RESUMO
Mathematical modelling of anaerobic digestion process has been used to give new insights regarding dynamics of the long chain fatty acids (LCFA) inhibition. Previously published experimental data, including batch tests with clay mineral bentonite additions, were used for parameter identification. New kinetics were considered to describe the bio-physics of the inhibitory process, including: i) adsorption of LCFA over granular biomass and ii) specific LCFA substrate (saturated/unsaturated) and LCFA-degrading populations. Furthermore, iii) a new variable was introduced to describe the state of damage of the acetoclastic methanogens in order to account for the loss of cell-functionality (inhibition) induced by the adsorbed LCFAs. The proposed model modifications are state compatible and easy to be integrated into the International Water Association's Anaerobic Digestion Model N°1 (ADM1) framework. Practical identifiability of model parameters was assessed with a global sensitivity analysis, while calibration and model structure validation were performed on independent data sets. A reliable simulation of the LCFA-inhibition process can be achieved, if the model includes the description of the adsorptive nature of the LCFAs and the LCFA-damage over specific biomass. The importance of microbial population structure (saturated/unsaturated LCFA-degraders) and the high sensitivity of acetoclastic population to LCFA are evidenced, providing a plausible explanation of experimental based hypothesis.
