Effects of manganese dioxides on the ultrafiltration membrane fouling by algal extracellular organic matter

Effects of manganese dioxides on the ultrafiltration membrane fouling by algal extracellular organic matter

Separation and Purification Technology 153 (2015) 29–36 Contents lists available at ScienceDirect Separation and Purification Technology journal hom...

585KB Sizes 5 Downloads 220 Views

Separation and Purification Technology 153 (2015) 29–36

Contents lists available at ScienceDirect

Separation and Purification Technology journal homepage: www.elsevier.com/locate/seppur

Effects of manganese dioxides on the ultrafiltration membrane fouling by algal extracellular organic matter Fangshu Qu, Zhongsen Yan, Wei Liu, Senlin Shao, Xiujuan Ren, Nanqi Ren, Guibai Li, Heng Liang ⇑ State Key Laboratory of Urban Water Resource and Environment (SKLUWRE), Harbin Institute of Technology, Harbin, 150090, PR China

a r t i c l e

i n f o

Article history: Received 1 June 2015 Received in revised form 21 August 2015 Accepted 22 August 2015 Available online 24 August 2015 Keywords: Water treatment Ultrafiltration Membrane fouling Manganese dioxide Extracellular organic matter

a b s t r a c t Membrane fouling caused by algal extracellular organic matter (EOM) is of great concern to the industrial implementation of ultrafiltration (UF) in algae-laden water treatment. Pre-oxidation with potassium permanganate (KMnO4) is thus widely utilized to tackle algae-related fouling issues. To verify the contribution of manganese dioxide (MnO2), which is the intermediate product in KMnO4 pre-oxidation, to fouling control, three types of manganese dioxides (commercial available, lab-prepared and in-situ formed MnO2) were adopted to treat EOM prior to filtration. The in-situ formed MnO2 exhibited a higher DOC removal rate (25% with 10 mg/L MnO2 used) than the commercial and lab-prepared MnO2 did. As far as adsorption selectivity was concerned, all MnO2 preferentially adsorbed the macromolecular and hydrophobic fractions of EOM, especially for protein-like substances. Filtration tests showed that the pretreatment with the in-situ formed MnO2 significantly alleviated the flux decline caused by EOM owing to the superb performance in removing macromolecular components. Other types of MnO2 brought about very minor improvements in membrane permeability. Moreover, reversibility analyses showed that the irreversible fouling was reduced to some extent by manganese dioxide adsorption, among which the in-situ formed MnO2 performed the best. However, the improvement in fouling reversibility was not prominent due to the minor removal in the hydrophilic fraction of EOM by manganese dioxides. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction An increasing demand of safe drinking water has necessitated developing and implementing more effective separation and purification techniques for potable water production. Membrane processes, including microfiltration, ultrafiltration (UF), nanofiltration and reverse osmosis, serve as appropriate alternatives to conventional techniques for water treatment like coagulation and sand filtration [1,2]. Of particular interest is the UF membrane that exhibits high pollutant removal and relatively low energy consumption, associated with the small pore size (0.05 lm) and low operating pressure (10–500 kPa) required, respectively. The UF membrane can totally retain particles, colloids, protozoa (3–15 lm), bacteria (0.5–10 lm), virus (0.02–0.08 lm) as well as algal cells (>2 lm) [3]. Among these pollutants, algal cells recently attract a considerable amount of attentions with regard to algal hepatotoxin and carcinogenic disinfection byproducts. Moreover, ⇑ Corresponding author. E-mail addresses: [email protected] (F. Qu), [email protected] (Z. Yan), [email protected] (W. Liu), [email protected] (S. Shao), [email protected] 163.com (X. Ren), [email protected] (N. Ren), [email protected] (G. Li), [email protected] (H. Liang). http://dx.doi.org/10.1016/j.seppur.2015.08.033 1383-5866/Ó 2015 Elsevier B.V. All rights reserved.

cell breakage and toxin release resulting from water treatment chemicals may bring about a higher safety risk to drinking water [4]. On the basis of size exclusion, UF can retain algal cells without rupturing them and hence substantially lower the health hazard related to algal cells and their metabolites. Despite a variety of advantages such as total cell removal and less toxin release, the application of UF in algae-laden water treatment is unavoidably restricted by membrane fouling which may boost energy demand and shorten membrane life-span. Both algal cells and extracellular organic matter (EOM) proved to cause severe flux decline as well as irreversible fouling [5]. As algal cell size is obviously larger than membrane pore size, algal cells only precipitate on membrane surface, forming a cake layer. Wicaksana et al. [6] studied the microfiltration membrane fouling by Chlorella cells using direct observation through the membrane technology (DOTM), and observed that cell deposition started to occur at a very low permeate flux. As to EOM, the fouling phenomenon and the related mechanisms, which are very complicated due to heterogeneous molecular weight (MW) and hydrophilicity [7], are fertile areas to explore for lots of researchers. To better understand EOM fouling, systematical characterization is usually undertaken to investigate EOM characteristics. It has been reported that EOM, which was mainly comprised of proteins and


