Accepted Manuscript Performance enhancement and fouling mitigation by organic flocculant addition in membrane bioreactor at high salt shock Haifeng Zhang, Zhongyu Gao, Lanhe Zhang, Lianfa Song PII: DOI: Reference:
S0960-8524(14)00558-6 http://dx.doi.org/10.1016/j.biortech.2014.04.053 BITE 13349
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
22 January 2014 15 April 2014 17 April 2014
Please cite this article as: Zhang, H., Gao, Z., Zhang, L., Song, L., Performance enhancement and fouling mitigation by organic flocculant addition in membrane bioreactor at high salt shock, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech.2014.04.053
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Performance enhancement and fouling mitigation by organic flocculant addition in membrane bioreactor at high salt shock Haifeng Zhanga, b, Zhongyu Gaoa, Lanhe Zhanga, Lianfa Songb,*
a
School of Chemistry Engineering, Northeast Dianli University, Jilin 132012, Jilin, P. R. China ；
b
Department of Civil and Environmental Engineering, Texas Tech University, 10th and Akron, Lubbock, TX 79409-1023, USA
* Corresponding Author. Tel.: +1 806 742 3598. E-mail: [email protected]
1
Abstract The main objective of this study was to investigate the effect of an organic flocculant (MPE50) addition on reducing membrane fouling and enhancing performance in membrane bioreactor (MBR) at the high salt shock. Results show that MPE50 addition is a reliable and effective approach in terms of both membrane fouling mitigation and pollutants removal improvement in the case of high salt shock. Compared to the control reactor, the MBR with MPE50 addition enhanced the average removal of COD, NH4+-N and TP by 4.1, 13.2 and 21.2%, respectively. Due to the effect of flocculation and adsorption by MPE50, a significantly reduction in the soluble microbial products (SMP) proteins amount was observed. As a result, the membrane fouling rate was mitigated successfully. Further, the increasing of mean particles size, Zeta potential and related hydrophobicity of the flocs would also have positive impacts on membrane fouling mitigation.
Key words: Membrane bioreactor; High salt shock; Related hydrophobicity; Modified fouling index; Soluble microbial products; Extracellular polymeric substances.
2
1. Introduction Among membrane processes, membrane bioreactor (MBR) technology is an innovative and promising option for wastewater treatment and reuse, which has been widely used for the treatment of various wastewater, including municipal, industrial and landfill leachate (Wintgens et al., 2005). However, membrane fouling is an unavoidable and probably the most serious challenge, which increases operational cost and shortens membrane life for MBR (Yang et al., 2006). Membrane fouling in MBR is mainly resulted from the physicochemical interaction between the suspended particles and membrane to form a cake layer and the adsorption of the dissolved species into membrane pores that lead to flux decline (Mutamim et al., 2013). The characteristics of the mixed liquor, such as morphological properties (mean floc size, floc fractal dimension), physical parameters (surface charge, relative hydrophobicity, dynamic viscosity), and the biochemical components of the mixed liquor (extracellular polymeric substances, soluble microbial products) govern the filterability and the membrane fouling (Ji et al., 2010).
The high or changing salinity imposes a challenge to activated sludge process (Yogalakshmi and Joseph, 2010), which is likely to occur in many biological wastewater treatment systems nowadays (Cortés-Lorenzo et al., 2012). For example, the coastal sewers are subjected to infiltration by seawater or when industrial effluents receive discharges from individual high-salinity processes. Additionally, the urban wastewater may also be affected by saline industrial wastewaters or other sources. For
3
example, seawater is used in Hong Kong for toilet flushing to save freshwater (Aslan and Simsek, 2012). Salt can also enter the urban wastewater flow when it is used in the cities for thawing snow and ice on the streets (Jang et al., 2013).
High salt concentration in wastewater can cause cell plasmolysis due to the dramatic increase in osmotic pressure and changes on microbial metabolism (Bassin et al., 2011; Vyrides and Stuckey, 2009). Moreover, salinity may significantly affect the physical and biochemical properties of the biomass, leading to a change of surface charge, hydrophobicity, filterability and bioflocculation of biomass (Sun et al., 2010). A limited number of studies have been conducted on MBR operation for saline wastewater shock, which indicated that salinity not only exerted adverse impact on the pollutant removal, biomass activity and microbial diversity, but also significantly deteriorated sludge filterability and resulted in serious membrane fouling (Reid et al., 2008; Yogalakshmi et al., 2010; Jang et al., 2013 ). Therefore, it is necessary to develop an effective method of reducing membrane fouling for MBR process at high salt shock.
Recently, application of inorganic and organic flocculants in MBR to modify the properties of mixed liquor artificially for alleviation of membrane fouling has been reported in several publications (Ji et al., 2010; Zhang et al., 2008). A novel organic flocculant called Membrane Performance Enhancer (MPE50) has been developed by Nalco Company and applied to MBRs for membrane fouling control (Nguyen et al., 2010; Ji et al., 2010). The enhanced filterability in MBR with the addition of MPE50 was attributed to an increase in floc size and to a decrease in concentration of soluble
4
foulants in the bulk phase (Ji et al., 2010). Moreover, it is reported that MPE50 does not negatively affect the overall aeration demand and bioactivity (Nguyen et al., 2010). Considering the significant impact of flocculant dosing on membrane fouling mitigation, it would be very interesting to study how MPE50 addition affects membrane fouling in MBR at high salt shock.
The main objective of this work was to study the effectiveness of flocculant on membrane fouling reduction in MBR at high salt shock. For this purpose, a control MBR and a hybrid MBR with MPE50 addition were operated in parallel. A comparison was made of the mixed liquor properties for both MBRs during the whole operation time, with parameters including the content of protein and carbohydrate in soluble microbial product (SMP) and extracellular polymeric substances (EPS), as well as the mean particle sizes, Zeta potential, and relative hydrophobicity of the flocs. The modified fouling index (MFI) was used to confirm the mixed sludge filterability under each condition.
2. Material and methods 2.1. MBR systems Two MBRs of identical size, MBR-A (the control reactor) and MBR-B (with flocculant addition), were operated for about 70 days. A schematic of the experimental setup is in Fig.1. Each reactor had a working volume of 8 L. The membrane module used in both systems was a bundle of U-shaped hollow fiber membranes (polyvinylidene fluoride, PVDF) with a pore size of 0.2 µm and a filtration area of 0.1 5
m2. Air was supplied to both reactors at 5 L/min. Permeate through the submerged membrane module was continuously withdrawn using a peristaltic pump at a constant flux of 10 L/m2·h, operated with an intermittent mode of 13 min on and 2 min off. Chemical cleaning procedure would be carried out if the trans-membrane pressure (TMP) reached about 28 kPa during the operation. The membrane modules were taken out and soaked in 0.5% (v/v) NaClO solution for about 2 h to recover their permeability. The seed sludge was taken from a lab-scale MBR for more than one year. Each MBR was seeded with 8 g/L of the acclimatized activated sludge. Both MBRs were operated under the same conditions and stabilized for 30 days. Hydraulic retention time (HRT) and sludge residence time (SRT) were maintained at 6 h and 30 days, respectively. Synthetic wastewater was used in order to avoid difficulties in quantification of metabolic products due to variation in the composition and concentration of organic pollutants in the real wastewater. The synthetic wastewater was prepared using glucose, NH4Cl, and KH2PO4 as a source of carbon, nitrogen, and phosphorous, respectively. The recipe could be obtained from our previous study (Zhang et al., 2008). During the steady state operation, both MBRs had similar mixed liquor suspended solids (MLSS) concentrations and mixed liquor volatile suspended solids (MLVSS) concentrations typically in the range of 10-13 g/L and 8-10.5 g/L, respectively. One simulation of salt shock was applied in this study. On day 31, 20 g/L of NaCl was administered in the feed for a period of 24 hours to simulate a shock salt loading. More frequent simulations are not appropriate in this study because salt tolerance active sludge would develop that may not adapt in the fresh water 6
environments in the normal conditions. Before starting the comparison tests between both MBRs, batch tests were conducted at various MPE50 (Nalco Company) dosages to determine the extent of SMP removals. According to the results of batch tests, 600 mg/L MPE50 was firstly added in MBR-B and 160 mg MPE50 was supplied daily to the reactor to compensate the loss by the sludge discharge. The addition of MPE50 can be stopped when membrane fouling is controlled and the performance of MBR is recovered. (Fig.1. Schematic of MBR-A and MBR-B)
2.2. Filterability
The filterability characteristics of mixed liquid were investigated in batch tests using Amicon 8400 dead-end cells (Millipore, Billerica, MA). The cells had a volume of 350 mL and an effective membrane filtration area of 41.8 cm2. All experiments were carried out at room temperature and under a pressure of 28 kPa. The permeation flux was determined by weighing permeates on an electronic top loading balance connected to a personal computer running an auto-reading program. A new membrane sheet (PVDF, 0.2 µm) was used for each trial. Under unstirred conditions, MFI was obtained from the plot of t/V versus V using the filtration equation at constant pressure (Ognier et al., 2002). MFI parameter gives an idea of the fouling potential characteristics of the mixed liquor. A higher MFI is result in a higher fouling rate for the mixed liquor. It is based on the cake filtration mechanism (Jang et al., 2006). MFI is defined as the gradient of the linear relationship between t/V and V. 7
t = V
µRm △P
MFI =
+
µαC 2△ P
(1)
V
µα C
(2)
2△P
