Influences of acid–base property of membrane on interfacial interactions related with membrane fouling in a membrane bioreactor based on thermodynamic assessment

Accepted Manuscript Influences of acid-base property of membrane on interfacial interactions related with membrane fouling in a membrane bioreactor based on thermodynamic assessment Leihong Zhao, Xiaolu Qu, Meijia Zhang, Hongjun Lin, Xiaoling Zhou, BaoQiang Liao, Rongwu Mei, Huachang Hong PII: DOI: Reference:

S0960-8524(16)30567-3 http://dx.doi.org/10.1016/j.biortech.2016.04.080 BITE 16439

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

5 February 2016 13 April 2016 16 April 2016

Please cite this article as: Zhao, L., Qu, X., Zhang, M., Lin, H., Zhou, X., Liao, B-Q., Mei, R., Hong, H., Influences of acid-base property of membrane on interfacial interactions related with membrane fouling in a membrane bioreactor based on thermodynamic assessment, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/ j.biortech.2016.04.080

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Influences of acid-base property of membrane on interfacial interactions related with membrane fouling in a membrane bioreactor based on thermodynamic assessment Leihong Zhao a,b, Xiaolu Qu b, Meijia Zhang a, Hongjun Lin a,*, Xiaoling Zhou a, Bao-Qiang Liao c, Rongwu Mei d, Huachang Hong a

a

College of Geography and Environmental Sciences, Zhejiang Normal University, Jinhua, 321004, PR China

b

c

Institute of Physical Chemistry, Zhejiang Normal University, Jinhua, 321004, PR China

Department of Chemical Engineering, Lakehead University, 955 Oliver Road, Thunder Bay, Ontario, P7B 5E1, Canada

d

Environmental Science Research and Design Institute of Zhejiang Province, Hangzhou 310007, PR China

*

Corresponding author. Tel.: +86 0579 82282485. E-mail address: [email protected]

1

Abstract Failure of membrane hydrophobicity in predicting membrane fouling requires a more reliable indicator and fouling control strategy. In this study, influences of membrane acid base (AB) property on interfacial interactions in two different interaction scenarios in a submerged membrane bioreactor (MBR) were studied according to thermodynamic approaches. It was found that both the polyvinylidene fluoride (PVDF) membrane and foulant samples in the MBR had relatively high electron donor (γ-) component and low electron acceptor (γ+) component. For both of interaction scenarios, AB interaction was the major component of the total interaction. The results showed that, the total interaction monotonically decreased with membrane γ-, while was marginally affected by membrane γ+, suggesting that γ- could act as a reliable indicator for membrane fouling prediction. This study suggested that membrane modification for fouling mitigation should orient to improving membrane surface γ- component rather than hydrophilicity.

Keywords Membrane bioreactor; membrane fouling; XDLVO theory; acid base interaction; interaction energy

2

Nomenclature D

closest distance between a sphere and a

κ

reciprocal Debye screening length (nm-1)

λ

decay length of AB interaction in water (0.6

planar surface (nm) h

separation distance between two planar surfaces (nm)

nm)

φ

contact angle (º)

θ

angle of the circular arc in the circular ring

e

electron charge (1.6×10-19 C)

∆G

interaction energy per unit area (mJ·m-2)

ξ

k

Boltzmann’s constant (1.38×10-23J· K-1)

Superscripts

p

roughness of membrane surface (nm)

zeta potential (mV)

AB

Lewis acid-base

R

radius of sludge floc (µ m)

EL

electrostatic double layer

r

radius of differential circular ring on

LW

Lifshitz-van der Waals

particle surface (µ m)

tol

total

interaction energy between membrane

+

electron acceptor

surface and particle (kT)

_

electron donor

U

z (r ,θ ) local amplitude directly below the

Subscripts

circular arc as a function of the position

f

foulant

of the differential circular arc defined by

h0

minimum equilibrium cut-off distance

r and θ Greek letters

ε rε 0

γ

(0.158 nm) l

liquid

permittivity of the suspending liquid

m

membrane

(C·V-1·m-1)

s

solid

surface tension parameter (mJ· m-2)

