Control of ultrafiltration membrane fouling caused by Microcystis cells with permanganate preoxidation: Significance of in situ formed manganese dioxide

Control of ultrafiltration membrane fouling caused by Microcystis cells with permanganate preoxidation: Significance of in situ formed manganese dioxide

Chemical Engineering Journal 279 (2015) 56–65 Contents lists available at ScienceDirect Chemical Engineering Journal journal homepage: www.elsevier...

2MB Sizes 0 Downloads 17 Views

Chemical Engineering Journal 279 (2015) 56–65

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Control of ultrafiltration membrane fouling caused by Microcystis cells with permanganate preoxidation: Significance of in situ formed manganese dioxide Fangshu Qu, Xing Du, Bin Liu, Junguo He, Nanqi Ren, Guibai Li, Heng Liang ⇑ State Key Laboratory of Urban Water Resource and Environment (SKLUWRE), Harbin Institute of Technology, Harbin 150090, PR China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Algal cell-related UF membrane

fouling control by KMnO4 pretreatment was studied.  Contributions of KMnO4 oxidation and in situ formed MnO2 adhesion were compared.  KMnO4 oxidation exhibited a minor effect on the characteristics of algal cells.  MnO2 adhesion promoted the aggregation of algal cells, improving their settleability.  MnO2 particles showed a superior ability in reducing cell-related fouling than KMnO4.

a r t i c l e

i n f o

Article history: Received 28 November 2014 Received in revised form 29 April 2015 Accepted 2 May 2015 Available online 8 May 2015 Keywords: Ultrafiltration Membrane fouling Microcystis Manganese dioxide Permanganate peroxidation

a b s t r a c t Control of the ultrafiltration (UF) membrane fouling caused by Microcystis cells using permanganate preoxidation was investigated with lab-cultured Microcystis aeruginosa. The contributions of two widely considered mechanisms, i.e. potassium permanganate (KMnO4) oxidation and in situ formed manganese dioxide (MnO2) adhesion, to the Microcystis cell fouling control were compared and discussed. Initially, effects of KMnO4 oxidation and in situ formed MnO2 adhesion on the characteristics of Microcystis cells, including viability, zeta potential, size distribution and settleability, were compared. Subsequently, filtration tests were undertaken to investigate the flux decline and fouling reversibility during UF of the untreated and treated cell solutions. The results indicated that KMnO4 oxidation exhibited a minor effect on the characteristics of Microcystis cells as well as the cell integrity under the tested KMnO4 exposure (1.0 and 2.0 mg/L). The adhesion of in situ formed MnO2 particles could promote the aggregation of the cells, apparently improving their settleability. With regards to membrane fouling, the in situ formed MnO2 particles displayed a superior capacity in alleviating both the flux decline and the irreversible fouling associated with Microcystis cells than KMnO4 oxidation did, owing to the reinforced aggregation of the cells and the adsorption of released extracellular organic matter (EOM). Moreover, the effect of membrane precoating with in situ formed MnO2 particles on the cell fouling was also studied. The precoating layer seldom reduced the flux decline by Microcystis cells, but improved the fouling reversibility probably by inhibiting the direct membrane-cell contact and retaining the released EOM. Ó 2015 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Tel./fax: +86 451 86283001. E-mail addresses: [email protected] (F. Qu), [email protected] (X. Du), [email protected] (B. Liu), [email protected] (J. He), [email protected] (N. Ren), [email protected] (G. Li), [email protected] (H. Liang). http://dx.doi.org/10.1016/j.cej.2015.05.009 1385-8947/Ó 2015 Elsevier B.V. All rights reserved.

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

1. Introduction How to guarantee the bio-safety of drinking water has become a worldwide concern since the discovery of the protozoa Giardia and Cryptosporidium, which have been the principal organisms controlling disinfection regulations [1]. Despite the removal efficiencies of the cysts of these protozoa up to 1.5–4.5 logs by traditional media filtration processes, the removal is not absolute [2]. In particular, Cryptosporidium is especially resistant to the commonly used disinfectant like chlorine. The deficiency of traditional water treatment processes in the bio-safety has promoted the application of low-pressure membrane filtration (i.e. ultrafiltration (UF) and microfiltration (MF)) in drinking water industry. The UF membrane, which has a pore size as small as 0.01 lm, can retain particles, colloids, protozoa (3–15 lm), bacteria (0.5–10 lm) as well as virus (0.02–0.08 lm) [2,3]. Till now, UF technology has been adopted in many waterworks; however, membrane fouling, which would increase energy consumption and shorten membrane life-span, hinders its further application [4,5]. As surface water is increasingly polluted by domestic sewage, Microcystis bloom frequently occurs in polluted reservoirs and lakes, bringing problems such as odors and toxins [6–9]. UF, which can completely retain Microcystis cells by size exclusion, becomes a promising technology to remove Microcystis cells from polluted water. Nevertheless, both Microcystis cells and their extracellular organic matter (EOM) proved to cause severe flux decline as well as irreversible fouling [10,11]. With regards to EOM, it was generally characterized and reported to have distinct features like high molecular weight and strong hydrophilicity [12]. Moreover, EOM fouling was also systematically investigated and was found to be closely related to both the organic content and the characteristics of EOM. As algal cell size is obviously larger than UF membrane pore size, Microcystis cells can only accumulated on the membrane surface, forming a cake layer [13]. Wicaksana et al. [14] studied the microfiltration membrane fouling by Chlorella cells using direct observation through the membrane technology (DOTM), and found that cell deposition started to occur even at a very low permeate flux. Babel et al. [15] reported that the cell layer formed by algal cell precipitation was compressible and that the fouling resistance would strongly aggravate when the cell layer was compressed. Qu et al. [11] compared the fouling potentials of Microcystis cells and their EOM, and concluded that the flux decline associated with the cells was even worse than that by EOM. Because cell fouling was usually considered to be less complicated than EOM fouling, the attention paid to the cell fouling was much less, leading to the lack of systematical understanding on the cell fouling mechanism and thus appropriate fouling control strategy. To control the membrane fouling by algae, pretreatments with ozone and chlorine are widely adopted. Hung and Liu [16] reported that ozone preoxidation improved the performance of MF during the separation of green algae by reducing cake compressibility and biomass loading. However, the oxidation by strong oxidants often results in the release of intracellular organic matter as well as toxins [17,18]. Thus, a weaker oxidant, potassium permanganate (KMnO4), is alternatively chosen for surface water treatment. Lin et al. [19] had demonstrated that permanganate preoxidation played a significant role in reducing the fouling during UF of surface water. There are two generally considered mechanisms for the algal fouling control by permanganate preoxidation. Liang et al. [20] put forwards that the weak oxidation provided by permanganate could inactivate the cells and make them easier separated. The other mechanism is associated with in situ formed manganese dioxide (MnO2) particles which are found to coat the cell surface and to change the characteristics of the cells, i.e. increasing the zeta potential and gravity [21]. Among these two