Where t is the filtration time (s), V the permeate volume per unit filtration area (m), µ the dynamic viscosity of permeate (Pa s), Rm the intrinsic membrane resistance (m-1), ∆P the applied transmembrane pressure (kPa), α the specific resistance (m/kg) and C is macromolecules concentration in bulk solution (mg/L). 2.3. Analytic methods Standard analytic methods were used to measure for chemical oxygen demand (COD), NH4+-N, total phosphate (TP), MLSS and MLVSS (APHA, 2005). The SMP samples were obtained by centrifuging the mixed liquor at 4000 rpm for 10 min, followed by filtration through a membrane of 0.45 µm. EPS were extracted by heat treatment (Morgan et al., 1990). Carbohydrates were determined according to the phenol-sulfuric acid method (Dubois et al., 1956) with glucose as standard. Proteins were determined by the modified Lowry method (Lowry et al., 1951) with bovine serum albumin (BSA) as standard. Relative hydrophobicity of the flocs was measured according to Rosenberg’s method (Rosengerg et al., 1980). Particle size distributions of flocs were measured with a Malvern Mastersizer (Malvern, UK). The Zeta potential was recorded using a Zetasizer Nano ZS Instrument (Malvern, UK). 2.4. Respirometry tests The biological activity was followed from respirometry tests used to measure autotrophic and heterotrophic activity by differentiating endogenous and exogenous
8
activity. The endogenous activity is defined as the rate of oxygen consumption in the absence of external substrate. Bacteria are forced to biodegrade their intracellular carbon as substrate to provide the energy necessary for cell maintenance and their respiration is minimal. The exogenous activity is defined as the rate of oxygen consumption after substrate addition and bacteria use the substrate available to supply their energy demand, and the resulting respiration is at its highest value (Villain and Marrot., 2013). The aerated reactor was filled with 1 L of sludge and every two minutes, 50 mL was sampled and injected into another reactor without oxygenation to measure the oxygen uptake rate (OUR) with a continuous dissolved oxygen probe (Mettler-Toledo, Columbus, OH). Then, specific nutrients or inhibitor were successively added to the aerated reactor according to four steps: (1) Endogenous respirations of autotrophic and heterotrophic micro-organisms were first measured over a period of 1 h; (2) Ammonium was added to measure the maximum activity of autotrophic micro-organisms; (3) Autotrophic micro-organisms were inhibited with allylthiourea to isolate heterotrophic endogenous respiration; (4) The maximum activity of heterotrophic micro-organisms was quantified after glucose addition. 3. Results and discussion 3.1 Comparison of conventional pollutant removal Fig. 2 and Table 1 showed the conventional pollutant removal efficiencies in both MBRs in different periods. As shown in Table 1, it could be seen that the average COD removal decreased from 96.2±0.8% (1-30 days run) to 89.6±0.9% (31-33 days run) for MBR-A, this result was also consistent with the temporal variation of heterotrophic 9
activities shown the Table 2. A decrease heterotrophic exogenous (from 19.4±4.7 to 17.2±2.4 mg O2 gVSS-1 h-1) and an increase endogenous (from 1.5±0.3 to 2.1±0.4 mg O2 gVSS-1 h-1) were observed in the MBR-A, which suggested that salt shock caused a slight diversion of carbon and/or energy from growth to adaptive responses and protection (Ray and Peters, 2008). In contrast, it was observed that average COD removal was slightly enhanced by 4.1% with MPE50 addition in MBR-B, reaching 93.7%. It was indicated that the cationic polymer addition could have a beneficial effect on organic matter removal when treating saline wastewaters.
(Table 1 Conventional pollutant removal efficiencies in both MBRs in different periods)
With regard to NH4+-N, the average removal was over 91% for both MBRs during the first 30 days. After salt shock, the average NH4 +-N removal dropped to 71.7±18.3 and 84.9±11.6% for MBR-A and MBR-B, respectively. The salt shock exhibited a significant negative impact on the autotrophic biomass in both reactors, for example, the autotrophic bacteria increased endogenous activities from 1.3±0.7 to 2.9±0.4 mg O2 gVSS-1 h-1 and simultaneously greatly decreased their exogenous activities from 4.7±0.5 to 0.7±0.2 mg O2 gVSS-1 h-1 in MBR-A. It took 10 d and 7 d to recover the stable NH4+-N removal efficiency for MBR-A and MBR-B, respectively. This result was consistent with the report (Jang et al., 2013) , which indicated that nitrifiers were very sensitive to high salinity and would be required a longer time to recover due to their slow growth rates. Compared with MBR-A, it was obviously that MPE50 caused a positive impact on nitrifiers in MBR-B, resulting in about 13.2% increase of average
10
NH4+-N removal efficiency and reduction of the recovery time by 3 days.
(Table 2 Respirometry activities in both reactors between before and after salt shock)
It was observed that the average TP removal efficiency in both MBRs were about 30-40% at the first 30 days. However, the TP removal during the days 31 and 41 was averaged at 9.1±3.2% in MBR-A, which might be explained that the phosphate accumulating organisms (PAO) were more sensitive to salinity than the carbon and nitrogen bacteria. PAO cells might have lost phosphate accumulation capability due to osmotic pressure (Uygur, 2006). Also, wash out of phosphate from unmetabolized substrate can also be other possible reason (Jang et al., 2013). In comparison, the average TP removal was enhanced by 21.2% when MPE50 addition operation period, which would be a result of the concurrent effect of chemical and biological phosphorus removal. It was also observed that the TP removal in MBR-B (36.1±3.3% ) was slight higher than MBR-A (33.8±1.4%) even after MPE50 dosing had been terminated during the period of 42-70 days, indicating that the MPE50 accumulation in sludge floc still had a positive effect on phosphorus removal.
(Fig. 2. Comparison between the control and the MBR with coagulant in removing conventional pollutants)
3.2 Membrane fouling rates and mixed sludge filterability
Typical TMP profiles depicting fouling trends for both MBRs were shown in Fig.3. The membrane fouling rates were computed based on TMP development each day
11
(dTMP/day) through the experimental time (see in Table 3). The increase rate of TMP is an important indicator for evaluating the development of membrane fouling in submerged MBRs (Le-Clech.et al., 2006). Based on the membrane fouling rate, the total operation time could be divided into three stages: the normal period, the sensitive period, and the recovery period. As could be seen from Fig. 3, it was not obviously different in TMP increase in the normal period for both MBRs, the membrane fouling rates were 1.26 and 1.31 kPa/d for MBR-A and MBR-B, respectively.
(Table 3 Membrane fouling rate (kPa/d) in different periods of both MBRs)
However, the membrane fouling rate significantly increased in the sensitive period for MBR-A. The fouling rate was so severe that membrane filtration had to be stopped for chemical cleaning just about 2 days operation, which was very apparently a sticky gel fouling layer was observed on the membrane surface when the fouled membrane was cleaned. On the contrary, significant improvement in filterability was observed in MBR-B by extending the filtration time by almost four-fold compared with MBR-A. The membrane fouling rate of MBR-B was 2.34 kPa/d, which was much lower than that of in MBR-A (12.17 kPa/d on average). The results suggested that MPE50 addition managed to control membrane fouling could be a useful strategy in case of salt shock. Otherwise, the MBR operation had to be terminated due to severe membrane fouling. The chemical cleaning frequency needs to be increased, which will reduce membrane lifetime and increase the operation costs (Le-Clech.et al., 2006).
(Fig.3. Profile of TMP rise up in the MBR-A and MBR-B) 12
In general, the TMP profiles exhibited two-stage process with linear gradual TMP rise followed by sudden increase in the rate of TMP rise (Cho and Fane., 2002). However, the MBR-A exhibited the only rapid increase stage in the sensitive period, which indicated that high salt shock significantly affect physical and biochemical properties of activated sludge, which then impacted upon mixed sludge filterability (Reid et al., 2006). The relationship between the fouling rates and the modified fouling indicate (MFI) was shown in Supplementary Fig. S1. The linear curve showed a strong correlation between the MFI and the fouling rate with an r-squared value of 0.98. Therefore, the MFI could be considered as a reliable mixed sludge filterability indicator to predict the extent of membrane fouling rate in MBR filtration process. A previous study showed a similar result (Zhang et al., 2008). 3.3 SMP in the supernatant Fig. 4 showed the proteins and carbohydrates concentration of SMP under different periods for both reactors. Compared with the normal period, it was noted that both the carbohydrate fraction of SMP (SMPc) and the protein fraction of SMP (SMPp) content increased obviously in the sensitive period for MBR-A, which was consistent with previous studies that high salt shock resulted in increased SMPc and SMPp concentrations in the supernatant of MBRs (Reid et al., 2006). It was explained that microorganisms respond to a salt shock acceleration of endogenous respiration (see in Table 2), accompanied by release of organic cellular constituents, increase the SMP by secretion and cells autolysis (Reid et al., 2006). Moreover, the increase in salinity also could increase the solubility of protein and carbohydrate of SMP, which might be 13
another reason for increasing the SMP content in the supernatant (Sun et al., 2010). Compared with the SMPc, the concentration of SMPp dramatically increased up to 5.58 mg/g SS in the sensitive period, which was almost 11 times higher than that in normal period in the MBR-A. The results were consistent with the findings of other studies showing that introduction of sodium caused a release of soluble proteins (Murthy et al., 1998).
(Fig. 4. Variations in the concentration of protein and carbohydrate under different operating conditions for SMP)
For the MBR-B, it could be seen that the SMPc and SMPp concentrations obviously decreased in the sensitive period compared with MBR-A. MPE50 exerted a greater impact on reducing the SMPp than SMPc, reflected by about 85.8% and 60.7% reduction in SMPp and SMPc, respectively. The role of additional MPE50 in MBR-B could be explained as followed: MPE50, like a flocculant, can weaken the repulsion among the SMP, via charge neutralization, allowing them to flocculate with each other by electrostatic attraction (Koseoglu et al., 2008). The reason of high SMPp removal could be attributed to the charged groups of proteins (Dizge et al., 2011).
