds such as carbohydrates, fats and proteins are hydrolyzed to monosaccharides, fatty acids, and amino acid s respectively by the action of extracellular enzymes produced by a group of facultative and obligator y anaerobic bac teria (Angelidaki et al 2011). The rate at which highly complex compounds are hydrolyzed is a function of the substrate and bacterial concentrations, as well as environmental factors such as pH and temperature (Braun, 2007). Hydrolysis can take place wit hin a few hours for carbohydrates and few days for proteins and lipids. More recalcitrant materials such as lignocellulosic biomass are only partially hydrolysable and may require some pretreatment (Zhang et al 2007). Hydrolysis is often considered the r ate limiting step in degradation of particulate organic matter such as manure, sewage sludge and crops residues ( Angelidaki et al 2011) Simple sugars, fatty acids and amino acids formed during the hydrolysis of macromolecules are catabolized by a group of ac id ogenic bacteria to produce short chain organic acids, C 1 C 2 molecules, alcohols, hydrogen and carbon dioxide. This step is linked to the methane formation phase by acetogenesis where the end products of the previous steps are converted into short c hain volatile fatty acids (VFAs) predominantly acetate by hydrogen producing acetogens (Yu and Shanbacher, 2010). This step is particularly sensitive to the hydrogen concentration, which favors
21 accumulation of undesirable compounds such as lactate, ethanol propionate, butyrate and longer chain volatile fatty acids. In general the higher the partial pressure of hydrogen, the less acetate is formed (Chandra et al 2012; Madsen et al 2011). Acetogenesis is a key step for the overall process since acetate and carbon dioxide are the main educts for methane formation. Methanogenesis is the final step of the overall process of methane formation. In this phase, obligate anaerobic archae a, known as methanogens convert reduced compounds into methane and carbon d ioxide Methanogen s use a limited number of substrate s including acetic acid, methanol, carbon dioxide and hydrogen (Yu and Shanbacher, 2010). This group of organism s use s carbon dioxide as an electron acceptor and produces methane as the major end product of metabolism (Angelidaki et al 2011). In an operating digester, methane is generated from two different pathways: in the aceticlastic pathway, acet ate is the single substrate and accounts for 70% of the total methane formation while the remaining 30% i s formed from the hydrogenotrophic pathway (Gerardi, 2003). Despite the importance of the aceticlastic methanogenesis, only members of the Methanosaetaceae and Methanosarcinacea e families are known to be able to metaboliz e acetate to methane ( Wilkie, 2005) While Methanosaetacea e only use acetate, members of Methanosarcinacea e are more versatile and can utilize other carbon compounds such carbon monoxide, carbon dioxide, and methylated C 1 compounds Hydrogenotrophic methanogens are responsible for the nonac eticlastic methane production via utilization of hydrogen as electron donor coupled to reduction of carbon dioxide (Wilkie, 2008)
22 Methanogenesis is the key step in the overall methane producing anaerobic fermentation. Besides the production of methane, m ethanogens help regulate and neutralize the pH by consuming volatile fatty acids which otherwise will accumulate and inhibit the growth of bacteria involved in previous stages. Furthermore, the conversion of H 2 to methane contributes to reduce the partial pressure of H 2 which is essential for acetate production from acetogenic bacteria (Braun, 2007). To ensure the reliability of the anaerobic digestion process, each subsequent group of organisms must process the organic intermediates immediately after thei r formation. In this regard, all bacterial species involved in the process must interact in a dynamic equilibrium for a proper fermentation and methanogenic process. The concentration of some metabolites such as hydrogen, propionate, ammonia, and sul f ide m ust be maintained fairly low since they can become inhibitory at high concentration s (Braun, 2007). In a stable fermentation process, a low hydrogen partial pressure is maintained by the activity of the hydrogen utilizing methanogens resulting in a regulat ion of the degradation of propionate and butyrate. By the conversion of acetic acid to methane and CO 2 acetate utilizing methanogens regulate the pH nearly to neutral. Thermodynamically, th e conversion of acetate to methane can only take place at very low partial pressures of hydrogen. Consequently, an increased partial pressure of hydrogen will result in accumulation of intermediate organic metabolites leading to a pH drop below 6 and the subsequent discontinu ation of methane generation ( Angelidaki et al 2011 ). Anaerobic Digesters and Operating Conditions The term a naerobic digester designates a wide variety of reactors that contain the microbial ecology necessary for the occurrence of anaerobic digestion and methane
23 production (Wilkie, 2005b) Regardles s of their individual scheme and construction, all anaerobic digester s are designed to maintain air exclusion, favor and maintain optimal conditions for adequate microbial activity, and collect methane gas (Yu et al 2010). B atch reactors, continuously st irred tank reactors (CSTR), up flow anaerobic sludge blanket (UASB) reactors, plug flow reactors, covered lagoons, two phase digesters, and fixed film digesters are all different types of anaerobic digester designs (Wilkie, 2005 a ) For the success of the o verall anaerobic digestion process, a series of operational parameters need to be maintained. Alkalinity and pH, temperature, organic loading rate hydraulic retention time and carbon/nitrogen ratio are among the most critical parameters that need to be m onitored. Alkalinity and pH Most microorganisms involved in the anaerobic digestion process are pH sensitive. While acid producing bacteria tolerate a wide range of pH between 4.0 and 8.5, methanogenic archaea function in a very narrow range of pH between 6.5 and 8.2. (Yu and Schanbacher, 2010). In a balanced and stable digester, the tendency of acidification produced by the accumulation of VFAs as a result of hydrolysis acidogenesis and acetogenesis is countered by the activity of methanogenic archaea, wh ich consume acid s and produce alkalinity in the form of ammonia and bicarbonate. The buffering capacity of the digester is maintained by the CO 2 concentration in the gas phase and the HCO 3 in the liquid phase (Appels et al 2008). Acidification can occur when the digester is loaded beyond the maximum metabolism capacity of the microbial consortia. In addition, when the hydrolysis rate is faster than methanogenesis,
24 accumulation of organic acids in the system causes a drop of pH and therefore an inhibition of methanogenesis. Temperature Methanogenic archae and bacteria involved in biomethanation are generally active in two temperature ranges. Mesophilic bacteria operate in the temperature range of 25 45C but exhibit the maximum activity when the process t emperature is nearly 35 o C. Most methanogenic microorganisms belong to this category. However a few species perform in the thermophilic (45 60C) range (Khalid et al 2011). Digesters operating under mesophilic condition are more stable and typically requi re less energy input. However, any sudden change in the operating temperature has to be avoided since methanogens are extremely sensitive to drastic and /or rapid thermal changes (Gerardi, 2003). Th e se microorganisms have the ability to ad a pt to slow changi ng conditions, but an abrupt change can momentarily disturb the fermentation process and subsequently require a longer recuperation period. Additionally, as temperature influences the growth rate and metabolism of microorganisms, digesters operating under thermophilic conditions have faster kinetics and methane production rates than mesophilic digesters and exhibit a better reduction of pathogens and sludge separation (Chandra, 2012). Hydraulic Retention Time Hydraulic retention time (HRT) is the amount of time required by a feedstock to be degraded into methane to a certain extent. Feedstock characteristics, environmental temperature, microbial population and type of digester generally dictate hydraulic retention time Under practical conditions, retention times vary from a few hours, in the case of easily degradable, highly soluble feedstocks ( e.g. monosaccharides), to several
25 months ( e.g. lignocellulosic feedstocks) (Braun, 2007). More recalcitrant material s or substrate s rich in particulate matter require a longer retention time. Hydraulic retention time needs to be taken into consideration when designing and monitoring a digester (Appels et al 2008). Higher hydraulic retention time facilitates a better volatile mass removal and offers a higher buffering capacity against the shock of overloading, but requires more volume. In a continuous reactor, a fraction of the microbial population is removed any time effluent is withdrawn from the digester. Consequently, hydraulic retention has to be long enough to al low the compensation of cell removal by cell growth to ensure a steady state microbial population and avoid process failure High rate digesters, such as fixed film reactors, have been developed to allow the increased growth of microbial populations o n gro wth media within the digester (Braun, 2007). Organic Loading Rate The organic loading rate is the amount of volatile solids fed or loaded, into a digester per unit of digester volume per unit of time The maximum loading rate is a funct i on of the substrat e properties (concentration and digestibility), environmental conditions (temperature and pH) and reactor performance. In principle, a digester should be fed based upon the microbial population capacity to metabolize the feedstock to metha ne. As the acid forming rate is generally faster than methanogenesis, overloading the digester favors accumulation of organic acids resulting in a consequent drop in pH and subsequently inhibits the methane formation rate. At this point, a positive feed back loop is create d where methanogenic inhibition further reduces organic acid conversion to methane and cause s further acidification. Good indicators of overloading are a drop in the methane content of biogas, an increase in the volatile fatty acid content in the effluent and a decrease of the pH value (Braun, 2007).
26 Carbon /Nitrogen Ratio The carbon / nitrogen (C/N) ratio of the feedstock is also a key factor for the success of the anaerobic digestion process. A carbon / nitrogen ratio ranging between 20 and 30 is considered t o be optimal for biogas production. When the C/N ratio is very high, microorganism involved in the process will try primarily to satisfy their protein requirements and when the feedstock nitrogen content is exhausted they are unable to degrade the leftover carbon and subsequently the cumulative methane yield will decrease. On the other hand, ammoni a is accumulated when the carbon C/N ratio is very low. Accumulation of ammoni a can be cytotoxic for the methanogenic community (Hansen et al 1998) Food waste a nd human excreta generally have a low C/N ratio while lignocellulosic materials have a high carbon nitrogen ratio. Co digestion of low and high C/N ratio materials is suggested for obtain ing a balanced composition of the feedstock. Feedstocks f or Anaerobic Digestion All biodegradable materials are suitable for anaerobic digestion to a certain exten t Appropriate substrates for anaerobic digestion can be categorized in three major groups depend ing upon their origin: agricultural, community based and indust rial feedstock s Agricultural feedstocks encompass a wide range of substrates including animal manure, energy crops, algal biomass and crop residues. The abundant supplies of agricultural feedstock s their high biogas yields, and their relat ive ly lo w cost make them appealing for anaerobic digestion. Historically, anaerobic digestion of animal manure has been mainly employed for waste stabilization and odor and pollution control. Characteristics of animal manures vary largely from species to species an d by management operation Animal manures have been reported to have relatively high
27 water contents, ranging from 75% (poultry manure) to 92% (beef cattle manure) and relatively high organic matter content, with volatile solids (VS) contents ranging from 72% (p oultry manure) to 93% (beef cattle manure) of the total solids (TS) (Yu, and Shanbacher, 2010) A nimal manures particularly poultry manure are rich in elemental nutrients such as nitrogen (N), phosphorus (P), and potassium (K) Some animal manure howeve r, have little readily fermentable substrates since most of the readily degradable substrates have been digested and assimilated by the animal (Yu, and Shanbacher, 2010). Moreover, animal manures are rich in microbial biomass, including bacteria and methan ogens. As a result, AD reactors operating with animal manures, especially cattle manures can be started without the addition of an external inoculum. Pig, cow and chicken manures are often treated anaerobically; however, more incentive has been directed to improv ing the energy recovery from anaerobic digestion by promoting co digestion of manure with other substrates to increase the organic content and obtain a more balanced feedstock. Methane potential s of major animal manures are presented in Table 1 1 Agricultural wastes and energy crops have also gained attention as promising feedstocks for anaerobic digestion. They are advantageous because they are available in consistent high quantit ies. In response to the food versus fuel debate, utilization of non edible agricultural products and dedicated energy crops that are capable of grow ing in wasteland s and require low agricultural inputs, ha ve been promoted as feedstock s for sustainable biogas production. Maize ( Zea mays L.) sunflower ( Helianthus annuus L.) grass es including straws from wheat ( Triticum aestivum L.) and rice ( Oryza sativa L.) and Sudan grass [ Sorghum bicolor nothosubsp drummond ii (Steud.) de We ex Davidse]
28 are among the most common energy crops that can be used for biogas production. Waste s from sugarcane ( Saccharrum spp.) processing and sweet sorghum ( Sorghum bicolor L.) plant materials have also been recognized as potential feedstock s for anaerobic digestion (Ward et al 2008) The suitability of an energy crop for producing biogas sustai nably depends upon some key factors such as methane yield per hectare, time of harvesting, mode of preservation and pre treatment requirements. An overview of methane potential of major agricultural wastes and energy crops is provided in Table 1 2. Althou gh the majority of crop residues are typically left in the field, a substantial amount of crop residues are recoverable and available for conversion to biogas in most cases. Crop residues typically contain relatively low water contents, high VS contents, a nd a large range of readily fermentable carbohydrates (Yu et al 2010) The methane potential production of crop residues varies from crop to crop, ranging from 161 to 241 m 3 CH 4 / dry ton (Yu and Shanbacher, 2010). One important shortcoming of biological degradation of agricultural wastes and energy crops is their relative ly high proportion of non biodegradable component s which hampers the digestible fraction from enzymatic action. Different pretreatment methods have been used to enhance the bio digestibi lity of these materials and consequently increase their methane yields. Pretreatment methods could be physical, chemical, and /or biological (Taherzadeh and Karimi, 2008). Fruit and vegetable wastes (FVW) are also generated in large quantity and are typical ly highly degradable. An overview of the methane yield of FVW is summarized in Table 1 2. Agricultural residues and energy crops are often co digested with other wastes to improve methane yield
29 Harvested aquatic plants have also received attention for the ir biogas production potential within the context of developing countries (Wilkie and Evans, 2010). Invasive aquatic plants have emerged as a major problem due to nutrient enrichment of receiving waters particularly in locations that suffer from a lack of wastewater management. Utilization of aquatic plants for anaerobic digestion has the potential to assist in ecological remediation of polluted water bodies while providing useful end products. The eff iciency of anaerobic digestion of aquatic weeds can be improved through co digestion in multistage reactors with manures (Wilkie and Evans, 2010). Community based feedstock s include the organic fraction of municipal solid waste (OFMSW), sewage sludge, food wastes, and institutional waste. The composition and physical features of municipal solid waste ( MSW ) vary dramatically depending on locality season, collection, and sorting In most societies, the OF MSW account s for more than 50 % of the total MSW stream, and can be partitioned into various categories inclu ding paper, grass clippings/yard waste and foods scraps all which can be converted to methane through AD (Yu et al 2010). It has been reported that OFMSW is relatively deficient in inorganic nutrient s an d contains little moisture or readily fermentable sugars, however, methane yield can be as high as 300 550m 3 CH 4 / dry ton if digested appropriately (Yu and Shanbacher, 2010) Waste activated sludge (WAS), the end product of municipal and industrial biological wastewater treatment plants, cause s serious d isposal problems because of the typically large volume generated. Disposal of these end products contributes to increasing the overall cost of treatment and the risk of pathogen contamination when not properly treated (Parawira, 2012). In this respect, ana erobic digestion of WAS has
30 been recognized not only for energy recovery from the WAS, which can reduce the disposal cost, but also for its capability to reduce pathogens and consequently change the quality of the final product which could be classed as f ertilizer rather than a waste. Waste activated sludge is a complex compound consist ing of partic ulate matter bacteria, and extracellular polymeric substances produced by bacteria. Even though the V S/ T S of WAS can range from 59 to 88%, only a small fractio n of organic matter is readily susceptible to microbial degradation. The remaining material is mostly cell ular biomass. The methane potential of WAS is reported to vary from 85 to 390 m 3 CH 4 /dry ton (Yu and Schanbacher, 2010). The recalcitrance of bacteria l cell walls to degradation constitutes the limiting factor of stabilization of WAS by anaerobic digestion. Different pre treatment methods have been developed to break the cell wall and expose the organic constituents to anaerobic degradation. Anaerobic d igestion of waste activated sludge is expected to gain popularity as world population increas es and more wastewater will undoubtedly need to be treated (Parawira, 2012) Even though Haiti does not currently have a function al sewage treatment system, AD co uld offer a viable solution to the lack of sanitation in both urban and rural areas. Proximate Analysis a nd Feedstock Composition Before feeding a digester, it is important to know the theoretical methane potential of a feedstock which is relevant for gau ging the operat ion of an anaerobic digestion system Currently, a number of methods have been used to predict methane yield including the methane index test (MIT), volatile solid (VS) content, and the chemical oxygen demand (COD). The MIT is the determinat ion of the methane potential production in a batch reactor using an active microbial inoculum. The VS represents the fraction of organic matter of the feedstock that can be eventually used for methane
31 production. The theoretical COD of a compound correspon ds to the amount of oxygen required to completely oxidize the organic contents of the material In this regard, the COD concentration is the equivalent of the volume of oxygen consumed in the combustion of the feedstock using a chemical oxidizing agent. Th is method assumes that methane production is directly related to organic matter degradation and 1 g of COD produces 350 mL of CH 4 (Speece, 2010). Therefore, the theoretical methane yield of a feedstock can be estimated when its COD value is known. Addition ally, the m ethane yield of a specific substrate is a function of organic content. Consequently, VS content is widely used to estimate the methane yield. Theoretical methane yield on a VS basis is 415 496 and 1014 mL /g of VS for carbohydrates, proteins an d fats respectively ( Table 1 3 ) showing the variability by substrate composition when using this method of prediction Another method that has been widely used for predicting methane production is the stoichiometric conversion of the feedstock based on i ts elemental composition. Under anaerobic condition s organic molecules are converted to CH 4 and CO 2 according the following equation: (1 1 ) The specific theoretical methane potential (B o ), commonly expressed as L CH 4 /g VS, can be calculated from equation (1 1 ), assuming 22.4 as the volume of 1 mole of gas under standard conditions (273 K and 1 atmosphere pressure) (Angelidaki et al 2011). Conversely, B o is a rough estimate of the amount of methane that can be generated from an organic compound; since details related to cell microbial generation influence the stoichiometry (Lyperatos and Pullammanappallul, 2010). However, the CH 4 volu me that can be practically obtained from a feedstock is always lower than B 0
32 because a fraction of the feedstock (2% to 5%) is synthetized as cell biomass, and a substantial portion of the feedstock is inert or non biodegradable (Angelidaki et al 2011) Overview o f Anaerobic Digester Effluent Anaerobic digester effluent (ADE) is the liquid residue after the microbial consum ption of the organic fraction by different microorganisms involved in the anaerobic digestion process (Gerardi, 2003). Based on the co mpositional characteristic s of ADE, many attempts have been made to explore the feasibility and benefits of its use in vegetable production. O nly the organic fraction of the feedstock is converted to biogas by the AD process, essential nutrients (N, P, K, Mg, and trace elements) required by plants, are retained in the residue (Arthurson, 2009). Furthermore as carbon is converted to methane, essential nutrients once bound in organic form are mineralized to inorganic plant available forms (Graunke and Wilki e, 2008) It has been reported that ADE contains higher NH 4 : total nitrogen ratio, decreased organic matter content, reduced biological oxygen demand and lower C/N ratio than raw organic residues (Moller and M u ller 2012). The NH 4 + N concentration of ADE d epends on the Total N concentration of the original feedstock. Anaerobic digester effluent from highly degradable materials such as cereal grains, manures, and food wastes generally have high NH 4 + : total nitrogen ratio and low carbon nitrogen ratio. Anaero bic digestion increases the availability of nitrogen in organic residues from 30 to 60% of the total nitrogen originally present in the feedstock while the total phosphate and potassium contents and availability remain unchanged during the digestion proces s (Polprasert, 1989).