57

mechanisms, which makes a greater contribution is still under debate. The potential of permanganate pre-oxidation for Microcystis fouling control raises two critical questions: (1) can permanganate pre-oxidation lead to cell lysis? If so, to what extent? (2) which mechanism can be reinforced to maximize the fouling control performance. The objective of this study was to verify the major mechanism of Microcystis cell fouling control by permanganate preoxidation, with KMnO4 oxidation and MnO2 adhesion compared. The effects of these two mechanisms on the characteristics of Microcystis cells, such as viability, zeta potential, size distribution and settleability, were investigated. Moreover, the flux decline and the fouling reversibility during UF of untreated and treated cell solutions were also compared and discussed. 2. Methods and materials 2.1. Algae culture and cell solution preparation Microcystis aeruginosa was purchased from the Culture Collection of Algae at the Institute of Hydrobiology, Chinese Academy of Sciences, China. The detailed instruction for algae cultivation had been presented by Qu et al. [22]. Axenic cultures were conducted with BG11 medium in 1 L conical flasks. The conical flasks were placed in an incubator at 25 °C with an illumination of 5000 lx provided for 14 h every day. Cultures were harvested at the stationary phase with the culture time of 42 d. To reduce the interference of EOM in fouling study, the cells were extracted from the culture solution and resuspended by the simulated surface water which was Milli-Q water spiked with 0.5 mM CaCl2, 1.0 mM NaHCO3, and 15.0 mM NaClO4 [23]. The chemicals were added to simulate the ionic strength of surface water and the pH was adjusted to 7.5 ± 0.1 with NaOH (0.1 M) and HCl (0.1 M) solutions. Specifically, the harvested algae solution was centrifuged at 4000g and 4 °C for 15 min, using a high speed refrigerated centrifuge (H2050R, Xiangyi, China). Subsequently, the supernatant was filtered through a 0.45 lm filter. The Microcystis cells remained in the centrifuge tube and those on the filter were collected and resuspended. Finally, the cell solution was diluted to the concentration of 1.0  106 cells/mL with the simulated surface water. The cell concentration was described by the optical density at the wavelength of 685 nm as demonstrated by Figs. S1 and S2 in the Supplementary information. 2.2. Membrane and experimental setup Flat polyethersulfone UF membranes (OM100076, Pall, USA) were used in current study. The molecular weight cutoff and surface area of the membrane were 100 kDa and 4.5  10 3 m2, respectively. A schematic diagram of UF system was illustrated by Qu et al. [22]. The system consisted of a stirred cell (Amicon 8400, Millipore, USA), a nitrogen gas cylinder, an electronic balance, a computer, gauges and pipes. UF experiments were performed in a dead-end filtration mode. The membrane was placed in the bottom of the stirred cell with its glossy side toward bulk solution. Nitrogen gas, at a constant pressure of 0.03 MPa, was utilized to drive feed solution across the membrane. Filtrate flowed into a beaker on an electronic balance connected to a personal computer and weighing data were automatically logged every five seconds. When the membrane was fouled, backwashing was conducted with Milli-Q water. The membrane was turned over and then the Milli-Q water (100 mL) was driven also by nitrogen gas through the membrane. 2.3. Experimental protocol This study was to verify the contributions of the two mechanisms, permanganate oxidation and MnO2 adhesion, to the control

58

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

of Microcystis cell fouling. KMnO4 stock solution was prepared by dissolving crystal KMnO4 in Milli-Q water (1.0 g/L) and filtering the solution through a 0.2 lm glass fiber filter. Then it was standardized by titration with sodium oxalate and diluted to required concentrations. To study the oxidation-related mechanism, KMnO4 was added into the cell solution (1.0  106 cells/mL) at two doses of 1.0 and 2.0 mg/L. The KMnO4 concentration could not exceed 2.0 mg/L, since excess KMnO4 would impart a purple color to treated water [24]. More details about the KMnO4 concentration selection were shown in Table S1 in the Supplementary information. The solution was rapidly mixed with a magnetic stirrer at 300 rpm for 2 min, following a slow mixing (100 rpm) of 28 min. Then samples were collected for characterization and filtration tests. With regard to the MnO2-involved mechanism, MnO2 particles were generated by the redox reaction between KMnO4 and stoichiometric Na2S2O3 [25]. The concentrations of in situ formed MnO2 particles were controlled at 0.55 and 1.1 mg/L, corresponding to the ultimate amounts of MnO2 in the reduction processes of 1.0 and 2.0 mg/L KMnO4, respectively [26]. When the reagents were simultaneously added into the cell solution, the solution was rapidly mixed at 300 rpm for 2 min, to ensure the formation of the MnO2 particles and the contact between these particles and the cells, followed by a slow mixing (100 rpm) of 30 min. After that, samples were collected for further characterization and filtration tests. Note that permanganate oxidation and MnO2 generation would simultaneously occur in the same process. It was determined in this experiment that the residual KMnO4 took a percentage of 70.3% of the initial amount (2.0 mg/L) after being mixed with the cell solution for 30 min (as shown in Fig. S3). Thus, in the scenario without adding Na2S2O3, both permanganate oxidation and in situ formed MnO2 adhesion would take effect in fouling control. In respect to the scenario with Na2S2O3, KMnO4 would preferentially react with Na2S2O3, making the in situ formed MnO2 adhesion the only reason responsible for fouling control. So contributions of the in situ formed MnO2 adhesion could still be verified despite that these two scenarios coincided to some extent. To make the description clearer, the former scenario was still referred to permanganate oxidation. To evaluate the effects of KMnO4 and MnO2 on the Microcystis cell fouling, a series of 3-cycle filtration tests were performed as instructed by Qu et al. [22]. The flux decline resulted from the deposit of the cells was investigated, and the fouling reversibility was also analyzed. The pH values of tested solutions were adjusted to 7.5 ± 0.1 beforehand. Moreover, fouling control by precoating the in situ formed MnO2 on membrane surface was also investigated. The MnO2 particles were formed in 300 mL Milli-Q water at a certain concentration (1.10 mg/L), and the solution was spontaneously filtered through a UF membrane under a constant pressure of 0.10 MPa to make the particles rapidly precipitate on the membrane surface. Then, the precoated membrane was used to filter the untreated Microcystis cell solution. In each filtration, the membrane was re-coated with the in situ formed MnO2, since the coating layer might be washed away during backwashing. The UF membrane fouling was evaluated in this study by the initial clean water flux, the reversible fouling and the irreversible fouling, representing the intrinsic membrane resistance, the cake layer resistance and the internal fouling, respectively. The procedure of filtration test and the calculation of different types of membrane fouling were illustrated by Qu et al. [22]. The filtration tests were all performed in triplicate. 2.4. Characterization of MnO2 particles, Microcystis cells and fouled membranes To evaluate the characteristics of MnO2 particles, certain amounts of KMnO4 and Na2S2O3 solutions were added into

