Maoshui Zhuo*ab,
Olusegun K. Abassab and
Kaisong Zhanga
aCAS Key Laboratory of Urban Pollutant Conversion, Institute of Urban Environment, Chinese Academy of Sciences, 1799 Jimei Road, Xiamen 361021, China. E-mail: mszhuo@iue.ac.cn; Tel: +86-13159256518
bUniversity of Chinese Academy of Sciences, Beijing 100049, China
First published on 4th April 2018
Treatment of N,N-dimethylacetamide (DMAC) wastewater is an important step in achieving the sustainable industrial application of DMAC as an organic solvent. This is the first time that treatment of a high concentration of DMAC in real wastewater has been assessed using membrane bioreactor technology. In this study, an anoxic–oxic membrane bioreactor (MBR) was operated over a month to mineralize concentrated DMAC wastewater. Severe membrane fouling occurred during the short-term operation of the MBR as the membrane flux decreased from 11.52 to 5.28 L (m2 h)−1. The membrane fouling was aggravated by the increased amount of protein fractions present in the MBR mixed liquor. Moreover, results from the excitation–emission matrix analysis identified tryptophan and other protein-like related substances as the major membrane-fouling components. Furthermore, analysis of the DMAC degradation mechanism via high performance liquid chromatography (HPLC) and ion chromatography (IC) revealed that the major degradation products were ammonium and dimethylamine (DMA). Although the MBR system achieved the steady removal of DMAC and chemical oxygen demand (COD) by up to 98% and 80%, respectively at DMAC0 ≤ 7548 mg L−1, DMA was found to have accumulated in the treated effluent. Our investigation provides insight into the prospect and challenges of using MBR systems for DMAC wastewater degradation.
Concentrations of DMAC in wastewater of as high as 20000 mg L−1 are characteristic of the discharge water from large-scale polymeric membrane factories. Interestingly, until now, only a few research works have focused on the treatment of DMAC wastewater. Also, DMAC recovery is an effective energy-saving alternative and the waste DMAC could be converted into energetic materials, as carbon derived from DMAC waste has already been utilized in power generation for the production of 100 MW L−1 power density in microbial fuel cells at a potential of 0.45 V.6 Moreover, the DMAC removal efficiency was between 15% and 50% with a hydraulic retention time (HRT) of 12 min. An internal microelectrolysis process has been applied to treat DMAC wastewater at an influent concentration of 50 mg L−1, resulting in a DMAC removal rate of 95%.7 Nevertheless, technologies for DMAC removal at higher concentrations are currently lacking.
As a pure culture, the isolated Rhodococcus sp. strain B83 has been confirmed to biodegrade DMAC without the need for any extra source of carbon and nitrogen, reaching a degradation efficiency of 96.1% in 120 hours when the initial DMAC concentration was 15000 mg L−1.5 However, at a DMAC concentration of greater than 15000 mg L−1, a significant adverse effect on the pH of the culture was observed, which was harmful to the growth of the Rhodococcus sp. strain B83.5 Hence, it is indispensable to assess alternative technologies able to reclaim wastewater containing high levels of DMAC.
Membrane bioreactors (MBRs) are thought to be suitable for the treatment of DMAC wastewater due to their advantages over conventional candidates, which include the ability to process higher biomass concentrations, a smaller carbon footprint, less sludge generation, and better membrane permeability.8–10 Membrane bioreactors are generally used for both municipal and industrial wastewater treatment.11,12 For instance, a laboratory-scale submerged anoxic–oxic membrane bioreactor has been operated continuously to treat simulated wastewater contaminated with DDA (dianilinodithiophosphoric acid), an organic toxic flotation reagent, and the chemical oxygen demand (COD) removal efficiency rose up to 80% only after the system reached stability within a HRT of 4 h.13 The most challenging issue in the application of MBRs is the widely known problem of fouling.14 Research on membrane fouling during the treatment of wastewater with high concentrations of pollutants has been well documented.15,16
In this work, the potential of MBR has been exploited for the treatment of DMAC-containing wastewater. The study set out to achieve the following goals: (1) to assess the performance of the anoxic–oxic-MBR over a range of DMAC concentrations, (2) to explore the effects of DMAC loading rates on membrane fouling, and (3) to study the degradation mechanism of DMAC based on the identification of catabolic intermediates. To the best of our knowledge, this is the first time that DMAC removal by MBR has been assessed, while the fouling implications remain largely unknown. Results from this study will be beneficial for the further development of a MBR process for the treatment of DMAC polluted wastewater.