To confirm the effect of the proteins and carbohydrates content on the mixed sludge filterability, the MFI values were correlated with the concentrations of SMPc and SMPp in different stages for both MBRs, respectively (see in Table 4). In the normal period, the carbohydrate SMP concentration should be the key factor contributing to mixed sludge filterability for both reactors( R2 = 0.84 for MBR-A , R2 = 0.86 for
14
MBR-B), which is consistent with the findings of other studies that had addressed SMPc as the major cause of the membrane fouling phenomenon (Arabi and Nakhla, 2009). In comparison, the MFI values showed better correlation with the SMPp concentration (R2 = 0.97, compared to R2 = 0.53 for SMPc) in the sensitive period for MBR-A. On the contrary, the MFI values were better correlated with the SMPc concentrations (R2 = 0.92, compared to R2 = 0.58 for SMPp) for MBR-B, which implied that major foulants in both reactors were different depending on MPE50 addition. In the recovery period, the concentrations of SMPc and SMPp had an important effect on the MFI for MBR-A. However, the SMPc concentration was still more important than SMPp in MBR-B.
(Table 4 Linear correlation between MFI and the SMP content)
3.4 EPS and the suspended flocs properties Fig. 5 presented that the proteins and carbohydrates of EPS concentrations in both reactors. From MBR-A, we observed a decrease of EPS in the sensitive period and an increase of EPS in the recovery period compared to the normal period. It could be seen that the protein fraction of EPS (EPSp) concentration decreased dramatically from 23 mg/g SS in the normal period to 10 mg/gSS in the sensitive period, which was explained that the high monovalent cations content could help in the extraction of EPS from the flocs to form the new SMP (Remy et al., 2011). In general, carbohydrates are extracellular components synthesized for specific function, while proteins can exist in the extracellular polymer network due to the excretion of intracellular proteins/enzymes 15
or cell lysis (Zhao and Gu, 2006). Therefore, it was indicated that the EPSp had a more sensitive than the carbohydrate fraction of EPS (EPSc) at a high salt shock, which was also consistent with the varieties of SMP in the sensitive period for MBR-A (see in Fig. 4).
(Fig. 5. Variations in the concentration of protein and carbohydrate under different operating conditions for EPS)
For MBR-B, the EPSp concentration increased dramatically from 17 mg/g SS in the normal period to 44 mg/g SS in the sensitive period. Increase in EPSp could be attributed to the decrease in concentrations of SMPp by the help of MPE50 addition. Significant amount of SMPp was removal from supernatant and transferred to the microbial flocs, which was likely due to formation of loose and more porous cake structures on the membrane surface (Dizge et al., 2011).
The EPS played a significant role in sludge or bacterial adhesion onto membrane surfaces by altering the physicochemical characteristics such as charge, hydrophobicity, and the polymeric properties (Tansel et al., 2006). The decreases in EPS with salt shock load would change their negative charges as well as hydrophobic regions, which resulted in floc deterioration and deflocculation (Le-Clech et al. 2006).
Table 5 showed the mean particles size, Zeta potential and relative hydrophobicity (RH) in different periods for both reactors. It could be seen that the high salt shock decreased the mean particles size of floc from 78.5 µm in the normal period to 70.3 µm in the sensitive period for MBR-A. Previous studies have demonstrated that high 16
concentrations of monovalent cations (such as Na+) have a detrimental effect on sludge flocculation and floc strength (Sobeck and Higgins, 2002; Arabi and Nakhla, 2009). Floc strength is partly determined by EPS that are responsible to maintain the floc structure and by divalent cations (such as Ca2+), forming bridges between the polymers. The divalent cations are replaced by monovalent cations under a high monovalent cation concentration, causing weaker intrapolymer bridges, and resulting in a disintegration of the floc structure into small individual flocs. It could be seen that the mean particles sizes increased 14.3 µm by the aid of MPE50, indicating that part of SMP and also smaller biological colloids had flocculated and larger aggregates had formed. Similar phenomena were also observed in MBRs with other flocculants addition (Ji et al., 2010; Zhang et al., 2008). It was note that the MPE50 addition increased the Zeta potential form -14.3 mV to -2.3 mV, which indicated that the negative surface charge of the microbial flocs became more neutral. The neutralized flocs would attract each other to produce larger flocs by a charge neutralization mechanism (Lee et al., 2007). The larger floc size was in favor of the reduction of membrane fouling, which agreed with the variations of TMP (see in Fig. 3)
(Table 5 Parameters relevant to the characteristics of the flocs for both MBRs)
Table 5 showed the variation RH with the different stages. It was been observed that decrease in RH (from 27.4% to 18.6%) for MBR-A and increase in RH (from 29.7% to 52.5%) for MBR-B occurred from the normal period to the sensitive period. Since proteins are more hydrophobic than carbohydrates (Ji et al., 2010), a reduction of
17
EPSp for MBR-A and a remarkable increase of EPSp for MBR-B led to the RH variations of flocs. It was reported that the increase of RH would enhance bioflocculation and result in larger more permeable flocs and reduce membrane fouling (Arabi and Nakhla, 2009). Moreover, the RH increase could improve filtration by reducing the interaction between the hydrophobic flocs and the hydrophilic membrane (Le-Clech et al., 2006).
4. Conclusions
High salt shock decreased the conventional pollutant removal efficiency for MBR process, which exhibited a significant negative impact on the autotrophic than heterotrophic biomass. The MPE50 addition enhanced the average removal of COD, NH4+-N and TP by 4.1, 13.2 and 21.2%, respectively. The concentration of SMP significantly increased at high salt shock, especially the content of SMP proteins, which correlated well with the sludge filterability. The mitigating membrane fouling by MPE50 could be attributed to a charge neutralization mechanism, increase in floc size and related hydrophobicity, decrease in concentration of SMP due to entrapment/ sorption onto flocs. Acknowledgments The authors wish to thank the Jilin Province Scientific and Technological Planning Project of China (No. 20120404 & No. 20130206061GX), the Science and Technology Development Program of Jilin city in China (No. 20106306 & No.201232403) and Chinese Society of Electrical Engineering for Power Technology Youth Innovation 18
Project (No. 2009NETU) for the partial support of this study.
References
1.
APHA. 2005. Standard Methods for the Examination of Water and Wastewater. twentieth ed. American Public Health Association, American Water Works Association, Water Environmental Federation, Washington, USA.
2.
Arabi, S., Nakhla , G., 2009. Impact of cation concentrations on fouling in membrane bioreactors. J. Membr. Sci. 343, 110-118.
3.
Aslan, S., Simsek, E., 2012. Influence of salinity on partial nitrification in a submerged biofilter. Bioresour. Technol. 118, 24-29.
4.
Bassin, J. P., Dezotti, M., San, A., 2011. Nitrification of industrial and domestic saline wastewaters in moving bed biofilm reactor and sequencing batch reactor. J. Hazard. Mater. 185, 242-248.
5.
Cho, B. D., Fane, A. G., 2002. Fouling transient in nominally sub-critical flux operation of a membrane bioreactor, J. Membr. Sci. 209, 391-403.
6.
Cortés-Lorenzo, C., Rodríguez-Díaz, M., López-Lopez, C., Sánchez-Peinado, M., Rodelas, B., González-López, J. 2012. Effect of salinity on enzymatic activities in a submerged fixed bed biofilm reactor for municipal sewage. Bioresour. Technol. 121, 312-319.
7.
Dizge, N., Yuksel, D., Koseoglu-Imer, Karagunduz, A., Keskinler B., 2011. Effects of cationic polyelectrolyte on filterability and fouling reduction of submerged membrane bioreactor (MBR). J. Membr. Sci. 377, 175-181.
8.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356.
9.
Jang, D., Hwang, Y., Shin, H., Lee, W., 2013. Effects of salinity on the characteristics of biomass and membrane fouling in membrane bioreactors. Bioresour. Technol. 141, 50-56.
10.
Jang, N., Ren, X., Cho, J., Kim, I. S., 2006. Steady-state modeling of bio-fouling 19
potentials with respect to the biological kinetics in the submerged membrane bioreactor (SMBR). J. Membr. Sci. 284 , 352-360. 11. Ji, J., Qiua, J., Wai, N., Wong, F, Li, Y., 2010. Influence of organic and inorganic flocculants on physical-chemical properties of biomass and membrane-fouling rate. Water Res. 44, 1627-1635. 12.
Koseoglu, H., Yigit, N. O., Iversen, V., Drews, A., Kitis, M., Lesjean, B., Kraume, M., 2008. Effects of several different flux enhancing chemicals on filterability and fouling reduction of membrane bioreactor (MBR) mixed liquors. J. Membr. Sci. 320, 57-64.
13.
Le-Clech, P., Chen V., Fane, A. G., 2006. Fouling in membrane bioreactors used in wastewater treatment. J. Membr. Sci. 284, 17-53.
14.
Lee, W. N., Chang I.S., Hwang, B.K., Park, P.K., Lee, C. H., Huang, X., 2007. Changes in biofilm architecture with addition of membrane fouling reducer in a membrane bioreactor. Process Biochem. 42, 655-661.
15.
Lowry, O. H., Rosebrough N. J., Farr, A. L., Randall R. J. 1951. Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem. 193, 265-275
16.
Morgan, J. W., Forster, C. F., Evison L., 1990. Comparative study of the nature of biopolymers extracted from anaerobic and activated sludges. Water Res. 24, 743-750.
17.
Murthy, S. N., Novak, J. T., Dehaas R. D., 1998. Monitoring cations to predict and improve activated sludge settling and dewatering properties of industrial wastewaters, Water Sci. Technol. 38, 119-126.
18.