33 Application of ADE to the soil has the ability to enhance soil microorganism activities as a result of increased inorganic nutrients, soil organic matter, and reduction of carbon/nitrogen ratio. Odlare et al (2008) investigated the effects of organic wastes (compost and ADE) on soil chemical and microbiological characteristics. They found that ADE enhanced microbial biomass and the proportion of metabolically active microorganisms compared to untreated control. Moreover, the rate of ammonia oxidation, and nitrogen mineralization capacity of the soil were stimulated by ADE application (Arthurson, 2009). Additionally, anaerobic digeste r effluent can be used as soil amendment and therefore contribute to improv ing the overall structure of the soil. Depending on the performance of the AD process, ADE can contain organic fractions that can contribute to increase SOM and induce positive effects on soil biological, chemical, and physical soil characteristics However, if the nutrients present in the ADE are in a useable form ADE can be classified as a fertilizer In a two year field study investigating the potential of ADE on crop yields, Montemurro et al (2008) found no significant difference in the cumulative plant dry weight of alfalfa sub jected to ADE and mineral fertilizers. Kocar (2008) compared the fertilization value of ADE from cattle slurry with those of commercially available organic and chemical fertilizers on safflower yield and recorded a higher yield with ADE application Furth ermore, Marchain (1992) reported that utilization of ADE induced a 6 20% increase in vegetable production suggesting that a large range of plants including cereals and vegetables can exhibit a significant response to ADE. Because ADE has a greater fraction of mineralized N, its application is more effective for crops possessing a short and intensive period of N uptake (Arthurson, 2009). Anaerobic digeste r effluent
34 was reported to be comparable to chemical fertilizers in terms of N uptake, fresh yield, and N uptake at harvest of spinach and komatsuma (Furukawa and Hasegawa, 2006). Sweet Sorghum as a Multipurpose Crop Sweet sorghum is a C 4 crop native to the tropic s with a wide adaptability to an array of environmental condition s (Walter and Monti, 2012) Th is crop has emerged as a potential energy crop and is considered to be superior to other energy crops because of its versatility yield potential and growth characteristics (Zegada Lizarazu and Monti, 2012) The high biomass yield of sorghum and the large amount of readily fermentable sugars stored in its stalk make it a promising energy crop candidate Sweet s orghum is capable of producing several products simultaneously ; sugar, starch protein and cellulose. The soluble solids in the j uice of the stalk ac counts for anywhere from 20 to 50% of the whole plant dry weight The fermentable sugars are mostly sucrose, glucose and fructose and are suitable for most bioprocessing technologies (Whitfield et al 2012) Furthermore, sweet sorghum has been recognized for its drought resistance and high water and nutrient use efficiency (Zedada Lizarazu and Monti, 2012) Compare d to sugarcane, it has a shorter growing cycle and can be incorporated into existing management practices (Erickson et al 2011, Walter and Mon ti, 2012). Regarding fermentable sugar production, sweet sorghum is tolerant to poor soil conditions (Almodares and Hadi, 2009). Although nitrogen fertilization may affect the total biomass production, little or no effects were observed on the dry weight s ugar fraction. Similarly, phosphorus application appeared to have negligible effects on sugar concentration (Whitfield et al 2012). As a drought resistant crop, sweet sorghum has the ability to become relatively dormant in response to water stress and th en resume growth under adequate moisture conditions (Miller and Ottman, 2010).
35 The suitability of sweet sorghum as a feedstock for ethanol production has been well described Miller and Ottman (2010) reported an average ethanol yield of 2,726 L/ha from the direct measurement of the juice sugars by high performance liquid chromatography (HPLC). When utilizing the total stalk nonstructural sugars, ethanol yield of M 81E, a cultivar developed for energy production, can reach 3,533 to 5,414 L/ha (Zhao et al 2 009). Interestingly, the average ethanol yield of sweet sorghum is comparable to the maize ethanol yield which is 3 751 L/ha for the United States (FAO, 2008) However, for a similar yield, sweet sorghum requires less water and nutrient s In addition, the amenability of sweet sorghum juice to fermentation compared to starch makes it a better candidate than corn for bioconversion. Harnessing the potential of sweet sorghum for the production of other biofuel s including biogas, is at its infantile stage. The Relevance o f Anaerobic Digestion Technology t o Haiti Located in the Northern hemisphere within the Caribbean basin, Haiti has a tropical climate with an average annual temperature of 25 C. The country receives an annual rainfall of 1400 mm, which is unev enly distributed among different regions. The history of the country is marked by political instability and non governance, which contribute in keeping the country in a cycle of poverty. Deterioration of living conditions in rural areas has led to a popula tion shift to urban areas, accelerating the urbanization process with se vere environmental consequences (UNEP, 2010). As one of the most populated countries of the region, human stress on land and demand for wood and charcoal as fuel have caused severe def orestation leading to extensive erosion. As depicted in F igure 1 4 7 0 % of the energy consumption in Haiti comes from biomass.
36 Every year, an estimate d 12 to 30 millions trees are cut down to provide energy to households and traditional industries (IBRD/WB 2007). Figures describing water and sanitation situations in Haiti are alarming. According to WHO (2010 ), approximately 56% and 28% of people living in urban areas ha ve access to improved water source s and improved sanitation, respectively. On the other hand, only 14% of people living in rural a reas have access to improved sanitation while access to improved drinking water is only 60%. The remaining population use shared sanitation facilities, unimproved facilities, or practice open defecation. The world health organization defines an improve d sanitation facility as one that hygienically separate s human excreta from human contact An improved drinking water source is conceived to be adequately protected from outside contamination particularly from fecal ma tter. Haiti also suffers from a deficit of municipal waste management. O nly 35% of the 4 100 m 3 of daily household waste generated in the largest Haitian cities are systematically collected (UNEP, 2010). Moreover, Haiti does not currently have a functionin g sewage system. This lack of waste management impacts the local environment and leads to widespread waterborne disease. In addition to waste management issues within Haiti the Haitian agricultural sector also struggles with providing adequate services w hile protecting the local environment. Account ing of rural households, the agriculture sector face s compelling challenges (Dolisca and Jolly, 2008). The continuous cultivation of the soil in absen ce of any nutrient restoration practices coupled with accelerated erosion due to unsustainable land use practices have resulted in diminished agricultural productivity. Additionally, lack of access to
37 mineral fertilizers, lack of agricultural infrastructur e, post harvest losses and poor economic and technic al assistance are among the various barriers that need to be addressed for the enhancement of agricultur al stability in Haiti. Understandably, there is no magical solution to the complex situation of Hai ti; however, any approach that aims at improving the environment while promot ing economic growth should be favored. Implementation of anaerobic digestion technology in Haiti could provide a sustainable solution to the sanitation problem and produce a clean burning fuel that can replace wood and charcoal. The biogas generated from farm and household wastes can be used directly for cooking and lighting or converted to electricity. Replacing woody biomass by biogas will not only contribute to decrease deforest ation but also reduce the occurrence of smoke related diseases due to the burning of unprocessed biomass. Furthermore, utilization of ADE as fertilizer will contribute to restor ing soil fertility and increas ing vegetable production yields. Thesis Rational e Food and energy security along with adequate treatment of domestic wastes are longstanding issues for Haitian households A holistic approach that simultaneously addresses the need for sustainable fo od production, clean burning fuels, and eco sanitation is of prime importance to promote prosperity in rural communities. Anaerobic digestion technology offers a sustainable solution to environmental problems of waste management provides clean burning fuel, and sustains food production This technology is par ticularly suitable for remote areas in Haiti that suffer from an ab sence of energy infrastructure. Therefore, a naerobic digestion can help reduce poverty and support sustainable development.
38 Utilization of anaerobic digester effluent as a nutrient source offers a sustainable way to produce low cost fertilizer that can supplement or displace high cost conventional fertilizer s Application of ADE to the soil offers multiple benefits that can be integrated into sustainable agricultural practices. However, ADE is not well known and its agronomic application remains a n under explored field in modern agricultural research. The few studies related to the fertilization value of ADE reported in the literature ha ve not fully examined whether ADE is more appropriate as a fertilizer soil amendment or both. Additionally, data related to the effects of ADE on growth and yield of most common vegetables are limited and inconsistent Integration of food and energy production can also be realized by the use of multipurpose c rops to simultaneously produce feed, food and fuels. In an integrated scenario, anaerobic digester effluent can also be used a s a fer tilizer to grow multipurpose crops and contribute to close the food and energy production cycle. Sweet sorghum is an exampl e of such a crop. Sweet sorghum is among the top biomass and sugar yielding crops (Yu et al 2012). Its biomass production can be fractionated into food and fuel components. Utilization of the nutritional part of the plant as food/ feed and the sugar and fibrous structural elements of the plant for bioenergy production maximizes the biomass production and helps address both energy and food production issues. There is an extensive literature describing e thanol production from sweet sorghum juice and various research projects are under way to maximize yield by producing ethanol from diff erent components of the plant. However, recent literature about the methane potential of different sorghum plant materials is limited. Different studies suggest that anaerobic digestion is one of the most energy efficient, as well as
39 environmentally sound ways of converting biomass to biofuels (Frigon and Guiot, 2010) The composition of diverse plant materials in terms of structural and non structural carbohydrates needs to be elucidated for assessing their sui tability for biogas and for feed or food production. In addit ion, determining the methane index of different sweet sorghum plant materials is an e ssential step in determining the possibility of using th is crop to produce b iofuel in the form of methane gas Hypothes e s A naerobic digeste r effluent contains plant nutrients in a useable form that can be used as a substitute to conventional fertilizers for sustainable production of onion ( Allium cepa L.) maize ( Zea mays L. ) and common bean ( Phaseolus vulgaris L.) production. S weet sorghum biomass components will have a favorable methane index so that th is crop can be used to simultaneously generate renewable fuels and protein for Haiti and other developing countries in need of b oth of these resources. Objectives The purpose of this study is to explore the applicability of anaerobic digestion of biomass into the Haitian farming system to generate bioenergy and sustain food production. This study was primarily aim ed at: 1) A ssessin g the suitability of sweet sorghum as a multipurpose crop for biofuels and food production in the context of Haitian farming systems and 2) Characterizing ADE and assessing its effect on the growth and yield of common vegetables grow n in Haiti such as onio n, corn, and bean. The effluent of an anaerobic digester treating organic residues was analyzed for major plant nutrients. The yields of bean, corn and onion were compared in a complete randomized field study under different fertilization regimes: anaerobi c digester effluent ( ADE ) urea, fish emulsion, and a control group receiving no fertilizer.
40 Table 1 1. Methane yield of animal manures and other organic wastes Feedstock Methane yield (mL/g of VS) Reference Pig manure 356 Moller et al 2004 Sow manure 275 Moller et al 2004 Poultry 245 Moody et al 2011 Paper (office) 369 C h ynoweth, 1999 Primary sludge 590 C h ynoweth, 1999 Food waste 396 Zhang et al 2011
41 Table 1 2 Methane yield of selected agricultural wastes Feedstock Methane yield (mL/g of VS) Reference Banana peels 227 Nallathambi, 2004 Mango peels 372 Nallathambi, 2004 Lemon pressings 473 Nallathambi, 2004 Rotten tomato 298 Nallathambi, 2004 Potato peel 267 L e steur et a l 2011 Onion culls 400 Nallathambi, 2004 Maize silage a nd straw 312 Mumme et al 2012 Cabbage 277 Ward et a l. 2008 Coffee waste 255 L e steur et a l 2011 Cassava 259 Zhang et al 2011 Wheat straw 227 249 Raposo et al 211 Table 1 3 Theoretical characteristics of typical substrate components. Adapte d from Angelidaki and Sanders, 2004 Substrate Composition VS/COD ratio CH 4 Yield mL/g VS CH 4 yield mL/g COD Carbohydrate (C 6 H 10 O 5 ) n 1.19 0.415 0.35 Protein C 5 H 7 NO 2 1.42 0.496 0.35 Lipids C 57 H 104 O 6 2.90 1.014 0.35 Ethanol C 2 H 4 O 2 2.09 0.730 0.35 Acet ate C 2 H 4 O 2 1.07 0.373 0.35 Propionate C 3 H 6 O 2 1.51 0.530 0.35
42 Figure 1 1. Environmental and social benefits from implementing anaerobic digestion. Adapted from Rajendra et al 201 2.
43 Figure 1 2. Integration of a naerobic digestion into farming sys tems, providing both fertilizers and energy. Adapted from Wilkie, 2008.
44 Figure 1 3. Sequential metabolic phases in anaerobic digestion Long chain fatty acids, intermediates Carbohydrates, lipid s, proteins Sugars, fatty acids, amino acids Acetate Methane, CO2 CO2, H2 1. Hydrolysis 2. Acidogenesis 3. Acetogenesis 4. Methanogenesis Acetate
45 Figure 1 4. Energy consumption by fuel type in Haiti. Adapted from IBRD/WB 2007. 55% 11% 4% 2% 6% 19% 1% 2% Firewood Charcoal Bagasse LPG Petrol Kerosene Diesel Fuel Oil
46 CHAPTER 2 MATERIALS AND METHOD S Characterization o f Anaerobic Digested Effluent Anaerobic digested effluent (ADE) was obtained from a n anaerobic digester fed organic residues (kitchen wastes and food), located at the Energy Research and Education Park of the University of Florida, Gain esville, Florida. The anaerobic digester was fed food wastes at a standard loading rate of 2 kg/m 3 /day and operated under mesophilic conditions with a hydraulic retention time of 20 days. Chemical oxygen demand, TS, VS, pH and conductivity of the ADE were measured using the same procedures aforementioned for these parameters. Oxidation reduction potential (ORP) of ADE was also determined prior application according to the standard methods (APHA, 2005) Nutrient Analysis of Anaerobic Digester Effluent Total nitrogen Total nitrogen was analyzed using the persulfate method, in coherence with standard methods (APHA, 2005). This method converts all nitrogen contained in the sample to nitrate. A colorimeter was used to measure the nitrogen concentration, which is proportional to the intensity of the yellow color formed at the end of the reaction. A representative sample of ADE was obtained and fully homogenized using a 360 mL stainless steel blender (Waring, Torrington, CT) for 1 minute on high speed. The entire h omogenized sample was poured into 600 mL glass beakers and fully homogenized with magnetic stir bars. Using a transfer pipette and volumetric flask, a 50 mL aliquot was obtained from each sample and diluted to 500 mL with DI water in a 500 mL volumetric fl ask for a dilution of 1 mL of ADE/10 mL. From the diluted sample a 50 mL
47 aliquot was obtained and further diluted to 500 mL with DI water in a 500 mL volumetric flask for a final dilution of 1 mL of ADE/100 mL. A 0.5 mL aliquot of the 1/100 diluted sample was added to Hach TN reagent tubes (10 150 mg N/L) along with potassium persulfate and sodium hydroxide as the digestion agents. Tubes were digested for 30 minutes at 103C in a Hach Model 45600 COD reactor (Hach Company, Loveland, CO). Digested tubes were cool ed to room temperature. Two reagents containing sodium metabisulfite were successively added to the digested tubes. Furthermore, 2 mL of the treated samples were transferred to Hach TN reagent tubes (HR 10 150 mg/L) containing sulfuric acid. At the en d of this reaction the sample turns yellow and the color concentration was determined using a Hach 890 colorimeter (Hach Company, Loveland, CO). Total ammonia nitrogen Total ammonia nitrogen was used to determine the TAN: TN ratio as an indication of the r eadily available nitrogen for plant uptake. Total ammoniacal nitrogen was measured using an ion selective electrode (Orion 95 12) with an Orion IonAnalyser 701A meter in accordance with standard methods (APHA, 2005). The procedure raised the pH of the samp le to >11.0 with a sodium hydroxide/EDTA solution (Orion ISA pH adjusting solution). The probe was calibrated using three standard solutions of 10, 50 and 100 mg NH 3 N. A standard curve was developed by plotting mV read by the probe in the y axis and NH 3 N concentration on the x axis. For measuring NH 3 N concentration of ADE, 100 l of ISA solution were added per 5ml sample. The probe was placed in the sample and allowed to stabilize before recording the mV concentration, which was used to determine the NH 3 N of the ADE.
48 Total phosphorus Total phosphorous was analyzed using the vanadomolydophosphoric acid colorimetric method in accordance with the standard method (APHA, 2005). In this method, all phosphate is converted into orthophosphate, which reacts with vanadium to form yellow heteropolymolybdophosphoric acid. The yellow concentration of this complex is proportional to phosphate concentration. A representative sample of ADE was obtained and was fully homogenized using a 360 mL stainless steel blender (War ing, Torrington, CT) for 1 minute on high speed. The entire homogenized sample was poured into 600 mL glass beakers and fully homogenized with magnetic stir bars. Using a transfer pipette and volumetric flask, a 100 mL aliquot was obtained from each sampl e and diluted to 500 mL with DI water in a 500 mL volumetric flask for a final dilution of 1 mL of ADE/5 mL. A 5 mL aliquot of the 1/5 diluted sample was added to Hach Phosphate reagent tubes (HR 0 100 mg N/L) along with one potassium persulfate powder pil low. Tubes were digested for 1 h at 150C in a Hach Model 45600 COD reactor (Hach Company, Loveland, CO). Digested tubes were cooled to room temperature. Two mL of 1.54 N sodium hydroxide and 0.5 mL of molybdovanate reagent were successively added to each digested sample. Treated tubes were read on a Hach 890 colorimeter (Hach Company, Loveland, CO). Elemental potassium Elemental potassium was measured using an ion selective electrode (Clean Grow Ltd. Cork, Ireland) according to standard methods (APHA, 2005 ). The probe was calibrated using standard solutions of 10, 50, and 100 mg/L of potassium. The calibrated probe was used to determine the potassium concentration of the ADE.