F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36

polysaccharides, possessed distinct features like high MW and strong hydrophilicity [8,9]. Chiou et al. [10] has demonstrated that algal species with more bound EOM caused worse fouling. To be more specific, size fractionation and XAD resin fractionation were employed to divide EOM into a variety of fractions, and the highMW and hydrophobic fractions were found to be responsible for the severe membrane fouling [11,12]. Moreover, when investigating the ceramic membrane fouling by EOM, Zhang et al. [13]characterized the foulants in the outer, middle and inner fouling layers, and got to know that the small hydrophilic molecules could lead to pore restriction. To address the fouling issues related to algae, chemical oxidants like ozone and chlorine are commonly adopted to pretreat algaeladen water. Ozone pre-oxidation was reported to considerably reduce the microfiltration fouling by Chlorella via reducing cake compressibility and biomass loading [14]. However, exposure of algal cells to strong oxidants definitely results in the release of toxins [15,16] as well as intracellular organic matter (IOM), which is of even higher membrane fouling potential than EOM [17]. Considering the trade-off relationship between cell inactivating performance and cell integrity preservation, potassium permanganate, which is a weak oxidant, becomes one of the most attractive chemicals for controlling the membrane fouling related to algal cells and organics. In a study on algae-laden water treatment by oxidationaided UF, Liang et al. [18] attributed the improvements in membrane permeability and water quality to the involvement of immediate reaction products of KMnO4. Lin et al. [19] had demonstrated that KMnO4 pre-oxidation mitigated the TMP buildup during surface water filtration by means of effectively adsorbing organics by in-situ formed MnO2. As intermediate products are usually very unstable, the direct evidence for fouling control by the in-situ formed MnO2 has not been provided in the related literature, leaving room for disputation. The objective of this study was to verify the contribution of manganese dioxides to EOM fouling control. With commercial available (C-MnO2), lab prepared (P-MnO2) and in-situ formed (IMnO2) MnO2 compared, effects of MnO2 adsorption on the EOM characteristics were investigated, and the contribution of MnO2 adsorption to fouling control was discussed. 2. Materials and methods 2.1. Algae culture and EOM extraction Microcystis aeruginosa was purchased from the Freshwater Algae Culture Collection, Institute of Hydrobiology, Chinese Academy of Sciences. The instruction on algae cultivation was described by Qu et al. [11]. Axenic cultures were grown in batches in 1 L conical flasks using BG11 medium. The conical flasks were cultured in an incubator (25 °C) and a light/dark cycle (14 h on, 10 h off) was provided to simulate natural light conditions. Algae cultures were harvested at the stationary growth phase (40–42 d). For EOM extraction, the harvested cell suspensions were initially centrifuged at 10,000g and 4 °C for 15 min using a high speed refrigerated centrifuge (H2050R, Xiangyi, China). Subsequently, EOM was obtained by filtering the supernatant through a 0.45 lm mixed cellulose filter (Taoyuan Co., Ltd, China). Prior to using, EOM solution was stored at 4 °C in a refrigerator with the storage time not exceeding 48 h. To prepare feed solution, the EOM solution was diluted to a DOC concentration of 5.0 ± 0.2 mg/L with Milli-Q water. 2.2. MnO2 preparation A stock suspension of C-MnO2 was prepared by dispersing 0.1 g MnO2 in 200 mL Milli-Q water under rapid mixing (200 rpm)

conditions provided by a magnetic stirrer. Because MnO2 particles were easy to deposit, the stock suspension was carefully mixed to make the particles uniformly suspend before dosing. P-MnO2 and IMnO2 were both prepared by reducing KMnO4 with a stoichiometric amount of Na2S2O3 [20]. The P-MnO2 stock suspension was prepared at a concentration of 500 mg/L, and it was rapidly mixed for 24 h before characterization or being added into EOM solution. To generate I-MnO2, KMnO4 was firstly added into EOM solution under rapid-mixing conditions, followed by a stoichiometric amount Na2S2O3. A small amount of soluble MnO 4 was left during I-MnO2 preparation and imparted the treated EOM solution slightly pink color which would gradually disappear over time. The chemicals such as commercial MnO2, KMnO4, Na2S2O3, HCl, NaOH and acetone were of analytical grade and supplied by Bench Chemicals (Tianjin, China). 2.3. Experimental protocol 2.3.1. EOM adsorption on MnO2 particles In this work, adsorption experiments were performed to investigate the impacts of the three types of MnO2 on organic contents as well as characteristics of EOM. MnO2 was added into EOM solutions under the concentrations of 0, 1, 2, 4, 6, 8, and 10 mg/L. The adsorption of EOM on MnO2 was performed by means of blending (100 rpm) the solutions in an air bath shaker (ZHWY-211B, Zhicheng, China) for 600 min. After that, the blended solution was filtered through a 0.45 lm mixed cellulose filter to separate MnO2 particles from aqueous solution, followed by measuring the ultraviolet absorbance at 254 nm (UV254) and dissolved organic carbon (DOC) concentration of the filtrate. In addition, EEM fluorescence spectrum, MW distribution and hydrophobicity/hydrophilicity of EOM before and after adsorption (MnO2 10 mg/L) were also determined. 2.3.2. Filtration experiments Filtration tests were performed to investigate the UF membrane fouling caused by raw and treated EOM, using 2 400-mL stirred UF cells (Amicon 8400, Millipore Corp., USA) as described by Qu et al. [11]. The experimental setup included a nitrogen cylinder, 2 stirred cells, and 2 electronic balances connected to a computer. Nitrogen gas at a constant pressure of 0.03 MPa was utilized to drive feed solution through the membrane. Permeate flowed into a beaker on the electronic balance and weighting data were automatically logged every five seconds. Flat polyethersulfone UF membranes (OM100076, Pall, USA) were used in current study. The MW cutoff and surface area of the membrane were 100 kDa and 4.5  103 m2, respectively. To remove preservatives, new membranes should be carefully rinsed beforehand via soaking in MilliQ water for 48 h. 2.3.3. Membrane fouling assessment In this study, membrane fouling was evaluated via flux decline and fouling reversibility. Prior to filtering EOM solution, the Milli-Q water flux of each membrane was measured and designated J0. Three continuous filtration cycles were performed in every filtration test. For each cycle, three steps were carried out: (1) filter 300 mL out of the 350 mL feed solution, (2) turn over the membrane and backwash it for 2 min with Milli-Q water, and (3) filter 100 mL Milli-Q water. The flux during the filtration of EOM solution was denoted J (L m2 h1), and the normalized flux could be described by J/J0. The flux at the end of filtration process was denoted Je (L m2 h1). Jm refers to the Milli-Q water flux in the third step in each filtration cycle. Then, the reversible fouling (RF, m1) in each filtration cycle and the accumulative irreversible fouling (IF, m1) could be calculated using the following equations [21]. The filtration test was duplicated for each type of feed water.


