Resilient performance of an anoxic biotrickling filter for hydrogen sulphide removal from a biogas mimic: Steady, transient state and neural network evaluation

Abstract Biological hydrogen sulphide (H2S) removal from a biogas mimic (pH = ∼7.0) was tested for 189 days in an anoxic biotrickling filter (BTF) inoculated with a pure culture of Paracoccus versutus strain MAL 1HM19. The BTF was packed with polyurethane foam cubes and operated in both fed-batch and continuous modes. The H2S inlet concentration to the BTF was varied between ∼100 and ∼500 ppmv during steady-state tests, and thereafter to ∼1000, ∼2000, ∼3000 and ∼4000 ppmv during shock-load (i.e. transient state) tests. The H2S removal efficiency (RE) ranged between 17 and 100% depending on the operational mode of the BTF and the presence of acetate as a carbon source. The maximum elimination capacity (ECmax) of the BTF reached 113.5 (±6.4) g S/m3 h with 97% RE during H2S shock-load experiments at ∼4000 ppmv which showed the robustness and resilient capacity of BTF for the large fluctuations of H2S concentrations. The results from polymerase chain reaction denaturing gradient gel electrophoresis (PCR–DGGE) revealed that P. versutus remained dominant throughout the 189 days of BTF operation which can imply the crucial role of this bacterium to remove H2S and upgrade to clean biogas. The analysis using artificial neural networks (ANNs) predicted the H2S and NO3−-N REs and SO42− production in the anoxic BTF. Consequently, this study revealed that a BTF can be used to treat H2S contamination of biogas under anoxic conditions.