Milli-Q water to generate the MnO2 particles. Beforehand, the weakly basic environment was maintained with the phosphate buffer solution. The particle size distribution and zeta potentials of the MnO2 particles were evaluated by a Zetasizer (Nano ZS90, Malvern, UK) with a He–Ne laser at a wavelength of 633 nm [27]. In order to study the surface functional groups of MnO2 particles, the prepared MnO2 suspension was filtered with a 0.2 lm glass fiber filter, and the sample was dried overnight. Then, the MnO2 particles were carefully collected and grinded. The Fourier transform infrared spectrum (FTIR) of the MnO2 particles was obtained with KBr pellets on a Perkin-Elmer spectrometer (Spotlight 400, USA). The measured wavelength wave number was between 4000 and 400 cm 1. The measured characteristics of Microcystis cells included cell concentration, zeta potential, size distribution and settleability. The cell concentration was depicted by UV absorbance at the wavelength of 685 nm obtained on a UV spectrometer (T6, Puxi, China) [16]. The zeta potential was determined by a zetasizer (Nano S90, Malvern, Britain) at the pH level of 7.5 ± 0.1. The size distributions of Microcystis cells were investigated by means of a laser particle analyzer (Mastersizer 2000, Malvern, Britain), with which a valid measure range of 0.02–2000 lm could be accessed. The settleability of Microcystis cells was evaluated by the settling test according to Chen and Yeh [21]. The cell samples were collected and measured in triplicate. Microcystis cell integrity was investigated by a flow cytometer (Accuri C6, BD Biosciences, Franklin Lakes, USA) equipped with an argon laser emitting at a fixed wavelength of 488 nm for fluorescence measurement. Fluorescent filters and detectors were all standard with the green and red fluorescence collected in channel FL1 (530 nm) and channel FL3 (630 nm), respectively. SYTOX green nucleic acid stain (Invitrogen, Life Technologies, Grand Island, USA) was used to determine the cell integrity according to Fan et al [28]. The SYTOX could permeate Microcystis cells which have lost cell wall integrity and stain the nucleic acid; hence the cells were classified into the viable type (SYTOX negative) and nonviable (SYTOX positive) type [23]. The SYTOX was added into samples at a concentration of 0.2 lM. After a development incubation time of 10 min, samples were injected into the flow cytometer, and CFlow software (version 264.15) was employed to collect and analyze the data. To further understand the impact of MnO2 adhesion on the Microcystis cell fouling, scanning electron microscopy (SEM)/energy-dispersive X-ray spectroscopy (EDS) was employed to analyze the morphologies of the fouled and hydraulically cleaned membranes. During membrane sample preparation, 1.0  106 cells/mL cell solution and 1.1 mg/L MnO2 were adopted. Before observation, the membrane samples were vacuum-dried and sputter-coated with approximately Au/Pd (20 Å). The elementary EDS analyses (Inca 300, United Kingdom) were conducted immediately after the SEM images were collected. The membrane surface microstructure was visualized using a field emission scanning electron microscope (LEO 1530 VP, Germany).

3. Results and discussion 3.1. Characteristics of MnO2 particles The characteristics of MnO2, including zeta potentials, particles size and surface functional groups, govern the interaction between the MnO2 particles and Microcystis cells. The zeta potential of the MnO2 particles under the weakly basic environment was measured as 24.5 ± 1.4 mV (data not shown), indicating a negatively charged surface. Fig. 1(a) shows the particle size distribution of the MnO2 particles (0.1 mmol/L) in the aqueous solution. It can

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

(a)

25

Number (%)

20

15

10

5

0 1

10

100

1000

10000

Particle size (nm) 50

(b) 3386

Adsorbance (%)

576

Mn-O

45

O-H 40 35 1621

30

1046

3000

2500

2000

1500

1000

account for 94.0% of the total cell count, while that the damaged cells take a percentage of 5.6%. Since parts of cells would naturally die owing to metabolism, it is reasonable to find a small part of damaged cells in the raw cell solution. When the cells were treated by KMnO4 at the doses of 1.0 mg/L and 2.0 mg/L for 20 min, proportions of intact cells were 93.3% and 94.9%, respectively. In addition, the variation of cell integrity during a 20-min KMnO4 preoxidation process under the doses of 1.0 and 2.0 mg/L was also investigated, and the results are shown in Fig. 2(d). It can be observed that no obvious decrease in the proportion of intact cells is found during the preoxidation process under the KMnO4 dose of 1.0 mg/L. When the KMnO4 dose increases to 2.0 mg/L, the proportion of intact cells even slightly increases. The results indicate that KMnO4 did not cause a visible cell integrity damage, probably due to its weak oxidation capacity under the neutral condition. Xie et al. [23] compared the damage of Microcystis cells resulted from three oxidants, and found that KMnO4 (0–2.0 mg/L) exhibited much less influence to cell integrity in comparison with ozone and chlorine. Fan et al. [28] reported that more than 98% of Microcystis cells would maintain the cell integrity when treated by less than 3 mg/L KMnO4, and that cell lysis only occurred under high-concentration (>5 mg/L) KMnO4 exposure. However, for a similar oxidant potassium ferrate, the visible cell damage was demonstrated under the oxidant dose of 1.0– 7.0 mg/L [33]. Therefore, in order to remove the cells without causing cell breakage, the concentration of KMnO4 should be keep under the appropriate level. 3.3. Effects of KMnO4 oxidation and MnO2 adhesion on zeta potentials of Microcystis cells and EOM release