F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36

IF n ¼ ðJ 0  J m Þ=J 0


RF ¼ ðJ 0  J e Þ=J 0  ðJ 0  J m Þ=J 0 ¼ ðJm  J e Þ=J 0


2.4. Analytical methods 2.4.1. Methods for characterizing MnO2 particles The attempt to understand the effect of MnO2 adsorption on EOM fouling necessitated the systematical characterization of different types of MnO2 particles. Zeta potential, particle size distribution, functional groups, surface area and surface morphology were measured and compared. The zeta potential representing the surface charge of MnO2 particles was analyzed by a zetasizer (Nano S90, Malvern, UK) in a broad pH range (1.86–11.53). The particle size distribution of MnO2 particles was determined using a particle size analyzer (Mastersizer 2000, Malvern, UK) by which a size range from 0.02 lm to 2 mm was accessible. The surface area and pore size distribution were measured by the BET method with nitrogen gas absorbed in the dry state [22], using a physisorption analyzer (ASAP2020, Micrometrics, USA). The surface microstructure of MnO2 particles was visualized using a field emission scanning electron microscope (SEM, LEO 1530 VP, Germany). Fourier transform infrared (FTIR) spectroscopy could provide reliable information about the functional groups on MnO2 particles. Samples of MnO2 particles were initially separated from aqueous solution by centrifugation at 3000 rpm. After centrifugation, the slurry was freeze-dried and stored at a desiccator for FTIR detection. The FTIR spectra of MnO2 particles was obtained by a Spotlight 400 FTIR spectrometer (PerkinElmer, USA) using the potassium bromide (KBr) pellet method. The measured wavelength range was between 4000 and 600 cm1. 2.4.2. Methods for EOM characterization DOC concentrations were measured using a total organic matter analyzer (multi N/C 2100S, Analytic Jena, Germany). Potassium hydrogen phthalate was adopted for calibration and a linear range between 0 and 20 mg L1 was obtained with an R2 value of 0.999. The ultraviolet absorbance at 254 nm (UV254) was determined using a UV spectrometer (T6, Puxi, China). The fluorescence spectra of EOM were obtained using a fluorescence spectrophotometer (F7000, Hitachi, Japan) and a 1 cm path length quartz cuvette. Fluorescence intensities were measured in triplicate at excitation (Ex) wavelengths of 220–450 nm in 5 nm increments and emission (Em) wavelengths of 250–550 nm in 1 nm increments. Spectral correction was taken to minimize the instrumental bias potentially including the wavelengthdependent variability in the transmission efficiency of monochromators and the fluctuations in spectrometer light intensity [23]. To prevent the inner filter effect which might result in a nonlinear relationship between the fluorescence intensity and foulant concentration, all samples were diluted to a UV254 value not exceeding 0.3 cm1 [23]. A controlled fluorescence spectrum of Milli-Q water was subtracted from all spectra to eliminate water Raman scattering. The effects of MnO2 adsorption on fluorophores were evaluated via the peak picking method on a basis of the fourcomponent model developed in a previous study, in which a dataset of 96 EOM-associated EEM fluorescence data was modeled using parallel factor analysis [24]. Sample-specific matrices of correction factors for inner filter effects calculated from UV absorbance scans were applied to the corrected data, which were then normalized to RU [25]. The normalized intensities of predetermined peaks (Table 1) were utilized to distinguish the EOM removal performance by different types of MnO2. MW distribution was analyzed by the UF fractionation method [11]. By filtering EOM solution through regenerated cellulose