Mutamim, N., Noor, Z., Hassan, M., Yuniarto, A., Olsson, G., 2013. Membrane bioreactor: Applications and limitations in treating high strength industrial wastewater. Chem. Eng. J. 225, 109-119.
19.
Nguyen, T. T., Guo, W., Ngo, H. H., Vigneswaran S., 2010. A new combined inorganic-organic flocculant (CIOF) as a performance enhancer for aerated submerged membrane bioreactor. Sep. Purif. Technol. 75, 204-209.
20.
Ognier, S., Wisniewski, C., Grasmick, A., 2002. Influence of macromolecular adsorption during filtration of amembrane bioreactor mixed liquor suspension. J. 20
Membr. Sci. 209, 27-37. 21.
Ray, S., Peters, C.A., 2008. Changes in microbiological metabolism under chemical stress. Chemosphere. 71, 474-483.
22.
Reid, E., Liu X., Judd, S.J., 2006. Effect of high salinity on activated sludge characteristics and membrane permeability in an immersed membrane bioreactor. J. Membrane Sci. 283, 164-171.
23.
Remy, M., Temmink, H., Brink, P., Rulkens W., 2011. Low powdered activated carbon concentrations to improve MBR sludge filterability at high salinity and low temperature. Desalination. 276, 403-407.
24.
Rosengerg, M., Gutnick, D., Rosengerg, E., 1980. Adherence of bacteria to hydrocarbons: a simple method for measuring cellsurface hydrophobicity. FEMS Microbiol. Lett. 9, 29-33.
25.
Sobeck, D. C., Higgins M .J., 2002. Examination of three theories for mechanisms of cationinduced bioflocculation. Water Res. 36, 527-538.
26.
Sun, C., Leiknes, T., Weitzenbok, J., Thorstensen, B., 2010. Salinity effect on a biofilm-MBR process for shipboard wastewater treatment. Sep. Purif. Technol. 72, 380-387.
27.
Tansel, B., Sager, J., Garland, J., Xu, S., Levine, L., Bisbee, P., 2006. Deposition of extracellular polymeric substances (EPS) and microtopographical changes on membrane surfaces during intermittent filtration conditions. J. Membr. Sci. 285, 225-231.
28.
Uygur, A., 2006. Specific nutrient removal rates in saline wastewater treatment using sequencing batch reactor. Process Biochem. 41, 61-66.
29.
Villain, M., Marrot, B., 2013. Influence of sludge retention time at constant food to microorganisms ratio on membrane bioreactor performances under stable and unstable state conditions. Bioresour. Technol. 128 , 134-144.
30.
Vyrides, I., Stuckey, D. C., 2009. Adaptation of anaerobic biomass to saline conditions: Role of compatible solutes and extracellular polysaccharides. Enzyme Microb. Technol. 44, 46-51.
31.
Wintgens, T., Melin, T., Schafer, A., Khan S., Muston M., Bixio, D., Thoeye, C., 21
2005. The role of membrane processes in municipal wastewater reclamation and reuse. Desalination. 178, 1-11. 32.
Yang, W., Cicek N., Ilg, J., 2006. State-of-the-art of membrane bioreactors: worldwide research and commercial applications in Aorth America. J. Membr. Sci. 270, 201-211
33.
Yogalakshmi, K. N., Joseph, K., 2010. Effect of transient sodium chloride shock loads on the performance of submerged membrane bioreactor. Bioresour. Technol. 101, 7054-7061.
34.
Zhang, H. F., Sun, B. S., Zhao, X. H., Gao, Z. H., 2008. Effect of ferric chloride on fouling in membrane bioreactor. Sep. Purif. Technol. 63, 341-347.
35.
Zhao, Y., Gu, P., 2006. Effect of powdered activated carbon dosage on retarding membrane fouling in MBR. Sep. Purif. Technol. 52, 154-160.
22
Figure captions:
Fig. 1. Schematic of MBR-A and MBR-B
Fig. 2. Comparison between the control and the MBR with coagulant in removing conventional pollutants
Fig. 3. Profile of TMP rise up in the MBR-A and MBR-B
Fig. 4. Variations in the concentration of protein and carbohydrate under different operating conditions for SMP
Fig. 5. Variations in the concentration of protein and carbohydrate under different operating conditions for EPS
23
Pressure gauge
Peristaltic pump
Agitator
Membrane Feeding pump
Rotameter
Level sensor
Air pu mp
MBR-A
Timer Pressure gauge
Peristaltic pu mp
Level controller MPE50
Wastewater Tank Membrane
Feeding pump
Level sensor
Rotameter Air pu mp
MBR-B
Fig. 1. Schematic of MBR-A and MBR-B.
24
100
80
70
MBR-A (COD) MBR-B (COD) + MBR-A(NH4 -N)
+
COD or NH4 -N removal (%)
90
60
+
MBR-B((NH4 -N)
50
40
(a)
50
(b)
MBR-A MBR-B
Phosphate removal(%)
40
30
20
10
0 0
10
20
30
40
50
60
70
Time(days)
Fig. 2. Comparison between the control and the MBR with coagulant in removing conventional pollutants
25
Normal period
Senstive period Recorvery period
25
MBR-A MBR-B TMP(kPa)
20
15
10
5 15
20
25
30
35
40
45
50
55
Time(days)
Fig.3. Profile of TMP rise up in the MBR-A and MBR-B
26
6.0
MBR-A
5.5
MBR-B SMPp SMPc
5.0 4.5
SMP/(mg/gSS)
4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Normal
Sensitive
Recovery
Normal
Sensitive
Recovery
Period
Fig. 4. Variations in the concentration of protein and carbohydrate under different operating conditions for SMP
27
60
MBR-A
55
MBR-B
EPSp EPSc
50 45
EPS(mg/mgSS)
40 35 30 25 20 15 10 5 0 Normal
Sensitive
Recovery
Normal
Sensitive
Recovery
Period
Fig. 5. Variations in the concentration of protein and carbohydrate under different operating conditions for EPS.
28
List of tables Table 1 Conventional pollutant removal efficiencies in both MBRs in different periods
Table 2 Respirometry activities in both reactors between before and after salt shock.
Table 3 Membrane fouling rate (kPa/d) in different periods for both MBRs.
Table 4 Linear correlation between MFI and the SMP content.
Table 5 Parameters relevant to the characteristics of the flocs for both MBRs
29
Table 1 Conventional pollutant removal efficiencies in both MBRs in different periods Parameter
COD
NH4+-N
TP
Removal efficiency (%)
Period
MBR-A
MBR-B
1-30 d
96.2±0.8
96.6±0.9
31-33 d
89.6±4.1
93.7±2.1
34-70 d
96.4±0.9
96.5±1.0
1-30 d
91.7±1.1
91.6±1.1
31-40 d
71.7±18.3
84.9±11.6
41-70 d
91.6±0.8
91.5±0.9
1-30 d
33.5±1.4
34.2±1.5
31-41 d
9.1±3.2
30.3±3.4
42-70 d
33.8±1.4
36.1±3.3
30
Table 2 Respirometry activities in both reactors between before and after salt shock Respirometry activities (mg O2 g VSS-1 h-1) Heterotrophic Autotrophic
Before salt shock
After salt shock
MBR-A
MBR-B
MBR-A
MBR-B
Exogenous
19.4 ± 4.7
20.7± 5.3
17.2 ± 2.4
17.9 ± 2.6
Endogenous Exogenous
1.5 ± 0.3
1.7 ± 0.4
2.1 ± 0.4
1.8 ± 0.2
4.7 ± 0.5
4.2± 0.3
0.7 ± 0.2
1.5 ± 0.4
Endogenous
1.3 ± 0.7
1.4 ± 0.5
2.9 ± 0.4
2.3 ± 0.5
31
Table 3 Membrane fouling rate (kPa/d) in different periods of both MBRs Reactors
Normal period
Sensitive period
Recovery period
MBR-A
1.26
12.17
3.82
MBR-B
1.31
2.34
2.31
32
Table 4 Linear correlation between MFI and the SMP content Period Normal
Sensitive
Recovery
SMP SMPp
MBR-A
MBR-B
Fitting equation
R2
Fitting equation
R2
y=0.0864x+0.276
0.39
y=0.0871x+0.2821
0.29
SMPc y=0.2523x+0.9874
0.84
y=0.5247x+0.2809
0.86
SMPp y=0.0384x+4.6748
0.97
y=0.0731x+0.4895
0.58
SMPc
0.92
0.53
y=0.1512x+0.7534
SMPp y=0.3512x+0.1969
y=0.018x+2.9306
0.94
y=0.1235x+0.2599
0.39
SMPc y=0.1756x+1.6374
0.89
y=0.5927x-0.5441
0.83
33
Table 5 Parameters relevant to the characteristics of the flocs for both MBRs. MBR-A
MBR-B
Period PSD(μ m)
Zeta (mV)
RH (%)
PSD(μ m)
Zeta (mV)
RH (%)
Nomal
78.5
-14.4
27.4
75.3
-14.6
29.7
Sensitive
70.3
-11.8
18.6
89.6
-2.3
52.5
Revovery
89.4
-15.3
32.5
89.3
-4.4
48.7
34
Supplementary Figure
Fig. S1. Correlation between MFI and membrane fouling rate for the both MBRs
35
18
Fouling rate(kPa/d)
16 14 y = 0.7216x - 1.041 R 2 = 0.9757
12 10 8 6 4 2 0 0
2
4
6
8
10
12
14 5
16
18
20
22
24
26
2
MFI(×10 s/L )
Fig. S1. Correlation between MFI and membrane fouling rate for the both MBRs.
36
Highlights Impact of salt shock on MBR performance was assessed. Autotrophs showed more sensitive than heterotrophs at salt shock. Higher SMP protein release was observed under high salt condition. Changes in sludge properties by MPE50 addition significantly reduced membrane fouling.
