AbstractBiofiltration of solvent and fuel vapors may offer a costeffective way to comply with increasingly strict air emission standards. An important step in the development of this technology is to derive and validate mathematical models of the biofiltration process for predictive and scaleup calculations. For the study of methanol vapor biofiltration, an 8‐membered bacterial consortium was obtained from methanol‐exposed soil. The bacteria were immobilized on solid support and packed into a 5‐cm‐diameter, 60‐cm‐high column provided with appropriate flowmeters and sampling ports. The solid support was prepared by mixing two volumes of peat with three volumes of perlite particles (i.e., peat–perlite volume ratio 2:3). Two series of experiments were performed. In the first, the inlet methanol concentration was kept constant while the superficial air velocity was varied from run to run. In the second series, the air flow rate (velocity) was kept constant while the inlet methanol concentration was varied. The unit proved effective in removing methanol at rates up to 112.8 g h−1 m−3 packing. A mathematical model has been derived and validated. The model described and predicted experimental results closely. Both experimental data and model predictions suggest that the methanol biofiltration process was limited by oxygen diffusion and methanol degradation kinetics. © 1993 John Wiley & Sons, Inc.