membranes (PLHK07610, PLTK07610, PLGC07610, PLBC07610, Millipore Corp., USA) under a constant pressure (0.1 MPa), EOM was fractionated into five fractions (>100, 30–100, 10–30, 3–10 and <3 kDa). The fractionation experiment was performed in a parallel mode with identical samples fed for the membranes. The stirrer was kept running (200 rpm) to prevent foulant deposition and to reduce concentration polarization. For each fraction, 100 mL EOM solution was fed into the stirred cell and 40 mL permeate was produced with the initial 10 mL discarded. After that, DOC concentrations of the feed solution and effluents were measured and the ratios of different fractions were calculated. The hydrophobic, transphilic and hydrophilic fractions of EOM were fractionated by nonionic macro-porous resins as described by Carroll et al. [29]. The organics adsorbed onto XAD-8 and XAD-4 resins (Sigma, USA) were regarded as hydrophobic (HPO) and transphilic (TPI) fractions, respectively, whereas the residual organics were defined as the hydrophilic (HPI) fraction. The detailed fractionation steps and resin cleaning processes were described by Qu et al. [11]. The fractionation experiments were repeated in triplicate. 3. Results and discussion 3.1. Characteristics of different manganese dioxides To understand EOM adsorption on MnO2 particles, it is significant to know the characteristics of various types MnO2 particles. C-MnO2 had a mean particle size of 40.5 ± 2.7 lm (shown in Table S1), much bigger than that of P-MnO2 (6.7 ± 1.5 lm). Hence, the specific surface area of C-MnO2 (33.7 ± 2.4 m2/g) was naturally smaller than that of P-MnO2 (157 ± 5.4 m2/g) (Table S1 in the Supplementary Material). I-MnO2, which are the intermediate product during the reduction process of permanganate, are very unstable. The particle size and surface area of I-MnO2 were not measured in this study. As reported by Huangfu et al. [20], I-MnO2 possessed the diameters of 24–105 nm with an average of 55.86 nm, which was almost one order of magnitude smaller than that of C-MnO2 in this study. As shown in Table S1, the iodine adsorption capacities were 1105, 1742 and 2529 mg/g for C-MnO2, P-MnO2 and IMnO2, respectively. In addition, the methylene blue adsorption capacity was also in the order of C-MnO2 (0.06 mg/g) < P-MnO2 (0.9 mg/g) < I-MnO2 (9.3 mg/g). The adsorption results correlate well with the determined particle sizes and surface areas of the three types of MnO2 particles. The zeta potentials of C-MnO2, P-MnO2 and I-MnO2 were 35.2, 30.2 and 20.4 mV at the pH of 6.86 (Fig. S1 in the Supplementary Material), respectively, indicating negatively charged surfaces of MnO2 particles under the natural aquatic environment. The isoelectric points of three types of MnO2 particles all located in the pH range of 1.86–3.86. The SEM images (Fig. S2 in the Supplementary Material) illustrate the surface morphology of MnO2 particles. Small I-MnO2 particles were loosely stacked, forming lots of big voids on the surface, whereas C-MnO2 and P-MnO2 particles were more tightly compacted. For C-MnO2 and P-MnO2, the mesopores were mainly in the range of 3–5 nm (Fig. S3 in the Supplementary Material). Moreover, P-MnO2 exhibited much higher pore volumes Table 1 Ex/Em pairs of components detected in fluorescent EEM spectrum of EOM [24]. Components

Ex wavelength (nm)

Em wavelength (nm)

Probable source

1 2 3 4

275 270/330 260/360 250/290

330 415 440 360

Tryptophan-like [26] Humic fluorophore [27] Humic-like substances [27] Amino acids, protein-like substances [28]


F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36

than C-MnO2 in this mesoporous range. As a type of intermediate product, I-MnO2 is so unstable that it is impossible to determine its PSD via the physisorption analyzer. Generally, I-MnO2 possesses some superb characteristics like small size, high surface area and rough surface, and these characteristics endow I-MnO2 much higher capacity in adsorbing EOM than C-MnO2 and P-MnO2. 3.2. Removal of EOM by different types of manganese dioxides Algal EOM which arises extracellularly via metabolic excretion is comprised of glycolic acid, carbohydrates, polysaccharides, amino acids, peptides, organic phosphorus, enzymes, vitamins, hormonal substances, inhibitors, and toxins [9]. For M. aeruginosa used in this work, the amount of EOM can reach 0.00095 ng cell1 [8]. To understand the role of manganese dioxides on EOM fouling, it is essential to investigate the effect of different types of manganese dioxides in EOM removal. As shown in Fig. 1, C-MnO2 exhibited the lowest removal performance, with the UV254 removal rate increasing from 7% to 10% as the MnO2 dosage increased from 1 mg/L to 10 mg/L. P-MnO2 showed better EOM removal with the UV254 removal rate of 18% at the dose of 10 mg/L. For I-MnO2, the UV254 removal rates were the highest under all the tested dosages. As shown in Fig. 1(b), the DOC removal by MnO2 particles was in a similar trend as the UV254 removal. I-MnO2 still led in the EOM removal, with the DOC removal rate as high as 25% under the dosage of 10 mg/L. For P-MnO2, the DOC removal rate increased from 6% to 17% with the dosage increasing from 1 mg/L to 10 mg/L. C-MnO2 lagged behind other types of MnO2 particles in





UV254 removal rate (%)


