Recipient Organization
STATE UNIV OF NEW YORK
(N/A)
SYRACUSE,NY 13210
Performing Department
Environmental Resources Engineering
Non Technical Summary
Approximately 25% of all food prepared in the US is wasted and food waste is the largest fraction of disposed municipal solid wastes in the US [1]. Treatment of food waste through anaerobic digestion would largely decrease the need for landfills. Anaerobic digestion not only solves the disposal issue, but also produces biogas which offers a great potential for energy recovery from methane (CH4). Although anaerobic digestion has been increasingly used to treat food waste in different parts of the world [2-5], the composition of food waste can vary greatly from country to country and from source to source (household, industrial, commercial, and institutional) within the same region [6]. Little is known about the types of food waste that are more suitable for anaerobic digestion or have a higher methane yield and, consequently, will result in greater economic benefits.Specific methane yield, which is calculated as the daily methane produced divided by the daily amount of volatile solids (VS) fed to a digester, is commonly used to indicate the efficiency of anaerobic digestion. Specific methane yield has been reported to be in a range of 0.3-0.6 L CH4/g VS of food waste [2,7-11]. Direct comparison among studies and extrapolation of a successful design to food waste with different compositions are difficult, given the variety of digester type and operational conditions used. C/N ratio is typically used to indicate applicability of anaerobic digestion for a given substrate, including food waste [10,12]. C/N ratio, however, is too general to distinguish various food wastes in their methane production potentials because of the highly varied composition of food waste [1]. For a given C/N ratio of food waste, its composition of carbohydrates, proteins, and lipids can be quite different. Lipid-rich waste such as fat and oil has high methane potentials [10]. Nevertheless,anaerobic digestion of high-lipid wastes may lead to a quick pH drop and sludge flotation, resulting in operational problems [10,13]. Anaerobic digestion of protein-rich food waste may generate too much ammonia that inhibits methanogens, while carbohydrate-rich food waste may have unfavorable C/N ratios due to nutrient limitation and fast acidification [2,10]. Our hypothesis is that specific methane yield in anaerobic digestion of food waste is regulated by the ratio of carbohydrate, protein, and lipid contents. Food waste is sorted at source into categories such as fats and oils, fruits and vegetables, and meat and fish. Therefore, carbohydrate : protein : lipid ratio can be estimated for a given combination of food waste and used for design of anaerobic digesters or adjustment of digester operational conditions. Our goal is to determine how food waste composition affects methane production and the economy of anaerobic digestion. Our specific objectives are to:Examine the effect of food waste composition on methane yield, thus determining the optimum composition of carbohydrate, protein and lipid in food waste for efficient anaerobic digestion;Determine the structure of microbial granules and kinetic parameters of methane production in anaerobic digestion of food waste with three different compositions, respectively;Determine the microbial community structure and optimum design and operational parameters for anaerobic digestion of food waste with different compositions;Explore the potential of co-digesting food waste with other organic wastes based on the optimum carbohydrate : protein : lipid ratio; andEvaluate the costs and benefits of anaerobic digestion for treatment of food waste with different compositions.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
Our goal is to determine how food waste composition affects methane production and the economy of anaerobic digestion. Our specific objectives are to:Examine the effect of food waste composition on methane yield, thus determining the optimum composition of carbohydrate, protein and lipid in food waste for efficient anaerobic digestion;Determine the structure of microbial granules and kinetic parameters of methane production in anaerobic digestion of food waste with three different compositions, respectively;Determine the microbial community structure and optimum design and operational parameters for anaerobic digestion of food waste with different compositions;Explore the potential of co-digesting food waste with other organic wastes based on the optimum carbohydrate : protein : lipid ratio; andEvaluate the costs and benefits of anaerobic digestion for treatment of food waste with different compositions.
Project Methods
The PI has used three 2-L completely stirred tank reactors (Figure 1) for co-digestion of food waste and dairy manure. The project proposed for the seed grant program will operate the three reactors in parallel for mesophilic anaerobic digestion (36 oC) of food waste with carbohydrate: protein: fat ratios set at 70:15:15, 55:15:30 and 55:30:15, respectively. The ratio of 70:15:15 represents the typical composition of food waste in the US [14]. The other two ratios represent fat-rich and protein-rich food waste. Sorted food waste will be collected from Sheraton Syracuse University Hotel or ESF's new Trailhead Café. Each sort of food waste will be analyzed witha NIRS - Foss NIRSystems Model 6500 at Dairy One, Ithaca, NY (http://www.dairyone.com) for proximate contents of water soluble carbohydrates, crude proteins, and fats, following AOAC International standards and methods. The three target compositions of feedstock will be prepared for 1-year use and stored frozen in small plastic bags.The digesters will be operated semi-continuously, or exchanging digestate with feedstock every two days at a hydraulic retention time of 30 days. Organic loading rate will be set at 4 g VS/L/d, which is a higher rate reported in the peer-reviewed publications for long-term stable operation of mesophilic digestion of food waste [2,7,11]. The initial 60 days of operation will be considered as a transition period from co-digestion of food waste and dairy manure to mono-digestion of food waste. After the transition period, the three digesters will be monitored weekly for biogas production rate using in-line gas meters, methane and carbon dioxide percentages in biogas samples using a Greenhouse Gas Chromatograph at ESF's AT&S, total solids and VS concentrations in waste digestate using the gravimetric methods, and ammonia concentration in waste digestate using a Lachat QuickChem autoanalyzer. Head space temperature, mixed liquor temperature, and mixed liquor pH will be measured in situ following biogas sampling and pressure recording. The size, compactness, porosity, and cell morphology of the microbial granules will be observed with a scanning electron microscope at the N.C. Brown Center for Ultrastructure Studies, following the methods described by Baloch et al. [15], once a month for three months. The differences between the digesters will be assessed with one-way analysis of variance in biogas production rate, specific methane yield, solids concentrations, ammonia concentration, and pH. The difference in microbial granular structure will be compared semi-quantitatively among the three digesters to reveal microbial response to the change in food waste composition.At the end of the semi-continuous operation of the digesters, feeding will be stopped for batch experiments to determine digestion kinetics. Cumulative biogas produced, mixed liquor pH, mixed liquor ammonia concentration, and mixed liquor concentrations of total solids, VS and volatile suspended solids will be measured periodically until stabilization. The biogas production dynamics will be analyzed using the first-order kinetics expression Y = Ym(1 - e-kt), where Y is the cumulative specific methane yield for a given time t, Ymthe ultimate specific methane yield, and k the first-order decay rate constant [16]. Excel Solver will be used to estimate the values of Ymand k with the time series of Y measurements using the least squares method.