(793d) Variable Reaction Order During Anaerobic Digestion of Industrial Wastewaters

Berson, R. E., University of Louisville
Ghorbanian, M., University of Louisville
Satyavolu, J., University of Louisville

Kinetic modeling of anaerobic digestion processes can be especially difficult when the exact composition of the feed stream is unknown, for example when the feed to the digester enters directly from a complex industrial chemical process, which may contain inhibitors of unknown nature and quantity.  There are a number of published models, but any given one is seemingly only accurate or valid for certain substrates, operating conditions, etc.  A new approach is considered here, which may challenge conventional thinking about kinetic rate constants. The hypothesis being that the rate constants may change over time due to the changing interactions between microorganisms, remaining substrate, and possibly inhibitors as substrate is consumed. 

Bench-scale respirometry tests were conducted to assess the impact of substrate consumption, which includes known and potentially unknown inhibitors, over the course of anaerobic digestion on the rate law parameters associated with the anaerobic digestion process. Three different industrial wastewaters (WW) were tested at mesophilic temperature (35 °C): soybean processing WW, brewery WW, and recycled beverage WW. The procedure involved adding wastewater at four COD concentrations (6, 8, 10, and 12 g/L) to bacterial sludge (12 g/L volatile suspended solids), continuously measuring methane production, and regressing the rate law model for varying initial substrate concentrations. Control reactors using ethanol as a toxicant-free substrate were tested as a basis for comparison. Culture seed blank reactors (no feed) were also included to correct for background gas production.

A first order rate equation is one commonly used approach to modeling the overall methane generation in the literature. However, it was observed here that the order of reaction increased and the rate constant decreased throughout the course of the AD reaction.  Results show that the rate order increased most rapidly when substrate conversion was most rapid since the reaction becomes prolonged at lower substrate concentrations and hence less contact between substrate and inoculum.  In the control reactor with no toxicants present, the reaction order reached 5.6 within the first ~20 hours of the process. In the brewery WW and beverage WW streams where sulfate content was low, the reaction order reached 3.1 and 2.5 within the first ~20 hours. In the soybean WW stream, where sulfates were extremely high, the reaction order reached just 1.5 within the first ~20 hours. The results show that rate law parameters are a function of time, which is attributed to increasing inhibitor-to-substrate ratio as substrate is consumed throughout the reaction.