When on day 23 of the operating period the mixing duration was changed from intermittent to continuous mixing, continuous mixing resulted in a higher biomass level in the effluent (12gVS/L) compared with an effluent biomass level of 8 gVS/L for the control reactor (Figure 4.5B) due to an increase in the SVI of the washed out biomass from 27 to 37 mL/gTS (the SVI of the biomass in the reactor increased from 15 to 19 mL/gTS; Angenent et al. 2001). Even though some biomass loss became apparent, the biomass concentration in the reactor stabilized to a level of 22gVS/L compared with 29gVS/L (Figure 4.5B) and stable reactor conditions prevailed with similar soluble COD and total volatile fatty acid (VFA) concentra­tions compared with the control reactor. After a steady- -tate biomass concentration was achieved in the bioreactors, the VMPR was slightly lower for the continuously mixed biore­actor compared with the intermittently mixed control (1.7 vs. 1.9 L/L/day; Figure 4.5A) due to a slightly lower VS removal efficiency in the continuously mixed bioreactor (related to a higher VS level in the effluent). In addition, the relative 16S rRNA levels for the predominant acetoclastic methanogens Methanosaeta concilii and the predominant hydrogenotrophic methanogens from the order Methanomicrobiales had only decreased somewhat in the con­tinuously mixed digester (but had stabilized) compared with the control reactor (Figure 4.5C, D). Thus, the change in mixing duration for the ASBR deteriorated the performance slightly with lower VS removal efficiencies, but did not jeopardize the methanogenic com­munity structure or the stability. This result shows that intermittent gentle mixing is advanta­geous, and indicates that continuous mixing is not required for efficient conversion in high-rate anaerobic digesters, and, in fact, reduced the methane yield. Noteworthy is that during the gentle mixing period the concentrations of the methanogenic populations were much lower in the effluent biomass compared with the reactor biomass (Figure 4.5C, D). This is another advantage of high-rate anaerobic digester systems.

During the 24-hour feeding cycle (ASBRs were fed instantly at t = 0), the shape of the cumulative biogas production curves were similar between the intermittently and continuously mixed bioreactors at gentle mixing conditions with a lower slope at the end of the cycle (Figure 4.6A). This confirms the previous conclusion that reactor stability was not

jeopardized during continuous/gentle mixing. The shapes of the cumulative production in both bioreactors show a lag period with a relatively low biogas production rate just after feeding and then a maximum rate (i. e., slope) between 1 and 3 hours after feeding, corre­sponding to a higher acetate concentration in the mixed liquor between 600 and 1000mg/L (Figure 4.6B, C). The lowest biogas production rates were observed at the end of the 24-hour cycle when the acetate concentration was ~100mg/L in both systems (Figure 4.6B, C). Some differences were noticed between the cumulative curves of the two bioreactors (no difference was observed when both reactors were intermittently and gently mixed), because the rate of biogas production was faster ~1 hour after feeding for the continuously mixed bioreactor versus the control bioreactor (Figure 4.6A), indicating minor diffusion limitations in the intermittently mixed reactor. After 4 hours in the cycle, however, the cumulative biogas production in the intermittently mixed bioreactor had caught up with the continuously mixed bioreactor due to a higher maximum rate between 1 and 3 hours after feeding swine waste (Figure 4.6A). The maximum biogas production rates (slope) matched the location of the maximum acetate concentration, which occurred for the continuously mixed reactor after ~1 hour (660 mg/L) and for the intermittently mixed control reactor after ~2 hours (1015mg/L; Figure 4.6B, C). This result shows that mixing duration affects the function of the anaerobic food web for high-rate anaerobic digestion.