Hydrolysis, methanogenesis and bioprocess performance during low-temperature anaerobic digestion of dilute wastewater

Abstract

Anaerobic digestion (AD) technologies represent sustainable and appropriate systems for the treatment of organic wastewaters. Lettinga et al (1987) stated, "A satisfactory application to raw domestic sewage would represent the maximum possible accomplishment for high-rate anaerobic treatment systems". The harnessing of this technology for municipal wastewater treatment, primarily using upflow anaerobic sludge blanket (UASB) reactors is now well established in tropical regions. A challenge remains, however, with regard to the full-scale application of these systems in countries with moderate- to low- ambient temperatures, and such applications have yet to be realised. To be successfully implemented at low temperatures, AD must overcome other physicochemical challenges and reportedly low levels of microbial activity. However, microbial populations are prolific in cold environments and no microbial barrier exists to the possibility of anaerobic digestion at low temperatures. This has been further demonstrated by successful lab and pilot-scale trials for a multitude of wastewaters over a range of low temperatures. The microbial communities underpin the entire AD pathway, thus the stability and sustainability of these systems depends on the microbial consortia that are still widely uncharacterised "black box" systems. The tools utilised in molecular microbial ecology are fundamental to understanding population dynamics. Linking molecular microbial ecology techniques to process data may help to provide essential insight into this complex and dynamic process. The overall themes of this thesis were associated with the hydrolysis, feasibility, stability and reproducibility of low-temperature anaerobic digestion (LtAD) of sewage. Essential to this understanding was the integration of process parameters, biomass characterisation in terms of hydrolysis kinetics and molecular characterisation using 16S rRNA gene analyses. Initially, the first step of the study (Chapter 2) commenced with a sludge-screening step in order to elucidate the factors influencing hydrolysis and biogas production. The activity of a sludge is an important factor to be taken into consideration during reactor design. It is essential to have sludge with a high activity on the substrate to be treated for efficient start up and avoidance of subsequent operational issues. In this study three anaerobic sludges (S1, S2 and S3) were chosen to degrade a range of complex, cellulose-derived, soluble and methanogenic substrates across three temperatures (10, 15 and 37°C). Activity was assessed through measuring the biogas production and from this the specific biogas activity was calculated. Hydrolysis was found to be rate limiting during the degradation of particulate complex substrates. Hydrolysis was positively correlated with increases in temperature. The greatest degradation of complex substrates was achieved using sludge originating from a full-scale reactor treating sewage sludge at 37°C. This sludge was non-granular in nature, while the remaining granular sludges proved more effective in the degradation of soluble compounds. The high level of variation in the biogas activity between sludge types highlights the importance of the long-term development of microbial consortia, particularly at low temperatures where doubling times are significantly lower. The two granular sludges from Chapter 2, which had displayed greater activity on soluble substrates, were chosen to seed a hybrid sludge bed-fixed-film reactor (Chapter 3). This trial was to assess the possibility and feasibility of long-term (732 days) anaerobic treatment of synthetic sewage at 12°C. Maximum organic loading rate (OLR) achieved was 1.5 kg COD m-3 d-1 with total COD concentrations within discharge limits for Ireland (125 mg l-1). Carbohydrate and protein removal rates routinely reached 100%. Phosphate removal was observed in the system (~78%), scanning electron microscopy (SEM) and energy dispersive X-ray (EDX) analysis revealed phosphate accumulation in the biofilm on the fixed-film section. Hydrolysis kinetics (Amax, Km, Vmax and k) indicated the development of an active proteolytic community with increased functioning at lower temperature. Specific methanogenic activity (SMA) revealed a preference for hydrogenotrophic methanogenesis at 12°C. Denaturing gradient gel electrophoresis (DGGE) of 16S rRNA gene sequence analysis revealed temporal changes in the microbial community, with cDNA analysis demonstrating greater banding patterns. Proteobacteria, Firmicutes, Chloroflexi, Bacteroidetes and Methanosaeta were demonstrated to be the dominant microbial communities through next generation sequencing (NGS) using Illumina technology. Psychrotolerant mesophilic Trichococcus species and Flavobacterium associated with low temperature protein degradation developed throughout the trial. Polyphosphate accumulating organisms (PAOs) such as Rhodocyclus, Chromatiales, Actinobacter and Acinetobacter were also discovered. A single perturbation event during the trial may have been as a result of an increase in the Firmicutes and hydrogenotrophic methanogenic population and subsequent decline in the acetoclastic community and related activity. In the final phase of this study the reproducibility of long term (889 days) low-temperature anaerobic digestion (LtAD) of synthetic sewage was demonstrated. This study focused on the community development based on the mesophilic inoculum utilised. For the evaluation two identical hybrid anaerobic sludge bed-fixed-film reactors were operated at 12°C for the treatment of SYNTHES. Chemical oxygen demand (COD) removal efficiencies were routinely >78% for total and soluble fractions. Hydrolysis kinetics indicated psychrophilic adaptation to the substrate in both reactors. K had increased by 20 times in Phase 4, compared to the seed inoculum. Km indicated a decrease in substrate affinity at mesophilic temperatures. qPCR indicated that the bacterial and archaeal numbers and relative composition were reproducible and stable between the systems. NGS demonstrated that the adaptation of the mesophilic inoculum and concurrent microbial community development was reproducible to family level in both reactors. NMDS and PCoA plots illustrated samples grouping based on sampling time. ANOSIM revealed that each time period was significantly different (P < 0.05), with no significant difference when based on reactor identity (P > 0.05). The microbial consortium identified in each reactor comprised Proteobacteria, Firmicutes, Bacteroidetes, Clostridia, Synergistetes and the acetoclastic Methanosaeta with hydrogenotrophic methanogens also present. Process divergence was observed in Phase 5. This may be attributed to the development of a larger proportion of Leptotrichiaceae and Desulfomaculum and a decrease in Trichococcus and hydrogenotrophic methanogens in one system. This highlights the importance of understanding microbial community dynamics to improve the AD process. Overall through the various approaches employed the findings of this PhD provide new evidence-based knowledge on the feasibility, efficiency and reproducibility of the anaerobic digestion process for the treatment of sewage at low temperature. Moreover, the study identifies the microbial consortium that developed underpinning the successful degradation of sewage at low-temperature.