Performance assessment of gas-phase toluene removal in one- and two-liquid phase biotrickling filters using artificial neural networks.
Journal:
Chemosphere
Published Date:
Nov 1, 2019
Abstract
The main aim of this work is to study gas-phase toluene removal in one- and two-liquid phase biotrickling filters (O/TLP-BTF) and model the BTF performance using artificial neural networks (ANNs). The TLP-BTF was operated for 60 d in the presence of silicone oil at empty bed residence times (EBRTs) of 120, 60, and 45 s, respectively, and toluene concentrations in the range of 0.9-3.1 g m. A t-test analysis indicated that increasing the silicone oil volume ratio from 5 to 10% v/v, did not significantly improve the TLP-BTF performance (p-value = 0.65 > 0.05). The results from ANN modeling showed that toluene removal was more negatively affected by the inlet concentration (casual index, CI = -5.63) due to the kinetic limitation. The CI values for inlet concentration (+4.01) and liquid trickling rate (-2.45) indicated that the diffusion-limited regime controlled the removal process in the OLP-BTF.