w

water

3

1. Introduction

While membrane bioreactor (MBR) technology has been regarded as one of most promising technologies for wastewater treatment (Meng et al., 2009; Wang et al., 2014), and has been extensively studied for more than 40 years, investigating factors, mechanisms and control strategies of membrane fouling is still the largest issue for this technology (Meng et al., 2009; Lin et al., 2014; Yan et al., 2016). Foulant adhesion has been considered as the prominent cause of membrane fouling in MBRs (Chen et al., 2012; Wang et al., 2013; Lin et al., 2014). Adhesion of sludge foulants on membrane surface is determined by a complicated interplay of forces including the Lifshitz-van der Waals forces (LW), electrostatic double layer (EL) interactions, and acid-base (AB) interactions. The dependence of these interactions on separation distance is generally incorporated in the so-called extended Derjaguin-Landau-Verwey-Overbeek (XDLVO) approach (van Oss, 1995). The AB interaction is based on electron donor (γ-)/electron acceptor (γ+) interactions between polar moieties in aqueous medium (van Oss, 2003). It was reported that AB interaction between biopolymers and substratum surfaces occurring in water is quantitatively more dominant than either LW or EL interactions (van Oss, 2003; Nguyen et al., 2011), as they are up to 10-100 times higher than LW and EL interactions (van Oss, 2003). Therefore, to reduce the adhesive fouling in MBRs, it is essential to understand the properties and effects of AB interaction on membrane fouling. Hitherto, XDLVO approach has been applied to interpret and predict membrane fouling phenomena in MBRs (Chen et al., 2012; Hong et al., 2013; Wang et al., 2013). In this approach, the AB property of a substance could be experimentally determined based on contact angle measurement (CAM) combined with the equations developed by Van Oss et al. (1988). Consequently, XDLVO approach may provide a useful tool to explore the exact roles 4

of AB interaction in membrane fouling. However, XDLVO approach is only valid for the interactions between two smooth planar surfaces (van Oss, 1993), in contrast, the real membrane surface is significantly rough as reported in lots of previous studies (Hoek and Agarwal, 2006; Mahendran et al., 2011; Chen et al., 2012). Meanwhile, the prevailed foulants in MBRs can be generally classified into three categories: sludge flocs, colloids and solutes (mainly soluble microbial products (SMPs) (Meng et al., 2009; Wu et al., 2012). These foulants have different chemical and morphological properties. The interactions between rough membrane surfaces and varies of foulants should differ from each other, and should be significantly different from those between two smooth planar surfaces. This situation suggested the inapplicability of XDLVO approach for these interaction scenarios, and also called for a proper approach to assess effects of AB interaction. Recently, the authors of this study have developed a novel approach which allows to modeling rough membrane surface and quantitatively assessing interaction energy between sludge flocs and rough membrane surfaces (Zhao et al., 2016). The new developed approach, combining the XDLVO approach may provide a methodological background for tracking the exact roles of AB property of membranes in membrane fouling in different interaction scenarios. Moreover, current fouling control strategy based on membrane modification mainly oriented to increasing membrane surface hydrophilicity. In XDLVO theory, surface hydrophilicity is quantitatively defined by the free energy of interaction between two identical surfaces immersed in water (△Gsws) (van Oss, 1995). However, it has been frequently reported that surface hydrophilicity failed to well predict membrane fouling (Choo and Lee, 1996; Chen et al., 2012; Subhi et al., 2012; Zhang et al., 2015). The limitations of membrane hydrophobicity require a more reliable indicator for membrane fouling prediction and mitigation. Considering the strength of AB interaction, AB property 5

of membrane may be able to serve for this requirement. However, to our knowledge, there is no specific study investigating this issue based on thermodynamic analysis. This study, therefore, aimed to investigate influences of AB property of membrane on membrane fouling in a submerged MBR (SMBR). Surface properties of membrane and foulants sampled from the SMBR apparatus were characterized. The interfacial interactions in different interaction scenarios were calculated. Influences of membrane AB property on interaction energies and membrane fouling were assessed by sensitivity analyses. Finally, implications for membrane fouling control were discussed.

2. Material and methods 2.1 Experimental MBR setup

A lab-scale SMBR apparatus with 60 L effective volume was continuously run. The simulated municipal wastewater was pumped to the reactor as influent. The influent was composed with 300 mg COD/L glucose plus the mineral medium. A flat sheet polyvinylidene fluoride (PVDF) membrane model consisting of five membrane elements was immersed vertically into the reactor. Each element had an effective filtration area of 0.1 m2. The membrane had an average pore size of 0.1µm. Underneath the membrane module, coarse bubble aeration of 180 m3air /m3permeate was supplied. Membrane flux of 30 L

m-2 h-1 was maintained with two calibrations conducted each day. The sludge retention time (SRT) and hydraulic retention time (HRT) for the SMBR were 45 d and 5.5 h, respectively. Gel layer was typically formed at start-up period of SMBR operation, whereas, cake layer was periodically formed at stable-operation period. These foulants were sampled in the relative period for study.