25 3500

59

500

-1

Wave number (cm ) Fig. 1. Particle size distribution (a) and FTIR spectrum (b) of in situ formed MnO2 particles.

be observed that the size of the MnO2 particles ranges between 20 and 100 nm, with an average size of 52.79 nm. Huangfu et al. [27] also investigated the size distribution of MnO2 colloids, and reported that the particles distributed around 55.86 nm within a very narrow range. The result implies that the newly generated MnO2 particles are nano-scale in size. Fig. 1(b) shows the FTIR spectrum of the MnO2 particles. It can be observed that the FTIR spectrum is characterized by four distinct peaks around 3386, 1621, 1046 and 576 cm 1. The strong absorption bands at 1046 and 576 cm 1 are attributed to Mn–O bond [29]. The strong absorption bands at 3386 and 1621 cm 1 indicate the occurrence of hydroxyl groups on the surface of the MnO2 particles [30]. The hydroxyl groups, which endow the MnO2 particles with strong hydrophilicity, may interact with organics as well as Microcystis cells by forming hydrogen bonds [30]. 3.2. Effects of KMnO4 oxidation on the integrity of Microcystis cells As the removal of intact Microcystis cells without causing cell lysis significantly avoids the addition of dissolved toxins and intracellular organics into the treated water [31], it is significant to evaluate the influence of KMnO4 oxidation on the cell integrity. Fig. 2 shows Flow cytometry results of the raw and KMnO4-treated cells. In Fig. 2(a)–(c), FL3-A corresponds to chlorophy II autofluorescence detected at 630 nm, and FL1-A corresponds to SYTOX green stain fluorescence detected at 530 nm. P1 and P11 are regions used to define lysed and live populations, respectively [28,32]. It can be observed in Fig. 2(a) that the intact cells in the original cell solution

Table 1 presents the zeta potentials of the untreated and treated Microcystis cells. It can be found that the surface of Microcystis cells is also negatively charged under the weakly basic condition with the zeta potential of untreated cells of 33.2 ± 3.6 mV. When pretreated by KMnO4, the cells do not exhibit apparent changes in the zeta potential. However, the pretreatment with MnO2 at the concentrations of 0.55 and 1.1 mg/L make the zeta potentials of Microcystis cells increase to 29.3 ± 3.7 and 25.2 ± 3.9 mV, respectively. Since the MnO2 particles are less negatively charged ( 24.5 ± 1.4 mV), the adhesion of MnO2 would slightly reduce the surface charge of Microcystis cells. As EOM cannot be totally separated by the centrifugation method, there may exist some residual EOM in the prepared cell solution. Thus, the residual EOM was determined and the result is shown Table 1. It can be found that the average concentration of the residual EOM in the controlled trial is 0.45 mg/L. KMnO4 preoxidation slightly increases the EOM concentrations to 0.78 and 0.89 mg/L under the doses of 1.0 and 2.0 mg/L, respectively. This implies that KMnO4 preoxidation probably stimulates the release of bound EOM [34], since the Flow cytometry results demonstrated the integrity of cells under the tested KMnO4 exposure. In presence of the MnO2 particles, the concentration of EOM is lowered in comparison with that in the raw cell solution. As the MnO2 particles are of strong adsorption capacity, parts of the released EOM may be adsorbed by the MnO2 particles. Huangfu et al. [27] also reported the observation of organic layer adsorbed on the MnO2 particles when studying the aggregation of MnO2 particles. 3.4. Effects of KMnO4 oxidation and MnO2 adhesion on the settleability and size distribution of Microcystis cells Fig. 3 presents the particle-size distributions of the untreated and treated cells. It can be observed in Fig. 3(a) that the size of Microcystis cells mainly distribute in the range of 2–7 lm, which well agrees with the single cell size reported by Kwon et al. [13].

60

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

Fig. 2. Flow cytometry results of Microcystis cells treated by KMnO4: (a) control, (b) 1.0 mg/L, 20 min, (c) 2.0 mg/L, 20 min and (d) variation of intact cells over time.

Table 1 Effects of KMnO4 and MnO2 on the surface charge of Microcystis cells and the release of EOM (n = 3). Pretreatments Untreated KMnO4 (1.00 mg/L) KMnO4 (2.00 mg/L) MnO2 (0.55 mg/L) MnO2 (1.10 mg/L)

Zeta potential (mV) 33.2 ± 3.6 33.4 ± 3.4 32.7 ± 2.6 29.3 ± 3.7 25.2 ± 3.9

EOM (mg/L) 0.41 ± 0.03 0.78 ± 0.05 0.85 ± 0.04 0.33 ± 0.02 0.28 ± 0.03

Meanwhile, there also exists some cell aggregates with the size of 10–300 lm, probably formed during the growth process. The oxidation at the KMnO4 dose of 1.0 mg/L does not change the size distribution of the cells, and the size distribution of the treated cells almost coincides with that of the raw cells. As the KMnO4 dose increases to 2.0 mg/L, the ratio of single cells increases accompanied by a decrease in the amount of cell aggregates. As shown in Table 1, KMnO4 can stimulate the release of the bound EOM from the cells [21,35], and the EOM proved to inhibit the aggregation of Microcystis cells [36]. Therefore, the amount of single cells increases in the presence of KMnO4. It can be also found in Fig. 3(b) that MnO2 pretreatment apparently reduces the ratio of single cells and increases the amount of cell aggregates. A higher MnO2 concentration leads to a more distinct reduction in the single cells. Since the in situ formed MnO2 particles are nano-sized, they are prone to adhering onto the cell surface. As shown in Table 1, the adhesion of MnO2 particles reduces the negative zeta potential of cell surface and thus lowers the electrostatic repulsion among the cells, promoting the formation of cell aggregates. Fig. 4 shows the effects of KMnO4 and MnO2 pretreatments on the residual UV685 values of Microcystis cell solution during a 2-h sedimentation process. It can be observed that the controlled cell