49 Economic analysis The nutrient concentration of ADE was used to determine the econ omic return if ADE is used as a fertilizer. Since nitrogen and phosphorus are the two key elements and are in different proportion within the ADE, the application could be formulated either on a nitrogen or phosphorus basis. Fertilizer recommendations for o nions, beans and corn (Table 2 1 and 2 2 ) and current fertilizer prices were used to determine the economic value of ADE (USDA, 2013). The commercial value of N, phosphorus and potassium was estimated using urea ($550/ton), superphosphate ($665/ton), and potassium chloride ($647/ton). Field Experiment The field experiment was conducted at the Energy Research and Education Park of the University of Florida, Gainesville FL. The tested crops in this experiment were onion ( Allium cepa cv ean ( Phaseolus vulgaris cv corn ( Zea ma ys cv 2012 on a raised bed of 1.2m width and 6.7m length. Bulbs were sown at 15 cm between plants in rows 20 cm apart, leading to a p lant density of 30,000 plants per ha. Bean seeds were planted one month later in a similar raise d bed divided into three rows. Beans were sown in raised beds in rows 60 cm apart, and 8 cm within plants while corn seeds were sown in rows 90 cm apart and 60 cm within plants. A fter the germination, new seeds were sown to fill the gaps within the plant population. Corn was planted on April 15th on a similar raised bed divided into two rows. A complete randomized design consisting of 4 treatments and 3 replicati ons were used to assess the effect of ADE on the yield and growth of the vegetables used in this experiment.
50 Soil samplings Soil samples were collected prior to and after the experiment for routine soil test analyses. Before the experiment, soil samples (0 20 cm) were taken in ten different random locations around the field and were combined to form a composite sample. The soil samples were dried in a forced air oven at 50 o C, pass ed through a 2mm screen, and analyzed for TN, TP, and K. At the end of the e xperiment, a composite soil sample of each treatment was taken for postharvest analysis. Soil testing results served to assess potential changes that occurred in the soil status by different treatments Fertilization Fertilization treatments included anaer obic digester effluent (ADE), fish emulsion (FE), urea, and a control treatment receiving no fertilizer. Commercially available fish emulsion with a guaranteed analysis of 5 1 1 (5% N, 1% P 2 O 5 1% K 2 O) (Alaska Fish Fertilizer, Lilly Miller Brands, Walnut C reek, Ca.), and analytical grade urea (46 0 0) (46% N, 0% P 2 O 5 0% K 2 O) (Fisher Scientific Corp., Fair Lawn, NJ) were used as comparative treatments. The fish emulsion is composed of 15% of water insoluble nitrogen, 10% ammonium nitrogen and 75% other wate r soluble nitrogen (Lilly Miller, 2013) Results from the soil testing prior to planting and IFAS recommendations for onion, bean and corn were used to dictate the application rate Nitrogen recommendations for bean, corn and onions are 120, 168 and 168 kg /ha, respectively. All fertilizers were split into two applications. The first half was applied four weeks after emergence of onion plants and three weeks after emergence of bean and corn. The second half was applied two months later for onions and 1 month later for bean and corn. Since ADE was in a liquid form, the urea was dissolved in water and the fish emulsion was diluted in water in order to have a uniform volumetric application. The
51 control treatment received an equivalent amount of water to e nsure s imilar soil moisture content. The amounts of ADE, FE, and urea applied to each crop are presented in Tables 2 1 and 2 2 Crop management followed agronomic practices commonly used for vegetables. Irrigation water was supplied two days a week during first t wo weeks following sowing and adjusted to one day per week for the rest of the experiment, using a drip irrigation system. The irrigation system was setup to deliver approximately 30 mm of water per week for all tested crops. Weed removal was regularly car ried out by hand when necessary. Plant sampling Onion Growth parameters including plant height and number of leaves per plant were measured two months after emergence. Plant height was measured from the soil surface to the top of the longest leaf of the plant. All plants were harvested when more than 50% of tops had dropped. Harvested plants were transported to the laboratory and were brushed to remove all soil particles and dried for two days. Fresh weight and bulb diameter were measure d and recorded. O nion plants were separated into bulbs and leaves for the measurement of total biomass and bulb yield. Bulbs were further divided into marketable yield and non marketable yield. Marketable yield were onions equal to or greater than 20mm in diameter. Non ma rketable yield were onions less than 20mm in diameter and those that were rotten or damaged. All collected data were computed for each plot and reported in Mg/ha. After the two day drying, a random bulb sample of each treatment was selected for the solub le content and tissue analysis. Bulbs were split in two halves. Tissues of one half were blended and the juice was used to determine soluble contents (% brix)
52 using a refractometer ( Atago Co. LTD, Japan ) (Russo, 2008). The other half of the bulb was cut in to small pieces and placed in paper bags and dried to constant weight at 70 o C. Corn and bean At maturity all bean and corn plants were harvested and fresh total biomass was recorded. All bean samples were dried at 70 o C until constant weight to determine dr y weight. D ry yield was partitioned in pods; grain and residues for each treatment. After recording the total fresh weight of corn, all plants were partitioned into ears and stalks. A sample of stalks and all the ears were dried at 70 o C until constant weig ht to determine the total dry yield. Corn grains were manually removed from the ears and were recorded to determine the grain yield. All collected data were computed for each plot and reported in Mg/ha. Tissue analysis Samples of dried onion bulb s, bean s, and grains of corn were ground with a hammer mill (Arthur H. Thomas Co, Philadelphia, PA, USA) w ith 2 mm and 0.85 mm mesh screen, successively Ground samples were sent to the analytical research laboratory of the University of Florida and analyzed for TK N, TP, and K, according the above mentioned procedures. Total nutrient concentration of onion b ulb s bean grains and corn grain s was calculated by multiplying dry matter yield by nitrogen concentration of each sample.
53 Characterization of Sweet Sorghum as a Multipurpose Crop Sweet sorghum was planted in May 201 1 at the energy grass field located at the Energy Research and Education Park of the University of Florida Gainesville F lorida The cultivar selected was M 81E, a late maturing cultivar developed i n Mississippi for use as potential bioenergy crop (Erickson et al 2011). Sorghum seeds were sown at a depth of 5 cm in rows with an approximate within row spacing of 5 to 8 cm and 91 cm between rows. The field consisted of 5 rows that were 18.2 m long. A fter emergence, sorghum plants were thinned to an average plant population at harvest of about 13.7 plants per m 2 After planting, sweet sorghum plants were irrigated every week using a drip irrigation system set up to provide 20 mm of water per week durin g the establishing period and were watered as needed once plants were established Weeds were removed manually using a garden hoe two times during the growing season. All plants were fertilize d using a blended granular 6 6 6 fertilizer (6%N, 6% P 2 O 5 and 6 %K 2 O (Sunniland Corporation, Sanford, FL) at a rate of 100 kg N /ha Nitrogen fertilizer was in ammonium form. F ertilization rate was split into two applications: the first half was applied at planting and the second half 60 days after emergence. Sweet sor ghum plants were harvested when plants reached hard dough stage At this stage, the grains are fully mature and cannot be squeezed between fingers. Four subplots of 4 by 0.91 m in the inner rows of the central part of the field were delineated for sample c ollection to avoid edge effects Individual plants of each subplot were measured for total height, stem height, average internode length, and average internode diameter. All plants in the subplots were harvested and weighed in the field to measure fresh bi omass yield. Harvested plants were divided into leaves, stems and heads. Stems harvested in each subplot were pressed using a mill (Moenda de Cana,
54 Vencedora Maqtron, Brazil) for the juice extraction. The weight of the juice was recorded and used to determ ine juice weight/plant fresh weight and juice weight/plant dry weight ratios. The juice was filtered using a 0.25mm sieve to remove plant residues, and frozen at 20 o C until further analysis. Subsample s of each plant component w ere collected, weighed, and oven dried at 50 o C for 5 days Dry weight was used to calculate the fraction of each plant component of the total biomass and summed to give the total biomass on a dry weight basis Juice Biomass, a nd Sugar Yield Determination The total biomass, juice, an d sugar yield were determined following the procedures used by Houlou and Stevens (2012). Juice extraction ratio and j uice yield were determined by the following equations: (2 1 ) (2 2 ) Similarly, the extracted stalk content and the extracted stalk yield can be calculated by: (2 3 ) ] (2 4 ) The dry weights of plant components were determined by multiplying the percent dry weight of each component by its fresh weight. S ugar yield was determined by multiplying the juice yield by the Brix concentration Sweet sorghum plant materials including leaves, heads, extracted stalks, chipped stalks and whole plants were used to characterize sweet sorghum as a potent ial candidate for biogas and food/feed production. Triplicate samples of each plant material were initially dried in an oven at 50 0 C for 5 days. Samples were further ground in a
55 cutting mill (Arthur H. Thomas Co, Philadelphia, PA, USA) w ith 2 mm and 0.85 mm mesh screen s successively. Ground samples were sealed in plastic bags at room temperature until their utilization Physicochemical Parameters Total Solids a nd Volatile Solids Ground samples were analyzed for total solid (TS), volatile solid (VS), and fiber analysis. Total solids (TS) and volatile solids (VS) were measured according to standard methods ( APHA, 2005 ) Triplicate representative samples (7 to 20 g) of different plant components were weighed into pre ashed, pre weighed 200 mL disposable aluminum dishes. Samples were dried at 103C in a drying oven (Precisi on Model STG 80, Thermo Fisher Scientific, Waltham, MA) for 24 h. Dried samples were placed in a desiccator to cool to room temperature, and then weighed to record TS. Volatile solids were measured by ashing the dried samples for 2 h using an ashing furnac e (Thermolyne 30400, Thermo Fisher Scientific, Waltham, MA) at 550C. Ashed samples were placed in a desiccator to cool to room temperature and then weighed. Volatile solids were calculated by subtracting ash weight from TS. Fiber Analysis Samples of each dried, ground non juice plant component were sent to the forage evaluation support laboratory of the University of Florida for fiber analysis, in vitro organic matter digestibility (IVOMD), crude protein (CP), and total phosphorus (TP). C rude fiber (CF) wa s analyzed in accordance to the ANKOM Technology Method 10. Crude fiber is the organic residue remaining after digesting the sample with 0.255N H 2 SO 4 and 0.313N NaOH. The compounds removed are predominantly protein, sugar, starch, and lipids and portions o f both structural carbohydrates and lignin. Crude fiber
56 contains mostly cellulose, and a small portion of the lignin but no ash. The CF content is an indication of the fibrous part of the feedstock and generally represents the indigestible fraction of carb ohydrates in the feed. However, some of the crude fiber components can be partially digested by microorganisms in the rumen. The lower the CF, the more suitable the feedstock is for both anaerobic digestion and animal feed. The non fiber content was calcul ated by subtracting the crude fiber content from the dry matter. Neutral Detergent Fiber Neutral detergent fiber (NDF) was determined using the method described by Golding et al (1985). The procedure consists of boiling a 1 g sample in a neutral detergent solution for 1 h. The residue or insoluble fraction left in the neutral detergent solution is the cell wall constituent or NDF. The NDF gives a better estimation of the total fiber constituents of the sample, since it measures cellulose, hemicellulose, li gnin, silica, tannins and cutins. Although some portions of the NDF can be digestible, NDF concentration is in general negatively correlated with feed intake and therefore can be used to predict the quality of the feedstock for animal feed and for anaerobi c digestion. Acid Detergent Fiber Acid detergent fiber was determined according the ANKOM Technology Method 8. This procedure uses an acid detergent solution composed of 20 g cetyl trimethylammonium bromide (CTAB) and 1 L 1.00N H 2 SO 4 to digest the sample. The remaining residue is the acid detergent fiber component of the feedstock and represents the least digestible fraction. Acid detergent fiber includes lignin, cellulose, silica, and insoluble form s of nitrogen but not hemicellulose. Neutral detergent fi ber and acid detergent fiber were used to determine hemicellulose and cellulose/lignin content.
57 Hemicellulose content was obtained by subtracting ADF from NDF while cellulose/ligni n content was calculated by subtracting hemicellulose from ADF. In V itro O rganic Matter Digestibility In vitro organic matter digestion was conducted using a modification of the two stage technique (Moore and Mott, 1974). In this procedure, the sample was incubated with rumen microorganisms for 48 h followed by 44 h incubation w ith acid pepsin. The IVOMD represents the portion of the dry matter in a feed that can be digested by animals at a specified level of feed intake. A higher IVOMD value indicates a better feedstock quality for both animal consumption and anaerobic digestion Nitrogen For total nitrogen analysis, non juice components were digested using a modification of the aluminum block digestion procedure of Gallaher et al. (1975). This method is a modification of the standard Kjeldahl procedure, which determines the tot al N of the sample, including organic (e.g., protein and non protein) and inorganic (e.g., nitrate) nitrogen. A 0.25 g sample was digested for at least 4h at 375C using 6 ml of H 2 SO 4 and 2 ml H 2 O 2 the catalyst used was 1.5 g of 9:1 K 2 SO 4 : CuSO 4 Nitrogen in the digestate was determined by semiautomated colorimetry (Hambleton, 1977). The total N present in the sample was used to determine the total amount of protein by multiplying the total N by 6.25. Total Kjedahl nitrogen of the juice was measured accord ing to EPA method section 351.2. Ammonium nitrogen (NH 4 N) was determined by a modification of EPA Method section 305 1. Total Phosphorus For the TP determination, ground samples were digested using a modification of the aluminum block digestion procedur e of Gallaher et al. (1975). A 0.25 g sample was
58 digested for at least 4h at 375C using 6 ml of H 2 SO 4 and 2 ml H 2 O 2 the catalyst used was 1.5 g of 9:1 K 2 SO 4 : CuSO 4 Phosphorus in the digestate was determined by semiautomated colorimetry (Hambleton, 1977 ). Total phosphorus and of the juice were measured according to EPA methods section 365 1. Characterization o f Sweet Sorghum Juice pH, Electrical Conductivity and Brix The juice obtained from pressing stems of harvested sweet sorghum was mixed thoroughl y a nd sieved through a 20 particles and submitted to basic analysis. Electrical conductivity and pH were measured using an Accumet Model 30 conductivity meter (Thermo Fisher Scientific, Waltham, MA), and Accumet Mo del 10 pH meter (Thermo Fisher Scientific, Waltham, MA), respectively. The Brix content (%) of the juice was measured using a po cket refractometer (Atago Co LTD, Japan) (Houlou and Stevens, 2012). Between two different sample readings, the refractometer w as cleaned with deionized water and dried with a paper towel. Juice samples were kept at 20 o C to preserve the sample for additional analysis Stored sweet sorghum juice was thawed and analyzed for TN, T P, and ammonium nitrogen according to above described methods. Alkalinity Alkalinity was measured in SS juice following APHA (2005) procedures. A 0.1N solution of H 2 SO 4 was s tandardized to a value of 0.11 N by titrating a known concentration of Na 2 CO 3 to a pH of 4.5 Alkalinity was calculated in mg CaCO 3 eq uivalents by the following equation : (2 6)
59 Potassium Elemental potassium was measured using an ion selective electrode (Clean Grow Ltd. Cork, Ireland) according to standard methods (APHA, 2005). The probe was calibrated using standard solutions of 10, 50, and 100 mg/L of known potassium concentration The calibrated probe was used to determine the potassium concentration of the ADE Chemi cal Oxygen Demand Total COD was measured using a modification to standard methods ( APHA, 2005 ) A representative sample of the juice was obtained and fully homogenized using a 360 mL stainless steel blender (Waring, Torrington, CT) for 1 minute on high speed. The entire homogenized sample was poured into 600 mL glass be akers and stirred on a magnetic stir plate. Using a transfer pipette and volumetric flask, a 5 0 mL aliquot was obtained from each sample and diluted to 500 mL with de ionized water in a 500 mL volumetric flask for a dilution of 10%(v/v). The diluted sample was further poured into 900 mL glass beakers and stirred on a magnetic stir plate. Using a transfer pipette and volumetric flask, a 5 0 mL aliquot was obtained from each sample an d diluted to 500 mL with de ionized water in a 6 00 mL volumetric flask for a final dilution of 1 00 %(v/v). Furthermore, a 50 mL aliquot was obtained from the 1/100 and diluted to 250 mL with de ionized water to obtain a final dilution of 1/500%(v/v). A 2 mL aliquot of the 1/ 500 diluted sample was added to Hach COD reagent tubes (HR 20 1500 mg COD/L). Tubes were digested for 2 h at 150C in a Hach Model 45600 COD reacto r (Hach Company, Loveland, CO).
60 Methane Index Test A methane index test ( MIT ) was conducted to assess the methane yield of different biomass materials used in this expe riment. The MIT assay was performed in accordance to the method describe by Wilkie et al (2004). The inoculum used in the MIT assay was a mixture of flushed dairy manure obtained from the University of Florida Dairy Unit in Hague, Florida and effluent of an operating anaerobic digester fed food waste and fibers. The resultant inoculum was composed of 90% flushed dairy manure and 10% food waste digester effluent by volume. In order to reduce the fraction of large particles and have a homogeneous inoculum, t he inoculum was completely mixed and screened through a 20 Newark W ire C loth C o., Newark, N.J). The pH of the mixed screened inoculum was 7.6. Samples of dried, ground material was added to individual 250 mL serum bottles along wi th 200 mL of the inoculum solution, which corresponded to a loading rate of 2g of volatile solid per liter. The amounts of sample added in the MIT bottles were 0.405 0.002 g, corresponding to 0.366 0.006 g of VS. In order to quantify endogenous methan e production from the inoculum, a negative control (blank) was used with inoculum only (no substrate addition). The MIT of the juice was conducted using three different OLR: 2 g COD/l, 1 g COD/L, and 0.5 g COD/L. The amount of juice sample added for each O RL is presented in Table 2 3 Additionally, a MIT assay was performed with avicel cellulose and glucose serving as positive control s Serum bottles were fitted with a rubber stopper and sealed with aluminum caps to maintain airtight anaerobic conditions All triplicate samples were inverted to prevent possible leaking and placed in an incubator at 35 C. All samples were periodically m easured for gas production for 60 days.