DOC removal. Generally, the EOM removal by MnO2 particles boosted with the increasing doses and the priority of performance is in the order of I-MnO2 > P-MnO2 > C-MnO2. There are two probable interpretations for the superb EOM removal performance of IMnO2. On one hand, the smaller particle size and higher surface area may endow I-MnO2 much stronger adsorption capacity than C-MnO2 and P-MnO2. On the other hand, some unsaturated components of EOM (humic-like components) containing benzene ring and carbon–carbon double band are likely to be degraded when KMnO4 is present in EOM solution [30]. To investigate the performance of different types of manganese dioxides in removing fluorescent components of EOM, the fluorescent spectra of raw and treated EOM solutions were obtained (see Fig. S4 in the Supplementary Material). On a basis of the predetermined excitation and emission wavelength pairs (As shown in Table 1), the normalized fluorescent intensities of the four components were obtained and the results are shown in Fig. 2. Components 1 and 4 are related to tryptophan-like substances and freeamino acids, respectively, whereas components 2 and 3 are both associated with humic-like substances [26,28]. For raw EOM, the fluorescent intensities of components 1 and 4 were 1.3 and 0.6 RU, respectively, much higher than that of humic-like components (0.3 RU). This result confirms the dominating component of protein-like substances in EOM. All the three types of manganese dioxides could remove parts of protein-like components of EOM, and I-MnO2 achieved the best performance with the fluorescent intensities of components 1 and 4 reduced by 68% and 62%, respectively. In terms of humic-like components, I-MnO2 only obtained a comparative removal performance to C-MnO2, whereas P-MnO2 exhibited the largest removal capacity with the fluorescent intensities of components 2 and 3 reduced by 59% and 62%, respectively. The results indicate that I-MnO2 preferentially adsorb the proteinlike substances in EOM and that P-MnO2 has higher capacities in removing the humic-like components of EOM.


3.3. Effects of different types of manganese dioxides on the EOM characteristics



0 0






MnO2 dose (mg/L)




MW distribution and hydrophilicity are two significant characteristics of EOM. In terms of MW, EOM possesses a bimodal distribution with the high-MW (>100 kDa) and low-MW (<1 kDa) fractions accounting for 45% and 42%, respectively (as shown in Fig. 3). Because EOM contains biopolymers such as polysaccharides and proteins, the hydrophilic fraction takes the highest percentage (57%) of total DOC, followed by the hydrophobic fraction (38%) and then the transphilic fraction (5%) (as shown in Fig. 4).




Fluorescence intensity (RU)

DOC removal rate (%)





0 0






MnO2 dose (mg/L) Fig. 1. Performance of different manganese dioxides in EOM removal: (a) UV254 and (b) DOC. Error bar indicates the standard error (n = 3).


Raw EOM C-MnO4 treated EOM


P-MnO4 treated EOM I-MnO4 treated EOM

0.8 0.6 0.4 0.2 0.0 1




Components Fig. 2. Performance of different manganese dioxides in removing fluorescent components of EOM.


F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36


EOM C-MnO2 treated EOM


P-MnO2 treated EOM


Ratio (%)

As shown in Fig. 3, slight reductions were observed in 100 kDa– 0.45 lm and 30–100 kDa fractions when EOM was treated by CMnO2. P-MnO2 showed better performance in removing the macromolecular components of EOM than C-MnO2, with the ratio of the high-MW fraction reduced to 36%, probably attributed to more mesopores in P-MnO2 (Fig. S3 in the Supplementary Material). In the case of I-MnO2, the ratio of the high-MW fraction decreased to 14% whereas the low-MW fraction accounted for 51% of total organics in the treated EOM solution. In accordance with the EEM spectra, these results imply that MnO2 particles preferentially adsorbed the high-MW fraction of EOM with I-MnO2 leading in the removal performance. As the main components of EOM, the protein-like substances are strong in hydrophobicity and thus easier to be adsorbed by I-MnO2 which has plenty of voids on the surface as illustrated by SEM images (Fig. S2 in the Supplementary Material) [31]. When EOM was treated by MnO2 particles, obvious reductions in hydrophobic ratios and augments in the proportions of transphilic and hydrophilic parts could be observed. Specifically, the proportions of hydrophobic fractions were reduced to 35%, 30% and 15% by C-MnO2, P-MnO2 and I-MnO2, respectively (As shown in Fig. 4). The results indicate the preferential adsorption of hydrophobic EOM fractions on MnO2 particles, perhaps due to the stronger hydrophobic interaction. Owing to the smaller size and higher surface area, I-MnO2 were more prone to adsorbing the hydrophobic components of EOM than other two types of MnO2 particles. Zhang et al. [32] also demonstrated the high adsorption capacity of the MnO2 particles formed in-situ in removing trace organics from aqueous solution.

I-MnO2 treated EOM

40 30 20 10 0 Hydrophobic



Fig. 4. XAD resin fractionation results of raw and manganese dioxide-treated EOM. Error bar indicates the standard error (n = 3).

1630, 1045 and 525 cm1. As the broaden peaks around 525 cm1 is assigned to the stretching vibration of Mn-O band [32,33], all the samples showed the characteristic peak of Mn-O band whether EOM adsorption occurred or not. Different from other MnO2 particles, I-MnO2 showed a strong absorbance around 3300 cm1 which refers to the bending vibration of adsorbed molecular water and the stretching vibration of hydroxyl group [34]. Hence, It can be inferred that there are much more hydroxyl groups on the surface of I-MnO2. Comparing the FTIR spectra of the MnO2 particles, no obvious peak shift occurred after EOM adsorption, but the peak intensity of hydroxyl group augmented to some extent, verifying the adhesion of EOM on the MnO2 particles.