Characteristics | Measurements |
---|---|
a COD: Chemical oxygen demand, BOD5: biochemical oxygen demand for 5 days, TOC: total organic carbon, TN: total nitrogen. | |
pH | 5.4 ± 0.05 |
COD (mg L−1) | 18924 ± 2553 |
BOD5 (mg L−1) | 1930 |
DMAC (mg L−1) | 9910 ± 684 |
TOC (mg L−1) | 5375 ± 249 |
TN (mg L−1) | 2067 ± 96 |
NH4+–N (mg L−1) | <0.08 |
Li (mg L−1) | 23.9 ± 0.5 |
Conductivity (mg L−1) | 466.5 ± 3.5 |
Turbidity (NTU) | 0.694 |
Parameter | Stage 1 | Stage 2 | Stage 3 |
---|---|---|---|
a HRT: Hydraulic retention time, SRT: sludge retention time, MLSS: mixed liquid suspended solids, MLVSS: mixed liquid volatile suspended solids, DO: dissolved oxygen. | |||
Time (days) | 1–13 | 14–25 | 26–37 |
HRTMBR (h) | 24 | 24 | 24 |
HRTAnoxic (h) | 0 | 10 | 10 |
SRT (d) | — | 55 | 55 |
Recycle ratio | — | 200% | 200% |
DMAC (mg L−1) | 1500–1700 | 800 | 3000 |
MLSSMBR (mg L−1) | 8558 ± 433 | 11897.5 ± 922.5 | 15898.8 ± 303.8 |
MLVSSMBR (mg L−1) | 5421 ± 24 | 7913.8 ± 1171.3 | 12496.3 ± 98.8 |
MLSSAnoxic (mg L−1) | — | 8353.87 ± 46.3 | 8610 |
MLVSSAnoxic (mg L−1) | — | 5606.3 ± 856.3 | 6627.5 |
pH | 8.3 ± 0.3 | 7.2 ± 0.3 | 8.6 ± 0.3 |
DOMBR (mg L−1) | 0.24–5.64 | 0.24–6.14 | 0.25–0.32 |
DOAnoxic (mg L−1) | — | ≤0.5 | ≤0.5 |
Air flow rate (L min−1) | 15 | 15 | 15 |
Suction time | Run: 10 min, pause: 2 min |
Membrane process | Micro-filtration, flat-sheet membrane plates |
Membrane area (m2) | 0.25 |
Membrane pore size (μm) | 0.35 |
Membrane material | Polyvinylidene fluoride (PVDF) |
Sampling from the MBR bulk sludge was performed once every three days. As for extracellular polymeric substances (EPS) on the membrane surfaces, the fouling layer materials were carefully scraped off from two different areas of membrane surfaces at the end of stages 1 and 3 and were then dissolved in 40 mL of demineralized water for subsequent EPS extraction procedures. Extraction of the soluble microbial product (SMP) and EPS was conducted using a modified thermal method.19 40 mL of activated sludge taken from the mixed liquid and membrane surface was first centrifuged at 6000g for 10 min. The supernatant was considered as the SMP. The remaining procedures of EPS extraction were conducted.20 The polysaccharide (PS) and protein (PT) content in both the SMP and EPS were measured using a sulfuric acid anthrone colorimetric method21 and BCA Protein Assay Kit,22 respectively. Besides this, the SMP and EPS were characterized using excitation–emission matrix (EEM) fluorescence spectroscopy. Different peaks in the EEM appeared at corresponding intersections of the excitation–emission wavelengths depending on the different types of functional groups present. A 3D scan fluorescence spectrophotometer (F-4600, HITACHI) with a PMT voltage of 650 V was applied for measuring the EEM spectra. The excitation (Ex) and emission (Em) sampling interval was 3.0 nm with a slit of 5.0 nm.
Identification of the DMAC degradation products was conducted using of ion chromatography (Agilent 7890A ICS and Dionex ICS-3000) and HPLC. All of the N,N-dimethylformamide (DMF), dimethylamine (DMA), N-methylacetamide (MMAC), acetamide, acetaldehyde and acetate standards were prepared in advance to verify the existence of acetate, dimethylamine, N-methylacetamide, and acetamide. Samples prepared using real wastewater contaminated with N,N-dimethylacetamide were tested to exclude instrument interference in the DMAC degradation.