61 Methane Production Measurement The hydraulic displacement method described by Wilkie et al (2004) was used to measure the volume of methane production from the MIT bottles using an apparatus consisting of 250 mL bottles filled with a KOH solution connected to a 50 mL pipette (Figure 2.2). For the reading of the methane production, MIT bo ttles were connected to a 250mL serum bottle filled with 5M KOH. This solution dissolved the carbon dioxide content of the biogas passing through it, so that only the methane proportion of the biogas generat ed pressure on the headspace of the MIT bottle A s the pressure increased, an equivalent volume of liquid was displaced proportional to the methane produced. The volume of methane is therefore the reading of the volume of liquid displaced. A concentration of 1g/L of alizarin was added to the KOH solution as a pH indicator to monitor for carbon dioxide saturation. Methane production from each sorghum component was adjusted by subtracting the methane volume of each bottle from the mean of inoculum control and was converted to standard temperature and pressu re (STP) (0 0 C and 760 mm Hg). Daily production was cumulated and adjusted per g of VS loaded. Statistical Analysis Cumulative methane production was calculated for each plant component. Mean and standard deviation were determined for each sample. Cumulat ive methane production was fit to a first order kinetic model using non linear fitting. ( 2 7) Where: CH 4 = cumulative methane production at time t (mL/g VS loaded @STP) CH 4f =ultimate cumulative methane production (mL/g VS loaded @STP)
62 k=ki netic rate constant (day 1 ) The solver application in Microsoft Excel 2011 was utilized to solve for CH 4f and k by minimizing the residual sum of squares of triplicate data points. Standard error of each parameter was calculated using the square root of th e variance. Significant differences between mean values of different parameters were determined using a T using JMP statistical software from SAS. T he methane yield of each plant component obtained from the MIT was used to determine the methane yield of that component per hectare. Methane yield of all components was summed to calculate the met hane yield of the whole plant. Additionally, the methane potential of sweet sorghum as a dedicated energy crop was evaluated using three different scenarios. In the first approach all the plant biomass were dedicated to anaerobic digestion. In the second a pproach, the heads are used as animal feed or human food, the leaves are left in the field for increasing soil organic matter and nutrient content, the stem (juice and extracted stalk) was used for methane production. In the last scenario, only the juice i s used for methane production, the extracted stalk is used for animal feed and the other parts are used as described in the second scenario.
63 Table 2 1 Fertilizer application volumes and total N application rate for onion and corn Fertilizer N content ( %) Application rate ( k g/ha) Experimental unit ( s f ) Total N added (g) Fertilizer added Urea 46 1 60 6.7 10 .2 22 (g) FE 5 1 60 6.7 10 .2 204 (g) ADE 0.5 1 60 6.7 10 .2 5 (gal) FE Fish emulsion, ADE anaerobic digester effluent Table 2 2 Fertilizer appli cation volumes and total N application rate for bean Fertilizer N content (%) Application rate ( k g/ha) Experimental unit ( s f ) Total N added (g) Fertilizer added Urea 46 1 20 6.7 8 .0 17 .4 (g) FE 5 120 6.7 8.0 160 .0 (g) ADE 0.5 120 6.7 8 .0 4.2 (gal) FE Fish emulsion, ADE anaerobic di gester effluent. Table 2 3. Amounts of feedstock and inoculum used in the MIT for Sweet Sorghum juice Sample ID ORL (g COD/L ) Volume inoculum (mL) Sample added Juice 1 200 1.07 mL Juice 2 200 2.14 mL Juice 0.5 400 1 07 mL Glucose 2 200 0.4 g
64 Figure 2 1. Partitioning of fiber in feedstock. Adapted from (Galyean 2010).
65 Figure 2 2. Apparatus for measuring methane production through hydraulic displacement. Components are: A) MIT Bottle, B) tube for biogas f low, C) needles in septa, D), 5M KOH with alizarin in 250 mL serum bottle, E) roller clamps for tube closure, F) tube for displaced KOH solution, G) 50mL pipette with 0.5 mL graduations, H) stand Adapted from (Wilkie et al 2004)
66 CHAPTER 3 CHAR ACTERIZATION OF ANAEROBIC DIGESTER EFFLUENT AS A FERTILIZER SOURCE FOR HAITI In Haiti, agricultural production accounts for 28 % of the gross domestic product (GDP) and involves mostly family own ed farms that supply more than 50% of national food production (Dolisca and Jolly, 2008) In the last two decades, Haitian farmers have not successfully maintained sufficient productivity because of accelerated degradation of the Haitian environment. A s a result of an unprecedented population explosion and severe exp loitation of the environment, agricultural productivity in Haiti suffers from constant soil quality depletion through human accelerated erosion processes. This situation, coupled to a lack of access to basic agricultural inputs such as fertilizers, improve d seeds, irrigation water and appropriate farming practices has contributed to keep ing the performance of the agriculture sector unsatisfactory (USAID, 2011). Agriculture in Haiti encompasses both cash and subsistence crops. Cash crops include plantain, rice, onion, and are mostly cultivated in irrigated areas and humid zones. Subsistence crops are mainly grown for family consumption and involve cereals (maize and sorghum), legumes (black beans, pigeon peas, and lima beans), t ubers (cassava and sweet pota to) plantain, fruits and fresh vegetables It is worthy to note that this categorization is not mutually exclusive. E x cess production of subsistence crops is often sold in the market and the farmer consumes cash crops as well. In the production systems, t hese two categories are intertwined together and commonly found on the same p iece of land. In a context of limited land resources, households tend to diversify their crops as much as possible to minimize their risk and ensure food production (Baro, 2002).
67 Many attempts have been made to address the weakness of the agricultural sector in Haiti. For instance, the government has been subsidizing fertilizers at 80% for more than a decade making the product more affordable to farmers. Despite this assistance, ch emical fertilizer is limited to a small number of farmers operating in irrigated land and humid hillsides where water is abundant. An average of 20,000 Mg of mineral fertilizers are used every year by only 5% of Haitians farmers mainly in rice producing ar eas and for high value vegetable crops (USAID, 2011). L ittle attention is being given to small acreage rain fed agriculture. Rain fed agriculture account s for 80% of agricultural production particularly traditional crops including beans ( Phaseolus vulgari s L), maize ( Zea mays L), cassava ( Ma n ihot esculenta ) and sweet potatoes ( Ipomoea batatas ). Organic and biological sources of nutrients have not been widely promoted in Haiti. Animal manures ha ve been used in the past but utilization of these resources h a s decreased considerably as a result of the declin e in livestock production. Although various institutions have been promoting composting of organic residues and utilization of nitrogen fixing tree s ( e.g. Leucena ) as methods to replenish nitrogen and orga nic contents of the soil, adoption of such alternative sources to sustain small farm vegetable production is limited. However, in the context of continuous increase s in the cost of synthetic fertilizers and pesticides, smallholder farmers must find alterna tives to unsustainable inputs that heavily depend on fossil fuels to successfully feed their communities. To sustain the food production system, farmers must adopt sustainable practices using alternative sources of soil nutrients; reduc e soil erosion, us e animal and green manures, mulch and compost to maintain soil fertility. Recycling of organic waste
68 through anaerobic digestion helps maintain soil nutrient status, stimulating various aspects of soil fertility. Anaerobic digestion (AD) of organic residues produces anaerobic digester effluent (ADE), a biofertilizer with elevated mineral nitrogen concentrations and organic matter (Tambone et al 2010). Along with other benefits, AD simultaneously supports crop production and environmental protection. Despit e its attractive benefits, implementation of AD is at its nascent stage in Haiti. In contrast to other organic fertilizers, ADE is not well known and its agricultural use remains an unexplored field in research. The efficient and appropriate use of ADE in crop production requires more in depth knowledge in terms of physical and chemical characteristics and fertilizer value. The purpose of this study was to assess the effectiveness of ADE as a nitrogen fertilizer. The effect of ADE on the yield and growth of onion, bean, and maize was compared in a field study with the response of a conventional synthetic fertilizer (urea) and an organic fertilizer (fish emulsion). The tested crops included maize, bean, and onion. Crops were chosen based on their importance i n the Haitian agricultural sector. Maize is among the top three cereals grown in Haiti. In fact, maize is grown by almost all Haitian farmers across all agro ecological zones. Maize production averaged 204,989 Mg per year representing about 40% of the tota l cereal production of the country. Farmers rarely apply fertilizers to corn and consequently obtain yields as low as 0.7 Mg/ha (USAID, 2011). Bean is extensively cultivated in humid and sub humid mountainous regions and in the irrigated plains of Haiti. A lthough its production has substantially increased in the last decades, bean yields are relatively low, averaging 0.7 Mg/ha. Onion is a cash crop and mostly cultivated in high elevation humid zones. Its yield is dependent on the usage of fertilizers.
69 Chara cterization and Nutrient Composition o f ADE Physico chemical characteristic s of ADE are presented in Table 3 1 The moisture content of the ADE was 98.98 0.05%. With a relatively high moisture content, the ADE used in this experiment can be classified as liquid material and can be directly used for fertigation to provide both water and nutrients. The pH (7.2 0.02 ) was slightly alkaline and corresponded to typical pH ranges of most commercial digesters (Alburquerque et al 2012). In addition, the low EC (4.5 0.06 mS/cm) of the ADE increases its suitability for land application with reduced risk of salinity accumulation and subsequent plant inhibition. The nutrient composition of the ADE is presented in Table 3 1 Total nitrogen (TN) total phosphorus ( TP ), and elemental potassium (K) were 502.5 16.7, 102.0 8.54, and 374.0 4 ppm respectively. Nutrient concentrations of the ADE were lower than values of most commercially digesters (Moller and Muller 2012). However, total ammoni acal nitrogen (TAN) was 301 9.13 ppm indicating that 60% of the total nitrogen was in a readily available form for plant uptake. While the ammonium fraction of the TN will be rapidly converted to nitrate in most soil conditions or assimilated directly by the plant roots, the organi c fraction will be mineralized overtime representing a nutrient reserve for the growing season. At an application rate of 25.4 mm/ha, the ADE can provide 120 k g N/ha, 56 k g P 2 O 5 /ha and 108 k g K 2 O/ha. At this rate, ADE supplies 100% of N 81% of K, and 42 % of the phosphorus recommended for beans An increase of the application rate to 57.76 mm/ha could provide all recommended nutrients for beans, maize and onions ( Tables 3 2 and 3 3) The cost of fertilizer for producing beans, maize, and onions is evaluated at $ 480, $540, and $625 US dollars/ha respectively Utilization of ADE will cover 70%, 88%, and 75% of the total
70 fertilizer cost for bean, maize and onion production, respectively, when the fertilizer application is formulated o n a nitrogen basis (Table 3 2) I f the application rate is formulated on a phosphorus basis, the ADE will cover all fertilizer cost s However, when the application rate is based on the phosphorus concentration of the ADE, nitrogen and potassium are over applied ( Table 3 3) In this experiment, soil testing prior to planting revealed that the site could provide the amounts of potassium and phosphorus essential to grow beans, maize and onions. Consequently, the fertilization rate was formulated to meet the nitrogen requirement only. E ffects o f ADE o n Vegetable Production Effects on Onion Yield The effects of different fertilizer treatments on onion yield are presented in Table 3 4 These results suggest a significant response (p<0.05) to all fertilized treatments when compared to the c ontrol (no fertilization). The maximum values of fresh total biomass (22.2 3.2 Mg /ha), total bulb (17. 5 2.4 Mg /ha) and marketable yield (12.0 2.5 Mg /ha) were found in onion plots treated with FE. A similar result was found in onions treated with ADE (P< 0.05) with 20.03 3.22 Mg /ha of total biomass, 15.21 1.22 Mg /ha of total bulb weight, and 11.03 Mg /ha of marketable bulb yield (Table 3 4 ). These results indicated that ADE stimulated an increase of 33.5%, 39.3% and 41% for the total biomass, total bulb a nd marketable bulb yield respectively, compared to the control group (Figure 3 1 ). The results of this experiment support the effectiveness of the application of ADE in increasing onion growth and yield. N o significant difference was found between the oth er types of fertilizer for all the response variables considered in this study.
71 Onions treated with ADE resulted in a greater concentration of TKN and phosphorus in bulb tissue ( F ig ure 3 2 ) compared to the other fertilizer s while those fertilized with fish emulsion contained higher concentration s of potassium in bulb tissue In terms of nutrient uptake (Figure 3 3 ), ADE and FE led to a greater P and K uptake as compared to urea and the control Onions treated with FE had significantly higher nitrogen uptake than the other treatments This finding suggests that a fraction of the yield obtained from ADE treatment can be explained by the K and P uptake. Nutrient content of the onion juice and characteristic s of the juice in terms of pH, EC and brix are presente d in F igure s 3 4 and 3 5 respectively. No significant different was found between the soluble content of onion bulbs suggesting that fertilizer sources were not a differentiating factor. Effect on Bean And Maize Yield Tables 3 5 showed the effects of ADE on the yield of beans. Bean total dry biomass ranged between 2.6 0.1 and 4.5 0.5 Mg/ha in all treatments tested. ADE yielded higher total biomass but not significantly higher than yields than either FE or urea. Similarly, the grain yield was identical am ong all fertilized plots. Compared to the control, bean yield increased by 40% with ADE, 30% with FE, and 45% with urea. Likewise, total biomass increased by 40% with ADE, 38% with FE, and 28% with urea (Figure 3 6). These results indicated that ADE applic ation enhanced bean yield to the same ext ent as urea and fish emulsion. Table 3 6 showed the effects of ADE on the yield of maize. Total biomass and grain yield of maize obtained from plants treated with the ADE, FE and urea treatments were higher than yie lds obtained in the control plots. The difference was not significantly different between plots treated with FE and the control fertilized plots, however, the
72 difference was significant between urea and ADE. The grain yield ranged between 7.7 and 13.8 Mg/h a within the fertilized plots. The yield of the control was 4.81 and was not significantly different than plants treated with FE. This may have resulted from the variation of the soil fertility of the site. In addition, because only 10% of the Nitrogen con centration of the fish emulsion was in a plant available form, the rate of nitrogen release may not be proportional to the nitrogen demand of maize. The grain yield obtained from this experiment was consistent with work by Loria et al (2002) who reported significant response of corn to raw and digested manure. However, the authors reported lower yield (6.12 to 7.84 Mg/ha) than the yield recorded in this experiment. Similarly, Chantighy et al (2008) found that liquid swine manure treated by anaerobic diges tion had a similar fertilizer value when compared to mineral fertilizer in terms of grain yield. Compared to the control, maize grain yield increased by 62% with ADE, 31 % with FE, and 6 1% with urea (Figure 3 7). The highest total biomass yield was obtained with ADE, representing an increase of 60% compared to the control. With respect to macronutrient contents of bean and maize grain, plants treated with ADE had grain TKN, TP, and potassium concentrations that were similar to the control values. However, AD E treatments reached higher nutrient concentration for bean than the control, however; the N concentration was not significantly different among various treatments. (Figures 3 9 and 3 11). The total N concentration in maize plants (Figure 3 9) was similar to the rates reported in other studies (Biau et al 2012). In addition, fertilizer treatment did not affect soluble content of onion or grain nutrient content of either bean or maize
73 Postharvest Soil Properties Differences between chemical and physical s oil properties before and after the experiment were not significant. As it is depicted in Table 3 10, most soil parameters did not change when compared with the value prior to the experiment irrespective of the fertilizer treatment. However, s oil TKN conce ntration was slightly higher in plots treated with ADE than the control plots (Table 3 7) for all tested crops. Total soil organic matter followed the same tendency indicating that application of ADE ha d the capability to increase soil organic matter. Soi l treated with ADE had also higher Ca, Mg, TP, and K concentration than the control. The results of this study were in accordance with those found by Lee (2010) who reported higher organic matter content, higher soluble phosphorus and greater K concentrati on in plots treated with organic fertilizers compared to unfertilized soil. Th is low change of the soil TKN concentration observed in plots treated with ADE corroborates the fact that the nitrogen component of the ADE is mostly in ionic form that is readil y available for plant uptake. Additionally, the soil organic matter of plots treated with ADE was significantly higher than the control but not significant ly different than the other treatments. Alburquerque et al (2012) reported no statistically signific ant difference for the soil TN concentration when comparing ADE and mineral fertilize r. Discussion The effects of ADE on the vegetative growth and yield of major crops commonly grown in Haiti were evaluated in this study. Nutrient concentrations of ADE fo und in this experiment support the assumption that ADE contains readily available plant nutrient s. A naerobic digester effluent showed positive fertiliz ation effects in onion bean, and maize production. The response of those crops in terms of vegetative g rowth, total
74 biomass and marketable yields to ADE application was significantly higher than the control (no fertilization). The results of this experiment indicated that ADE increased the yield of onion, bean and maize by 42.5, 40.5, and 65%, respectively compared to the control. These results are in corroboration with those found by Alburquerque et al (2012) who reported significantly greater response of plant development (watermelon) to ADE and mineral fertilizer compare d to the control. Likewise, no s tatistical difference was recorded on marketable yield of lettuce between ADE and traditional organic and mineral fertilizers (Montemurro et al 2010). Rajendran et al (2012) reported that ADE increased potato yield by 27.5% and forage by 1.5% compared t o plants that did not receive any fertilizer For such types of crop s ADE contains enough plant available N for vegetable production and therefore can be used as a substitute for synthetic nitrogen. An economical comparison between conventional nitrogen f ertilizer and the ADE suggested a 100% saving s in fertilizer cost when application is made on phosphorus basis assuming that the farmer produces the ADE on his farm The results of this experiment showed that anaerobic digestion could help farmers maximiz e nutrient recycling from organic wastes and reduce their reliance on inorganic fertilizer. Like all fertilizer applications, ADE application depends on specific crop nutrient demand and the nutrient concentration of the anaerobic digester effluent. This s tudy demonstrated that ADE was able to supply nutrient to high demand nitrogen crops such as maize. The TKN concentration in the soil at the end of the experiment did not differ significantly between the three fertilizer sources; however, soils treated wit h ADE had higher TKN concentration than the control The slight increase of soil nitrogen
75 concentration can be explained by the fact that a certain fraction of the total nitrogen concentration of ADE is in organic form and may not be completely mineralized during the growing season. Furthermore, only a 7% increase of TKN was found in soil trea t ed with ADE compare to the control plots. The relative ly high ammonium concentration of ADE indicates that ADE may be most beneficial for high N demanding crops with a short growing period specifically in organic vegetable production where fast release fertilizers are generally lacking (Moller and Muller, 2012). Additionally, application of ADE was effective at increasing soil organic matter. This improvement in soil o rganic matter is essential to sustain productivity and enhance biological and chemical soil properties. This is in agreement with Tambone et al (2009) who reported that ADE could act as a soil amendment. Anaerobic digester effluent can be used to enhance organic farming where inorganic fertilizers cannot be applied (Walsh et al 2012). In areas with poor soil quality, the application of nutrients (organic or inorganic) is necessary to sustain food production. Heilig and Kelly (2012) reported that dry bean s growing under organic production with no fertilizer inputs had 19.8% lower yield than beans growing under conventional production with nitrogen inputs. This study suggested that even though beans have the potential to fix nitrogen, application of ADE cou ld substantially increase the yield. The result of this experiment showed that ADE could be used to supplement elemental nutrients needed to increase yields in bean production. Loria et al (2007) evaluated the use of anaerobically digested swine manure as a nitrogen source in corn production. The results of this study suggest that farmers in developing countries such as Haiti can use anaerobic digestion as a method to recycle farm wastes and generate biofertilizer at low cost. They can substantially
76 increa se their yield by using higher fertilization rates, which would depend on the nutrient concentration of the ADE and the growing crops. They can preferentially digest wastes high in nitrogen content such as animal manure, urine, human waste, and biomass fro m legumes. Agoramoorthy and Hsu (2008) reported that after the launch of The improvement in the liv ing conditions in terms of reduction of usage of firewood kerosene and fertilizer, and improvement of hygiene and human health support the potential of AD to promote local development and a cleaner environment.
