TOWARDS LOW-CARBON EMISSION BIOTRICKLING FILTRATION OF VOLATILE ORGANIC COMPOUNDS FROM AIR: AN ARTIFICIAL NEURAL NETWORK APPROACH

In this study, a classical biotrickling filter (based on compost microorganisms) and an upgraded biotrickling filter (based on a mixture of compost microorganisms and microalgae Arthrospira platensis PCC 8005) are evaluated in terms of carbon dioxide production, during their use for volatile organic compounds (VOCs) removal from air. The experiments were performed using acetic acid vapors as model VOC and the biotrickling filter (BTF) performance was observed at different VOC concentrations, gas flowrates and pH values. Although the removal of acetic acid vapors was maximum for the both biosystems, the carbon dioxide production was different. The influence of the microorganisms� types and of the operating parameters on the carbon dioxide production are correlated via artificial neural network algorithms, depicting the most favorable conditions towards a low-carbon emission biotrickling filtration process for VOCs removal from air.