Elsevier

Water Research

Volume 91, 15 March 2016, Pages 305-313
Water Research

Innovative sponge-based moving bed–osmotic membrane bioreactor hybrid system using a new class of draw solution for municipal wastewater treatment

https://doi.org/10.1016/j.watres.2016.01.024Get rights and content

Highlights

  • A mixture of MgCl2 and Triton X-114 can serve as a novel draw solution.

  • The reverse flux of novel draw solution was 4.5 times lower than that of only MgCl2.

  • Low salt accumulation was achieved during 90-day SMB-OsMBR operation.

  • Approximately 100% NH4–N and 98% PO4–P were removed by the SMB-OsMBR hybrid system.

  • Moving free sponge carriers in the bioreactor continuously cleaned the FO membrane.

Abstract

For the first time, an innovative concept of combining sponge-based moving bed (SMB) and an osmotic membrane bioreactor (OsMBR), known as the SMB-OsMBR hybrid system, were investigated using Triton X-114 surfactant coupled with MgCl2 salt as the draw solution. Compared to traditional activated sludge OsMBR, the SMB-OsMBR system was able to remove more nutrients due to the thick-biofilm layer on sponge carriers. Subsequently less membrane fouling was observed during the wastewater treatment process. A water flux of 11.38 L/(m2 h) and a negligible reverse salt flux were documented when deionized water served as the feed solution and a mixture of 1.5 M MgCl2 and 1.5 mM Triton X-114 was used as the draw solution. The SMB-OsMBR hybrid system indicated that a stable water flux of 10.5 L/(m2 h) and low salt accumulation were achieved in a 90-day operation. Moreover, the nutrient removal efficiency of the proposed system was close to 100%, confirming the effectiveness of simultaneous nitrification and denitrification in the biofilm layer on sponge carriers. The overall performance of the SMB-OsMBR hybrid system using MgCl2 coupled with Triton X-114 as the draw solution demonstrates its potential application in wastewater treatment.

Introduction

Advances in wastewater treatment technology have facilitated increasing the pollutant removal efficiency and meeting stringent effluent regulations. However, there are still many challenges faced in wastewater treatment processes, especially in relation to nutrient and trace organic removal, which necessitate improving existing wastewater treatment processes for achieving higher removal efficiency (Sayi-Ucar et al., 2015). Currently, membrane technology is employed to augment water supplies, and it is crucial for sustainable water production. Among the membrane processes, membrane bioreactor (MBR) technology has become one of the most effective options for improving water sustainability; this technology encourages wastewater reuse, requires less space and produces less sludge (Guo et al., 2012, Ramesh et al., 2006). However, conventional activated sludge-based MBRs pose operational and R&D problems such as membrane fouling, high energy consumption, and limited nutrient removal capability (Nguyen et al., 2012).

To overcome these problems, a novel osmotic membrane bioreactor (OsMBR) with the following unique features was developed: (i) osmotic pressure is used as the driving force instead of hydraulic pressure, (ii) forward osmosis (FO) membranes show high rejection for a wide range of contaminants, and (iii) the membranes have a low fouling tendency (Cornelissen et al., 2011, Gwak et al., 2015, Qiu and Ting, 2014, Tan et al., 2015). Nevertheless, a major technical challenge to OsMBR application was the lack of appropriate draw solutions that could reduce salt accumulation and membrane fouling during long-term operation (Ge et al., 2012, Kim, 2014). Yap et al. (2012) demonstrated that the reverse salt flux from the draw solution into the bioreactor and the high salt rejection by the FO membrane caused the build-up of salinity in the bioreactor. Increased bioreactor salinity can severely impact on microbial viability and membrane performance because some functional bacteria are more sensitive to high salinity conditions (Moussa et al., 2006, Osaka et al., 2008). Kinetics studies have suggested that nitrogen and phosphorus removal efficiency dropped to 20% and 62%, respectively, when salt concentration was 5% NaCl in the bioreactor (Dinçer and Kargi 2001, Uygur and Kargı 2004). In addition, the salinity stress enhanced the release of both soluble microbial products and extracellular polymeric substances, leading to severe membrane fouling (Park et al., 2015).