77 Table 3 1. Physico chemical characteristic of anaerobic digester effluent (ADE) Parameter Unit Value pH 7. 2 0.02 Moisture % 98.93 0.05 Total solids % 1.07 0.05 Total ash %TS 23.36 0.73 TOM %TS 76.64 0.73 COD mg/L 15000 1000 EC ms/cm 4.51 0.06 ORP mV 298 15 Nitrogen (total) mg/L 502.5 16.7 Phosphorus (total) mg/L 102.0 8.54 Elemental potass ium mg/L 374 40 TAN mg/L 301 9.13 TOM: Total organic matter, COD: Chemical oxygen demand, EC: electrical conductivity ORP: Oxidation reduction potential, TAN: total ammonia Nitrogen
78 Table 3 2. Economic value of ADE for bean, maize and onions at ap plication rate based on total nitrogen Nutrient Crops Recommended (kg/ha) Application rate (mm/ha) Supplied ( k g/ha) Additional needed ( k g/ha) Cost needed (US dollars) Economic value of ADE (US dollars) % Saving N Bean 120 25.4 120 0 143 143 100 P 2 O 5 Bean 134 25.4 56 79 194 81 42 K 2 O Bean 134 25.4 108 26 145 117 81 Total 105 482 341 70.7 N Corn 168 33.67 168 0 201 201 100 P 2 O 5 Corn 134 33.67 78 56 194 113 58 K 2 O Corn 134 33.67 152 0 145 163 100 Total 56 540 477 88 N Onion 168 33.67 168 0 201 201 100 P 2 O 5 Onion 168 33.67 78 90 243 113 46 K 2 O Onion 168 33.67 152 16 181 163 90 Total 106 625 477 76 Fertilizer cost was estimated using current market price of urea, superphosphate and potassium chloride
79 Table 3 3. Economic value of ADE for bean, maize and onion production at application rate s based on total phosphorus Nutrient Crops Recommended (kg/ha) Application rate (mm/ha) Supplied (kg/ha) Additional needed (kg/ha) Cost needed (US dollars) Economic value of ADE (US dollars) % Saving N Bean 120 57.76 288 0 143 345 100 P 2 O 5 Bean 134 57.76 134 0 194 194 100 K 2 O Bean 134 57.76 260 0 145 311 100 Total 0 482 850 100 N Corn 168 57.76 288 0 201 417 100 P 2 O 5 Corn 134 57.76 134 0 194 145 100 K 2 O Corn 134 57.76 260 0 145 280 100 Tot al 0 540 842 100 N Onion 168 72.4 361 0 201 432 100 P 2 O 5 Onion 168 72.4 168 0 243 243 100 K 2 O Onion 168 72.4 326 0 181 351 100 Total 625 1026 100
80 Table 3 4. Effect of fertilizer treatment on onion ( Allium cepa ) growth and yiel d Values are given as treatment mean followed by 1 standard deviation. Values followed by different letters are significantly different at p<0.05. EU: experimental unit Marke table yield are onion of diameter equal or higher than 20mm Table 3 5. Effect of fertilizer treatment on bean ( Phaseo lus vulgaris ) growth and yield Treatment Total biomass (g/EU) Grain yield (g/EU) Total biomass (Mg/ha) Grain yield (Mg/ha) ADE 280.0 131.5 4.5 2.11 FE 270.0 139.9 4.3 2.25 Urea 233.3 112.3 3. 7 1.80 Control 164.3 76.9 2.6 1.24 Values are given as treatment mean followed by 1 standard dev iation. Values followed by different letters are significantly different at p<0.05 EU: experimental unit
81 Table 3 6. Effect of fertilizer treatment on maize growth and yield Treatment Total biomass (kg/EU) Grain yield ( k g/EU) Total biomass (Mg/ha) Grain yield (Mg/ha) ADE 1.74 0.86 27.9 13.9 FE 1.01 0.48 16.3 7.7 Urea 1.87 0.80 30.1 12.9 Control 0.64 0.30 10.3 4.8 Values are given as treatment mean followed by 1 standard deviation. Values followed by different letters are significantly different at p<0.05 EU: experimental unit Table 3 7. Soil characteristic at time of planting Parameter Unit Value pH 6.25 0.15 SOM % 3.61 0.14 P mg/kg 110.5 20.5 K m g/kg 61 18 Mg mg/kg 1047 170.5 Ca mg/kg 212 6
82 Table 3 8. S oil properties after harvesting as affected by different fertilizer treatment (mg/kg) Treatment Crop P K Ca Mg TKN SOM (%) ADE Onion 88.84 37.6 847.2 226.4 1478 4.77 a Urea Onion 76.6 3 7.36 720.4 282 1469 4.34 b FE Onion 84.32 67.16 581.6 161.8 1373 4.40 b Control Onion 79.28 39.64 696 182.7 1445 4.00 c ADE Beans 115 71.2 1073 282 1285 4.46 a Urea Beans 88.84 37.6 847.2 226.4 1356 4.52 a FE Beans 92.04 31.28 845.6 219.2 1431 4.39 a ADE C orn 51 38 821 164 1353 3.68 a Urea Corn 48.24 25.11 732.42 159.64 1176.47 3.49 ab FE Corn 51.16 49.01 801.34 173.05 1530.36 3.63 a Control Beans 84.32 67.16 581.6 161.8 1373 4.34 b Control Corn 41.30 25.91 587.73 126.00 795.85 3.25 c
83 Figure 3 1. Incre ase in onion yield by fertilization treatment compared to the control (no fertilization) Figure 3 2. Nutrient concentration in dry onion bulbs by fertilization treatment Total Kjeldahl nitrogen (TKN), Total phosphorus ( T P), potassium (K). Values repres ent the average and error bars are 0 10 20 30 40 50 60 70 ADE Urea FE % Increased compared to the control Total Biomass Marketable yield 0 5 10 15 20 25 30 35 40 45 50 Urea ADE FE Control Tissue analysis of bulb (g/kg) TKN TP K
84 Figure 3 3. Nutrient uptake by onion bulbs by fertilization treatment Total Kjeldahl nitrogen (TKN), Total phosphorus ( T P), potassium (K). Values represent the average with one standard deviation Figure 3 4. Characteristics of onion juice by fertilization treatment 0 10 20 30 40 50 60 70 ADE FE Urea Cont Nutrient Concentration in onion tissues (kg /ha) TP K TKN 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Urea ADE FE Control pH Brix (%) C. E. (ms/cm)
85 Figure 3 5. Nutrient contents of onion juice by fertilization treatment in mg/L Figure 3 6. Percentage increase in bean y ield compared to the control (no fertilization) 0 500 1000 1500 2000 2500 3000 3500 4000 Urea ADE FE Control TN TP K 0 10 20 30 40 50 60 ADE Urea FE % increased compared to the control Total biomass Grain yield
86 F igure 3 7. Percentage i ncrease in maize yield compared to the control (no fertilization) Figure 3 8. Tissue a nalysis of bean ( grain ) by fertilization treatment 0 10 20 30 40 50 60 70 80 ADE Urea FE % Increased compared to the control Total biomass Grain yield 0 5 10 15 20 25 30 35 40 ADE FE Urea Cont Nutrient content (g/kg) TP K TKN
87 Figure 3 9. N utrient uptake of bean (grain) by fertilization treatment Figure 3 10. Tissue analysis of maize ( grain ) by fertilization treatment 0 10 20 30 40 50 60 70 80 ADE FE Urea Cont Grain nutrient uptake (kg/ha) TP K TKN 0 2 4 6 8 10 12 14 16 18 ADE FE Urea Control Nutrient content in corn grain (g/kg) TP K TKN
88 Figure 3 11. Nutrient uptake of maize (grain) by fertilization treatment 0 50 100 150 200 250 300 350 ADE FE Urea control Grain nutrient concentration (kg/ha) TP K TKN
89 CHAPTER 4 CHARACTERIZATION OF SWEET S ORGHUM FOR BIOFUELS AND FOOD PRODUCTION Sweet sorghum (SS) has been recog nized as an attractive bioenergy crop because of its high productivity, drought tolerance, and the versatility of its components. Its richness in directly fermentable sugars makes it a valuable alternative feedstock to most common bioenergy crops (Stefaniac et a l 2012) Sweet Sorghum stem contains approximately equal quantities of soluble (glucose and sucrose), and insoluble (hemicellulose and cellulose) carbohydrates For a comparable yield, SS consumes less water and fertilizer than most common bioenergy crop s According to Miller and Ottman (2010), irrigation frequencies did not affect the yield of the Cultivar M 81E, which ranged from 20 to 31 Mg/ha. However, Wortmann et al (2010) reported yields of 8 to 48 Mg/ha across different N fertilization rates. Swee t sorghum has been extensively evaluated for its sugar content and ethanol yield. Miller and Ottman (2010) reported an average of ethanol yield of 2 726 L/ha from the fermentation of juice sugars. Furthermore, Zhao et al (2009), evaluated the ethanol yie ld of both non structural carbohydrates and simple sugars, and reported estimated ethanol yield ranged between 3 533 to 5 414 L ha 1 Although the science and technology for ethanol production are technically available, implementation of fuel ethanol produ ction has not been economically feasible yet in many countries around the world. Many developing countries including Haiti lack the necessary infrastructures and cannot afford the high investment cost required to embrace ethanol production for transportati on fuel. Adoption of sweet sorghum in the Haitian farming system is at its nascent stage. Currently, it is bei n g explored as an alternative to boost local agriculture and supp ly a reliable animal feed source. Despite these attractive feat ures, characteriza tion of SS
90 plant biomass for biofuels in the form of methane coupled with food/feed production has not been explored. Therefore, it is necessary to investigate t he amenability of SS biomass for biofuels and feed/food production in terms of biomass partitio ning, proximate analysis, and methane production potential. Sweet sorghum plant biomass obtained from the experimental field at the Energy Research and Education Park at the U niversity of Florida, Gainesville FL., was partitioned into distinctive componen ts and analyzed for its potential methane p r o duction and its chemical characteristics. Crop Yield a nd Biomass Partitioning The characterization of various SS plant component s is presented in Table 4 1. Weights of fresh leaves, fresh heads, fre sh extracted stem, and juice were 0.084 0.02, 0.096 0.03, 0. 487 0.09, and 0.215 0.07 g/plant, res pectively. These results show that the stem accounted for 80% of the total fresh biomass of the plant, which was 0. 882 g while the leaves and the heads represented 9.5% and 10.5% of the total plant biomass, respectively. T he total dry biomass was 0.307 g/plant, suggesting that the total solid content of the plant was approxima tely 35 %. Total fresh biomass yield and dry biomass yield averaged 74.2 12.1 Mg/ha and 29. 0 4.9 Mg/ha respectively, and were in the range reported by Erickson et al (2011). Dry biomass partitioning followed a simila r pattern to the fresh biomass (Figure 4 1) The fresh stalk yielded higher biomass ( 41.04 1.4 Mg/ha) than the juice (1 8.04 4.5 Mg/ha), heads ( 8.04 1.51 Mg/ha), and leaves ( 7.08 0. 95 Mg/ha ) (Table 4 2 ). The m orphologic al characteristics of Sweet S orghum used in this experiment reflected the general value reported in the literature. Average plant height, internode length, a nd internode diameter were 310 2, 20 1 and 1.5 0.3 cm respectively (Table 4 3). T he plant population 13.2 plants /m 2 was in
91 the density range of 12 to 20 plants/m 2 reported for SS ( Zegada Lizarazu and Monti 201 2 ) Feedstock C omposition Organic Mat ter, Fiber C ontent and Chemical Oxygen Content Structural and non structural carbohydrate content of various plant components reflects the suitability of the plant biomass to be used as a feedstock for an aerobic digestion and for food or feed. Non fiber (N F), hemicellulose content (%DM), cellulose/lignin content (%DM), and crude fiber (%DM ) of various SS biomass ar e presented in Table 4 4. The OM varied from 94. 4 % to 98. 4 % ; extracted stalk yielded the highest value. R esults p ortrayed the typical range of OM reported in the literature for sorghum and other grasses. The chemical oxygen demand (COD) varied from 190 5 to 1032 30 g/kg (Table 4 5). These results indicated that all SS plant biomass contained sufficient organic matter for conversion in to methane However, the organic fraction of the feedstock is not sufficient to predict its digestibility. In this regard, the nature and the composition of different structural element s are key parameters in evaluating a feedstock. As it is depicted in Table 4 4, t he fiber content on a dry weight basis of extracted stalk s heads and leaves averaged 33.6 1.1, 8 .0 0.2, and 32.1 1.3% respectively Fiber values suggest ed that sweet sorghum heads had a larger proportion of readily digestible materials available to be converted to methane. Chipped stalks and whole plants contained 26. 4 1.5 and 27.5 2.3% fiber respectively. The lower fiber content of chipped stalks and whole plant compared to leaves and extracted stalk s can be explained by the addition of leftov er non structural sugars from the juice, which contributes to decreas ing the relative fiber concentration. Hemicellulose content was higher for the leaves (24.9 1.0 %) and was lower for heads (4.7 0. 4 %)
92 However, the whole plant had higher cellulose/ l ignin content. High hemicellulose and low cellulose/lignin contents indicate that a large fraction of the fiber can be hydrolyzed without any chemical treatment and therefore is available for microbial conversion through anaerobic digestion. These findings were in accordance with those found by Madibela (2002) who reported hemicellulose values for different sweet sorghum plant materials of 28.9 to 34.7 %. Godin et al (2010) evaluated cellulose, hemicelluloses, lignin, and ash content in various lignocellulo sic crops for second generation bioethanol production and reported that cellulose, hemicellulose, lignin, and ash content of 29.7 1.0, 25.1 2.3, 3.2 0.5, and 9.5 1.1% for SS plant component s, respectively. The composition of sweet sorghum biomass s uggested that sweet sorghum plant components contain relatively low fiber and support their suitability for both biofuel and feed production. In Vitro Organic Matter Digestibility In vitro organic matter digestibility (IVOMD) is an estimation of the solubl e cell constituents and various nonstructural carbohydrates, nitrogenous compounds, cellulose and hemicellulose that can be degraded by microorganisms (Burner et al 2009). The IVOMD percentage was 46,7 2.1, 66,3 1.8, 60.1 0. 5 58.1 6, 2 and 53.1 2 9 % for extracted stalks, heads, leaves, chipped stalks, and whole plants (Figure 4 2) The relatively high value of IVOMD in the h eads probably results from its relatively low fiber and high protein content compared to the other plant components. The hi gher value of IVOMD of the heads suggests that heads have fodder qualities and therefore can be used as animal feed.
