Elsevier

Bioresource Technology

Volume 269, December 2018, Pages 153-161
Bioresource Technology

Granulation process in an expanded granular sludge blanket (EGSB) reactor for domestic sewage treatment: Impact of extracellular polymeric substances compositions and evolution of microbial population

https://doi.org/10.1016/j.biortech.2018.08.100Get rights and content

Highlights

  • An EGSB reactor was used for the treatment of low-strength domestic sewage.

  • The sludge granulation process and mechanism were systematically investigated.

  • 3D-EEM is used to identify the EPS compositions and distribution during sludge granulation.

  • A feasible start-up strategy for the treatment of low-strength domestic sewage was developed.

  • High-throughput sequencing results suggest a shift of methanogen metabolism during granulation.

Abstract

In this study, an expanded granular sludge blanket (EGSB) reactor was used for the treatment of low-strength domestic sewage and the sludge granulation process was systematically investigated. At an optimized hydraulic retention time (HRT) of 5 h, up-flow velocity (Vup) of 1.9 m/h, and organic loading rate (OLR) of 2.16 kg COD/m3/d, the average COD removal efficiency was 71.5 ± 2.3%. Completely granular sludge can be observed after 107 d of continuous operation. Analysis of the distribution and composition of the extracellular polymeric substances (EPS) indicates that the tightly bound EPS (TB-EPS) content shows an increasing trend, while the loosely bound EPS (LB-EPS) content did not significantly alter after the granular sludge was formed. The three-dimensional excitation-emission matrix technique (3D-EEM) confirms that aromatic protein-like substances are of key importance to sludge granulation. High-throughput sequencing analysis indicates that the metabolism shifted from hydrogenotrophic (Methanobaterium) to aceticlastic methanogens (Methanosaeta) during sludge granulation.

Introduction

Anaerobic granular sludge technology is a promising method for the treatment of organic pollutants due to its energy effectiveness, limited nutrients requirements and low sludge production (Chi et al., 2018, Chong et al., 2012, Huang et al., 2016). Numerous advanced anaerobic reactors, including internal circulation (IC) reactors and expanded granular sludge blanket (EGSB) reactors, have been developed and applied for the treatment of various high-concentration industrial wastewaters (Stazi and Tomei, 2018, Teng et al., 2018). However, these anaerobic reactors usually require a long-term and high-concentration of substrates to start-up (Ghangrekar et al., 2005, Hulshoff Pol et al., 2004, Zhang et al., 2017), resulting in limited engineering applications of anaerobic processes for the treatment of low-strength wastewater such as domestic sewage (Liu et al., 2018a).

The key to achieving start-up of an anaerobic reactor is to cultivate granular sludge with excellent settleability and high biomass concentration. This depends on the biodegradation function of the microorganisms in the sludge, as well as other operational parameters (e.g., hydraulic retention time (HRT) and up-flow velocity (Vup)) (Hulshoff Pol et al., 2004, Yang et al., 2017). Mounting evidence supports that extracellular polymeric substances (EPS) excreted by microorganisms play a key role during sludge granulation despite certain discrepancies (Bala Subramanian et al., 2010, Chen et al., 2016, Shi et al., 2017). One report shows that the acetotrophic methanogen Methanosaeta is responsible for sludge granulation (Hulshoff Pol et al., 2004), while other studies suggest that the granulation process is due to the excessive amount of EPS excreted by Methanobacterium (Suarez et al., 2018, Torres et al., 2018). Thus, the fundamental mechanisms responsible for sludge granulation have not yet been fully elucidated.

An often-overlooked fact is that granulation largely depends on a high level of biomass growth (Liu et al., 2018a). Current studies suggest that the biomass accumulation can be enhanced by adding multivalent cations (e.g., calcium and iron), and/or a support material (e.g., granular activated carbon) during start-up (Hulshoff Pol et al., 2004, Yang et al., 2018, Zhang et al., 2018). However, this will inevitably increase the cost of cultivation and the complexity of operation (Li et al., 2018, Wang et al., 2005). Alternatively, sludge granulation can also be achieved by increasing the organic loading rate (OLR) and controlling the operational parameters (e.g., HRT and Vup) of reactors (Liu et al., 2018a).

There are few reports available on the application of anaerobic granular sludge technology for the treatment of low-strength domestic sewage. Not surprisingly, all these anaerobic reactors were performed by direct inoculating mature granular sludge, rather than cultivating the granular sludge from seed sludge (Stazi and Tomei, 2018). Thus, several important questions remain. For example, which types of EPS are essential during sludge granulation? What are the key components of EPS during sludge granulation? What is the impact of other operational parameters? What is the feasible start-up strategy for the treatment of low-strength domestic sewage? Also, the spatial distribution of EPS during sludge granulation for the treatment of low-strength domestic sewage is unknown. Therefore, an in-depth study of the sludge granulation during the treatment of low-strength domestic sewage is warranted.

In this study, we adopted the strategy of increasing the OLR and Vup to accelerate the growth of sludge granules. The main objective was to evaluate the granulation and microbial population dynamics in an EGSB reactor by treating low-strength domestic sewage. To achieve the-above objectives, the change in volatile fatty acids (VFAs), specific methanogenic activity (SMA) and EPS contents were determined during sludge granulation. The microbial community distributions and EPS distributions of the microbial aggregates during the granulation process were correlated, and a corresponding granulation mechanism is proposed. This work may bring new insights into the mechanism of anaerobic sludge granulation.

Section snippets

Reactor set-up and operational conditions

A typical EGSB reactor with an effective working volume of 28.6 L was employed to treat the synthetic domestic sewage. The reactor consists of a bottom column (with an internal diameter of 10 cm and a height of 100 cm), an upper three-phase separator (25 × 25 × 25 cm), and an external circulation system. Three sampling ports (A, B, and C) were established along the EGSB column at a height of 25, 50 and 75 cm, respectively. One run lasted 135 days. The whole experiment can be divided into four

Performance of the EGSB reactor

Fig. 1 shows the change of influent and effluent COD concentrations, and the relevant COD removal efficiency with the operation time. During the operation, HRT was gradually reduced from 8 to 4 h. Consequently, the relevant OLR increased from 1.35 to 2.70 kg COD/m3/d. Influent and effluent COD concentrations were in the ranges of 322–554 mg/L and 76–280 mg/L, respectively. In Period I (days 1–24), COD removal efficiency increased linearly with time at an HRT of 8 h, due to increased biomass.

Conclusions

The granules in an EGSB reactor using domestic sewage as the influent were cultivated successfully for 107 days. The optimal HRT was 5 h at an OLR of 2.16 kg COD/m3/d. A corresponding average COD removal efficiency of 71.5 ± 2.3% was obtained. The EPS distribution and composition suggest that the TB-EPS content had a positive impact on the formation of the granular sludge. In addition, high-throughput sequencing analysis indicated that the metabolism shifted during the sludge formation from

Acknowledgements

This work was supported by Shanghai Pujiang Program (No. 18PJ1400400), the Shanghai Science and Technology Committee (No. 17DZ1202204), the Natural Science Foundation of Shanghai, China (No. 18ZR1401000), and the National Key Research and Development Program of China (No. 2018YFF0215703 and No. 2016YFC0400501). Y.L. thanks Donghua University for the start-up grant (No. 113-07-005710).

References (43)

Cited by (0)

1

Both authors contributed equally to this work.

View full text