solution does not display an obvious decrease in cell concentration with the UV685 value varying between 0.145 and 0.149 cm 1 and that KMnO4 oxidation can only slightly improve the settleability of Microcystis cells. However, the cell removal performance is dramatically enhanced by MnO2 pretreatment (0.55 mg/L) with the residual UV685 value reduced to 0.064 cm 1 at the sedimentation time of 60 min. When the MnO2 concentration increases to 1.10 mg/L, the UV685 value is lowered to 0.066 cm 1 at the sedimentation time of 30 min. The final UV685 value is as low as 0.028 cm 1. The results indicate that MnO2 particles make a much greater contribution than individual KMnO4 in improving the settleability of Microcystis cells. Chen and Yeh [21] performed a similar study and found that KMnO4 oxidation could enhance the cell settleability by means of MnO2 which would coat the cell surface and increase its gravity. In their study, the removal rate of Microcystis cells was improved from 5% to 8% at the KMnO4 dose of 3 mg/L and sedimentation time of 3 h, much lower than approximate 80% achieved by the in situ formed MnO2 particles in current study. This may be associated with the incomplete conversion of KMnO4 to MnO2 due to its weak oxidation ability [21]. Overall, MnO2 particles play a more significant role in improving the cell settleability than KMnO4 did. 3.5. Effects of MnO2 adhesion and KMnO4 oxidation on the flux decline by Microcystis cells To investigate the effects of MnO2 adhesion and KMnO4 oxidation on the flux decline by Microcystis cells, filtration tests of the treated cell solutions were performed in parallel with a controlled trial. It can be observed in Fig. 5(a) that the accumulation of untreated cells on the membrane leads to severe flux decline with the specific flux at the end of filtration as low as 0.108. In the

61

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

(a)

14

Untreated 1.0 mg/L KMnO4

1.0

MnO2 1.1 mg/L

MnO2 precoating1.1mg/L

0.8

8

0.6

6

0.4

4

0.2

0 0.1 12

1

10 100 Particle size (μm)

0.0

1000

0

(b)

(b)

10 8

Untreated 0.55 mg/L MnO2

1.0

1.1 mg/L MnO2

0.8

100 200 300 400 500 600 700 800 900 Filtration volume (mL) Untreated KMnO4 1.0 mg/L KMnO4 2.0 mg/L

0.6

6

J/J0

Volumn ratio (%)

MnO2 0.55 mg/L

2.0 mg/L KMnO4

10

2

0.4

4

0.2

2

0.0

0 0.1

1

10 100 Particle size (μm)

1000

Fig. 3. Effects of KMnO4 (a) and MnO2 (b) on the size distribution of Microcystis cells.

0.16

Control KMnO4(1.00mg/L)

0.12

KMnO4(2.00mg/L)

-1

UV685(cm )

Untreated

J/J0

Volumn ratio (%)

12

(a)

0.08

MnO2(0.55mg/L) MnO2(1.10mg/L)

0.04

0.00 0

20

40

60

80

100

120

Time (min) Fig. 4. Effects of KMnO4 and MnO2 on the settleability of Microcystis cells.

presence of 0.55 and 1.10 mg/L MnO2, the final specific fluxes are improved to 0.251 and 0.324, respectively. The results indicate that the adhesion of the in situ formed MnO2 particles on the cell surface dramatically reduces the flux decline associated with the cell precipitation. In another similar study, Liang et al. [20] also reported that the in situ formed MnO2 particles, the reduction product of KMnO4, was effective in controlling the membrane fouling

0

200

400 600 Filtration volume (mL)

800

Fig. 5. Effects of MnO2 adhesion (a) and KMnO4 oxidation (b) on the flux decline caused by Microcystis cells.

related to algae. There are two probably interpretations for this phenomena. On one hand, the in situ formed MnO2 particles, which proved to have a very large specific surface area and strong activity, may adhere to the cell surface, making the cells aggregate (as seen in Fig. 3(b)). As reported by Yu et al. [37], bigger particles could form a less homogeneous and more porous cake layer on the membrane surface, leading to less severe fouling. When these cell aggregates precipitates on the membrane surface during UF, a more porous cake layer is formed as illustrated in Fig. 6. On the other hand, the adhesion of the in situ formed MnO2 particles to Microcystis cells changes the characteristics of cell surface, i.e. to improve cell rigidity and reduce cell compressibility [20]. Thus, the presence of in situ formed MnO2 particles results in a more porous and less compressible cake layer during UF of Microcystis cell solution. Meanwhile, cake layer fouling was demonstrated to account for a great part of the flux decline during UF of algae-rich water [11]. Therefore, MnO2 adhesion can make a distinct contribution to alleviating the flux decline associated with Microcystis cells. To further verify the role of MnO2 in controlling the membrane fouling caused by Microcystis cells, the filtration performance of the membrane coated by MnO2 particles (1.10 mg/L) was investigated. As illustrated in Fig. 5(a), the precoated MnO2 layer slightly reduces the flux decline caused by Microcystis cells, but the effect is very limited with the final specific flux of 0.113, similar to that in the controlled trial. This result implies that the precoated MnO2 layer plays a minor role in reducing flux decline caused by

62

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

(a)

1

Reversible fouling

2

δMnO2 dose: 0.00mg/L

3

1

2 3 0.55mg/L

1

Irreversible fouling

1 2 2 3 3 1.10mg/L 1.10mg/L(Precoating)