6

2.2. Analytical methods

Membrane samples were prepared by the following procedure: 2 cm × 4 cm membrane pieces were first cut from a large piece of a virgin PVDF membrane, and then pressed tightly to flatten the surface within two glass slides and mounted with the glass slides. Thereafter, the mounted membranes were transferred to a desiccator for 24 h in order to get rid of excess water. The sludge floc samples were pretreated before measurements with the following process: the sludge suspension obtained from the MBR was filtered through a stirred cell (model 8200, Amicon) to form sludge lawns. The resulted sludge lawns were fixed within two glass slides to form relatively flat surface, and then dried in a desiccator for 24 h. The gelling foulant samples were directly obtained from gel layer, which were further fixed within two glass slides and then dried in a desiccator for 24 h. The prepared samples were subjected to the following measurements. Static contact angles of three probe liquids (ultra-pure water, glycerol and diiodomethane) on the surfaces of the prepared samples were measured using a contact angle meter (Kino industry Co., Ltd, USA) according to the sessile drop method. Surface zeta potential of the samples of sludge foulants was determined using a Zetasizer Nano ZS (Malvern Instruments Ltd., UK). Surface zeta potential of the membrane samples was measured by a Zeta 90 Plus Zeta Potential Analyzer (Brookhaven Instruments, UK). Membrane surface morphology was characterized by an atomic force microscopy (AFM) (NT-MDT). Triplicate measurements were conducted for each measurement item of a sample. Calculation of interactions between foulants and membrane surface was performed on the platform of MATLAB 2013. Sensitivity analysis was conducted by plotting change of AB property of membrane (from -100% to 100%) versus change of the parameters of

7

concern. As AB property of membrane will not change along with the MBR operation. Therefore, we changed the AB property artificially to study effects of these changes on the interfacial interactions.

2.3 Methods to calculate interfacial interactions 2.3.1 Interfacial interactions between two infinite planar surfaces

According to XDLVO approach (van Oss, 1995), the individual interaction energy per unit area (specific energy) between two infinite planar surfaces at separation distance (h) (∆GLW(h), ∆GAB(h) and ∆GEL(h)) can be obtained by the following equations:

∆G

LW

LW h0

(h) = ∆G

h0 2 h2

exp( ∆G AB (h) = ∆GhAB 0

(1)

h0 − h

λ

)

(2)

 ζ 2 + ζ f2 1  ∆G EL ( h ) = κζ mζ f ε r ε 0  m (1 − coth κ h) +   2ζ ζ κ h sinh m f  

(3)

where h0 is the hypothetical minimum equilibrium cut-off distance, generally equal to a constant of 0.158 nm. Two planar surfaces are assumed to contact each other at h0 (Meinders et al., 1995). ∆ Gh0 means the individual specific energy in contact (at distance of h0), which is obtained by Eqs.4-6, respectively (van Oss, 1995): ∆GhLW = −2 0

(

γ mLW − γ wLW

∆GhAB = 2  γ w+ 0  ∆GhEL = 0

ε 0ε rκ 2

(

)(

γ LW − γ wLW f

)

γ −f + γ m− − γ w− + γ w−

)

(

(4)

)

γ +f + γ m+ − γ w+ − γ −f γ m+ − γ +f γ m− 

  2ξ ξ (ξ m2 + ξ f2 ) 1 − coth(κ h0 ) + 2 m f 2 csch(κ h0 )  ξm + ξ f  



(5)

(6)

where γ LW , γ + and γ − are the LW, electron donor and electron acceptor surface tension components of a substance (subscripts m, f and w denote membrane, foulant and water, 8

respectively), respectively. Calculation of these components for a substance is achieved by measurements of the contact angle ( φ ) combined with the extended Young’s equation (Eq. 7) (Van Oss et al., 1988). Three liquids including ultrapure water, glycerol and diiodomethane were used as probe liquids in the current study.

(1 + cos φ ) tol γ l = γ lLW γ sLW + γ l− γ s+ + γ l+ γ s− 2

(7)

XDLVO approach provides a quantitative definition of surface hydrophilicity as follows: ∆Gsws = −2

(

2

γ sLW − γ wLW

) − 4(

γ s+γ s− + γ w+γ w− − γ s+γ w− − γ s−γ w+

)

(8)

The more positive ∆Gsws is, the higher hydrophilicity of a substance become. 2.3.2 Interfacial interactions between sludge floc and membrane surface

The individual interaction energy ( U fwm ( D ) ) between a sludge floc (with radius R) and the hypothetical smooth planar membrane surface at the closest distance (D) was obtained by Eqs.9-11, respectively (Hoek and Agarwal, 2006). LW U LW fwm ( D ) = 2π∆Gh0

h02 R D

(9)

h − D  AB U fwm ( D) = 2π RD∆GhAB exp  0  0  λ 

(10)