37
S0960-8524(14)00558-6 http://dx.doi.org/10.1016/j.biortech.2014.04.053 BITE 13349
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
22 January 2014 15 April 2014 17 April 2014
Please cite this article as: Zhang, H., Gao, Z., Zhang, L., Song, L., Performance enhancement and fouling mitigation by organic flocculant addition in membrane bioreactor at high salt shock, Bioresource Technology (2014), doi: http://dx.doi.org/10.1016/j.biortech.2014.04.053
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Performance enhancement and fouling mitigation by organic flocculant addition in membrane bioreactor at high salt shock Haifeng Zhanga, b, Zhongyu Gaoa, Lanhe Zhanga, Lianfa Songb,*
a
School of Chemistry Engineering, Northeast Dianli University, Jilin 132012, Jilin, P. R. China ；
b
Department of Civil and Environmental Engineering, Texas Tech University, 10th and Akron, Lubbock, TX 79409-1023, USA
* Corresponding Author. Tel.: +1 806 742 3598. E-mail: [email protected]
1
Abstract The main objective of this study was to investigate the effect of an organic flocculant (MPE50) addition on reducing membrane fouling and enhancing performance in membrane bioreactor (MBR) at the high salt shock. Results show that MPE50 addition is a reliable and effective approach in terms of both membrane fouling mitigation and pollutants removal improvement in the case of high salt shock. Compared to the control reactor, the MBR with MPE50 addition enhanced the average removal of COD, NH4+-N and TP by 4.1, 13.2 and 21.2%, respectively. Due to the effect of flocculation and adsorption by MPE50, a significantly reduction in the soluble microbial products (SMP) proteins amount was observed. As a result, the membrane fouling rate was mitigated successfully. Further, the increasing of mean particles size, Zeta potential and related hydrophobicity of the flocs would also have positive impacts on membrane fouling mitigation.
Key words: Membrane bioreactor; High salt shock; Related hydrophobicity; Modified fouling index; Soluble microbial products; Extracellular polymeric substances.
2
1. Introduction Among membrane processes, membrane bioreactor (MBR) technology is an innovative and promising option for wastewater treatment and reuse, which has been widely used for the treatment of various wastewater, including municipal, industrial and landfill leachate (Wintgens et al., 2005). However, membrane fouling is an unavoidable and probably the most serious challenge, which increases operational cost and shortens membrane life for MBR (Yang et al., 2006). Membrane fouling in MBR is mainly resulted from the physicochemical interaction between the suspended particles and membrane to form a cake layer and the adsorption of the dissolved species into membrane pores that lead to flux decline (Mutamim et al., 2013). The characteristics of the mixed liquor, such as morphological properties (mean floc size, floc fractal dimension), physical parameters (surface charge, relative hydrophobicity, dynamic viscosity), and the biochemical components of the mixed liquor (extracellular polymeric substances, soluble microbial products) govern the filterability and the membrane fouling (Ji et al., 2010).
The high or changing salinity imposes a challenge to activated sludge process (Yogalakshmi and Joseph, 2010), which is likely to occur in many biological wastewater treatment systems nowadays (Cortés-Lorenzo et al., 2012). For example, the coastal sewers are subjected to infiltration by seawater or when industrial effluents receive discharges from individual high-salinity processes. Additionally, the urban wastewater may also be affected by saline industrial wastewaters or other sources. For
3
example, seawater is used in Hong Kong for toilet flushing to save freshwater (Aslan and Simsek, 2012). Salt can also enter the urban wastewater flow when it is used in the cities for thawing snow and ice on the streets (Jang et al., 2013).
High salt concentration in wastewater can cause cell plasmolysis due to the dramatic increase in osmotic pressure and changes on microbial metabolism (Bassin et al., 2011; Vyrides and Stuckey, 2009). Moreover, salinity may significantly affect the physical and biochemical properties of the biomass, leading to a change of surface charge, hydrophobicity, filterability and bioflocculation of biomass (Sun et al., 2010). A limited number of studies have been conducted on MBR operation for saline wastewater shock, which indicated that salinity not only exerted adverse impact on the pollutant removal, biomass activity and microbial diversity, but also significantly deteriorated sludge filterability and resulted in serious membrane fouling (Reid et al., 2008; Yogalakshmi et al., 2010; Jang et al., 2013 ). Therefore, it is necessary to develop an effective method of reducing membrane fouling for MBR process at high salt shock.
Recently, application of inorganic and organic flocculants in MBR to modify the properties of mixed liquor artificially for alleviation of membrane fouling has been reported in several publications (Ji et al., 2010; Zhang et al., 2008). A novel organic flocculant called Membrane Performance Enhancer (MPE50) has been developed by Nalco Company and applied to MBRs for membrane fouling control (Nguyen et al., 2010; Ji et al., 2010). The enhanced filterability in MBR with the addition of MPE50 was attributed to an increase in floc size and to a decrease in concentration of soluble
4
foulants in the bulk phase (Ji et al., 2010). Moreover, it is reported that MPE50 does not negatively affect the overall aeration demand and bioactivity (Nguyen et al., 2010). Considering the significant impact of flocculant dosing on membrane fouling mitigation, it would be very interesting to study how MPE50 addition affects membrane fouling in MBR at high salt shock.
The main objective of this work was to study the effectiveness of flocculant on membrane fouling reduction in MBR at high salt shock. For this purpose, a control MBR and a hybrid MBR with MPE50 addition were operated in parallel. A comparison was made of the mixed liquor properties for both MBRs during the whole operation time, with parameters including the content of protein and carbohydrate in soluble microbial product (SMP) and extracellular polymeric substances (EPS), as well as the mean particle sizes, Zeta potential, and relative hydrophobicity of the flocs. The modified fouling index (MFI) was used to confirm the mixed sludge filterability under each condition.
2. Material and methods 2.1. MBR systems Two MBRs of identical size, MBR-A (the control reactor) and MBR-B (with flocculant addition), were operated for about 70 days. A schematic of the experimental setup is in Fig.1. Each reactor had a working volume of 8 L. The membrane module used in both systems was a bundle of U-shaped hollow fiber membranes (polyvinylidene fluoride, PVDF) with a pore size of 0.2 µm and a filtration area of 0.1 5
m2. Air was supplied to both reactors at 5 L/min. Permeate through the submerged membrane module was continuously withdrawn using a peristaltic pump at a constant flux of 10 L/m2·h, operated with an intermittent mode of 13 min on and 2 min off. Chemical cleaning procedure would be carried out if the trans-membrane pressure (TMP) reached about 28 kPa during the operation. The membrane modules were taken out and soaked in 0.5% (v/v) NaClO solution for about 2 h to recover their permeability. The seed sludge was taken from a lab-scale MBR for more than one year. Each MBR was seeded with 8 g/L of the acclimatized activated sludge. Both MBRs were operated under the same conditions and stabilized for 30 days. Hydraulic retention time (HRT) and sludge residence time (SRT) were maintained at 6 h and 30 days, respectively. Synthetic wastewater was used in order to avoid difficulties in quantification of metabolic products due to variation in the composition and concentration of organic pollutants in the real wastewater. The synthetic wastewater was prepared using glucose, NH4Cl, and KH2PO4 as a source of carbon, nitrogen, and phosphorous, respectively. The recipe could be obtained from our previous study (Zhang et al., 2008). During the steady state operation, both MBRs had similar mixed liquor suspended solids (MLSS) concentrations and mixed liquor volatile suspended solids (MLVSS) concentrations typically in the range of 10-13 g/L and 8-10.5 g/L, respectively. One simulation of salt shock was applied in this study. On day 31, 20 g/L of NaCl was administered in the feed for a period of 24 hours to simulate a shock salt loading. More frequent simulations are not appropriate in this study because salt tolerance active sludge would develop that may not adapt in the fresh water 6
environments in the normal conditions. Before starting the comparison tests between both MBRs, batch tests were conducted at various MPE50 (Nalco Company) dosages to determine the extent of SMP removals. According to the results of batch tests, 600 mg/L MPE50 was firstly added in MBR-B and 160 mg MPE50 was supplied daily to the reactor to compensate the loss by the sludge discharge. The addition of MPE50 can be stopped when membrane fouling is controlled and the performance of MBR is recovered. (Fig.1. Schematic of MBR-A and MBR-B)
2.2. Filterability
The filterability characteristics of mixed liquid were investigated in batch tests using Amicon 8400 dead-end cells (Millipore, Billerica, MA). The cells had a volume of 350 mL and an effective membrane filtration area of 41.8 cm2. All experiments were carried out at room temperature and under a pressure of 28 kPa. The permeation flux was determined by weighing permeates on an electronic top loading balance connected to a personal computer running an auto-reading program. A new membrane sheet (PVDF, 0.2 µm) was used for each trial. Under unstirred conditions, MFI was obtained from the plot of t/V versus V using the filtration equation at constant pressure (Ognier et al., 2002). MFI parameter gives an idea of the fouling potential characteristics of the mixed liquor. A higher MFI is result in a higher fouling rate for the mixed liquor. It is based on the cake filtration mechanism (Jang et al., 2006). MFI is defined as the gradient of the linear relationship between t/V and V. 7
t = V
µRm △P
MFI =
+
µαC 2△ P
(1)
V
µα C
(2)
2△P