Filtration cycle

Relative fouling

the cells. It is generally recognized that the permeability of cake layer depends on its porosity and compressibility [14,38]. Since the cells may precipitate on the top of precoated MnO2 layer, and form a cake layer of which neither porosity nor compressibility is changed, it is rational to find a very limited improvement in the membrane flux by precoating. In case of KMnO4 oxidation, the flux decline resulted from cell precipitation is only slightly alleviated as shown in Fig. 5(b), with the final specific fluxes of 0.115 and 0.141 at the KMnO4 doses of 1.00 and 2.00 mg/L, respectively. The results suggest that KMnO4 oxidation plays a very limited role in improving the flux decline associated with Microcystis cells. Since KMnO4 oxidation would simultaneously generate MnO2 particles, which are considered to be positive in the fouling control, several previous studies concluded that KMnO4 oxidation could effectively reduce the membrane fouling during surface water filtration [19,20]. In this study, the MnO2 particles also prove to alleviate the flux decline caused by Microcystis cells. However, as reported by Chen and Yeh [21], KMnO4 oxidation could also induce the release of bound EOM, which was of strong membrane fouling potential. Therefore, when KMnO4 preoxidation is employed to control the Microcystis cell fouling, the contribution to fouling control made by the in situ formed MnO2 particles may be compensated by the existence of EOM (as shown in Table 1). Moreover, it is demonstrated by the cell size distribution (Fig. 3(a)) that KMnO4 oxidation inhibits the aggregation of Microcystis cells, which may result in the formation of a dense cake layer (as shown in Fig. 6) during UF, leading to severe flux decline. Generally, KMnO4 oxidation exhibits a minor effect in the control of flux decline associated with Microcystis cells.

Relative fouling

Fig. 6. Schematic diagram of Microcystis cell fouling control by KMnO4 oxidation and MnO2 adhesion.

0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

(b)

1

Reversible fouling

2

KMnO4 dose: 0.00mg/L

3

2 3 1.00mg/L Filtration cycle 1

Irreversible fouling

1

2 3 2.00mg/L

3.6. Effects of KMnO4 oxidation and MnO2 adhesion on the reversibility of Microcystis cell fouling

Fig. 7. Effects of MnO2 pretreatment (a) and KMnO4 preoxidation (b) on the reversibility of fouling by Microcystis cells.

To further understand the effects of KMnO4 oxidation and MnO2 adhesion on the fouling behavior of Microcystis cells, the reversible and irreversible fouling caused by the untreated and treated cells were investigated and the results are shown in Fig. 7. It can be observed in Fig. 7(a) that the untreated cells cause both the reversible and irreversible fouling, and that the fouling reversibility gradually exacerbates as more filtration cycles are performed.

Since Microcystis cells are organisms in themselves, they can deposit on and even adhere to the membrane surface, leading to the reversible and irreversible fouling, respectively [14]. It can be also found in Fig. 6(a) that MnO2 pretreatment dramatically alleviates the Microcystis cell fouling, with the total fouling reduced by nearly 20% at the dose of 1.10 mg/L. Additionally, the irreversible

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

63

Fig. 8. Morphology characterization of membranes: SEM image (a) and EDS spectrum (b) of the membrane fouled by MnO2 (1.1 mg/L) pretreated cell solution; SEM image (c) and EDS spectrum (d) of hydraulically clean membrane (MnO2: 1.1 mg/L).

fouling ranges between 0.116–0.156 and 0.098–0.136 in the 3-cycle filtration processes under the MnO2 doses of 0.55 and 1.10 mg/L, respectively, less than that in the controlled trial (0.120–0.249). The results suggest that the in situ formed MnO2 particles are of great potential in alleviating the irreversible fouling related to Microcystis cells. The probable reason is ascribed to the adsorption of the released EOM, which was considered as the culprit responsible for severe irreversible UF membrane fouling [39,40]. Moreover, the aggregation of the cells, resulted from the adhesion of the MnO2 particles, may also contribute to the alleviation of the irreversible fouling. When the membrane is precoated with the in situ formed MnO2 particles, the irreversible fouling by Microcystis cells is visibly eased, despite the minor variation in the total fouling (as shown in Fig. 6). The result indicates that the MnO2 layer can alleviate the irreversible but reversible Microcystis cell fouling. In another study on the NOM fouling control by precoating, an iron oxide layer was found to prevent large-sized colloids from adhering to the membrane and thus to improve the fouling reversibility [41]. In current study, the layer of in situ formed MnO2 particles can effectively hinder the penetration of Microcystis cells and adsorb the released EOM, resulting in better fouling reversibility. With regards to KMnO4 oxidation, no apparent improvements are made in the fouling reversibility at the tested doses. The reversible fouling in the 3-cyle filtration process ranges between 0.64–0.75 and 0.66–0.77 at the KMnO4 dose of 1.0 and 2.0 mg/L, respectively, slightly lower than that in the controlled trial. In respect to the irreversible fouling, the performance of KMnO4 oxidation is not obvious in fouling control either. It is known that the reversible and the irreversible fouling are dependent on cake layer structure and interfacial interaction, respectively. The generated MnO2 particles may contribute to improve the cake layer characteristics and weaken the cell-membrane adhesion, improving the fouling reversibility; however, the amount of the MnO2 particles is much smaller than that in the scenario with reducing regents

added. Meanwhile, KMnO4 oxidation can induce the release of extracellular polymers which probably penetrates into the cake layer formed by Microcystis cells, making it less porous [11]. Moreover, the extracellular polymers are significant foulants for the irreversible membrane fouling [39,42]. Thus, KMnO4 oxidation only exhibits a very minor contribution to the improvement of both the reversible and irreversible fouling caused by Microcystis cells. 3.7. Morphology characterization of fouled and hydraulically cleaned membranes Fig. 8 shows the SEM images and EDS spectra of the fouled and hydraulically cleaned membranes which were used to filter Microcystis cell solution pretreated by the in situ formed MnO2 particles. In Fig. 8(a), a thick cake layer is observed on the membrane surface after filtering the cell solution. When the membrane is backwashed, most cake layer is washed away, but there are some residual cells on the membrane (as shown in Fig. 8(c)). This result is consistent with the irreversible fouling caused by Microcystis cells. For the individual KMnO4 scenario, a cake layer of cells and residual cells are observed in the membranes before and after hydraulic rinsing (as shown in Fig. S4). Based on the EDS analysis, several types of elements such as carbon, oxygen, permanganate, gold, silicon and iron are found in the cake layer. Gold element is the strongest in absorbance intensity, since gold spraying was adopt to endow the membrane samples better conductivity. Although the difference in manganese intensity is not obvious in the spectrum, the data show that the weight ratios of manganese are 1.40 ± 0.16% and 0.31 ± 0.09% for the fouled and rinsed membrane samples, respectively. It implies that the MnO2 particles deposit on the membrane as well as algal cells during filtration and that great part of them are washed away during hydrophilic rinsing. In addition, this result also verifies the existence of the residual manganese on the hydrophilic cleaned membrane.