Moreover, an increase in the total dissolved solid (TDS) concentration in the bioreactor tank can reduce the osmotic pressure difference across the FO membrane, causing the water flux to decrease rapidly (Uygur, 2006, Ye et al., 2009). For example, Holloway et al. (2014) used NaCl salt as the draw solution in an OsMBR system with mixed liquor suspended solids (MLSS) of 5 g/L and achieved high removal efficiencies for phosphate and chemical oxygen demand (96%) for a high water flux (5.72 L/(m2 h)). However, because monovalent ions (Na+ with a hydrated radius of 0.18 nm and Cl with a hydrated radius of 0.19 nm (Kiriukhin and Collins, 2002)) could easily pass through the FO membrane (membrane pore size: 0.37 nm) (Xie et al., 2012 (a)), the TDS concentration in the bioreactor increased by approximately 8 g/L after 40 days (Holloway et al., 2014). To minimize salt leakage, Qiu and Ting (2013) demonstrated that using a divalent salt such as MgCl2 (Mg2+ with a hydrated radius of 0.3 nm (Kiriukhin and Collins, 2002)) in the draw solution in a submerged OsMBR could help increase organic matter removal to 98% and reduce salt leakage compared with an NaCl draw solution. However, the mixed liquor conductivity in the OsMBR was still high, ranging from 2 to 17 mS/cm for a 80-day operation, because of the reverse transport of MgCl2 from the draw solution and the rejection of dissolved solutes in the feed by the FO membrane.

A mixture of Ethylenediaminetetraacetic acid disodium salt (EDTA-2Na) and Triton X-100 was used as the draw solution in an OsMBR in our previous study. Although it can reduce the reverse salt flux appreciably and minimize salt accumulation in the bioreactor for a 60-day operation (Nguyen et al., 2015a), the water flux was relatively low because of the limited solubility of EDTA-2Na salt in water. Meanwhile, the solubility of MgCl2 is high (up to 5 M) so as it can produce a high osmotic pressure and high water flux. Therefore, to achieve a high water flux and minimal salt leakage, a mixture of Polyethylene glycol tert-octylphenyl ether (Triton X-114) and MgCl2 was used as the draw solution in the current study. The advantage of using the non-ionic Triton X-114 surfactant is that it has a large structure involving a long straight carbon chain and a low critical micelle concentration (CMC) of 0.2 mM. This structure leads to the formation of second layers on the membrane surface, constricting the membrane pores and minimizing reverse salt diffusion. Moreover, the high water solubility of MgCl2 can produce high osmotic pressure as well as a high water flux in an OsMBR system.

Up to this date, the major technical challenges to OsMBR application are the build-up of salinity in the bioreactor, the membrane fouling in long-term operation and limited nutrient removal in single reactor, which motivated the author to carry out this work. To the best of our knowledge, a draw solution containing a mixture of Triton X-114 surfactant and MgCl2 salt has not been used for a sponge-based moving bed (SMB)-OsMBR hybrid system to simultaneously achieve a low salt accumulation, a low fouling and high nutrient removal efficiency. Hence, this study systematically investigated the performance of the mixture as the draw solution in an SMB-OsMBR system for municipal wastewater treatment. First, the effect of the Triton X-114 concentration on the water flux and reverse salt flux was evaluated using deionized (DI) water as the feed solution. Next, the variation of the water flux and amount of salt accumulation with the operating duration was examined using synthetic wastewater as the feed solution. The nutrient removal efficiency was then determined in the SMB-OsMBR hybrid system for the proposed draw solution. Finally, the membrane fouling characteristics were analyzed using scanning electron microscopy and energy dispersive x-ray spectroscopy (SEM–EDS), and fluorescence excitation-emission matrix (FEEM) spectrophotometry.

Section snippets

Description of SMB-OsMBR

A laboratory scale SMB-OsMBR system is shown in Fig. 1. The FO module with an effective membrane area of 120 cm2 was fabricated with a tube configuration and wrapped in OsMem™ cellulose triacetate with embedded polyester screen support (CTA-ES) flat sheet membranes (Hydration Technologies, Inc., Albany, OR, USA). It was then immersed in the vertical position in the bioreactor tank (6 L), with the active layer of the membrane facing the feed solution. Sponge biocarriers (Table 1) were added to

Effect of surfactant concentration on water flux and reverse salt flux

Fig. 2 shows the reverse salt fluxes and water fluxes for five draw solutions with various Triton X-114 concentrations and a fixed MgCl2 concentration of 1.5 M. FO experiments were conducted with the active layer of the membrane facing the feed solution, which was DI water. The reverse flux decreased considerably when Triton X-114 with concentrations ranging from 0.5 to 2.5 mM was coupled with the MgCl2 draw solution. Fig. 2 indicates that higher concentrations of Triton X-114 coupled with the

Conclusions

The study found that an optimal mixture of 1.5 mM Triton X-114 and 1.5 M MgCl2 as the draw solution simultaneously facilitated a high water flux (11.38 L/(m2 h)) and low reverse salt flux (2.03 g/(m2 h)). The SMB-OsMBR hybrid system showed excellent ability to remove ammonium (approximately 100%) and phosphorus (>98%) in single reactor. This was particularly the case when an ideal attached-growth medium (sponge) provided free mobile carriers for combining the active biomass with the OsMBR

Acknowledgment

This work was supported by the Ministry of Science and Technology of the Republic of China under the grant number of 101-2221-E-027 -061 -MY3.

References (38)

Cited by (0)

View full text