Fig. 2 Variation of treatment indicators in the MBR effluent including (a) COD, (b) DMAC, (c) TOC, and (d) NH4+, and the TN removal efficiencies. |
At the third stage, the DMAC removal efficiency still reached 100% regardless of the increase in the influent concentration (from 800 to 3346 mg L−1 and then continuously to 7548 mg L−1). Similar characteristic removal of a toxic constituent by MBR has been previously reported,23,24 where elevated concentrations of antibiotics were consistently removed and the shock loading effect of DMAC did not affect the removal performance.25 The TN removal rate also improved relative to the influent DMAC concentration (Fig. 2d). Interestingly, regardless of the variation in the influent DMAC concentration, its removal by the MBR system remains constant. Hence, our results demonstrates that treatment of DMAC wastewater using a MBR system could withstand the influence of influent fluctuation, which is a common characteristic of various industrial discharges, despite the low biodegradability index (BOD5/COD = 0.1) of the raw wastewater.
The initial high SMP concentration in the MBR bulk sludge is mainly due to the influent DMAC concentration (1500 mg L−1). A variety of microbial products are formed due to changes in the microbial activity in non-steady state conditions and the microorganisms will generally secrete SMP and EPS to protect their fragile membrane from damage in stress conditions.31,32 As described in Section 1, the microorganisms completely adapt to the new environment after three days. However, the TMP steadily increased, while the flux varied inversely with the TMP. In a similar study, it was reported that exposure of microorganisms to pharmaceutical compounds increased the production of SMPs.33 At stage 2, the membrane flux was recovered by physical cleaning and was re-inserted into the MBR set-up. In contrast, the TMP values were lower at stage 2 owing to a decrease in the influent DMAC concentration (800 mg L−1). The potential toxicity of DMAC has been well studied.4,34 Hence, it is speculated that the lower fouling rate corresponds to the low SMP and EPS produced during this stage (Fig. 4).
At stage 3, a sudden rise in the TMP and subsequent membrane flux reduction occurred after the influent DMAC increased to 3000 mg L−1, as shown in Fig. 3. Similarly, the membrane flux decreased from 11.52 to 5.28 LMH. Thus, effective membrane fouling control measures are paramount for MBR operation in constant flux mode when treating wastewater contaminated with high levels of DMAC. For example, mechanical cleaning using fluidized particles, such as beads and biofilm carriers, can be practically applied with potential, particularly for flat-sheet membrane modules.14 In the same vein, anti-fouling membrane materials could be considered.35 Similarly, addition of a pre-treatment unit such as hydrolysis and advanced oxidation treatment, could help to increase the biodegradability of DMAC contaminated wastewater. As such, soluble foulants released by microorganisms in response to DMAC toxicity will be reduced, thus mitigating membrane fouling. In recent work by J. K. Xue et al., it was shown that using ozone for the pretreatment of oil sands process-affected water prior to MBR treatment effectively mitigated membrane fouling.36
The SMP PS/PT ratio in the bulk sludge rose from 0.6 to 2, and finally decreased down to 0.9 at stage 3. Likewise, the EPS PS/PT ratio increased from 0.4 to 4.7 and then decreased down to 1.0. It was found that the SMP and EPS PS/PT ratios were lower than 1.0 when severe membrane fouling occurred during the treatment of raw oil sands process-affected water (OSPW) without pretreatment.36 The SMP and EPS PS/PT ratios on the fouled membrane surface were lower than 0.55, as shown in Fig. 4, which indicates that protein fractions contribute more to the fouling layer. Fouled membranes are dominated by biopolymers, including SMP-polysaccharides and EPS-proteins, in the early and late stages of fouling.14 Therefore, as shown in this work, protein fractions played a key role in the fouling of the membrane.
To demonstrate the role of inter-foulant species (e.g. polysaccharides, proteins and humic substances) on membrane fouling during the MBR operation, the EEM fluorescence spectra of the SMP and EPS were acquired (Fig. 5). A high intensity of peak C at an Ex/Em wavelength of 287/350 was found to be dominant among the three main peaks and has been identified in the literature as being due to the presence of tryptophan and other protein-like related substances.40–43 Peak A and peak B, located at Ex/Em wavelengths of 311/385 and 239/388, have been previously ascribed to being due to the presence of marine humic and fulvic acid-like species.33,42 These two peaks intermittently appeared during the analysis of the SMP and EPS compositions. Current research44,45 is focused on a potential approach for achieving effective fouling control through the selection and cultivation of polysaccharide-degrading bacteria or enzymes. Thus, bioaugmentation of the MBR with specific bacteria or enzymes capable of degrading the dominant protein-like foulant responsible for peak C will be an effective strategy for mitigating membrane fouling in DMAC treatment using a MBR.