64

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65

However, whether did the residual manganese contribute to the irreversible membrane fouling? It cannot be answered by this study and requires a further research.

[10] [11]

4. Conclusions [12]

The mechanisms of Microcystis cell fouling control by permanganate peroxidation were systematically investigated, with KMnO4 oxidation and in situ formed MnO2 adhesion compared. The following conclusions can be drawn. (1) The individual KMnO4 oxidation exhibited a minor effect on the surface charge and size distribution of Microcystis cells as well as the cell integrity under the tested KMnO4 exposure (1.0 and 2.0 mg/L). However, KMnO4 oxidation could stimulate the release of EOM. (2) The pretreatment with in situ formed MnO2 particles could promote the aggregation of Microcystis cells, improving their settleability, whereas KMnO4 oxidation resulted in more single cells. (3) The in situ formed MnO2 particles displayed a superior capacity in alleviating both the flux decline and irreversible fouling associated with Microcystis cells than KMnO4 oxidation did, owing to the reinforced aggregation of the cells and the adsorption of the released EOM. (4) The precoating layer of the in situ formed MnO2 particles could seldom reduce the flux decline associated with Microcystis cells, probably due to the limited impact on the cake layer, but it improved the fouling reversibility by inhibiting the direct contact between the cells and the UF membrane surface.

[13] [14]

[15] [16] [17]

[18]

[19]

[20]

[21] [22]

[23]

[24]

Acknowledgements This research was jointly supported by National Natural Science Foundation of China (Grants 51308146), China Postdoctoral Science Foundation funded project (Grants 2013M540293) and Heilongjiang Postdoctoral Fund (Grants LBH-Z13083). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2015.05.009.

[25]

[26]

[27]

[28]

[29]

References [30] [1] J.B. Rose, C.P. Gerba, W. Jakubowski, Survey of portable water supplies for Cryptosporidium and Giardia, Environ. Sci. Technol. 25 (1991) 8. [2] J.G. Jacangelo, R.R. Trussell, M. Watson, Role of membrane technology in drinking water treatment in the United States, Desalination 113 (1997) 9. [3] H. Huang, T.A. Young, K.J. Schwab, J.G. Jacangelo, Mechanisms of virus removal from secondary wastewater effluent by low pressure membrane filtration, J. Membr. Sci. 409–410 (2012) 1–8. [4] G. Amy, Fundamental understanding of organic matter fouling of membranes, Desalination 231 (2008) 44–51. [5] B. Van Der Bruggen, C. Van Decasteele, T. Van Gestel, W. Doyen, R. Leysen, A review of pressure-driven membrane processes in wastewater treatment and drinking water production, Environ. Prog. 22 (2003) 11. [6] R. Henderson, S.A. Parsons, B. Jefferson, The impact of algal properties and preoxidation on solid–liquid separation of algae, Water Res. 42 (2008) 1827–1845. [7] S. Merel, D. Walker, R. Chicana, S. Snyder, E. Baures, O. Thomas, State of knowledge and concerns on cyanobacterial blooms and cyanotoxins, Environ. Int. 59 (2013) 303–327. [8] L. Li, N. Gao, Y. Deng, J. Yao, K. Zhang, Characterization of intracellular & extracellular algae organic matters (AOM) of Microcystic aeruginosa and formation of AOM-associated disinfection byproducts and odor & taste compounds, Water Res. 46 (2012) 1233–1240. [9] M.B. Dixon, Y. Richard, L. Ho, C.W. Chow, B.K. O’Neill, G. Newcombe, A coagulation-powdered activated carbon-ultrafiltration–multiple barrier

[31]

[32]

[33]

[34]

[35]

[36]