Where t is the filtration time (s), V the permeate volume per unit filtration area (m), µ the dynamic viscosity of permeate (Pa s), Rm the intrinsic membrane resistance (m-1), ∆P the applied transmembrane pressure (kPa), α the specific resistance (m/kg) and C is macromolecules concentration in bulk solution (mg/L). 2.3. Analytic methods Standard analytic methods were used to measure for chemical oxygen demand (COD), NH4+-N, total phosphate (TP), MLSS and MLVSS (APHA, 2005). The SMP samples were obtained by centrifuging the mixed liquor at 4000 rpm for 10 min, followed by filtration through a membrane of 0.45 µm. EPS were extracted by heat treatment (Morgan et al., 1990). Carbohydrates were determined according to the phenol-sulfuric acid method (Dubois et al., 1956) with glucose as standard. Proteins were determined by the modified Lowry method (Lowry et al., 1951) with bovine serum albumin (BSA) as standard. Relative hydrophobicity of the flocs was measured according to Rosenberg’s method (Rosengerg et al., 1980). Particle size distributions of flocs were measured with a Malvern Mastersizer (Malvern, UK). The Zeta potential was recorded using a Zetasizer Nano ZS Instrument (Malvern, UK). 2.4. Respirometry tests The biological activity was followed from respirometry tests used to measure autotrophic and heterotrophic activity by differentiating endogenous and exogenous
8
activity. The endogenous activity is defined as the rate of oxygen consumption in the absence of external substrate. Bacteria are forced to biodegrade their intracellular carbon as substrate to provide the energy necessary for cell maintenance and their respiration is minimal. The exogenous activity is defined as the rate of oxygen consumption after substrate addition and bacteria use the substrate available to supply their energy demand, and the resulting respiration is at its highest value (Villain and Marrot., 2013). The aerated reactor was filled with 1 L of sludge and every two minutes, 50 mL was sampled and injected into another reactor without oxygenation to measure the oxygen uptake rate (OUR) with a continuous dissolved oxygen probe (Mettler-Toledo, Columbus, OH). Then, specific nutrients or inhibitor were successively added to the aerated reactor according to four steps: (1) Endogenous respirations of autotrophic and heterotrophic micro-organisms were first measured over a period of 1 h; (2) Ammonium was added to measure the maximum activity of autotrophic micro-organisms; (3) Autotrophic micro-organisms were inhibited with allylthiourea to isolate heterotrophic endogenous respiration; (4) The maximum activity of heterotrophic micro-organisms was quantified after glucose addition. 3. Results and discussion 3.1 Comparison of conventional pollutant removal Fig. 2 and Table 1 showed the conventional pollutant removal efficiencies in both MBRs in different periods. As shown in Table 1, it could be seen that the average COD removal decreased from 96.2±0.8% (1-30 days run) to 89.6±0.9% (31-33 days run) for MBR-A, this result was also consistent with the temporal variation of heterotrophic 9
activities shown the Table 2. A decrease heterotrophic exogenous (from 19.4±4.7 to 17.2±2.4 mg O2 gVSS-1 h-1) and an increase endogenous (from 1.5±0.3 to 2.1±0.4 mg O2 gVSS-1 h-1) were observed in the MBR-A, which suggested that salt shock caused a slight diversion of carbon and/or energy from growth to adaptive responses and protection (Ray and Peters, 2008). In contrast, it was observed that average COD removal was slightly enhanced by 4.1% with MPE50 addition in MBR-B, reaching 93.7%. It was indicated that the cationic polymer addition could have a beneficial effect on organic matter removal when treating saline wastewaters.
(Table 1 Conventional pollutant removal efficiencies in both MBRs in different periods)
With regard to NH4+-N, the average removal was over 91% for both MBRs during the first 30 days. After salt shock, the average NH4 +-N removal dropped to 71.7±18.3 and 84.9±11.6% for MBR-A and MBR-B, respectively. The salt shock exhibited a significant negative impact on the autotrophic biomass in both reactors, for example, the autotrophic bacteria increased endogenous activities from 1.3±0.7 to 2.9±0.4 mg O2 gVSS-1 h-1 and simultaneously greatly decreased their exogenous activities from 4.7±0.5 to 0.7±0.2 mg O2 gVSS-1 h-1 in MBR-A. It took 10 d and 7 d to recover the stable NH4+-N removal efficiency for MBR-A and MBR-B, respectively. This result was consistent with the report (Jang et al., 2013) , which indicated that nitrifiers were very sensitive to high salinity and would be required a longer time to recover due to their slow growth rates. Compared with MBR-A, it was obviously that MPE50 caused a positive impact on nitrifiers in MBR-B, resulting in about 13.2% increase of average
10
NH4+-N removal efficiency and reduction of the recovery time by 3 days.
(Table 2 Respirometry activities in both reactors between before and after salt shock)
It was observed that the average TP removal efficiency in both MBRs were about 30-40% at the first 30 days. However, the TP removal during the days 31 and 41 was averaged at 9.1±3.2% in MBR-A, which might be explained that the phosphate accumulating organisms (PAO) were more sensitive to salinity than the carbon and nitrogen bacteria. PAO cells might have lost phosphate accumulation capability due to osmotic pressure (Uygur, 2006). Also, wash out of phosphate from unmetabolized substrate can also be other possible reason (Jang et al., 2013). In comparison, the average TP removal was enhanced by 21.2% when MPE50 addition operation period, which would be a result of the concurrent effect of chemical and biological phosphorus removal. It was also observed that the TP removal in MBR-B (36.1±3.3% ) was slight higher than MBR-A (33.8±1.4%) even after MPE50 dosing had been terminated during the period of 42-70 days, indicating that the MPE50 accumulation in sludge floc still had a positive effect on phosphorus removal.
(Fig. 2. Comparison between the control and the MBR with coagulant in removing conventional pollutants)
3.2 Membrane fouling rates and mixed sludge filterability
Typical TMP profiles depicting fouling trends for both MBRs were shown in Fig.3. The membrane fouling rates were computed based on TMP development each day
11
(dTMP/day) through the experimental time (see in Table 3). The increase rate of TMP is an important indicator for evaluating the development of membrane fouling in submerged MBRs (Le-Clech.et al., 2006). Based on the membrane fouling rate, the total operation time could be divided into three stages: the normal period, the sensitive period, and the recovery period. As could be seen from Fig. 3, it was not obviously different in TMP increase in the normal period for both MBRs, the membrane fouling rates were 1.26 and 1.31 kPa/d for MBR-A and MBR-B, respectively.
(Table 3 Membrane fouling rate (kPa/d) in different periods of both MBRs)
However, the membrane fouling rate significantly increased in the sensitive period for MBR-A. The fouling rate was so severe that membrane filtration had to be stopped for chemical cleaning just about 2 days operation, which was very apparently a sticky gel fouling layer was observed on the membrane surface when the fouled membrane was cleaned. On the contrary, significant improvement in filterability was observed in MBR-B by extending the filtration time by almost four-fold compared with MBR-A. The membrane fouling rate of MBR-B was 2.34 kPa/d, which was much lower than that of in MBR-A (12.17 kPa/d on average). The results suggested that MPE50 addition managed to control membrane fouling could be a useful strategy in case of salt shock. Otherwise, the MBR operation had to be terminated due to severe membrane fouling. The chemical cleaning frequency needs to be increased, which will reduce membrane lifetime and increase the operation costs (Le-Clech.et al., 2006).
(Fig.3. Profile of TMP rise up in the MBR-A and MBR-B) 12
In general, the TMP profiles exhibited two-stage process with linear gradual TMP rise followed by sudden increase in the rate of TMP rise (Cho and Fane., 2002). However, the MBR-A exhibited the only rapid increase stage in the sensitive period, which indicated that high salt shock significantly affect physical and biochemical properties of activated sludge, which then impacted upon mixed sludge filterability (Reid et al., 2006). The relationship between the fouling rates and the modified fouling indicate (MFI) was shown in Supplementary Fig. S1. The linear curve showed a strong correlation between the MFI and the fouling rate with an r-squared value of 0.98. Therefore, the MFI could be considered as a reliable mixed sludge filterability indicator to predict the extent of membrane fouling rate in MBR filtration process. A previous study showed a similar result (Zhang et al., 2008). 3.3 SMP in the supernatant Fig. 4 showed the proteins and carbohydrates concentration of SMP under different periods for both reactors. Compared with the normal period, it was noted that both the carbohydrate fraction of SMP (SMPc) and the protein fraction of SMP (SMPp) content increased obviously in the sensitive period for MBR-A, which was consistent with previous studies that high salt shock resulted in increased SMPc and SMPp concentrations in the supernatant of MBRs (Reid et al., 2006). It was explained that microorganisms respond to a salt shock acceleration of endogenous respiration (see in Table 2), accompanied by release of organic cellular constituents, increase the SMP by secretion and cells autolysis (Reid et al., 2006). Moreover, the increase in salinity also could increase the solubility of protein and carbohydrate of SMP, which might be 13
another reason for increasing the SMP content in the supernatant (Sun et al., 2010). Compared with the SMPc, the concentration of SMPp dramatically increased up to 5.58 mg/g SS in the sensitive period, which was almost 11 times higher than that in normal period in the MBR-A. The results were consistent with the findings of other studies showing that introduction of sodium caused a release of soluble proteins (Murthy et al., 1998).
(Fig. 4. Variations in the concentration of protein and carbohydrate under different operating conditions for SMP)
For the MBR-B, it could be seen that the SMPc and SMPp concentrations obviously decreased in the sensitive period compared with MBR-A. MPE50 exerted a greater impact on reducing the SMPp than SMPc, reflected by about 85.8% and 60.7% reduction in SMPp and SMPc, respectively. The role of additional MPE50 in MBR-B could be explained as followed: MPE50, like a flocculant, can weaken the repulsion among the SMP, via charge neutralization, allowing them to flocculate with each other by electrostatic attraction (Koseoglu et al., 2008). The reason of high SMPp removal could be attributed to the charged groups of proteins (Dizge et al., 2011).
