How Inert COD Fractions Can Impact Methane Yield: A Food Waste Co-digestion Case Study

Process modelling is a valuable tool to design and optimize co-digestion system operation, however, co-digestion of solids is often challenging to model as the feedstock is often variable in quantity and quality. Proper feedstock characterization is important for co-digestion modelling, as uncertainties associated with the biodegradability of the waste could lead to inaccurate model predictions. Often, the reported total COD values do not differentiate between biodegradable and inert organic matter, and information on the inert matter is often lacking when reporting feedstock characterizations. A case study was conducted with the GPS-X Co-digestion and Biosolids Treatment Library using feedstock data from a public anaerobic digester modelling tool which is available from the U.S. Environmental Protection Agency. Each feedstock lists parameters such as total solids (%TS), volatile solids (%VS) and the methane yield (L/kg VS). The Co-digestion and Biosolids Treatment Library include a modified ADM1 model in which feedstocks needs to characterize with respect to carbohydrates, proteins, lipids, and inert material. It is found that without characterizing inert COD, the model tends to overpredict the methane yield from each feedstock, and modelling error could occur when inaccurate waste characterization is used. The inert COD is then calculated with theoretical COD to VS ratios for carbohydrates, protein, lipids and inert, as well as the chemical ratio of COD to methane. By determining the inert organic matter fraction, the model can predict a methane yield that is within 5% of the reported value.