Collins

Four combinations of litter and carcasses from broiler chickens were examined utilizing a thermophilic, stirredtank digester of demonstration size of approximately 10,000 gal. Under computed optimal loading rates, litter with paper bedding had the highest daily production of methane over an 8-day retention period. The greatest methane production per lb of volatile solids was achieved over 10 days with litter and paper bedding combined with carcasses. This research found that sufficient poultry litter is generated within 20 mi (32 km) of Moorefield, WV, to support a commercial-sized digester operation. However, anaerobic digestion of poultry waste cannot be financially supported by methane production alone. To be financially viable, anaerobic digestion requires a disposal fee for poultry waste and/or the sale of the digested solid effluent as an organic fertilizer to retail markets. INTRODUCTION The poultry industry is West Virginia’s largest agricultural enterprise, totaling 51% of the state’s agricultural cash receipts in 1997. Much of this industry is concentrated in the Potomac Headwaters region (Grant, Hampshire, Hardy, Mineral, and Pendleton counties). This concentrated production is supported almost exclusively by imported feed, IMPLICATIONS Investments in anaerobic digestion of poultry waste require more than monetary returns from methane gas alone to pay for themselves. When computed under optimal loading rates, present value calculations of the revenue generated from methane production did not cover the estimated investment cost of a digester facility. This facility would need additional revenues, such as a disposal fee for the poultry litter or revenue from the sale of digested litter as a soil amendment, to make up for the shortfall. resulting in a large surplus of nutrients contained in the waste products. Seven million lb of nitrogen and phosphorus are produced in the 160,000 tons (145,000 metric tons) of poultry litter generated annually in the five-county region. Numerous concerns have been raised about the potential for water quality impacts from land disposal of this litter, particularly in regard to nitrogen and phosphorus flowing into the Potomac River and the Chesapeake Bay.1,2 To address such concerns about poultry litter, anaerobic digestion technology has been promoted as one method to transform poultry litter into both energy and a fertilizer product with a lower potential for nutrient runoff. To encourage this technology, the West Virginia Department of Agriculture awarded a contract to Olin Corp. in September 1994 to conduct a demonstration project of Olin’s unique process for anaerobic digestion. Prior research has demonstrated the technical feasibility of anaerobic digestion of animal wastes, including poultry waste.3,4 Operational parameters have been established to provide for optimal production.5 However, the economic feasibility of digestion using poultry litter has not been favorable.6 Economics of size for methane production are very important. Large-scale digestion (for example, handling poultry manure from over 500,000 chickens) has a much higher rate of return on capital investment for methane production than do smaller operations.6 Given the above prior research, the objectives of this study are to (1) determine the optimal loading rates for poultry waste mixtures within a demonstration-size digester to maximize methane production from anaerobic digestion; and (2) project the poultry waste requirements for, and economic feasibility of, expanding this demonstration digester into a larger, more commercially viable size. Collins, Murphy, and Bainbridge 1038 Journal of the Air & Waste Management Association Volume 50 June 2000 PROJECT DESCRIPTION The demonstration digester developed by Olin was designed to process up to 1 ton of waste (loaded daily) and produce up to 2000 ft3 of methane daily. The demonstration digester tank holds 9635 gal (36,472 L). With 16.67% reserved for air space, this leaves 8030 gal (30,397 L) of liquid holding capacity. The Olin digester uses a state-of-the-art, fully automated microprocessor control operated by a programmable logic controller along with a process/recipe station for operator input.7 Limited operator intervention in the digestion process is required. Specific operations, such as digester loading, recirculation, and mixing, can be accomplished by remote access to the process computer. The technology is a thermophilic, stirred-tank design. Heat increases the rate of bacteria reproduction, destroys pathogens, and reduces the retention time needed to digest the slurry. Stirring releases gases trapped by scum layers that form in the digester and also mixes the anaerobic bacteria into the incoming slurry. This reduces the time required for the new material to begin methane generation. Prior to loading in the digester, the size and density of poultry waste are reduced by grinding to increase the surface area available for bacterial action. Water is added in a mixing tank to form a semi-liquid slurry with less than 7% total solids. This slurry is heated to 132 °F in a heater. Digestion occurs in a three-step process. First, enzymes convert complex organic compounds into simpler, soluble compounds. These compounds are then converted by acid bacteria into soluble simple organic acids, mainly acetic acid. Finally, methane bacteria convert the organic acids into methane and carbon dioxide. This process also generates small amounts of hydrogen sulfide and ammonia. Upon discharge, some of the gas and liquid effluent are reintroduced into the tank along with new material to continue the process. After separation, the remaining solid and liquid effluents can be used as fertilizers. Construction of the demonstration digester north of Moorefield began in February 1995 and was completed by June 1995. The digester was purged of oxygen, and a starter culture of microbes was introduced. Tests and pilot operations were conducted throughout 1995, until the digester reached design processing and methane generation capacity in early 1996. The digester was operated for a period of approximately 8 months during 1996. Four trials of broiler chicken waste material inputs were examined: (1) litter with paper bedding, (2) litter with wood chip bedding, (3) litter with paper bedding and chicken carcasses, and (4) litter with paper bedding and twice the amount of carcasses as in trial 3. Due to operational problems with the grinding and loading of poultry waste slurry at the demonstration digester, the loading rates varied considerably (from 0 to 870 gal) during this 8-month trial period. METHODS Objective 1 The variations in loading rates and mixtures during the four trials enabled determination of an optimal loading rate through estimation of multiple regression models. To explain methane production, a single model was developed to forecast the maximum gas production with the digester operating at optimal loading rates. Daily report data of loading and gas production from the demonstration digester were used in model estimation.7 The dependent variable for the regression model was daily methane production in cubic feet. During the first three trials of waste inputs, the volume of gas production from the digester was measured by a mechanical flow meter. During the fourth trial, an electronic flow meter was used in addition to the mechanical one. The electronic flow meter produced much higher and more accurate estimates of gas output than the mechanical meter. To adjust the previous three trials for projected electronic flow estimates, a linear relationship was derived between electronic and mechanical flow meter estimates in ft3/day. The estimated coefficients for this linear relationship were electronic flow = 151.81 + 1.5008 * mechanical flow. The adjusted R2 was 0.663. This linear relationship was corrected for autocorrelation between error terms (Rho = 0.3575), and two observations were dropped because they were obvious outliers in the data. To determine the quantity of methane gas produced, the electronic flow estimates were multiplied by the measured percentage of methane in the daily gas production. To explain methane production, various independent variables were examined. The primary variable utilized was reduction in volatile solids, RVS, measured as the daily reduction in pounds of volatile solids within the digester. The theoretical basis for this variable came from research by Morris et al.8 As shown, RVS was computed as a ratio of two calculations: (1) a reduction in volatile solids that were fed into the digester in the numerator, and (2) the average retention time (in days) in the denominator. % VS input * % TS input * loading rate * % reduction in VS * 8.345 lb/gal 8030 gal (digester capacity) / loading rate RVS =