To confirm the effect of the proteins and carbohydrates content on the mixed sludge filterability, the MFI values were correlated with the concentrations of SMPc and SMPp in different stages for both MBRs, respectively (see in Table 4). In the normal period, the carbohydrate SMP concentration should be the key factor contributing to mixed sludge filterability for both reactors( R2 = 0.84 for MBR-A , R2 = 0.86 for
14
MBR-B), which is consistent with the findings of other studies that had addressed SMPc as the major cause of the membrane fouling phenomenon (Arabi and Nakhla, 2009). In comparison, the MFI values showed better correlation with the SMPp concentration (R2 = 0.97, compared to R2 = 0.53 for SMPc) in the sensitive period for MBR-A. On the contrary, the MFI values were better correlated with the SMPc concentrations (R2 = 0.92, compared to R2 = 0.58 for SMPp) for MBR-B, which implied that major foulants in both reactors were different depending on MPE50 addition. In the recovery period, the concentrations of SMPc and SMPp had an important effect on the MFI for MBR-A. However, the SMPc concentration was still more important than SMPp in MBR-B.
(Table 4 Linear correlation between MFI and the SMP content)
3.4 EPS and the suspended flocs properties Fig. 5 presented that the proteins and carbohydrates of EPS concentrations in both reactors. From MBR-A, we observed a decrease of EPS in the sensitive period and an increase of EPS in the recovery period compared to the normal period. It could be seen that the protein fraction of EPS (EPSp) concentration decreased dramatically from 23 mg/g SS in the normal period to 10 mg/gSS in the sensitive period, which was explained that the high monovalent cations content could help in the extraction of EPS from the flocs to form the new SMP (Remy et al., 2011). In general, carbohydrates are extracellular components synthesized for specific function, while proteins can exist in the extracellular polymer network due to the excretion of intracellular proteins/enzymes 15
or cell lysis (Zhao and Gu, 2006). Therefore, it was indicated that the EPSp had a more sensitive than the carbohydrate fraction of EPS (EPSc) at a high salt shock, which was also consistent with the varieties of SMP in the sensitive period for MBR-A (see in Fig. 4).
(Fig. 5. Variations in the concentration of protein and carbohydrate under different operating conditions for EPS)
For MBR-B, the EPSp concentration increased dramatically from 17 mg/g SS in the normal period to 44 mg/g SS in the sensitive period. Increase in EPSp could be attributed to the decrease in concentrations of SMPp by the help of MPE50 addition. Significant amount of SMPp was removal from supernatant and transferred to the microbial flocs, which was likely due to formation of loose and more porous cake structures on the membrane surface (Dizge et al., 2011).
The EPS played a significant role in sludge or bacterial adhesion onto membrane surfaces by altering the physicochemical characteristics such as charge, hydrophobicity, and the polymeric properties (Tansel et al., 2006). The decreases in EPS with salt shock load would change their negative charges as well as hydrophobic regions, which resulted in floc deterioration and deflocculation (Le-Clech et al. 2006).
Table 5 showed the mean particles size, Zeta potential and relative hydrophobicity (RH) in different periods for both reactors. It could be seen that the high salt shock decreased the mean particles size of floc from 78.5 µm in the normal period to 70.3 µm in the sensitive period for MBR-A. Previous studies have demonstrated that high 16
concentrations of monovalent cations (such as Na+) have a detrimental effect on sludge flocculation and floc strength (Sobeck and Higgins, 2002; Arabi and Nakhla, 2009). Floc strength is partly determined by EPS that are responsible to maintain the floc structure and by divalent cations (such as Ca2+), forming bridges between the polymers. The divalent cations are replaced by monovalent cations under a high monovalent cation concentration, causing weaker intrapolymer bridges, and resulting in a disintegration of the floc structure into small individual flocs. It could be seen that the mean particles sizes increased 14.3 µm by the aid of MPE50, indicating that part of SMP and also smaller biological colloids had flocculated and larger aggregates had formed. Similar phenomena were also observed in MBRs with other flocculants addition (Ji et al., 2010; Zhang et al., 2008). It was note that the MPE50 addition increased the Zeta potential form -14.3 mV to -2.3 mV, which indicated that the negative surface charge of the microbial flocs became more neutral. The neutralized flocs would attract each other to produce larger flocs by a charge neutralization mechanism (Lee et al., 2007). The larger floc size was in favor of the reduction of membrane fouling, which agreed with the variations of TMP (see in Fig. 3)
(Table 5 Parameters relevant to the characteristics of the flocs for both MBRs)
Table 5 showed the variation RH with the different stages. It was been observed that decrease in RH (from 27.4% to 18.6%) for MBR-A and increase in RH (from 29.7% to 52.5%) for MBR-B occurred from the normal period to the sensitive period. Since proteins are more hydrophobic than carbohydrates (Ji et al., 2010), a reduction of
17
EPSp for MBR-A and a remarkable increase of EPSp for MBR-B led to the RH variations of flocs. It was reported that the increase of RH would enhance bioflocculation and result in larger more permeable flocs and reduce membrane fouling (Arabi and Nakhla, 2009). Moreover, the RH increase could improve filtration by reducing the interaction between the hydrophobic flocs and the hydrophilic membrane (Le-Clech et al., 2006).
4. Conclusions
High salt shock decreased the conventional pollutant removal efficiency for MBR process, which exhibited a significant negative impact on the autotrophic than heterotrophic biomass. The MPE50 addition enhanced the average removal of COD, NH4+-N and TP by 4.1, 13.2 and 21.2%, respectively. The concentration of SMP significantly increased at high salt shock, especially the content of SMP proteins, which correlated well with the sludge filterability. The mitigating membrane fouling by MPE50 could be attributed to a charge neutralization mechanism, increase in floc size and related hydrophobicity, decrease in concentration of SMP due to entrapment/ sorption onto flocs. Acknowledgments The authors wish to thank the Jilin Province Scientific and Technological Planning Project of China (No. 20120404 & No. 20130206061GX), the Science and Technology Development Program of Jilin city in China (No. 20106306 & No.201232403) and Chinese Society of Electrical Engineering for Power Technology Youth Innovation 18
Project (No. 2009NETU) for the partial support of this study.
References
1.
APHA. 2005. Standard Methods for the Examination of Water and Wastewater. twentieth ed. American Public Health Association, American Water Works Association, Water Environmental Federation, Washington, USA.
2.
Arabi, S., Nakhla , G., 2009. Impact of cation concentrations on fouling in membrane bioreactors. J. Membr. Sci. 343, 110-118.
3.
Aslan, S., Simsek, E., 2012. Influence of salinity on partial nitrification in a submerged biofilter. Bioresour. Technol. 118, 24-29.
4.
Bassin, J. P., Dezotti, M., San, A., 2011. Nitrification of industrial and domestic saline wastewaters in moving bed biofilm reactor and sequencing batch reactor. J. Hazard. Mater. 185, 242-248.
5.
Cho, B. D., Fane, A. G., 2002. Fouling transient in nominally sub-critical flux operation of a membrane bioreactor, J. Membr. Sci. 209, 391-403.
6.
Cortés-Lorenzo, C., Rodríguez-Díaz, M., López-Lopez, C., Sánchez-Peinado, M., Rodelas, B., González-López, J. 2012. Effect of salinity on enzymatic activities in a submerged fixed bed biofilm reactor for municipal sewage. Bioresour. Technol. 121, 312-319.
7.
Dizge, N., Yuksel, D., Koseoglu-Imer, Karagunduz, A., Keskinler B., 2011. Effects of cationic polyelectrolyte on filterability and fouling reduction of submerged membrane bioreactor (MBR). J. Membr. Sci. 377, 175-181.
8.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356.
9.
Jang, D., Hwang, Y., Shin, H., Lee, W., 2013. Effects of salinity on the characteristics of biomass and membrane fouling in membrane bioreactors. Bioresour. Technol. 141, 50-56.
10.
Jang, N., Ren, X., Cho, J., Kim, I. S., 2006. Steady-state modeling of bio-fouling 19
potentials with respect to the biological kinetics in the submerged membrane bioreactor (SMBR). J. Membr. Sci. 284 , 352-360. 11. Ji, J., Qiua, J., Wai, N., Wong, F, Li, Y., 2010. Influence of organic and inorganic flocculants on physical-chemical properties of biomass and membrane-fouling rate. Water Res. 44, 1627-1635. 12.
Koseoglu, H., Yigit, N. O., Iversen, V., Drews, A., Kitis, M., Lesjean, B., Kraume, M., 2008. Effects of several different flux enhancing chemicals on filterability and fouling reduction of membrane bioreactor (MBR) mixed liquors. J. Membr. Sci. 320, 57-64.
13.
Le-Clech, P., Chen V., Fane, A. G., 2006. Fouling in membrane bioreactors used in wastewater treatment. J. Membr. Sci. 284, 17-53.
14.
Lee, W. N., Chang I.S., Hwang, B.K., Park, P.K., Lee, C. H., Huang, X., 2007. Changes in biofilm architecture with addition of membrane fouling reducer in a membrane bioreactor. Process Biochem. 42, 655-661.
15.
Lowry, O. H., Rosebrough N. J., Farr, A. L., Randall R. J. 1951. Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem. 193, 265-275
16.
Morgan, J. W., Forster, C. F., Evison L., 1990. Comparative study of the nature of biopolymers extracted from anaerobic and activated sludges. Water Res. 24, 743-750.
17.
Murthy, S. N., Novak, J. T., Dehaas R. D., 1998. Monitoring cations to predict and improve activated sludge settling and dewatering properties of industrial wastewaters, Water Sci. Technol. 38, 119-126.
18.
Mutamim, N., Noor, Z., Hassan, M., Yuniarto, A., Olsson, G., 2013. Membrane bioreactor: Applications and limitations in treating high strength industrial wastewater. Chem. Eng. J. 225, 109-119.