   1 + e −κ D  U EL + (ξ m2 + ξ f2 ) ln (1 − e−2κ D )  fwm ( D ) = πε r ε 0 R  2ξ mξ f ln  −κ D   1− e   

(11)

U fwm ( D ) between a sludge floc and the modelled rough membrane surface at D was

calculated by the method developed by the authors of this study (Zhao et al., 2016). Fig. 1 depicts a sludge floc on the top of a rough surface. The sludge floc was treated as a sphere, and the rough membrane surface was modeled by assigning a periodic sinusoidal function to both x and y coordinates of asperity distribution (Eq. 12) (Fig. 1a and c). z ( x, y ) = p x cos(π x / 2wx ) + p y cos(π y / 2 wy )

(12) 9

where the terms (px, py) and (π/wx, π/wy) denote the scaled amplitudes and frequencies of asperities along with x and y coordinates, respectively. For simplification, the following assumptions were set: px = py = p, wx = wy = w, and p = w = 50 nm, according to the real average roughness of membrane surface. As shown in Fig. 1, U fwm ( D ) can be obtained by integrating the interaction energy between opposing differential planar elements over the entire sphere surface. U LW =∫ fwm ( D）

2π

AB U fwm ( D) = ∫

2π

U EL fwm ( D ) = ∫

2π

0

0

0

∫

R

∫

R

∫

R

0

0

0

∆G LW ( D + R + 2 p − R 2 − r 2 − z (r ,θ )) rdrdθ

(13)

∆G AB ( D + R + 2 p − R 2 − r 2 − z (r ,θ ))rdrdθ

(14)

∆G EL ( D + R + 2 p − R 2 − r 2 − z ( r ,θ ))rdrdθ

(15)

where, r is the circular ring radius; dr is the differential ring radius; dθ is the differential angle of the differential circular arc in the circular ring; z (r ,θ ) is local amplitude directly below the circular arc as a function of the position of the differential circular arc defined by r and θ . Based on the establishment of a cylindrical and a three-dimensional coordinate system, z (r ,θ ) can be obtained as:

z (r , θ ) = p cos(π r cos θ / 2 w) + p cos(π r sin θ / 2 w)

(16)

Fig. 1. Schematics of a particle on top of rough surface: (a) outline view, (b) side view, (c)

modeled rough membrane surface, and (d) top view.

Due to the impossibility to obtain the antiderivatives in Eqs. 13-15, U fwm ( D ) cannot be straightforwardly calculated. Herein, composite Simpson’s rule was adopted to estimate these double integrals. b

d

a

c

∫∫

m

n

f (r ,θ )drdθ = ∑∑ ∫ i =1 j =1

x2 i

x2 i− 2

∫

y2 j y2 j −2

f ( x, y )dxdy ≈

hk m n ∑∑ ( f 2i −2,2 j −2 + f2i ,2 j −2 + f2i,2 j + f2i−2,2 j ) 9 i =1 j =1

+4( f 2 i −1,2 j −2 + f 2 i ,2 j −1 + f 2 i −1,2 j + f 2 i − 2,2 j −1 ) + 16 f 2 i −1,2 j −1

(17)

where, sign “fi, j” represents f (ri, θj), r1=a, ri=r1+ih (i=1, 2, … , 2m+1), θ1=b, θj=θ1+jk (j=1, 10

2, … , 2n+1), and h= (b-a)/2m, k= (d-c)/2n, m and n are segment number for the interval [a, b] of variable r and interval [c, d] of variable θ, respectively. m and n values were set to be 2000 in order to ensure the negligible calculation error

3. Results and discussion 3.1. Surface properties and interaction scenarios

Adhesive foulants in MBRs can be generally classified into gelling foulants and sludge flocs. The formers are formed as a result of the gelation of soluble microbial products (SMPs) and colloids. Table 1 lists the surface properties with respect of zeta potential and contact angle of the PVDF membrane, gelling foulants and sludge flocs. It should be noted that this study focused on the relationships between these surface properties and interfacial interactions. While other properties or conditions, such as mixed liquor suspended solids (MLSS) and sludge retention time (SRT), certainly affect these surface properties, they do not affect these relationships. If effects of other properties on surface properties are of concerns, several references (Liao et al., 2001; Meng et al., 2006) can be referred to. Based on these data, the surface tensions were obtained by the extended Young’s equation, and the results are shown in Table 2. PVDF is one of the most prevailed membrane materials used in MBR applications. (Santos and Judd, 2010). As shown in Table 1 and 2, all the samples of PVDF membrane, gelling foulants and sludge flocs in the MBR have relatively high electron +

donor ( γ − ) surface tension component and low electron acceptor ( γ ) surface tension component. This is not surprising considering that surfaces of polymers such as SMPs and extracellular polymeric substances (EPSs) contained in foulant samples are not chemically inert, and involve significant effects of hydrogen bonds (Hermansson, 1999; Nguyen et al., 2011). These polymers are usually strong electron donors but have very little 11

electron-acceptor capacity (Van Oss et al., 1988). It should be noted that the AB surface tension component ( γ AB ) for gelling foulants and sludge flocs is negative. The negative

