A characterization of the three potential biofilter packing materials was performed. The highest water retention capacity (WRC) was found in vermiculite (65%), while the WRC for lava rock was 15%. Although vermiculite showed a higher WRC, lava rock favored water irrigation, which ensured that the desired moisture level was maintained and avoided sulfate accumulation at the same time.

The acetic, propionic, butyric and valeric acid could also be present in the biogas stream, their assimilation was determined in batch experiments. The biodegradation of different loadings of acetic and propionic acids as individual substrates were also evaluated in the lava rock biofilter as it as previously described elsewhere (Ramirez-Saenz et al., 2009).

As the MR proportion increased in the ADS, the H2S concentration in the biogas stream also increased. The biodegradation of H2S was determined in the lava rock biofilter under two different empty-bed residence times (EBRT). Results for H2S elimination capacity as a function of H2S inlet loading in the lava rock biofilter, operated at 85 sec and 31 sec EBRT, are depicted in Figure 2. As shown in Figure 2A, at an EBRT of 85 sec, the relationship between the inlet loading and the elimination capacity was linear, and the critical H2S elimination capacity defined by Devinny et al., 1999 (i. e., deviation from the 100% removal capacity) was not yet reached at an inlet loading of 144 g/ m3h. Under these operation conditions, the removal efficiency of H2S for loadings between 36 and 144 g/ m3h was always above 98 %. Furthermore, the EC reached a maximum of 142 g/ m3h when the H2S loading was 144 g/m3h.

For an EBRT of 31 sec (Figure 2B), the H2S elimination capacity was found to be linear with respect to H2S inlet loading up to 200 g/ m3h (100% removal efficiency). A higher inlet loading of 300 g/ m3h reduced the removal efficiency in the system to 85 %. An inlet loading of 400 g/ m3h (corresponding to 3000 ppmv) caused the removal efficiency to drop to 75%, which suggested inhibition of biological activity and/or insufficient mass transfer. In this case, the critical H2S EC was 200 g/ m3h, whereas a maximum H2S EC value of 232 g/ m3h was achieved in the biofilter. At the same time, however, the removal of VFAs present in the gaseous stream (approximately 10 ppmv) reached 99%.

For long operation times, the biofilter nearly eliminated all the H2S from the biogas stream. The H2S concentrations of the AD gas stream were previously diluted to maintain an inlet concentration of 1500 ppmv and to allow complete elimination (99% removal efficiency), but the high removal efficiency was maintained over 90 days and complete biodegradation of VFAs was also observed. Fifty days after the start-up period, a technical failure in the AD system blocked the feeding of the biofilter, no data was obtained during that time. After operation conditions were restored, an inlet H2S concentration was maintained at approximately 1500 ppmv from day 103 to 194 at an EBRT of 31 sec. Under these conditions, a removal efficiency of 95% was maintained for 90 days. Higher concentrations, around 3000 ppmv, caused a drop in the biofilter efficiency to 50% (Ramirez-Saenz et al., 2009). The biofilter was fed with the ADS gas stream every two weeks, which corresponded to the HRT of the AD system.

According to the stoichiometry of aerobic biological H2S oxidation (Eq. 1 and 2) and the sulfate determinations obtained between days 103 and 194 of the operational period of the biofilter, 51 to 60% of the H2S was completely oxidized to sulfate. These data are correlated with those reported by Fortuny et al., 2008 with respect to the H2S conversion to sulfate. The elimination of these compounds allowed the potential use of the biogas while maintaining the methane (CH4) content throughout the process.