Fig. 5 EEM fluorescence spectra of the SMP (1 and 2) and EPS (3 and 4) in mixed-liquor suspended sludge and the EPS on the membrane surface (5 and 6). |
To evaluate the membrane rejection efficiency of the polysaccharides and proteins in the MBR system, concentrations of the PS and PT in the sludge supernatant and permeate were measured and are presented in Table 3. Complete retention of the PT fractions was observed at stages 1 and 2, and only small amounts of the PS and PT fractions were detected in the permeate. Biomass commonly utilizes organic material along with the generation of SMPs in biological treatment reactors.46 In another report, it was observed that SMPs contribute to the majority of the unremoved COD in effluent wastewaters.46 However, few studies have shown the possibility of SMP retention by MBR.47,48 Thus, MBR is a promising technology for the treatment of DMAC contaminated wastewater, although membrane fouling is still a common shortcoming to be overcome.
Samples | PSMBR (mg L−1) | PSpermeate (mg L−1) | PTMBR (mg L−1) | PTpermeate (mg L−1) |
---|---|---|---|---|
S1 | 23.54 | 0.00 | 120.62 | 0.00 |
S2 | 41.71 | 0.00 | 126.57 ± 3.57 | 0.00 |
S3 | 22.91 | 0.00 | 89.11 ± 2.00 | 0.00 |
S4 | 37.56 | 2.28 | 62.68 ± 3.57 | 0.00 |
S5 | 31.21 | 0.00 | 55.40 ± 2.38 | 0.00 |
S6 | 29.12 | 6.81 | 58.39 ± 0.55 | 0.00 |
S7 | 63.27 ± 1.76 | 4.98 ± 0.39 | 29.51 ± 0.93 | 0.00 |
S8 | 78.82 ± 0.27 | 6.34 ± 0.63 | 36.50 ± 1.138 | 2.03 ± 0.34 |
S9 | 84.00 ± 1.87 | 7.22 ± 0.13 | 49.03 ± 2.91 | 2.76 ± 0.21 |
S10 | 100.20 ± 3.27 | 6.16 ± 0.57 | 76.56 ± 2.98 | 1.95 ± 0.41 |
S11 | 140.34 ± 0.19 | 4.91 ± 0.26 | 153.97 ± 18.49 | 2.59 ± 0.08 |
In order to study the DMAC degradation mechanism in pathway 1, dimethylamine (DMA) and ammonium concentrations were determined using ion chromatography (Agilent 7890A ICS). The accumulation of ammonium (Fig. 2d) and DMA were readily observed at treatment stage 3 (Fig. 8a). While in pathway 2, acetamide and N-methylformamide (MMAC) were not readily detectable, probably due to low concentrations. In pathway 1, the degradation of DMA to ammonia49 and the conversion of acetamide into ammonia have been previously confirmed.5,50 It has also been reported that the maximum accumulation of dimethylamine (DMA) was equal to 62% of the initial dimethylformamide (DMF) concentration.51 As can be seen in Fig. 6a, the acetate concentration was too low to accumulate in the A/O-MBR because it is readily biodegradable organic matter.52 Nevertheless, during DMAC utilization by the Rhodococcus sp. strain B83, the accumulation of acetate ions was detected.5 The permeate concentration of CDMA (carbon in DMA) and TOC are directly related, as shown in Fig. 8b. DMA was found to be the main organic component of the effluent. Similarly, accumulation of ammonium in the A/O-MBR treatment system resulted in limited TN removal efficiencies (Fig. 4d). Application of aerobic granular sludge for simultaneous nitritation–denitritation treatment53 and anammox processes combined with denitrifying anaerobic methane oxidation (DAMO)54 could be considered in order to alleviate the ammonium accumulation in future studies.
Fig. 8 (a) One of the DMAC transformation products (dimethylamine). (b) The linear relationship between the concentration of CDMA and TOC in the permeate. |
This journal is © The Royal Society of Chemistry 2018 |