The effect of temperature reduction on the performance and stability of UASB reactors treating synthetic sewage

Specific Methane Production (SMP) The objective of this investigation is to determine the effect of temperature reduction on the performance and stability of laboratory-scale up-flow anaerobic sludge blanket (UASB) treating synthetic sewage. The work was carried out using four 4litre continuously fed UASB reactors. The operation started at a constant hydraulic retention time of ~24 hours (up flow velocity ~0.02 m/hour) and an influent COD concentration of ~2250 mg/L, with an organic loading rate of ~2.25 g COD/L/day. During the experiment, the operating temperature was reduced in a single step from 35 to 30 0C, then stepped down to 25 0C. COD removal efficiency fell slightly from ~96 % in all reactors to ~93%. Effluent TSS concentrations remained below 20 mg l and TSS removal efficiency ranged between 93-98%. Gas production showed a slight disturbance following the initial drop in temperature to 30 0C, but stabilised at around 0.74 L CH4/L, 0.30 L CH4/g CODadded and 0.31 L CH4/g CODremoved. The subsequent drop to 25 0C produced a stronger disturbance, but both volumetric and specific methane production recovered to the previous values by about day 412. The average biogas methane content in all reactors was unchanged or marginally higher at 78-79%.


Introduction
Temperature has a considerable influence on the growth and survival of microorganisms. Below 30 ⁰C, the methanogenic activity in an (up-flow anaerobic sludge blanket) UASB may be affected and the digestion rate will decrease by about 11% for each ⁰C temperature decrease (van Haandel and Lettinga 1994;Gómez 2011). Although anaerobic treatment is possible in all three temperature ranges (psychrophilic, mesophilic and thermophilic), low temperature usually leads to a decline in the maximum growth rate and methanogenic activity (Bodik et al., 2000). However, long-term acclimation to lower temperatures is possible.
According to Seghezzo et al. (1998), the UASB process has been successfully applied for low strength wastewaters such as domestic sewage at 20 ⁰C or above, with chemical oxygen demand (COD) removal efficiencies in the range of 57% to 82% at hydraulic retention time (HRT) from 5 to 15 hours (Seghezzo et al., 1998). Lettinga et al. (1983) reported that when raw domestic sewage was treated in a 120-L reactor at temperatures in the range of 8-20 °C, 65-85% COD reduction was achieved (Lettinga et al., 1983). COD removal with 97% efficiency was achieved at 20 ⁰C and 90% COD reduction was still observed when the temperature dropped to 10 ⁰C. At lower temperatures (5-20 ⁰C), however, the performance of one-stage UASB reactors can be severely limited by the slow hydrolysis of entrapped solids that accumulate in the sludge bed when high loading rates are applied. Singh and Viraraghavan et al. (1996) used a UASB system to treat municipal wastewater in lowtemperature conditions and reported 70% COD removal at 11 ⁰C and 30-50% at 6 ⁰C. Lew et al. (2003) reported a gradual decrease in COD removal efficiency with decreasing temperature, from 82% at 28 °C to 72% at 20 ⁰C, 68% at 14 ⁰C and 38% at 10 ⁰C. Halalsheh (2002) treated a relatively high strength municipal sewage (COD = 1531 mg/L) from Jordan in a pilot UASB reactor operated at temperature (18-25 ⁰C). Total COD removal efficiencies were 62% and 51% in summer and winter, respectively at HRT of 24 hours. Singh and Viraraghavan (1998) used two UASB reactors, with total volume of 8 L, for the treatment of municipal wastewater at 20 ⁰C. Start-up was at a HRT of 48

ORIGINAL RESEARCH
Open Access hours, and was reduced to 10 hours by the 280 th day of operation. The total COD of the municipal wastewater varied in the range of 350 to 500 mg/L with soluble COD of 150-300 mg l -1 . The reactors achieved a treatment efficiency in terms of COD removal of the order of 60-75% (Singh and Viraraghavan, 1998). Takahashi et al. (2011) investigated treatment of raw sewage using a 1.15-m 3 pilot-scale UASB reactor operated at a HRT of 8 hours at wastewater temperatures ranging from 10.6 to 27.7 ⁰C for more than 1100 days. The stable removal efficiencies for total COD and SS were 63± 13% and 66 ± 20 %, respectively. After two years of operation, the average concentration of the retained sludge was 24.5 g SS/L. The solid retention time was evaluated as 293 ± 114 days, which was sufficient for mineralisation of solid organic matter, as indicted by a growth yield of 0.132 g VSS/g COD. In summer, the water temperature increased above 20 °C leading to enhanced biodegradation. The above results indicate considerable potential for UASB operating on low-strength municipal-type wastewater at ambient temperatures of 30 ⁰C or below, but also leave some questions unanswered.
The objective of this investigation is to determine the effect of temperature reduction on the performance and stability of laboratory-scale UASB in treating synthetic sewage.