93 Nutrient Content Figures 4 3 and 4 4 show the CP, ash, and phosphorus content of non juice components. Sweet s orghum heads had significan tly higher crude protein content (11.2 1.6%) than leaves (6.54 1.3%) and stalk (1.18 0. 6 %). The phosphorus content of the heads was 0.470 0.0 02 % and was significantly higher than the phosphorus content of the leaves (0.277 0.0 01 %), the stalks (0. 078 0.0 01 %), the chipped stalk (0.076 0.0 04 %), and the whole pl ant (0.190 0.0 02 %). These results were in accordance with Colombini et al (2012) who reported crude protein con tent of 12.3, 10.5, and 8.7 % of DM for whole plant grain sorghum silage, forage sorghum silage, and corn silage respectively The relatively high protein content of the heads suggests that they are a promising feedstock for animal feed and human consumption. Methane Yield a nd Methane Production f rom Non Juice Components Cumula tive methane production was measured for 60 d in the assay. Methane production measurements were normalized to per g COD loaded and STP conditions after subtracting methane production from the inoculum blank. Cumulative methane production was fitted to a first order kinetic equation with the kinetic rate constant (k) and ultimate cumulative methane yield (CH 4f ) as fitted parameters. Table 3 6 shows the methane yield and the kinetic rates of various SS biomass components The methane yield obtained for the fibrous components of SS was 3 44.9 1, 312.0 1. 7, and 303.1 0.6 mL CH 4 /g COD for heads, leaves, and extracted stalk, respectively. Different SS components showed variability in total methane yield Sweet sorghum heads had t he highest methane yield whi le extracted stalk had the lowest yield. Statistical analysis of the result (p<0.05) showed that methane yield of sweet sorghum heads was significantly different from the yields of other parts of the plant. The
94 difference among various methane yields seeme d to result from the diversity of the chemical composition of different plant components used in this experiment. On an organic matter basis, t hese values were in accordance with Gunaseelan (2007) who reported a typical range of 228 to 538 mL CH 4 /g of OM f or different sorghum cultivars Furthermore, the result of this study de monstrated that methane yield of sweet sorghum plant materials is comparable to the yield reported for Zea mays L. s ilage which is 280 420 m L CH 4 /g OM higher than winter straw (28 0 m L CH 4 /g OM), rice straw (230 mL CH 4 /g OM), rape straw (190 mL CH 4 /g OM) and Miscanthus X giganteus (200 mL CH 4 /g OM ) (Klimiuk et al 2010). Methane yield obtained from different SS components was within typical range for forage, which is generally rich i n cellulose content protected in most cases by a complex fiber structure making the carbon content inaccessible for microbial activity (Rincon et al 2010) The theoretical methane yield could be estimated at 350 mL CH 4 / g COD It is evident that t he metha n e yield obtained in this study ranged aro und the expected yield particularly for heads and leaves. Sweet sorghum heads reached 98% of the theoretical yield As it is depicted in Table 4 4 SS heads had higher non structural carbohydrates which probably c ontribute d to facilitating the degradation process. Methane Production Rate for Fibrous Components Of S weet Sorghum The methane production rate reflects the biodegradability characteristic of the substrate and production of inhibitory intermediate produc ts during the anaerob ic digestion process. Figures 4 5 and 4 6 display measured and fitted models of distinctive methane production patterns for SS stalks, heads an d leaves during the course of 60 day retention time. All SS components showed a monophasic c urve of methane production and more than 80% of the methane yield was achieved b y 20 days for heads
95 and between 25 to 40 days of incubation for other components (Table 4 6 ). The methane production rat e appears to approach zero at 45 days residence time. Le aves had a higher rate but overall lower methane yield than heads. The lower rate of the heads can be explained by the higher protein content that is more difficult to digest. Although the total methane yield of chipped stalk and the whole plan t (Figures 4 7 and 4 8) were significantly lower ( p< 0.05) than the heads, their estimated kinetic rates did not show any significant difference The higher methane yield of the chipped stalk and the whole plant compared with the leaves and extracted stalk can be expla ined by the presence of relatively high amounts of directly available sugar in both chipped stal k and the whole plant that contributed to increase the microbial activity. Proximate Analysis for M ethane Yield a nd Kinetic Rate Prediction Compositional chara cteristics of different SS biomass components were expected to predict potential methane production Although significant difference s were observed in fiber content, crude protein, and IVOMD of different feedstocks the difference found among the methane y ield s could be attributed to their chemical characteristics. Regression analysis between meth ane yield of different plant component s used in this experiment and the NF component was strong ( r 2 =0.90) (Figure 4 9) A positive relationship was also observed be tween methane yield and crude protein (r 2 =0.75 ) (Figure 4 10 ) As it was anticipated, samples with high non fiber content reached the highest methane yield. The non fiber component represents the fraction of the feedstock that are easy digestible. However, a negative correlation was found between hemicellulose and the methane yield (r 2 =0.82) (Figure 4 11) This could be explained by the fact that samples with low hemicellulose content have either high non fiber content or high cellulose/ lignin content. Non fiber content and cellulose are
96 more degradable than hemicellulose. In contrast, a poor correlation (r 2 =0.09 ) was found between the yield and the cellulose/lignin content of the component (Figure 4 12) A poor correlation was found between the crude prot ein and the kinetic rate (R 2 =0.07) (Figure 4 13). Contrarily a higher correlation was found between the kinetic rate and cellulose/lignin content (r 2 =0.72) (Figure 4 1 4 ). These findings are in accordance with Gunasealan (2007) who reported similar relati onships between methane yield and carbohydrate content, acid detergent fiber lignin, and cellulose content for sorghum and napier grass. However, a multiple regression using hemicellulose and crude protein as predictable variables revealed a very strong r elationship ( r 2 =0.97) with methane yield (Table 4 7). Likewise, a strong correlation (r 2 =0.99) was found between non fiber and cellulose/lignin and the kinetic rates For instance, SS heads and the whole plant had the lowest hemicellulose content and highe st crude protein; accordingly, their methane yield was higher than the other plant components Likewise, the methane production rate of SS leaves was significantly higher than the rate exhibited by the other substrates. Part of this r elatively high product ion rate can be attributed to their low cellulose/lignin. These findings indicated that a combination of multiple parameters is more adequate to predict methane production than single parameters. Characterization o f Sweet Sorghum Juice f or Methane Producti on Juice Yield and Chemical Characteristics The extracted juice averaged 18.0 4.5 M g/ha corresponding to a juice extraction ratio of 52.5 % (g of juice/g stalk) by weigt The extracted juice weight did not represent the actual juice yield of the plant sin ce 100% extraction efficiency is unachievable. Most industrial extraction reaches 80% efficiency (Wor tmann et a l
97 2010), but this study did not take into account the extraction efficiency since the leftover sugar in the stalk would be recovered by using t he stalk for bioprocessing. Figure 4 1 5 shows chemical characteristics of the juice. Electrical conductivity, pH brix content, chemical oxygen demand and alkalinity were 6.4 0.4 mS/cm, 5.1 0.1, 16.1 0.5 %, 189 5 g/kg and 403.33 30 m g as CaCO 3 res pectively. The brix was in the range of 10.7 and 18.9 % repo rted by Houl ou and Stevens ( 2010). The sugar yield was 2.9 Mg/ha b ased on actual extraction data and was in the range of 2 to 9.9 Mg/ha reported by Houlou and Stevens (2010). A theoretical methane yield of 66.9 mL of CH 4 can be obtained from the conversion of 1 gram of juice through anaerobic digestion. Additionally, the juice contains substantial amounts of major nutrients. Total nitrogen, ammonia nitrogen, TP and K were approximately 62.8 8.8, 2 4.9 6.3, 221.9 80.4, and 2 102.9 129.8 ppm, respectively. Total nitrogen concentration of the juice was lower than the value reported by Erickson et al (2012) who evaluated the nutrient partitioning in SS plant in response to different fertilization application rates. The presence of substantial amounts of nutrients in the juice is a beneficial feature of the juice as a candidate feedstock for anaerobic digestion. Anaerobic Digestion of Sweet Sorghum Juice Figures 4 1 6 and 4 1 7 show the methane prod uction and kinetic rate s of SS juice and the fitted model for the juice. The methane yield of SS juice ranged between 335.1 5.7 mL/g COD at an organic loading rate (OLR) of 1g COD / L and 320.60 20 mL/g COD at an OLR of 2g COD/L. Methane yields obtained from the two tested OLRs were not significantly different (p<0.05). However, the kinetic rate was significantly higher for the lowest loading rate. As the juice is rich in directly fermentable sugars, accumulation of fatty acids resulted in acidification from the rapid hydrolysis of sugars which were
98 generated at a rate that surpassed the metabolic capacity of the acetogen population. The accumulation of fatty acids leads to a drop in the pH and the subsequent inhibition of the overall process. This dilem ma was overcome by lowering the OLR to 1g COD/L. At this rate 85% of the ultimate methane yield was achieved at 6 days while only 57.8% was achieved at 2 g COD/L for a similar retention time. Methane Yield o f Sweet Sorghum Biomass The methane yield obtain ed from different SS plant components were used to determine the amounts of methane that can be obtained from 1 ha of sweet sorghum A simple scenario based on utilization of all sweet sorghum components for biodige stion is presented in Figure 4 1 8 To tal SS biomass of about 28.8 Mg DM/ h a pe r season were partitioned into 6.5, 3.4 and 16.3 Mg/ha of heads, lea ves, extracted stalk, respectively. The j uice yield was 18.0 4 Mg (wet weight)/ha. Based on the organic matter and the methane yield of SS plant compone nts, the methane yields ( m 3 CH 4 / ha per season) was 2 100, 1 000, 5 500, and 1 200 for heads, leaves extracted stalk and juice respectively. Anaerobic digestion of all SS plant biomass would reach a total yield of 9 800 m 3 /ha per season. An integrated s cheme based on utilization of the extracted stalk and juice for methane production is presented in Figure 4 1 9 According to this scenario, sweet sorghum heads would be use d for animal feed or food for human consumption, the leaves would stay in the field for increasing nutrients and organic matter of the s oil. As a result, 6 600 m 3 CH 4 / ha per season can be produce d coupled to the production of 6.5 and 3.4 Mg / ha of heads and leaves for each season respectively Since nitrogen and phosphorus concentrations of the leaves are relatively high compared to the stalk, allowing the leaves to stay in the field would improve soil fertility. The ash concentration
99 of the leaves (5.7 0.2%) was higher than other parts of the plant and constitutes another valuable reaso n for leaving the leaves in the field. Even though the methane yield of sweet sorghum in the second scenario would be lower in comparison to the first scenario, utilization of the heads and the leaves for other purposes strengthens the value of SS as a sus tainable crop. When SS heads are used as animal feed, animal manures are produced as feedstock for anaerobic digestion, which will probably increase the methane yield. In a more holistic scenario (Figure 4 20 ) the juice can be used for ethanol production while the extracted stalk is used for methane production. The stillage produced from the ethanol fermentation can also be diverted to the digester. Wilkie et al (2000) reported COD value for stillage produced after fermentation of SS as high as 79.9 g/L a nd therefore can be a carbon rich feedstock for anaerobic digestion. The diversity of biofuels that can be obtained from this approach highlights the versatility of sweet sorghum and the role it can play in an in tegrated energy and food production system. Application of Anaerobic Digestion o f Sweet Sorghum Proximate analysis, IVODM, and MIT of SS plant components show that SS can be digested for producing methane gas. If all SS biomass is dedicated to anaerobic digestion, approximately 9,800 m 3 CH 4 /ha can b e produced per season. This amount of methane gas is equivalent to 7 900 million BTU or 7 300 kg LPG per ha per season (Table 4 9). The methane yield of SS depends on a large extent to the biomass yield. Table 4 10 presents the energy yield per dry ton. De pending on the scenario, different considerations are needed when designing a biodigester and deciding how to digest SS plant components. In the event where all SS component are diverted to biodigestion, one approach may consist of digesting each component separately. This approach
100 would require a sequential process including harvesting, partitioning plant component s squeezing the stem, pretreatment, and feeding the digester. In this experiment, the pretreatment consisted only of size reduction. Differenc es between chemical characteristics of each SS component and the methane production rate suggested that different parameters are necessary for an adequate anaerobic process. The sweet sorghum juice is composed of directly fermentable sugars that require a relatively lower OLR but a lower retention time when compared to fibrous components. While the relatively short retention time required for the juice could be a potential advantage, the lower OLR would require a bigger reactor volume with economic conseque nces. On the other hand, the digestion of fibrous components of SS including heads, leaves, and stalks would necessitate a smaller digester volume at a longer residence time particularly for a batch digester. Another challenge that also needs to be address ed is the short harvesting time and poor storage qualities of SS component s requiring large extraction capacity for immediate processing. With the plenitude of directly fermentable sugars in the juice and the ubiquitousness of microorganisms, fermentation of the juice can even start in the field. Further treatment would be necessary to preserve the composition characteristics of the juice before processing. Additionally, storing a large volume of juice would be exceptionally challenging. Preservation of sug ars after harvest is a major concern for the efficiency and economy of bioprocessing sweet sorghum. Because of the high water content and the sugar content of the plant, a short distance between the processing unit and the agronomic production site is requ ired.
101 Since the ultimate methane yield of the chipped stalk (326.0 2.76 mL CH 4 /g COD) was not significantly different than the methane yields from the juice, and the kinetic rate from the juice was significantly higher than the stalk, simultaneous diges tion of the juice and the stalk would be the most plausible scenario. According to Figure 4 7, approximately 90% of the ultimate methane yield from the chipped stalk that includes both juice and extracted stalk was obtained at 30 days, which is between the 10 days for the juice and 45 days for the extracted stalk to generate equivalent methane yield percentages. This would facilitate the design of a reactor that would handle a practical OLR during a relatively short period of time. The chipped stalk can be loaded at a higher rate than the juice and would require less retention time than the extracted stalk. This model of digestion of the chipped stalk would corroborate the integrated scenario where the heads are used for animal and or human food and the leav es left in the field. Ensilage : A Potential Storage Technique of Sweet Sorghum Component for both Feed and Biogas Production Considering the results obtained from the digestion of the chipped stalk both in terms of methane yield and kinetic rates, ensilage would be a useful technique for SS bioprocessing and can be considered as a storage or a pretreatment method either for animal feed or anaerobic digestion. During the ensilage process, soluble sugars of SS components would be consumed by acid producing ba cteria that could be either native or inoculated. The regulation of the process is maintained by the pH, which will drop drastically with the production of organic acids resulting from the process and the subsequent inhibition of microorganisms. The richne ss of SS plant s in soluble sugars and its high moisture content increase the attractiveness of SS for silage production. Amer et al (2012) reported that
102 SS silage had a similar effect on dry matter intake, energy corrected milk, and milk efficiency on dai ry cows compared to alfalfa silage. Similarly, Sun et al (2010) found that silage production from chopped SS plant s and extracted stalk met standard requirements for good silage quality (pH not more than 4.2, volatile basic nitrogen to total nitrogen rati o no more than 12.5%, and little volatile fatty acids content). Successful silage production was attributed to the poor juice extraction ratio, which left large amounts of sugars in the stalk. If the juice were used for producing ethanol or biogas, a highe r extraction rate is desired, since 100% extraction rate is unachievable, the extracted stalk with the leftover sugar could be ensiled as an animal feed or later biogas production. Whitfield et al (2012) suggested that sweet sorghum silage could be the an swer to the short processing window after harvesting for biofuel production. However, the obvious disadvantages of ensilage for ethanol production specifically are the consumption of sugars by acid production bacteria, which are responsible for the ensilag e process. This conversion of sugars to lactic acids is not a loss for anaerobic digestion since lactic acids are also precursors of methane production. Depending on the efficiency of the juice extraction process, the amount of sugars required for a comple te ensiling may not be sufficient in the extracted stalk (Whitfield et al 2012) consequently ensiling the chipped stalk would be the most realistic approach (Figure 4 21) Sweet Sorghum : A Crop for Fuel a nd Food Production i n Haiti The results of this study indicated that SS is a multipurpose crop that can be used to produce fuel, feed, and food. Anaerobic digestion of all SS biomass or selected components can produce substantial amounts of renewable energy in the form of methane gas. Adoption of SS in an integrated energy and food production system can
103 contribute positively to an improved environment and promote agricultural and rural development. The development of such system would promote access to energy particularly in rural areas of developing cou ntries, support the growth of local economies and sustain long term improvement in food security (FAO, 2008). The decline of agricultural productivity in Haiti and the continuous degradation of the environment necessitate a multifaceted approach that simul taneously addresses food and energy security, environment restoration, hygiene and sanitation. Although utilization of SS in the Haitian farming system is at its nascent stage, its development could help bolster local agriculture. E lements that give ground s for the suitability of SS to the Haitian farming system include favorable climate, existing cultural practices for grain sorghum, and existing sugarcane infrastructure. Sweet sorghum can be grown all year round in Haiti with the possibility of two or mor e harvests a year. A dditionally, a s it has been reported in the literature, SS has low water and nitrogen requirements. Consequently, farmers can rely only on rainfall as water sources in most regions of the country and can supply only limited fertilizer n eeded. If the integrated scenario (Figure 4 20 ) is adopted, the effluent from the AD and the nutrients from the leaves can supply the fertilizer requirement for the crop. Since Haitian farmers have grown grain sorghum for centuries, they would not need to adopt any new technology or adjust their cultural practices for SS. Cultural practices for g rain s orghum are similar to those required for SS. However, they would need to harvest the stems and follow different steps for bioprocessing of SS. In the current grain sorghum production system, farmers usually collect the grains that are used for human consumption and or animal feed. The leaves and the stem are left in the field or fed to animals. The residues are burnt in the field
104 prior to new plantings. In the system with SS, each component of the plant would be sustainably used to produce added value products that would contribute to improve farmer livelihoods. The existing infrastructures for sugarcane such as extracting mills and distilleries can also be allo cated to sweet sorghum processing. Farmers can sell the juice to local distilleries to supplement incomes or they can digest the SS for producing biogas (Figure 4 22) Production of biogas from the stem would be the most sustainable approach. It would reco ncile food production to clean energy and soil quality protection. The heads can be targeted specifically to support livestock production, which is elementary to ensure food security. In addition, the leaves and the solid residues from the digestion of the chipped stalk can be used as a soil amendment and fertilizer to improve and sustain soil quality.
105 Table 4 1 Partitioning of sorghum plant biomass in g/plant. Mean values and standard deviation Plant components Fresh weight (g) Dry weight (g) % DM Lea ves 0.08 0.02 0.04 0.00 47.79 7.9 Heads 0.10 0.03 0.08 0.02 80.65 1.92 Juice 0.22 0.07 0.031 0.01 4.3 1.08 Extracted stem 0.49 0.09 0.12 0.08 38.62 1.69 Total 0.88 0.22 0.31 0.09 31.52 2.1 Table 4 2. Sweet sorghum biomas s yield Plant component Harvested FW FW (Mg/ha) DW (Mg/ha) Leaves 1.7 0.2 7.1 0.9 3.4 0.4 Heads 1.9 0.4 8.0 1.5 6.5 0.8 Juice 4.3 1.1 18.0 4.5 0.8 0.2 Extracted stem 9.9 0.7 41.0 2.9 16.2 8.7 Total biomass 16.8 0.6 74.2 12.1 26.9 8.7 FW : fresh weight, DW : dry weight
106 Table 4 3. Morphological characteristics of SS plant Plant parameters Average Standard deviation Plant height (cm) 311.3 19 Internode length (cm) 20.1 0.9 Plant population (plant/m 2 ) 13.2 1.8 Internod e diameter (cm) 1.5 0.3 Stem height (cm) 222.0 15.5 Table 4 4. Proximate analysis of fibrous components of SS Plant components Non fiber Hemicellulose Cellulose+Lignin Crude fiber Extracted stalk 58.9 0.9c 20.9 1.2b 21.3 1.4ab 33.6 1.1a Heads 83.00.4a 4.70.4d 12.2 0.8c 8.0 0.2c Leaves 60.41.1c 24.9 1.0a 10.40.7c 32.1 1.3a Chipped Stalk 64.4 1.4b 14.1 1.0c 18.42.4b 26.4 1.5b Whole plant 64.72.0b 12.81.1c 23.02.6a 27.5 2.3b Values are mean and standard deviation. Numbers foll owed by the same letter are not significantly different p<0.05.
107 Table 4 5 Chemical oxygen demand of Sweet Sorghum components Plant Components Mean (g/kg) Standard deviation Leaves 962.6 91.1 Heads 1005.3 11.3 Extracted stalk 1101.1 49.4 Juice 188.7 5.33 Chipped Stalk 1032.0 31.5 Whole plant 993.1 25.1 Table 4 6. Estimates of methane production kinetics of SS components Plant component Methane Yield (mL/g COD) K (day 1 ) Extracted stalk 303.1 0.6d 0.041 0.0013b Heads 344.9 1.03a 0.046 0.0019b Leaves 312.0 1.7c 0.058 0.0017a Chipped stalk 317.0 0.6 c 0.039 0.0013b Whole plant 326.0 6.3 b 0.040 0.0011b Values are mean and standard error. Numbers followed by the same letter are not significantly different P<0.05.