3.4. Changes of manganese dioxides after adsorbing EOM

3.5. Effects of different types of manganese dioxides on EOM fouling

To further verify the adhesion of EOM, MnO2 particles were extracted and characterized after treating EOM solution. The average particle sizes of MnO2 particles increased to 67.7, 24.2 and 70.5 lm for C-MnO2, P-MnO2 and I-MnO2, respectively. It can be easily noted that the adhesion of EOM made the MnO2 particles agglomerate to some extent, especially for I-MnO2. As shown by the XAD resin fractionation results, MnO2 particles preferentially adsorbed the hydrophobic fractions of EOM which might increase the potential and diversity of interactions between MnO2 particles [20]. Therefore, the agglomeration of MnO2 colloids was accelerated to some degree. Fig. 5 presents the FTIR spectra of the three types of MnO2 particles before and after EOM adsorption. As shown in Fig. 5, absorbance peaks can be observed at the wave numbers around 3300,

Fig. 6 presents the flux decline curves during the filtration of raw and pretreated EOM solutions. It can be observed that raw EOM caused severe flux decline with the final normalized flux of 0.11. As EOM is comprised of macromolecular proteins and polysaccharides, these organics can be retained and deposit on membrane surface during filtration, leading to the decrease in membrane permeability. Zhang et al. [35] performed a study on the microfiltration membrane fouling by algal organic matter, and found that the majority of the flux decline was attributed to the large amount of organic matter depositing on the membrane surface. When EOM was treated by C-MnO2 or P-MnO2, the flux decline curves almost coincided with that during the filtration of raw EOM (Fig. 6), with the finial normalized flux as small as 0.13. When EOM was treated by I-MnO2, the flux decline was significantly alleviated with the final normalized flux higher than 0.22. In a variety of related articles, high-MW proteins and polysaccha-

60 EOM



C-MnO2 treated EOM




I-MnO2 treated EOM


Absorbance (%)

DOC ratio (%)

P-MnO2 treated EOM

40 30 20 10


0 100k-0.45 μm






Molecular weight (Da)








Wave number (cm-1) Fig. 3. MW distribution of raw and manganese dioxide-treated EOM. Error bar indicates the standard error (n = 3).

Fig. 5. FTIR spectra of MnO2 particles before and after EOM adsorption.


F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36



C-MnO2 treated

P-MnO2 treated


I-MnO2 treated

Percentage (%)


Normarized flux (J/J0)







50 40 30 20 10


0 0.0

Raw EOM 0










Accumulative volume (mL)

C-MnO2 treated P-MnO2 treated

I-MnO2 treated

Fig. 8. Effects of manganese dioxide adsorption on the mass balance of EOM during filtration. Error bar indicates the standard error (n = 3).

Fig. 6. Effects of manganese dioxide adsorption on the flux decline by EOM.

rides were identified as the culprits for severe membrane fouling [36–38]. C-MnO2 and P-MnO2 could not effectively remove these organics as illustrated by the fluorescence EEM spectra, whereas I-MnO2 exhibited a high adsorption capacity for protein-like substances. Therefore, in the case of I-MnO2, less flux shrinkage occurred during the filtration of the treated EOM solution. To better understand the effects of MnO2 particles on EOM fouling, fouling reversibility was analyzed and the results are presented in Fig. 7. It can be observed that EOM caused not only reversible fouling but also irreversible fouling. As EOM contains high-MW organics which can be retained by UF membrane, a cake layer may form on the membrane surface during filtration, causing reversible fouling. Meanwhile, there are also low-MW organics in EOM that may be trapped into the membrane pores. As a consequence, the irreversible fouling took place in the filtration test. As illustrated in Fig. 7, with the help of manganese dioxide adsorption, both the total fouling and the irreversible fouling by EOM were alleviated to some extent, and I-MnO2 still led in the performance of fouling control. As far as the irreversible fouling is concerned, foulant identification has been done in a variety of studies, but the results are not always consistent with each other. In our previous study, the high-MW and hydrophobic fractions exhibited some contributions to the irreversible fouling [11]. As illustrated by the size fractionation and XAD resin fractionation results, I-MnO2 preferentially adsorbed the high-MW and hydrophobic fractions. Thus, I-MnO2 performed better in alleviating irreversible fouling than other types of MnO2. However, Kimura et al. [39] have demonstrated that the hydrophilic fraction of natural organic matter (NOM) took the major responsibility for the evolution of irreversible fouling, because hydrophilic NOM or

Reversible fouling Irreversible fouling


Normalized fouling

0.8 0.6 0.4 0.2 0.0 Raw EOM

C-MnO2 treated P-MnO2 treated I-MnO2 treated

Fig. 7. Effects of manganese dioxide adsorption on the reversibility of fouling by EOM. Error bar indicates the standard error (n = 3).

