Modeling process instability in anaerobic digestion by using a thermodynamically corrected model.

Mathematical modeling of anaerobic digestion requires consideration of many different aspects. One of these is the computation of biochemical reaction rates under varying conditions, where probing mechanisms for process instabilities are of special importance in finding optimal efficiency. This work implements a model with thermodynamic considerations applied to all aceto- and methanogenic processes on the basis of respective Gibbs free energies. It demonstrates their necessity to reflect the accumulation of longer-chain acids in some cases of process instability. For this purpose, the results of different computational setups are compared to experimental data for methane production and acid concentrations of a mesophilic and a thermophilic reactor under varying conditions. In the mesophilic setup with process instability, propionic acid of the thermodynamic simulation represented experimental values well (7.8 g/L compared to 7.1 g/L), while the reference computation without a thermodynamic approach represented less than 1% of the experimentally observed concentration (0.01 g/L compared to 7.1 g/L). An investigation into mechanisms for process collapse under conditions with high acid concentrations ( > 8 g/L) but relatively stable pH (6.4) suggested the incorporation of an inhibition term by overall acid concentration for both methanogenic pathways.