approach for removing toxins from two Australian cyanobacterial blooms, J. Hazard. Mater. 186 (2011) 1553–1559. T. Li, B.Z. Dong, Z. Liu, W.H. Chu, Characteristic of algogenic organic matter and its effect on UF membrane fouling, Water Sci. Technol. 64 (2011) 1685. F. Qu, H. Liang, J. Tian, H. Yu, Z. Chen, G. Li, Ultrafiltration (UF) membrane fouling caused by cyanobacteria: fouling effects of cells and extracellular organics matter (EOM), Desalination 293 (2012) 30–37. R.K. Henderson, A. Baker, S.A. Parsons, B. Jefferson, Characterisation of algogenic organic matter extracted from cyanobacteria, green algae and diatoms, Water Res. 42 (2008) 3435–3445. B. Kwon, N. Park, J. Cho, Effect of algae on fouling and efficiency of UF membranes, Desalination 179 (2005) 203–214. F. Wicaksana, A.G. Fane, P. Pongpairoj, R. Field, Microfiltration of algae (Chlorella sorokiniana): critical flux, fouling and transmission, J. Membr. Sci. 387–388 (2012) 83–92. S. Babel, S. Takizawa, Microfiltration membrane fouling and cake behavior during algal filtration, Desalination 261 (2010) 46–51. M.T. Hung, J.C. Liu, Microfiltration for separation of green algae from water, Colloids Surf., B: Biointerfaces 51 (2006) 157–164. E.C. Wert, J.A. Korak, R.A. Trenholm, F.L. Rosario-Ortiz, Effect of oxidant exposure on the release of intracellular microcystin, MIB, and geosmin from three cyanobacteria species, Water Res. 52 (2014) 251–259. H. Ou, N. Gao, C. Wei, Y. Deng, J. Qiao, Immediate and long-term impacts of potassium permanganate on photosynthetic activity, survival and microcystin-LR release risk of Microcystis aeruginosa, J. Hazard. Mater. 219– 220 (2012) 267–275. T. Lin, S. Pan, W. Chen, S. Bin, Role of pre-oxidation, using potassium permanganate, for mitigating membrane fouling by natural organic matter in an ultrafiltration system, Chem. Eng. J. 223 (2013) 487–496. L. Heng, Y. Yanling, G. Weijia, L. Xing, L. Guibai, Effect of pretreatment by permanganate/chlorine on algae fouling control for ultrafiltration (UF) membrane system, Desalination 222 (2008) 74–80. J.J. Chen, H.H. Yeh, The mechanisms of potassium permanganate on algae removal, Water Res. 39 (2005) 4420–4428. F. Qu, H. Liang, Z. Wang, H. Wang, H. Yu, G. Li, Ultrafiltration membrane fouling by extracellular organic matters (EOM) of Microcystis aeruginosa in stationary phase: influences of interfacial characteristics of foulants and fouling mechanisms, Water Res. 46 (2012) 1490–1500. P. Xie, J. Ma, J. Fang, Y. Guan, S. Yue, X. Li, L. Chen, Comparison of permanganate preoxidation and preozonation on algae containing water: cell integrity, characteristics, and chlorinated disinfection byproduct formation, Environ. Sci. Technol. 47 (2013) 14051–14061. H.G. Perterson, S.E. Hrudey, I.A. Cantin, T.R. Perley, S.I. Kenefick, Physiological toxicity, cell membrane damage and the release of dissolved organic carbon and geosmin by aphanizomenon flos-aquae after exposure to water treatment chemicals, Water Res. 29 (1995) 1515–1523. Z. Wang, S.W. Lee, J.G. Catalano, J.S. Lezama-Pacheco, J.R. Bargar, B.M. Tebo, D.E. Giammar, Adsorption of uranium(VI) to manganese oxides: X-ray absorption spectroscopy and surface complexation modeling, Environ. Sci. Technol. 47 (2013) 850–858. R. Liu, L. Sun, R. Ju, H. Liu, J. Gu, G. Li, Treatment of low-turbidity source water by permanganate pre-oxidation: in situ formed hydrous manganese dioxide as filter aid, Sep. Purif. Technol. 117 (2013) 69–74. X. Huangfu, J. Jiang, J. Ma, Y. Liu, J. Yang, Aggregation kinetics of manganese dioxide colloids in aqueous solution: influence of humic substances and biomacromolecules, Environ. Sci. Technol. 47 (2013) 10285–10292. J. Fan, L. Ho, P. Hobson, J. Brookes, Evaluating the effectiveness of copper sulphate, chlorine, potassium permanganate, hydrogen peroxide and ozone on cyanobacterial cell integrity, Water Res. 47 (2013) 5153–5164. L. Zhang, J. Ma, M. Yu, The microtopography of manganese dioxide formed in situ and its adsorptive properties for organic micropollutants, Solid State Sci. 10 (2008) 148–153. Q. Qin, Q. Wang, D. Fu, J. Ma, An efficient approach for Pb(II) and Cd(II) removal using manganese dioxide formed in situ, Chem. Eng. J. 172 (2011) 68– 74. F. Sun, H.-Y. Pei, W.-R. Hu, C.-X. Ma, The lysis of Microcystis aeruginosa in AlCl3 coagulation and sedimentation processes, Chem. Eng. J. 193–194 (2012) 196– 202. R.I. Daly, L. Ho, J.D. Brooks, Effect of chlorination on Microcystis aeruginosa cell integrity and subsequent microcystin release and degradation, Environ. Sci. Technol. 41 (2007) 4447–4453. S. Zhou, Y. Shao, N. Gao, S. Zhu, L. Li, J. Deng, M. Zhu, Removal of Microcystis aeruginosa by potassium ferrate(VI): impacts on cells integrity, intracellular organic matter release and disinfection by-products formation, Chem. Eng. J. 251 (2014) 304–309. F. Qu, H. Liang, J. He, J. Ma, Z. Wang, H. Yu, G. Li, Characterization of dissolved extracellular organic matter (dEOM) and bound extracellular organic matter (bEOM) of Microcystis aeruginosa and their impacts on UF membrane fouling, Water Res. 46 (2012) 2881–2890. J. Fan, R. Daly, P. Hobson, L. Ho, J. Brookes, Impact of potassium permanganate on cyanobacterial cell integrity and toxin release and degradation, Chemosphere 92 (2013) 529–534. M. Ma, R. Liu, H. Liu, J. Qu, W. Jefferson, Effects and mechanisms of prechlorination on Microcystis aeruginosa removal by alum coagulation: significance of the released intracellular organic matter, Sep. Purif. Technol. 86 (2012) 19–25.

F. Qu et al. / Chemical Engineering Journal 279 (2015) 56–65 [37] W.-Z. Yu, J.-H. Qu, J. Gregory, Pre-coagulation on the submerged membrane fouling in nano-scale: effect of sedimentation process, Chem. Eng. J. 262 (2015) 676–682. [38] L.-G. Shen, Q. Lei, J.-R. Chen, H.-C. Hong, Y.-M. He, H.-J. Lin, Membrane fouling in a submerged membrane bioreactor: impacts of floc size, Chem. Eng. J. 269 (2015) 328–334. [39] X. Zhang, L. Fan, F.A. Roddick, Influence of the characteristics of soluble algal organic matter released from Microcystis aeruginosa on the fouling of a ceramic microfiltration membrane, J. Membr. Sci. 425–426 (2013) 23–29.

65

[40] K. Kimura, K. Tanaka, Y. Watanabe, Microfiltration of different surface waters with/without coagulation: clear correlations between membrane fouling and hydrophilic biopolymers, Water Res. 49 (2014) 434–443. [41] P. Yao, K.H. Choo, M.H. Kim, A hybridized photocatalysis-microfiltration system with iron oxide-coated membranes for the removal of natural organic matter in water treatment: effects of iron oxide layers and colloids, Water Res. 43 (2009) 4238–4248. [42] Y.-T. Chiou, M.-L. Hsieh, H.-H. Yeh, Effect of algal extracellular polymer substances on UF membrane fouling, Desalination 250 (2010) 648–652.