108 Table 4 7 Prediction of methane yield of sweet sorghum components from different independent variables Response variable Predictable variables R 2 P value Constant rate 1 NF, cellulose and lignin 0.99 0.0074 Methane yield 2 CP, Hemicellulose 0.97 0.0322 1: P redi ction expression: 0.097 0.001*(cellulose and lignin) 0.0004NF 2: Prediction expression: 327+ 2.14*CP 1.19 *Hemicellulose NF: Non Fiber, CP: crude protein Table 4 8 Sweet s orghum juice characteristics Juice TP (mg/L) TN (mg/L) K (mg/L) NH3 N (mg/L ) Average 2 42.0. 62.8 2100.0 24.97 SDTV 91 8.81 129.80 6.34 Table 4 9. Methane yield and energy value of Sweet Sorghum on a per ha basis SS Components CH 4 yield (m 3 /ha) BTU (Million/ha) LPG ( k g/ha) Juice 1150 10.1 41 0.4 860. 7.5 Extracted stal k 5450 23.7 190 0.8 4060. 17.6 Heads 2240 11.4 80 0.4 1670 8.5 Leaves 1001. 7.8 35. 0.3 750 5.5 Total 9840 23.8 350 0.8 7340 17.8 BTU: British thermal unit. LPG: Liquefied petroleum gas
109 Table 4 10. Methane yield and energy val ue of Sweet Sorghum on Mg basis SS Components CH 4 (m 3 /dry Mg) BTU (10 3 / dry Mg) LPJ ( k g/dry Mg) Juice 26 0.01 920 0.3 20 0.01 Stalk 120 0.0 4350 0.0 100 0.00 Heads 51 0.0 1790 0.0 40 0.00 Leaves 23 0.2 798 6.2 16 0.13 Total 220 0.5 7860 19.0 170 0.40 Figure 4 1. Sweet sorghum biomass partitioning. Values are fresh weight in k g/ha 7.08 8.04 41.04 18.04 Leaves Heads Stalk Juice
110 Figure 4 2. Organic matter (OM) and in vitro organic matter (IVOMD) digestibility of SS components Figure 4 3. Crude protein and a sh content of SS component 0 20 40 60 80 100 Extracted stalk Leaves Heads Chipped Stalk Whole plant % DM OM(%) IVOMD(%) 0 2 4 6 8 10 12 14 Extracted stalk Leaves Heads Chipped Stalk Whole plant % DM CP (%DM) Ash (%DM)
111 Figure 4 4. Phosphorus content of SS component Figure 4 5 Cumul ative methane production of SS component measured from the MIT 0 0.1 0.2 0.3 0.4 0.5 0.6 Extracted stalk Heads Leaves Chipped stalk Whole plant % DM 0 50 100 150 200 250 300 350 0 20 40 60 80 Cumulative CH 4 (mL/g COD@STP) Elapsed time (day) Heads Extracted Stalk Leaves
112 Figure 4 6. Fitted cumulative methane production curve of SS component Figure 4 7. Cumulative methane production of whole plant versus chipped stalk 0 50 100 150 200 250 300 350 400 0 20 40 60 Cumulative CH 4 (mL/g COD@STP) Elapsed Time (day) CH 4f = CH 4t *(1 exp( kt)) Fit-Heads Fit-Leaves Fit-Extracted Stalk 0 50 100 150 200 250 300 350 400 0 20 40 60 Cumulative CH 4 (mL/g COD@STP) Elapsed Time (day Whole Plant Chipped Stalk
113 Figure 4 8 Fitted cumulative methane production of whole plant versus chipped stalk Figure 4 9. Correlation between methane yield and fiber content. 0 50 100 150 200 250 300 350 0 20 40 60 Cumulative CH 4 (mL/g COD@STP) Elapsed time (day) CH 4f = CH 4t *(1 exp( kt)) FitChipped Stalk Fit Whole Plant y = 1.5592x + 217.26 R = 0.8998 300 305 310 315 320 325 330 335 340 345 350 50 60 70 80 90 Methane yield (mL/g COD @STP) NF (%DM)
114 Figure 4 10. Correlation me thane yield and crude protein Figure 4 11. Correlation methane yield and crude protein y = 3.5557x + 301.49 R = 0.7491 300 305 310 315 320 325 330 335 340 345 350 0 5 10 15 Methane yield (mL/g COD@STP) Crude protein (%) y = 1.807x + 348.37 R = 0.8211 300 305 310 315 320 325 330 335 340 345 350 0 10 20 30 Methane yield (mL/g COD @STP) Hemicellulose (%DM)
115 Figure 4 12. Correlation methane yield and crude protein Figure 4 1 3 Correlation between constant rate and crude protein y = 0.874x + 335.53 R = 0.0929 300 305 310 315 320 325 330 335 340 345 350 5 10 15 20 25 Methane yield (mL/g COD@STP) Cellulose and Lignin(%) y = 0.0005x + 0.044 R = 0.0722 0.03 0.035 0.04 0.045 0.05 0.055 0.06 0 5 10 15 Constant rate par day Crude protein(%)
116 Figure 4 1 4 Correlation between constant rate and cellulose/lignin Figure 4 1 5 Sweet sorghum juice characteristics. Alkalinity values are in g/L CaCO 3 10) y = 0.0011x + 0.0648 R = 0.718 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 5.0 10.0 15.0 20.0 25.0 Constant rate per day Cellulose and Lignin (%) 0 5 10 15 20 25 pH E.C (mS/cm) Brix (%) TCOD(g/Lx10) Alknality
117 Figure 4 1 6 Cumulative methane production from SS juice Figure 4 1 7 Fitted cumulative methane production of SS juice 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 0 5 10 15 20 Cumulative CH4 mL/g COD Elapsed time (day) OLR 2g COD/L OLR 1g COD/L 0 50 100 150 200 250 300 350 0 5 10 15 20 25 Cumulative CH 4 (mL/g COD@STP) Elapsed time (day) CH 4f = CH 4t *(1 exp( kt)) Fit @ 2g COD/L Fit @ 1 g COD/L
118 Figure 4 1 8 Methane of sweet sorghum plant. Estimates are for 1 ha. All data are in dry weight basis except for the juice Figure 4 1 9 Methane of 1 ha sweet sorghum plant. Only the stem is used for digestion. All data are in dry weight ba sis except for the juice
119 Figure 4 20 Integration of sorghum sorghum for fuel and feed production. Extracted stem and juice go to ethanol and/or biogas production, heads are used as animal or human food; leaves are left in the field
120 Figure 4 21. Ensiling of sweet sorghum for feed and fuel production Figure 4 22. Diversity of sweet sorghum ends products
121 CHAPTER 5 CONCLUSION Fundamental needs for affordable, clean energy and increased food production resulting from an unprecedented grow th of world population are grand challenges that developing countries currently face. A pproximately up to 90% of rural households in developing countries currently rely on raw biomass as their primary source of energy (Bru ce et al 2000; FAO, 2010). Utilization of raw biomass as an energy source has severe impacts in terms of both public health and environmental degradation. In Haiti, utilization of biomass as the primary energy source has led to intensive deforestation wi th subsequent decline in soil fertility and agricultural productivity, due primarily to widespread erosion. As a result, new methods that simultaneously address the need of energy and food production, while nurturing the soil must be developed. Maximizing synergies between food and fuel production can be achieved by the utilization of technologies that allow maximum utilization of agricultural residues and foster sustainable stewardship of agricultural lands (FAO, 2010). Among current bioenergy crops, s wee t s orghum has been recog nized as an attractive energy crop because of its versatility, yield potential and growth characteristics (Zegada Lizarazu and Monti, 2012). Extensive efforts have been made to evaluate the potential of SS for ethanol production; ho wever, little is known about its potential methane production and the dual role this crop can play to meet food and fuel demands, particularly in Haiti. This study explore d the applicability of the anaerobic digestion of biomass into the Haitian farming sy stem to generate bioenergy and sustain food production. Th e two objectives of this study were: 1) c haracteriz e anaerobic digester effluent ( ADE ) and assess its effect on the growth and yield of common
122 vegetables grow n in Haiti specifically onion, corn, an d bean and 2) a ssess the suitability of SS a s a multipurpose crop for biofuels and food production Fertilizer Value of Anaerobic Digester Effluent Addition of anaerobic digester effluent to the soil had a positive effect on the growth and yield of the veg etables used in the described experiments The yield s of crops treated with ADE were significantly higher than those that did not receive any fertilizer but were not significantly different from treatments that were fertilized with urea and fish emulsion. ADE fertilization experiment s indicated that ADE had both nutritive value for plant growth as well as increasing soil organic matter a nd thus might be useful for both organic and conventional vegetable production. M ineralization appears to not be the limi ting factor for plant growth after ADE application. A large fraction of total nitrogen (60% in this study) was in the soluble ammonium form, which can be directly assimilated by plants and is also readily converted to nitrate in favorable soil conditions. Additionally, t he organic fraction gradually mineralize s to become plant available and therefore constitute s a slow release nutrient reserve for the growing period. However, timing and rate of ADE application are essential to avoid nutrient losses and sy nchronize plant needs to nutrient availability. Utilization of ADE as a fertilizer source is very attractive in promoting sustainable agriculture particularly in developing countries such Haiti Anaerobic digestion not only offer s an environment ally friend ly fuel producing method of management for organic wastes generated on farm but also recycles plant nutrients that can be used to offset the need for synthetic fertilizers. This is particularly important for nitrogen demanding crop s like onions and corn where fertilizer cost represents a large fraction of the operating cost.
123 Characterization of Sweet Sorghum for Food and Fuel Production Sweet Sorghum plant components obtained from the energy grass field trial at the Energy Research and Education Pack, Uni versity of Florida, Gainesville were analyzed according to standard forage quality tests. Triplicate samples of each component were analyzed for crude protein, fiber, acid detergent fiber, neutral detergent fiber, and ash. Results from the proximate analys is were used to determine the non fiber, hemicellulose and cellulose/lignin content of SS components. A methane index test (MIT) was performed to evaluate the potential methane production. Sweet Sorghum heads had lower fiber content (8.0
124 can be use d for animal feed or food for human consumption, the leaves would stay in the field for increasing nutrients and organic matter of the s oil. Since nitrogen and phosphorus concentrations of the leaves are relatively high compared to the stalk, allowing the leaves to stay in the field would improve soil fertility. In addition, whe n the heads and others components of the plants are used for animal feed, animal manure are produced, which can be anaerobically digested. The solid residue of the anaerobic digestion process can be used as soil amendment while the liquid effluent would be used as fertilizer. Practical Implication and Recommendations The primary motivation of this research was rooted in the need to provide clean energy and affordable fertilizer sources necessary to improve the living conditions of Haitian rural households. The results of this study indicate the existence of promising potential for co production of energy and food Implementation of anaerobic digestion and the use of Sweet Sorghum as a multipurpose crop would eliminate possible competition between food and fu el production. This synergy between food and fuel production is particularly crucial for Haiti where food production is scarce, an d unsustainable energy sources are used to prepare food on a daily basis A zero waste system can be created using AD as a mea ns to recycle biomass, organic wastes and plant nutrients to produce renewable energy biofertilizer and food In this system, the
125 balance between soil fertility management, energy and animal feed can be achieved through the tradeoffs highlighted by the re sults of this study. Integration of food and energy production is a developing area of study and extensive research and development are needed to elucidate the role that such systems may play within local development. For instance, the ADE obtained from th e digestion of SS plant components can be harnessed as a fertilizer source for growing SS, while the production of SS biomass generates food (grain), biogas for cooking, and additional biofertilizer for the next SS crop. In the Haitian context, implementat ion of AD on a farm scale would be more profitable. Farm demonstration systems that could show the feasibility of producing and processing s weet s orghum into renewable energy through anaerobic digestion should be encouraged at local levels. Similarly, pilo t digestion systems that demonstrate the effectiveness of anaerobic digester effluent as a fertilizer should be promoted at the individual farm level. Without successfully integrating renewable energy, biofertilizer and food production, it is unlikely that farmers may be attracted to practice such systems. Adoption of anaerobic digestion in an integrated energy and food production system can contribute positively to an improved environment and promote agricultural and rural development. The development of s uch system would promote access to energy particularly in rural areas, support the growth of local economies and sustain long term improvements in food security.
126 LIST OF REFERENCE S Agoramoorthy, G., and Hsu, M.J. 2008. Biogas plants ease ecological stre remote villages. Human ecology 36 (3) : 435 44 1. DOI 10.1007/s10745 008 91638 Alburquerque J .A D e la Fuente C Ferrer Costa A Carrasco L Cegarra J Abad M Bernal M P ., 2012. Assessment of the fertiliser potential of digestates from farm and agroindustrial residues. Biomass Bioenerg y 40:181 18 9. D OI :10.1016/j.biombioe.2012.02.018 Almodares, A., Hadi, M.R. 2009. Production of bioethanol from sweet sorghum: a review. Afr ican J ournal Agric ultural Res ources 4 (9) : 772 780 http://www.academicjournals.org/pdf Am er, S., Seguin P., and Mustafa, A.F. 2012. Effects of feeding sweet sorghum silage on milk production of lactating dairy cows Journal of Dairy Sciences. 95 (2) : 86 94 D OI : 10.3168/jds.2011 4884 Angel idaki, I. and Sanders, W. 2004. Assessment of the anaerobic biodegradability of macropollutants. Reviews in Environmenta l Science and biot echnology. 3: 117 129. http://download.springer.com/pdf Angelidaki I., Dimitar K., Damien J.B., Caroline M.P., and Alfons J.M.S. 2011. Biomethanation and its potential Methods in Enzymology 494: 327 351. DOI: 10.1016/B978 0 12 385112 3.00016 0 APHA 2005 Standard Methods for the Examination of Water and Effluent. American Public Health Association, American Water Works Association and Water En vironment Federation, Washington, DC. Appels L. Baeyens J., Degrece J., Dewil R. 2008. Principles and potential of the anaerobic digestion of waste activated sludge. Progress in Energy and Combustion Science 34 (6) : 755 781. DOI: 10.1016/j.pecs.2008.06.002 Arthurson, V. 2009 Closing the global energy and nutrient cycles through application of biogas residue to agricultural land potential benefits and drawbacks. Energies 2 (2) : 226 246. DOI: 10.3390/en20200226 Baro, M. 2002. Food Insecurity and livelihood systems in Northwest Haiti. Journal of political ecology 9: 1 34 http ://jpe.library.arizona.edu.pdf Biau, A., Santiveri, F., Mijangos, I., and Lloveras, J. 2012. The impact of organic and mineral fertilizers on soil quality parameters and the productivity of irrigated maize crops in semiarid regions. European Journal of S oil Biology 53: 56 61. DOI: 10.1016/j.ejsobi.2012.08.008
127 Burner, D.M., Tew, T.L., Harvey J.J., and Belesky D.P. 2009. Dry matter partitioning and quality of Miscanthus, Panicum, and Saccharu m genotypes in Arkansas, USA. Biomass and Bioenergy 33 (4) : 610 619. DOI: 10.1016/j.biombioe.2008.10.002 Braun, R. 2007. Anaerobic digestion: a multi faceted process for energy, environmental management and rural development .I n P. Ranalli (ed.), Improvement of Crop Plants for Industrial End Uses, 335 416 http://download.springer.com/static/pdf/ Bruce, N., Perez Padilla, R., Albalak, R. 2000. Indoor air pollution in developing countries: a major environmental and public health challenge. Bulletin of the World Health Organization 78(9): 1078 1092. http://uh7qf6fd4h.search.serialssolutionsh Cornejo, C., and Wilkie, A. C. 2010. Greenhouse gas emissions and biogas potential from livestock in Ecuador Energy for Sustainable Development. 14: 256 266 DOI: 10.1016/j.esd.2010.09.008 Chandra R., Takeuchi H., Hasegawa T ., 2012. Methane production from lignocellulosic agricultural crop wastes: A review in context to second generation of biofuel production. Renewable and S ustainable Energy Reviews 16 (3) : 1462 1476 DOI: 10.1016/j.rser.2011.11.035 Chantigny, M.H., Angers, D.A., Belanger, G., Rochette, P., Eriksen Hamel, N., Bittman S., Buckley, K., Masse, D., and G aser M. 2008. Yield and nutrient export of grain corn fertilized with raw and treated liquid swine manure. Agronomy Journal 100(5): 1303 1309 DOI: 10.2134/agronj2007.0361 Chynoweth, D. P., Turick C.E., Owens J. M., Jerg er, D.E., and Peck, M.W. 1993. Biochemical methane potential of biomass and waste feedstock. Biomass and Energy 5 (1): 95 111 DOI: 10.1016/0961 9534(93)90010 2 CIAT 2010 S eed System Security Assessment, Haiti A study funded by the United States Agency for International Development, Office of Foreign Disaster Assistance http://sssa haiti main report.pdf Colombi, S., Galassi, G., Crovetto, G.M., and Rapetti, L. 2012. Milk production, nitrogen balance, and fiber digestibility prediction of corn, whole plant grain sorghum, and forage sorghum silages in the dairy cow. Journal of Dairy Sciences 9 5 (8) : 4457 4467. DOI: 10.3168/jds.2011 4444 Dolisca F., and Jolly C. M., 2008. Technical efficiency of traditional and non traditional crop production. A case from Haiti World Journal of Agricultur al Sciences 4(4): 416 426. http://www.idosi.org/wjas/wjas4(4)/2.pdf Erickson J. E., Woodard K. R., and Sollemberger L.E. 2012. Optimizing Sweet Sorghum Production for Biofuel in the Southeastern USA through Nitrogen Fertilization and Top Removal Bioenergy Res sources 5 (1) : 86 94 DOI: 10.1007/s12155 011 9129 3
