Modelling phosphorus (P), sulphur (S) and iron (Fe) interactions during the simulation of anaerobic digestion processes

This paper examines the effects of different model formulations when describing sludge stabilization processes in wastewater treatment plants by the Anaerobic Digestion Model No. 1 (ADM1). The proposed model extensions describe the interactions amongst phosphorus (P), sulfur (S), iron (Fe) and their potential effect on total biogas production (CO₂, CH₄, H₂ and H2S). The ADM1 version, implemented in the plant-wide context provided by the Benchmark Simulation Model No. 2 (BSM2), is used as the basic platform (A0). Four (A₁ – A₄) different model extensions are implemented, simulated and evaluated. The first approach (A₁) considers P transformations by accounting for the kinetic decay of polyphosphates (X_PP) and potential uptake of Volatile Fatty Acids (VFA) to produce Polyhydroxyalkanoates (X_PHA) by Phosphorus Accumulating Organisms (X_PAO). The second model formulation (A₂) describes biological production of sulfide (SH2S) by means of Sulfate-Reducing Bacteria (X_SRB). This approach also considers potential SH2S inhibition effect on biomass and mass transfer phenomena (aqueous-gas). The third evaluated model (A3) considers chemical iron (III) (S_Fe+3) reduction to iron (II) (S_Fe+2) using hydrogen (S_H2) as the electron donor. Finally, the last evaluated approach (A₄) is based on accounting for Multiple Mineral Precipitation. The ADM1 thereby switches from a 2-phase (aqueous-gas) to a 3-phase (aqueous-gas-solid) system. Simulation results show that the implementations of A₁ and A₂ lead to a reduction in biogas production. This reduction is attributed to two factors. Firstly, there is a fierce competition for substrate (S_H2, VFA) between the existing and the new groups of microorganisms. Secondly, there is a decrease of aceticlastic and hydrogenotrophic methanogenesis due to SH2S inhibition. Models A₃ and A₄ reduce the free S_H2S (and consequently inhibition) plus cationic load and soluble P availability due to ion pair formation and metallic carbonate/phosphate precipitation. The final version of the manuscript will provide a deeper analysis of the different model assumptions, the effect that operational/design conditions might have on the model predictions, a detailed description of the weak acid-base chemistry and practical implications in view of plant-wide modelling/development of resource recovery strategies.