19.
Nguyen, T. T., Guo, W., Ngo, H. H., Vigneswaran S., 2010. A new combined inorganic-organic flocculant (CIOF) as a performance enhancer for aerated submerged membrane bioreactor. Sep. Purif. Technol. 75, 204-209.
20.
Ognier, S., Wisniewski, C., Grasmick, A., 2002. Influence of macromolecular adsorption during filtration of amembrane bioreactor mixed liquor suspension. J. 20
Membr. Sci. 209, 27-37. 21.
Ray, S., Peters, C.A., 2008. Changes in microbiological metabolism under chemical stress. Chemosphere. 71, 474-483.
22.
Reid, E., Liu X., Judd, S.J., 2006. Effect of high salinity on activated sludge characteristics and membrane permeability in an immersed membrane bioreactor. J. Membrane Sci. 283, 164-171.
23.
Remy, M., Temmink, H., Brink, P., Rulkens W., 2011. Low powdered activated carbon concentrations to improve MBR sludge filterability at high salinity and low temperature. Desalination. 276, 403-407.
24.
Rosengerg, M., Gutnick, D., Rosengerg, E., 1980. Adherence of bacteria to hydrocarbons: a simple method for measuring cellsurface hydrophobicity. FEMS Microbiol. Lett. 9, 29-33.
25.
Sobeck, D. C., Higgins M .J., 2002. Examination of three theories for mechanisms of cationinduced bioflocculation. Water Res. 36, 527-538.
26.
Sun, C., Leiknes, T., Weitzenbok, J., Thorstensen, B., 2010. Salinity effect on a biofilm-MBR process for shipboard wastewater treatment. Sep. Purif. Technol. 72, 380-387.
27.
Tansel, B., Sager, J., Garland, J., Xu, S., Levine, L., Bisbee, P., 2006. Deposition of extracellular polymeric substances (EPS) and microtopographical changes on membrane surfaces during intermittent filtration conditions. J. Membr. Sci. 285, 225-231.
28.
Uygur, A., 2006. Specific nutrient removal rates in saline wastewater treatment using sequencing batch reactor. Process Biochem. 41, 61-66.
29.
Villain, M., Marrot, B., 2013. Influence of sludge retention time at constant food to microorganisms ratio on membrane bioreactor performances under stable and unstable state conditions. Bioresour. Technol. 128 , 134-144.
30.
Vyrides, I., Stuckey, D. C., 2009. Adaptation of anaerobic biomass to saline conditions: Role of compatible solutes and extracellular polysaccharides. Enzyme Microb. Technol. 44, 46-51.
31.
Wintgens, T., Melin, T., Schafer, A., Khan S., Muston M., Bixio, D., Thoeye, C., 21
2005. The role of membrane processes in municipal wastewater reclamation and reuse. Desalination. 178, 1-11. 32.
Yang, W., Cicek N., Ilg, J., 2006. State-of-the-art of membrane bioreactors: worldwide research and commercial applications in Aorth America. J. Membr. Sci. 270, 201-211
33.
Yogalakshmi, K. N., Joseph, K., 2010. Effect of transient sodium chloride shock loads on the performance of submerged membrane bioreactor. Bioresour. Technol. 101, 7054-7061.
34.
Zhang, H. F., Sun, B. S., Zhao, X. H., Gao, Z. H., 2008. Effect of ferric chloride on fouling in membrane bioreactor. Sep. Purif. Technol. 63, 341-347.
35.
Zhao, Y., Gu, P., 2006. Effect of powdered activated carbon dosage on retarding membrane fouling in MBR. Sep. Purif. Technol. 52, 154-160.
22
Figure captions:
Fig. 1. Schematic of MBR-A and MBR-B
Fig. 2. Comparison between the control and the MBR with coagulant in removing conventional pollutants
Fig. 3. Profile of TMP rise up in the MBR-A and MBR-B
Fig. 4. Variations in the concentration of protein and carbohydrate under different operating conditions for SMP
Fig. 5. Variations in the concentration of protein and carbohydrate under different operating conditions for EPS
23
Pressure gauge
Peristaltic pump
Agitator
Membrane Feeding pump
Rotameter
Level sensor
Air pu mp
MBR-A
Timer Pressure gauge
Peristaltic pu mp
Level controller MPE50
Wastewater Tank Membrane
Feeding pump
Level sensor
Rotameter Air pu mp
MBR-B
Fig. 1. Schematic of MBR-A and MBR-B.
24
100
80
70
MBR-A (COD) MBR-B (COD) + MBR-A(NH4 -N)
+
COD or NH4 -N removal (%)
90
60
+
MBR-B((NH4 -N)
50
40
(a)
50
(b)
MBR-A MBR-B
Phosphate removal(%)
40
30
20
10
0 0
10
20
30
40
50
60
70
Time(days)
Fig. 2. Comparison between the control and the MBR with coagulant in removing conventional pollutants
25
Normal period
Senstive period Recorvery period
25
MBR-A MBR-B TMP(kPa)
20
15
10
5 15
20
25
30
35
40
45
50
55
Time(days)
Fig.3. Profile of TMP rise up in the MBR-A and MBR-B
26
6.0
MBR-A
5.5
MBR-B SMPp SMPc
5.0 4.5
SMP/(mg/gSS)
4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Normal
Sensitive
Recovery
Normal
Sensitive
Recovery
Period
Fig. 4. Variations in the concentration of protein and carbohydrate under different operating conditions for SMP
27
60
MBR-A
55
MBR-B
EPSp EPSc
50 45
EPS(mg/mgSS)
40 35 30 25 20 15 10 5 0 Normal
Sensitive
Recovery
Normal
Sensitive
Recovery
Period
Fig. 5. Variations in the concentration of protein and carbohydrate under different operating conditions for EPS.
28
List of tables Table 1 Conventional pollutant removal efficiencies in both MBRs in different periods
Table 2 Respirometry activities in both reactors between before and after salt shock.
Table 3 Membrane fouling rate (kPa/d) in different periods for both MBRs.
Table 4 Linear correlation between MFI and the SMP content.
Table 5 Parameters relevant to the characteristics of the flocs for both MBRs
29
Table 1 Conventional pollutant removal efficiencies in both MBRs in different periods Parameter
COD
NH4+-N
TP
Removal efficiency (%)
Period
MBR-A
MBR-B
1-30 d
96.2±0.8
96.6±0.9
31-33 d
89.6±4.1
93.7±2.1
34-70 d
96.4±0.9
96.5±1.0
1-30 d
91.7±1.1
91.6±1.1
31-40 d
71.7±18.3
84.9±11.6
41-70 d
91.6±0.8
91.5±0.9
1-30 d
33.5±1.4
34.2±1.5
31-41 d
9.1±3.2
30.3±3.4
42-70 d
33.8±1.4
36.1±3.3
30
Table 2 Respirometry activities in both reactors between before and after salt shock Respirometry activities (mg O2 g VSS-1 h-1) Heterotrophic Autotrophic
Before salt shock
After salt shock
MBR-A
MBR-B
MBR-A
MBR-B
Exogenous
19.4 ± 4.7
20.7± 5.3
17.2 ± 2.4
17.9 ± 2.6
Endogenous Exogenous
1.5 ± 0.3
1.7 ± 0.4
2.1 ± 0.4
1.8 ± 0.2
4.7 ± 0.5
4.2± 0.3
0.7 ± 0.2
1.5 ± 0.4
Endogenous
1.3 ± 0.7
1.4 ± 0.5
2.9 ± 0.4
2.3 ± 0.5
31
Table 3 Membrane fouling rate (kPa/d) in different periods of both MBRs Reactors
Normal period
Sensitive period
Recovery period
MBR-A
1.26
12.17
3.82
MBR-B
1.31
2.34
2.31
32
Table 4 Linear correlation between MFI and the SMP content Period Normal
Sensitive
Recovery
SMP SMPp
MBR-A
MBR-B
Fitting equation
R2
Fitting equation
R2
y=0.0864x+0.276
0.39
y=0.0871x+0.2821
0.29
SMPc y=0.2523x+0.9874
0.84
y=0.5247x+0.2809
0.86
SMPp y=0.0384x+4.6748
0.97
y=0.0731x+0.4895
0.58
SMPc
0.92
0.53
y=0.1512x+0.7534
SMPp y=0.3512x+0.1969
y=0.018x+2.9306
0.94
y=0.1235x+0.2599
0.39
SMPc y=0.1756x+1.6374
0.89
y=0.5927x-0.5441
0.83
33
Table 5 Parameters relevant to the characteristics of the flocs for both MBRs. MBR-A
MBR-B
Period PSD(μ m)
Zeta (mV)
RH (%)
PSD(μ m)
Zeta (mV)
RH (%)
Nomal
78.5
-14.4
27.4
75.3
-14.6
29.7
Sensitive
70.3
-11.8
18.6
89.6
-2.3
52.5
Revovery
89.4
-15.3
32.5
89.3
-4.4
48.7
34
Supplementary Figure
Fig. S1. Correlation between MFI and membrane fouling rate for the both MBRs
35
18
Fouling rate(kPa/d)
16 14 y = 0.7216x - 1.041 R 2 = 0.9757
12 10 8 6 4 2 0 0
2
4
6
8
10
12
14 5
16
18
20
22
24
26
2
MFI(×10 s/L )
Fig. S1. Correlation between MFI and membrane fouling rate for the both MBRs.
36
Highlights Impact of salt shock on MBR performance was assessed. Autotrophs showed more sensitive than heterotrophs at salt shock. Higher SMP protein release was observed under high salt condition. Changes in sludge properties by MPE50 addition significantly reduced membrane fouling.
37
