Biofiltração e biopercolação de metano presente em gases residuais gerados em processos anaeróbios

This study investigated the use of biofilters and biotrickling filters for methane (CH4) abatement in air mixtures representatives of fugitives emissions from the desorption of dissolved gases in effluents from anaerobic reactors treating domestic sewage (0.3 to 3.3 %v/v), also named waste gases. The biofilters and biotrickling filters were operated in bench scale (useful volume = 7.9 L), being subjected to different inlet loads (approx. 3 to 152 gCH4.m-3.h-1) and empty bed residence times (42.8 to 7.4 min). The study considered the use of the following novel packing media for improving CH4 abatement, mixed or not with organic fraction of composted leaves: Biobob®, a sponge-based material consisting of polyethylene rings filled with polyurethane sponges; blast furnace slag; and expanded vermiculite. Additionally, to overcome limitations related to the solubilization of CH4 gas in the aqueous phase of the biofilm of the reactors, the strategy of adding mass transfer vectors in this aqueous phase was also evaluated. The following substances and volumetric fractions were tested: surfactant Tween 20 (0 to 0.5 %v/v); propylene glycol (0 to 60 %v/v); silicone oil (0 to 50 %v/v); soybean oil (0 to 50 %v/v). A pre-selection step based on tests of CH4 solubility in mixtures of the mass transfer vectors and on methanotrophic activity tests performed in enriched bacterial cultures submitted to varying concentrations of the vectors was carried out. Then, silicone and soybean oil (which showed the greatest potential for use as vectors) were tested in long-term operation of the biotrickling filters. The biofilter packed with composted leaves and expanded vermiculite showed high CH4 conversion efficiencies (> 90%) when operated at an empty bed residence time higher than or equal to approx. 30 min and when it was fed with a waste gas containing CH4 concentrations less than or equal to approx. 1.2 % v/v. For the other packing media tested (composted leaves with blast furnace slag or sponges), the long-term operation of the biofilters at the same operating conditions resulted in much lower CH4 conversion efficiencies, ranging from around 25 to 55%. The operation of the biotrickling filters (packed only with sponges or expanded vermiculite, and added by a liquid phase containing nutrients) showed very similar performance to the biofilters, proving to be a less attractive alternative, since its operation is much more complex and costly than the biofilters. Despite the promising results obtained in the solubility tests and in the methanotrophic activity tests performed for silicone oil and soybean oil, in operational scale the beneficial effect of using these vectors has not been confirmed. A complete failure of the biotrickling filter packed with expanded vermiculite and added by soybean oil (50 %v/v) was observed, as well as poor performance in CH4 conversion (< 20%) for the biotrickling filter packed with sponges and added by silicone oil (5 to 50 %v/v) (submitted to different CH4 inlet concentrations, approx. 0.3 to 3.3 %v/v and to an empty bed residence time of 29.5 min). In the case of the biotrickling filter filled with expanded vermiculite and added by silicone oil (5 %v/v), there was a greater removal efficiencies (38 to 64%), but still lower than the efficiency observed for more simplified systems (biofilters and biotrickling filters without the addition of silicone oil).