biopolymers might be selectively trapped on/in the membranes. In the study on the ceramic membrane fouling by algal organics, Zhang et al. [13] further verified the contribution of the hydrophilic components of EOM to the irreversible fouling. In this study, all MnO2 particles preferentially adsorbed the hydrophobic fractions and let off the hydrophilic molecules which could reach the inner membrane pores and cause pore restriction. Therefore, the irreversible fouling was not dramatically reduced even in the case of I-MnO2. Fig. 8 presents the mass balance results during the filtration of raw and pretreated EOM solutions. Permeate, retentate and deposit refer to the organics penetrating across the membrane, remaining in the retentate and depositing on the membrane surface, respectively. During the filtration of raw EOM, the permeate, retentate and deposit accounted for 42%, 25% and 33%, respectively, well agreeing with the size fractionation results. When EOM was pretreated by MnO2 particles, the ratios of the organics passing through the membrane apparently increased accompanied by a decrease in the ratios of deposit. Specifically, the ratios of organics in permeate were 54%, 59% and 63% when EOM was treated by CMnO2, P-MnO2 and I-MnO2, respectively. As MnO2 particles preferentially adsorbed the high-MW fraction of EOM (as illustrated in Fig. 3), the low-MW fractions, which could penetrate through the membrane, took relatively higher proportions of total organics in treated EOM solutions, especially in the case of I-MnO2. When CMnO2, P-MnO2 and I-MnO2 were involved in the pretreatment, the ratios of the organics depositing during filtration were 22%, 20% and 11%, respectively. The preferential removal of high-MW and hydrophobic fractions of EOM by MnO2 particles took the main responsibility for the reduced deposition of organics. Moreover, in consideration of the significant role of zeta potential in adsorption, the zeta potentials of EOM solutions before and after MnO2 treatment were determined. An increase in negative zeta potentials, which helped to augment the electrostatic repulsion and reduce the deposition of EOM on membrane surface, was observed after the treatment (Table S1 in the Supplementary Information). Overall, the pretreatment with MnO2 particles helped to reduce the deposition of EOM and the priority of performance was in the order of I-MnO2 > P-MnO2 > C-MnO2, which is in line with the priority in fouling alleviation. Overall, to reinforce the fouling control, it is reasonable to promote the conversion of permanganate into IMnO2 as much as possible. Some reducing compounds can be incorporated into the process for higher membrane permeability. 4. Conclusion Effects of the pretreatment with MnO2 particles on the characteristics of EOM and the membrane fouling by EOM were investi-

F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36

gated with three types of MnO2 particles (commercial available, lab-prepared and in-situ formed MnO2) compared. The following conclusions can be drawn. (1) I-MnO2 exhibited a stronger capacity in removing EOM than C-MnO2 and P-MnO2 did, with UV254 and DOC removal rates of 18% and 25%, respectively, under the dose of 10 mg/L. (2) All types of MnO2 particles preferentially adsorbed the highMW and hydrophobic fractions of EOM, with I-MnO2 leading in removal performance. Moreover, I-MnO2 particles preferentially adsorbed the protein-like components of EOM. (3) The pretreatment with C-MnO2 and P-MnO2 were unable to effectively reduce the severe flux decline caused by EOM, whereas the membrane permeability substantially recovered with the help of I-MnO2 due to their preferential adsorption of the protein-like components of EOM. (4) I-MnO2 exhibited better performance in alleviating the irreversible fouling by EOM than C-MnO2 and P-MnO2 did, probably due to the reduced deposition of high-MW organics and the reduced adhesion of hydrophobic components of EOM. However, the improvement in fouling reversibility was not prominent due to the minor removal in the hydrophilic fraction of EOM.

Acknowledgements This work was jointly supported by National Natural Science Foundation of China (Grants 51308146), Open Project of State Key Laboratory of Urban Water Resource and Environment (ES201511-02), China Postdoctoral Science Foundation funded project (Grants 2015T80360) and Heilongjiang Postdoctoral Fund (Grants LBH-Z13083). Anonymous referees are greatly acknowledged for their constructive and helpful suggestions. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.seppur.2015.08. 033. Reference [1] 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. [2] M.P. Lopes, C.T. Matos, V.J. Pereira, M.J. Benoliel, M.E. Valério, L.B. Bucha, A. Rodrigues, A.I. Penetra, E. Ferreira, V.V. Cardoso, M.A.M. Reis, J.G. Crespo, Production of drinking water using a multi-barrier approach integrating nanofiltration: a pilot scale study, Sep. Purif. Technol. 119 (2013) 112–122. [3] J.G. Jacangelo, R.R. Trussell, M. Watson, Role of membrane technology in drinking water treatment in the United States, Desalination 113 (1997) 119– 127. [4] E.C. Wert, F.L. Rosario-Ortiz, Intracellular organic matter from cyanobacteria as a precursor for carbonaceous and nitrogenous disinfection byproducts, Environ. Sci. Technol. 47 (2013) 6332–6340. [5] B. Kwon, N. Park, J. Cho, Effect of algae on fouling and efficiency of UF membranes, Desalination 179 (2005) 203–214. [6] 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. [7] W. Zhang, W. Zhang, X. Zhang, P. Amendola, Q. Hu, Y. Chen, Characterization of dissolved organic matters responsible for ultrafiltration membrane fouling in algal harvesting, Algal Res. 2 (2013) 223–229. [8] 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. [9] N. Her, G. Amy, H.R. Park, M. Song, Characterizing algogenic organic matter (AOM) and evaluating associated NF membrane fouling, Water Res. 38 (2004) 1427–1438. [10] Y.T. Chiou, M.L. Hsieh, H.H. Yeh, Effect of algal extracellular polymer substances on UF membrane fouling, Desalination 250 (2010) 648–652.