Research Methods
An experimental investigation was carried out using four 4-litre continuously fed UASB reactors ( Figure  1). The synthetic sewage feed was used and prepared daily from frozen pre-prepared concentrate by dilution with tap water to obtain the desired organic loading rate (OLR). The composition of the synthetic wastewater used is given in Table 1. The operation started at a constant HRT of ~24 hours (up flow velocity ~0.02 m/hour) and an influent COD concentration of ~2,250 mg/L, with an OLR of ~2.25 g COD/L/day. During the experiment, the operating temperature was reduced in a single step from 35 to 30 ⁰C, then stepped down to 25 ⁰C as indicated in Table 2. Temperature reduction was achieved by adjusting the thermostat on the thermo-circulator.

Characteristics of synthetic sewage
The average characteristics of the synthetic wastewater are shown in Table 3.

Reactor performance
The main performance parameters are summarised in Table 4 and Figure 2.

Treatment performance
Effluent COD concentrations showed little or no change on reduction to 30 ⁰C, but rose sharply from around 100 mg/L to 150 mg/L as the temperature was further reduced. COD removal efficiency fell slightly from ~96% in all reactors to ~93%. Effluent TSS concentrations remained below 20 mg/L and TSS removal efficiency ranged between 93-98%. a Volumetric methane production b Specific methane potential (SMP) per g COD added c Specific methane potential (SMP) per g COD removed d ratio of actual SMP per g COD removed to the theoretical value of 0.35 L/g COD Figure 2. Volumetric methane production, biogas methane content, specific biogas and methane production, COD removal, TSS removal and actual/theoretical methane for R1-4.The vertical dotted lines indicate a change in temperature.

Biogas production
Gas production showed a slight disturbance following the initial drop in temperature to 30 ⁰C, but stabilised at around 0.74 L CH4/L, 0.30 L CH4/ g CODadded and 0.31 L CH4/g CODremoved. The subsequent drop to 25 ⁰C produced a stronger disturbance, but VMP and SMP recovered to the previous values by about day 412. The average biogas methane content in all reactors was unchanged or marginally higher at 78-79%. Table 5 and Figure 3 show the results at 30 and 25 ⁰C together with the values for 35 ⁰C. A slight fall in VMP can be seen, as well as the increase in biogas methane content. The performance was better than expected as it showed relatively little change despite the large change in operating temperature.

Specific experimental findings
Step changes in operating temperature to 30 ⁰C led to some short-term disturbance in performance, seen in fluctuating gas production parameters, which stabilised after a period of ~30 days at values slightly below those at the original operating temperature of 35 ⁰ C but stabilised at around 0.74 L CH4/L, 0.30 L CH4/g COD added and 0.31 L CH4/g COD removed.
A similar disturbance and further reduction in specific and volumetric gas production was seen on reducing the temperature to 25 ⁰ C but VMP and SMP recovered to the previous values after ~20 days. A marked change in effluent COD concentration from ~100 to ~200 mg/L occurred when the temperature was reduced from 30 to 25 ⁰ C. This recovered to ~150 mg/L after ~20 days, but showed no further decline in the experimental period.
Despite the relatively short acclimatisation and operating period, performance in terms of both treatment and energy conversion remained good, indicating that the system had the potential to operate effectively at lower temperatures without extended acclimatisation.

Conclusions
This experiment showed that the reactors experienced some disturbance after a reduction in temperature, which could be seen in fluctuating gas production parameters for a period of ~30 days, after which stabilisation appeared to occur. There was a marked change in COD removal and effluent COD concentration when the temperature was reduced from 30 to 25 ⁰C. Performance in terms of both treatment and energy conversion remained high.