128 Erickson, J.E., Helsel, Z.R., Woodard, K.R., Vendramini, J.M.B. Wang, Y., Soll enberger, L.E, and Gilbert, R.A. 2011. Planting date affects biomass and Brix of Sweet Sorghum Grown for Biofuel across Florida. Agronomy Journal 103 (6): 1827 1833 DOI: 10.2134/agronj2011.0176 FA O, 2008. The state of food and agriculture. Biofuels: Prospects, risks and opportunities. Food and Agriculture Organization of the United Nations Rome. ftp://ftp.fao.org/docrep/fao/011/i0100 e/i0100e.pdf FAO, 2010. Making integrated food energy systems work for people and climate: A review. Environment and Natural Resources Management working paper. Food and Agriculture Organization of the United Nations Rome. http://www.fao.org/docrep/013/i2044e/i2044e.pdf Frigon, J. and Guiot, S. R. 2010. Biomethane production from starch and lignocellulosic crops: a comparative review. Biofuels, Bioprod ucts and Bioref inning Biopr 4 (4) : 447 458. DOI: 10.1002/bbb.229 Furukawa, and Y. Hasegawa, H. 2006. Response of spinach and komatsuma to biogas effluent made from source separated kitchen garbage. Journal of Environment Quality.35 (6) : 1939 194 7. http://www.ncbi.nlm.nih.gov/pubmed/16973635 Gallaher, R. N., Weldon, C. O. and Futral, J. G. 1975. An aluminum block digester for plant and soil analysis. Soil Sci ence Soc iety of Amer ica. 39 ( 4) :803 806 https://www.soils.org/publications/sssaj/abstracts/39/4/SS0390040803 Godin, B., Ghysel, F., Agneessens, R., Schmit, T., Gofflot, S., Lamaudire, S., Sinnaeve,G ., Goffart, J. P., Grin, P.A., Stilmant, D., Delcarte, J., 2010. Cellulose, hemicelluloses, lignin, and ash contents in various lignocellulosic crops for second generation bioethanol production. Biotechnol ogie Agron omie Soc iete Environ nement 14 (2): 54 9 560 http://www.bib.fsagx.ac.be/base/text/v14ns2/549.pdf Golding, E. J., Carter, M.F., and Moore, J. E. 1985. Modification of the neutral detergent fiber procedure for hay. J ournal of Dairy Sci ence 68 (10) : 2732 2736. http://download.journals.pdf Gerardi, M.H. 2003. The Microbiology of Anaerobic Digesters; John Wi ley & Sons: Chich ester, UK. Gu naseelan, V.N. 2007. Regression models of ultimate methane yields of fruits and vegetable solid wastes, sorghum and napiergrass on chemical composition. Biore source Technology 98 (6) : 1270 1277 DOI: 10.1016/j.biotech.2006.05.014 Graunke, R. E and Wilkie, A. C 2008. Converting food waste to biogas: sustainable gator dinning. Sustainability. 1: 391 394. DOI: 10.1089/SUS.2008. 9914
129 Hambleton, L. G. 1977. Semiautomated method for simultaneous determination of phosphorus, calcium and crude protein in animal feeds. J.A.O.A.C. 60:845 852. Hansen, K. H., Ang elidaki, I., and Ahring, B. K., 1998 Anaerobic digestion of swine manure: inhibition by ammonia. Water research 32 (1), 5 12. DOI: 10.1016/S0043 1354(97)00201 7 Heilig, J. A. and Kelly J. D. 2012. Performance of dry bean genotypes grown under organic and convent ional production systems in Michigan. Agronomy Journal 104(5): 1485 1492. DOI: 10.2134/agronj2012.0082 Holou, R. A. Y. and Stevens, G. 2010. Juice, sugar, and bagasse response of sweet sorghum (So rghum bicolor (L.) Moench cv. M81E) to N fertilization and soil type. Global Change Biology and Bioenergy 4 (3) : 302 310 DOI: 10.1111/j.1757 1707.2011.01126.x IBRD/WB, 2007, Haiti: Strategy to alleviate the pressure of fuel demand on national woodfuel resources. Energy sector Management programs. International Bank for Reconstruction and Development/ Work bank, Washington, United States. https://www.esmap.orgpdf Katukiza, A.Y., Ronteltap, M., Niwagaba, C.B., Foppen, J.W.A., Kansiime, F. and Lens, P.N.L. 2012. Sustainable sanitation technology options for urban slums. Biotechnology advances 30(5): 964 978. DOI: 10.1016/j.biotechadv.2012.02.007 Khalid A., A rshad M., Anjum M., Mahmood T. and Dawson L. 2011. The anaerobic digestion of solid organic waste. Waste Management 31 (8) : 1717 1744. http://dx.doi.org/10.1016/j.wasman.2011.03.021 Kli miuk, E., Pokoj, T., Budzynski, W., an d Dubis, B. 2010. Theoretical and observed biogas production from plant biomass of different fiber contents. Bioresource Technology 101 (24) : 9527 9535. DOI: 10.1016/j.biortech.2010.06.130 Ko car, G. 2008 Anaerobic digesters: from waste to energy crops as an alternative energy source Recov ery Util ization and Environ mental Effects. 30 (7) : 660 669. DOI: 10.1080/00908310600628404 Lee J. 2010. Effect of application methods of organic fertilizer on growth, soil chemical properties and microbial densities in organic bulb onion production. Scientia Horticulturae 124 (3) :299 305. DOI: 10.1016/j.scienta.2010.01.004 Lesteur, M., Bellon Maurel V Gon zalez, C., Latrille, E., Roger, J.M., Junqua, G., Steyer, P. 2010. Alternative methods for determining anaerobic biodegradability: A review. Process Biochemistry 45 (4) : 431 440 DOI :10.1016/j.procbio.2009.11.018
130 Lilly Miller, 2013. Alaska fish emulsion fertilizer. http://www.lillymiller.com/ fish Loria, E.E., Sawyer, J. E., Barker, D. W., Lundwall, J. P., and Lorimo r J.C. 2007. Use of anaerobically digested Swine manure as a nitrogen source in Corn Production. Agronomy journal 99 (4) : 1119 1129. DO I: 10.2134/agronj2006.0251 Ly p eratos, G., and Pullammanappallul, P.C. 2010. Anaerobic digestion in suspended growth bioreactors. Handbook of environmental engineering/ Environmental biotechnology 10: 395 438. http://link.springer.com/chapter Mardsen M., Holm Nielsen J.B., Espensen K.H. 2011. Monitoring of anaerobic digestion processes: A review perspective Renewable and Sustainable Energy Reviews 15 (6) : 3141 315 5. DOI: 10.1016/j.rser.2011.04.026 Marchain, U. 1992 Biogas process for sustainable development. In FAO Agricultural service bulletin 9 5. Food and Agricultural Organization. Rome, Italy. http://www.fao.org/docrep/T0541E/T0541E00.HTM Mihelcic, J. R., Fry, L. M., Myre, E. A., Phillips, L. D., and Barkdoll, B. D. (2009). Field guide to environmental engineering for development worker s: Water, sanitation, and indoor air. Reston, VA: American Society of Civil Engineers Miller, A. N. and Ottman, M. J. 2010.Irrigation frequency effects on growth and ethanol yield in Sweet Sorghum. Agronomy Journal 102(1): 60 70 DOI:10.2134/agronj2009.0191 Montemurro, F., Canali, S., Convertini, G., Ferri, D., Tittarelli, F., Vitti C 2008 Anaerobic digestates application on fodder crops: effects on plant and soil Agrochemica 52 (4) : 297 712. http://cat.inist.fr/ Moller, K., and Mller, T. 2012. Effects of anaerobic digestion on digestate nutrient availability and crop growth: A review. Eng. Life Sci. 12 (3), 242 257. http://cat.inist.fr/?aModele=afficheN&cpsidt=20812635 Moller HB, Sommer S G Ahring B K ., 2004. Methane productivity of manure, straw and solid fractions of manure. Biomass Bioenergy 26 (5) : 4 85 495 DOI:10.1016/j.biombioe.2003.08.008 Moody, L.B., Burns, R. T., Bishop, G., Sell S. T and Spajic, R. 2011. Using biochemical methane potential assays to aid in co substrate selectio n for co digestion. Applied Engineering in Agriculture. 27(3): 433 439. http://elibrary.asabe.org/2011 Moore, J. E., and Mott G.O. 1974. Recovery of residual organic matter from in vitro digestion of forages. J ournal of Dairy Sci ence 57 (10) :1258 1259. DOI:10.3168/jds.S0022 0302(74)85048 4
131 Mumme, J., Linke, B., Tlle, R., 2010. Novel upflow anaerobic solid state (UASS) reactor. Bioresour ce Technol ogy 101 (2): 592 599. DOI:10.1016/j.biortech.2009.08.073 Nallathambi, G. V. 2004. Biochemical methane potential of fruits an d veg etable solid waste fe edstocks. Biomass Bioenergy. 26 (4) :389 99 DOI:10.1016/j.biombioe.2003.08.006 Odlare, M., Pell, M., and Svensson, K. 2008 Changes in soil chemical and microbiolo gical properties during 4 years of application of various organic residues Waste Management 28: 1246 1253. DOI:10.1016/j.wasman.2007.06.00 Parawira W. 2012. Enzyme research and applications i n biotechnological intensification of biogas production. Critical Reviews in Biotechnology 32(2): 172 186. DOI: 10.3109/07388551.2011.595384 Polprasert, C., 1989. Organic wastes recycling. J ohn Wil ey & Sons, Chichester, UK Pretty, J., Sutherland, W., Ashby, J., Auburn, J., Baulcombe, D., Bell, M., Bentley, J., Bickersteth, S., Brown K., Burke,J. et al 2010. The top 100 questions of importance to the future of global agriculture. Internatio nal Journal of Agricultural Sustainability 8(4) 219 236. DOI: 10.3763/ijas.2010.0534 Rajendran, K., Aslanzadeth S., and Taherzadeh M. 2012. Household biogas digesters. Energies 5 (8) : 2911 2942. DOI: 10.3390/en5082911 Raposo, F., De la Rubia, M.A., Fernndez Cegr, and Borja, V. R. 2011. Anaerobic digestion of solid organic substrates in batch mode: An overview relating to methane yields and exper imental procedures. Renewable and Sustainable Energy Reviews 16 (1) : 861 877 DOI: 10.1016/j.rser.2011.09.008 Ri ncon, B., Banks, C.J. and Heaven, S. 2010. Biochemical methane potential of winter wheat ( Triticum aestivum L.): Influence of growth storage practice Bioresource Technology. 101 (21) : 8179 8184. DOI: 10.1016/j.biortech.2010.08.062 Roberts T. L 2009. The role of fertilize Better crops 93(2): 12 15. http://www.ipni.net/ppiweb/bcrops.pdf Russo, V.M. 2008. Plant density and nitrog en fertilizer rate on yield and nutrient content of onion developed from greenhouse grown transplants. HortScience 43(6): 1759 1764. http://hortsci.ashspublications.org/cont ent/43/6/1759.full.pdf+html Speece, R.E. 2010. Anaerobic Biotechnology and Odor/Corrosion Control for Municipalities and Industries Archae Press, Nashville, TN.
132 Stefaniak, T.R., Dahlberg J.A., Bean B. W., Dighe, N., Wolfrum, E.J., and Rooney, W.L. 2012 Variation in biomass composition components among forage, biomass, sorghum sudangrass, and Sweet Sorghum types. Crop Sciences 52 (4) : 1949 1954. DOI: 10.2135/cropsci2011.10.0534 Sun, X.Z., Yama na, N., Dohi, M., Nakata, N., 2010. Deve lopment of a roller belt extrac tor for chop harvested sweet sorghum. Trans. ASABE 53 (5) 1631 1638. Taherzadeh, M., and Karimi, K. 2008. Pretreatment of Lignocellulosic Wastes to Improve Ethanol and Biogas Productio n: A Review. Int ernational Journal of Molecular Sciences 9 (9): 1621 1651. DOI: 10.3390/ijms9091621 properties of digesta te by studying the organic matter composition and the degree of biological stability during the anaerobic digestion of the organic fraction of MSW. Bioresour e Technol ogy 100 (12) : 3140 3142. DOI:10.1016/j.biortech.2009.02.012 Tilman D. Cassman K.G., Matson P.A., Naylor R., Polasky S. 2002. Agricultural sustainability and intensive production practices Nature 418: 671 677. DOI:10.1038/n ature01014 UNEP, 2010. The Geo Haiti: State of the Environment. Report 2010. United Nations Environment Program, Haiti. http://www.unep.org/publications/search USAID, 2011. Best market analysis Haiti. Office of food for peace. United States Agency for International Development. Port au Prince Haiti. http://pdf.usaid.gov/ USDA, 2013. Fertilizer use and price. United S tates Department of Agriculture ww.ers.usda.gov/data products/fertilizer use and price.aspx#.UWsN9b8cu0s US EPA, 1994. Determination of metals and trace elements in water and wastes by inductively coupled plasma atomic emission spectrometry Method 200.7. US environmental protection agency file://localhost/. http/::www.caslab.com:EPA Method 200 _7: US EPA 1993. Determination of total kjeldahl nitrogen by semiautomated colorimetry. Method 351.2. US environmental protection agency file://localhost/. http/::www.caslab.com:EPA M ethod 351_2: Walsh, J.J., Jones D. J., Edwards Jones, G., and Williams A.P. 2012. Replacing inorganic fertilizer with anaerobic digestate may maintain agricultural productivity at less environmental cost J ournal of Plant Nutr ition Science.175 (6): 84 0 845 DOI: 10.1002/jpln.201200214 Walter, Z. and Monti A. 2012. Are we ready to cultivate sweet sorghum as a bioenergy feedstock? A review on field management practices. Biomass and Bioenergy 40: 1 12. http://dx.doi.org/10.1016/j.biombioe.2012.01.048
133 Ward A.J., Hobbs P.J., Holliman P.J., Jones D.L. 2008. Optimization of the anaerobic digestion of agricultural resources. Bioresourc e Technology 99(17): 7928 7940 DOI: 10.1016/j.biortech.2008.02.044 Whitfield, M.B., Chinn M.S., and Veal M.W. 2012. Processing of materials derived from sweet sorghum for biobased produc ts Industrial Crops and P roducts 37 (1) : 362 375. DOI: 10.1016/j.indcrop.2011.12.011 WHO 2010 Progress on sanitation and drinking water: 2010 update. World Health Organization, Geneva, Switz erland. http://whqlibdoc.who.int/publications Wilkie, A.C., Riedesel, K.J., and Owens, J.M. 2000. Stillage characterization and anaerobic treatment of ethanol still age from conventional and cellulosic feedstock. Biomass and Bioenergy 19 (2) : 63 102 DOI: 10.1016/S0961 9534(00)00017 9 Wilkie, A.C., Smith, P.H., Bordeaux, F.M. 2004. An economical bioreacto r for evaluating biogas potential of particulate biomass. Bioresource Technology 92(1), 103 109. DOI: 10.1016/j.biortech.2003.08.007 Wilkie, A.C. 2005 a Anaerobic Digestion: Biology and Ben efits. in: Dairy Maure Management: Treatment, Handling, and Community Relations Natural Resource, Agriculture, and Engineering Service. Ithaca, NY, pp. 63 72. http://biogas.ifas.u fl.edu/pdf Wilkie, A. C. 2005b. Anaerobic digestion of dairy manure: design and process considerations. in: Dairy Maure Management: Treatment, Handling, and Community Relations Natural Resource, Agriculture, and Engineering Service. Ithaca, NY, pp. 301 3 12. http://biogas.ifas.ufl.edu/Publs/NRAES2005.pdf Wilkie, A.C. 200 6 The other bioenergy solution : the case for converting organics to biogas. Resource: Engineering& Technolo gy for a sustainable wolrd. 13(8): 11 12 http://biogas.ifas.ufl.edu/pdf Wilkie, A.C. 2007. Eco Engineering a sustainable society. Resource Resource: Engineering& Technology for a sustainable wor l d 19 20. Wilkie, A.C. 2008. Biomethane from biomass, biowaste, and biofuels. in: Bioenergy (Eds.) J.D. Wall, C.S. Harwood, A. Demain, American Society for Microbiology. Washington http://biogas.ifas.ufl.edu Wilkie, A.C. and Evans J. 2010. Aquatics plants: an opportunity feedstock in the age of bioenergy. Biofuels 1 (2), 311 321 http://biogas.ifas.ufl.edu/Publs/Biofuels 1(2)311 321(2010).pdf Wortmannn, C.S., Liska, A., Ferguson, R.B., Lyon D.J., and Klein R.N. 2010. Dry land performance of Sweet Sorghum and grain crops for biofuel in Nebraska. Agronomy Journal 102(1): 319 3 26 DOI: 10.2134/agronj2009.0271
134 Yu, z., Morrison M., and Schanbacher F.L., 2010 Production and utilization of methane biogas as renewable fuel. In : Biomass to biofuels : strategies for global in dustries (Eds.) Vertes A.A., Qureshi N., Blaschek H.P., and Yukawa H. John Wiley, Ltd DOI: 10.1002/9780470750025.ch20 Yu Z., and Shanbacher F.L., 2010. Production of methane biogas as fuel throu gh anaerobic digestion. In Singh O.V., and Harvey S.P., sustainable biotechnology http://download.springer.com/pdf Zegada Lizarazu, W and Monti A. 2012. Are we ready to cultivate sweet sorghum as a bioenergy feedstock? A review on field management practices. Biomass and Bioenergy 40: 1 12. Doi: 10.1016/j.biombioe.2012.01.048 Zhao Y L Dolat A Steinberger Y Wang X Osman A Xie G H. 2009. Biomass yields and changes in chemical composition of sweet sorghum cultivars grown for biofuel. Field Crops Res 111 (1 2) : 55 64 DOI: 10.1016/j.fcr.2008.10.006 Zhang, L., Lee, Y.W., Jahng, D., 2011. Anaerobic co digestion of food waste and piggery wastewater: focusing on the role of trace elements. Bioresource Tech nology 102 (8) : 5048 5059 DOI: 10.1016/j.biortech.2011.01.082 Zhang, Q., He J., Tian, M., Mao, Z. Tang L., Zhang J., and Zhang, H. 2011. Enhancement of methane production from cassava residues by biological pretreatment using a constructed microbial consortium. Bioresource Technology 102 (19) : 8899 8906 DOI: 10.1016/j.biortech.2011.06.061
135 BIOGRAPHICAL SKETCH Reginald Tou ssaint was born at Petit trou de Nippes, Haiti. He received his agricultural s c iences with a concentration in natural resources and e nvironment from the State University of Haiti in 2005. Thereafter, he worked for the Pan American Deve lopment Foundation as a local development facilitator. In 2011, he was enrolled in the Graduate School at the University of Fl interdisciplinary ecology with a concentration in soil and water s cience. Reginald worked as a rese arch assistant at the Bioenergy and Sustainable Technology Laboratory under application of anaerobic digestion for producing renewable energy and biofertilizer for developing countries including Haiti. Reginald earned his Master of Science degree in i nterdiscipli nary e cology in spring of 2013