[11] 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. [12] X. Zhang, L. Fan, F.A. Roddick, Feedwater coagulation to mitigate the fouling of a ceramic MF membrane caused by soluble algal organic matter, Sep. Purif. Technol. 133 (2014) 221–226. [13] X. Zhang, L. Fan, F.A. Roddick, Understanding the fouling of a ceramic microfiltration membrane caused by algal organic matter released from Microcystis aeruginosa, J. Membr. Sci. 447 (2013) 362–368. [14] M.T. Hung, J.C. Liu, Microfiltration for separation of green algae from water, Colloids Surf., B 51 (2006) 157–164. [15] 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. [16] 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. [17] L. Li, Z. Wang, L.C. Rietveld, N. Gao, J. Hu, D. Yin, S. Yu, Comparison of the effects of extracellular and intracellular organic matter extracted from Microcystis aeruginosa on ultrafiltration membrane fouling: dynamics and mechanisms, Environ. Sci. Technol. 48 (2014) 14549–14557. [18] H. Liang, Y.L. Yang, W.J. Gong, X. Li, G.B. Li, Effect of pretreatment by permanganate/chlorine on algae fouling control for ultrafiltration (UF) membrane system, Desalination 222 (2008) 74–80. [19] 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. [20] 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. [21] D. Jermann, W. Pronk, R. Kagi, M. Halbeisen, M. Boller, Influence of interactions between NOM and particles on UF fouling mechanisms, Water Res. 42 (2008) 3870–3878. [22] H. Jiang, L. Yang, C. Li, C. Yan, P.S. Lee, J. Ma, High-rate electrochemical capacitors from highly graphitic carbon-tipped manganese oxide/mesoporous carbon/manganese oxide hybrid nanowires, Energy Environ. Sci. 4 (2011) 1813–1819. [23] R.K. Henderson, A. Baker, K.R. Murphy, A. Hambly, R.M. Stuetz, S.J. Khan, Fluorescence as a potential monitoring tool for recycled water systems: a review, Water Res. 43 (2009) 863–875. [24] H. Yu, F. Qu, H. Liang, Z.-S. Han, J. Ma, S. Shao, H. Chang, G. Li, Understanding ultrafiltration membrane fouling by extracellular organic matter of Microcystis aeruginosa using fluorescence excitation–emission matrix coupled with parallel factor analysis, Desalination 337 (2014) 67–75. [25] R.K. Henderson, N. Subhi, A. Antony, S.J. Khan, K.R. Murphy, G.L. Leslie, V. Chen, R.M. Stuetz, P. Le-Clech, Evaluation of effluent organic matter fouling in ultrafiltration treatment using advanced organic characterisation techniques, J. Membr. Sci. 382 (2011) 50–59. [26] Y. Shutova, A. Baker, J. Bridgeman, R.K. Henderson, Spectroscopic characterisation of dissolved organic matter changes in drinking water treatment: from PARAFAC analysis to online monitoring wavelengths, Water Res. 54 (2014) 159–169. [27] S.K. Ishii, T.H. Boyer, Behavior of reoccurring PARAFAC components in fluorescent dissolved organic matter in natural and engineered systems: a critical review, Environ. Sci. Technol. 46 (2012) 2006–2017. [28] S.A. Baghoth, S.K. Sharma, G.L. Amy, Tracking natural organic matter (NOM) in a drinking water treatment plant using fluorescence excitation–emission matrices and PARAFAC, Water Res. 45 (2011) 797–809. [29] T. Carroll, S. King, S.R. Gray, B.A. Boltom, N.A. Booker, The fouling of microfiltration membranes by NOM after coagulation treatment, Water Res. 32 (2000) 2861–2868. [30] D. He, X. Guan, J. Ma, X. Yang, C. Cui, Influence of humic acids of different origins on oxidation of phenol and chlorophenols by permanganate, J. Hazard. Mater. 182 (2010) 681–688. [31] R.H. Peiris, C. Halle, H. Budman, C. Moresoli, S. Peldszus, P.M. Huck, R.L. Legge, Identifying fouling events in a membrane-based drinking water treatment process using principal component analysis of fluorescence excitation– emission matrices, Water Res. 44 (2010) 185–194. [32] 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. [33] 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. [34] M. Kumar, M. Ulbricht, Low fouling negatively charged hybrid ultrafiltration membranes for protein separation from sulfonated poly(arylene ether sulfone) block copolymer and functionalized multiwalled carbon nanotubes, Sep. Purif. Technol. 127 (2014) 181–191. [35] 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. [36] Y. Shen, W. Zhao, K. Xiao, X. Huang, A systematic insight into fouling propensity of soluble microbial products in membrane bioreactors based on hydrophobic interaction and size exclusion, J. Membr. Sci. 346 (2010) 187– 193.


F. Qu et al. / Separation and Purification Technology 153 (2015) 29–36

[37] 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. [38] H. Yamamura, K. Okimoto, K. Kimura, Y. Watanabe, Hydrophilic fraction of natural organic matter causing irreversible fouling of microfiltration and ultrafiltration membranes, Water Res. 54 (2014) 123–136.

[39] K. Kimura, T. Maeda, H. Yamamura, Y. Watanabe, Irreversible membrane fouling in microfiltration membranes filtering coagulated surface water, J. Membr. Sci. 320 (2008) 356–362.