γ AB means that the foulants tend to increase their surface area through deformation (Chakraborty and Zachariah, 2007). Similar phenomenon has been also reported in literature (Hwang et al., 2011). It can be seen from Table 2, the surface properties of the PVDF membrane are comparable to those reported in literature regarding MBR research (Chen et al., 2012; Wang et al., 2013). The gelling foulants possess similar surface tensions as data reported by Tian et al. (2013), and sludge floc foulants also have surface tensions close to those reported by other researchers (Khayet, 2004; Feng et al., 2009). These comparisons demonstrate the representativeness of both PVDF membrane and foulant samples in this study for MBRs. Therefore, the conclusions regarding effects of AB property obtained in the following sections should have universal significance for MBRs. The PVDF membrane used in this study was characterized to be very rough with an average roughness of about 50 nm by AFM scans.

Table 1 Contact angle and zeta potential data for membrane, gelling foulants and sludge

flocs

Table 2 Measured surface tension parameters (mJ·m-2) for membrane, gelling foulants and

sludge flocs

Fig. 2 shows the schematics of the potential interaction scenarios associated with rough membrane surface and various foulants. As shown in Fig. 2a, the gelling foulants mainly consisting of SMPs and colloids are very flexible and small. They can fit membrane surface morphology when they adhere to membrane surface. Therefore, the interaction scenario 12

shown in Fig. 2a can be regarded as the scenario of interactions between two infinite planar surfaces. The rough membrane surface just increases the area of interaction between gelling foulants and membrane surface. Fig. 2b illustrates the interaction scenario regarding a sludge floc and rough membrane surface. It is a general operation for interaction calculation that, sludge floc is treated as a sphere, and the rough membrane surface is modeled by a mathematical equation. Fig. 2. Schematics of the interaction scenarios: (a) associated with gelling foulants and rough

membrane surface, and (b) associated with sludge floc foulants and rough membrane surface.

3.2. Effects on interfacial interactions regarding gelling foulant adhesion

Fig. 3 shows the dependence of specific energy on the separation distance for the scenario shown in Fig. 2a. It is observed that, contrary to EL interaction, AB and LW interactions are attractive in the whole separation distance. AB interaction is much long-ranged than LW and EL interaction. Moreover, AB interaction is the predominant component of total interaction. van Oss (2003) also reported that AB interaction between biopolymers or cells was extremely high. The interaction strength between two planar surfaces can be generally quantified by the specific energy in contact (separation distance = 0.158 nm). As illustrated by this parameter, AB specific energy in contact accounts for about 66% of total specific energy in contact, indicating that AB interaction plays a key role in adhesion of gelling foulants. As a result of these features, the total interaction is continuously attractive in the whole separation distance. Continuous attraction of gelling foulants interacting with rough membrane surface suggests the easy adhesion of gelling foulants. Lots of studies have reported that gelling foulants were difficult to remove by aeration once it formed on membrane surface (Wang et al., 2008; van den Brink et al., 2009; Wang and Waite, 13

2009). The underlying cause of this phenomenon has not been well explored in literature. It was revealed that the high strength of AB interaction stemmed from the rich γ − component of both gelling foulants and membrane (Table 2). From thermodynamic viewpoint, this fouling phenomenon could be plausibly explained. Fig. 3. Profiles of the specific energies with separation distance.

The total and AB interaction would change with AB surface tensions. Effects of AB surface tensions on the total interaction were investigated by sensitivity analysis, and the results are presented in Fig. 4. Under conditions in this study, γ m− and γ m+ components of membrane exert different impacts on the total specific energy in contact. Starting from the zero point, 100% increase in γ m− corresponds to 79% decrease in the total specific energy in contact, whereas, 100% increase in γ m+ only leads to 0.7% decrease in the total specific energy in contact. This suggests that γ m− rather than γ m+ plays important role in total energy and adsorptive fouling. The result clearly shows that the absolute value of specific energy in contact monotonically decreases with γ m− value, indicating membrane surface γ m− can act as an effective indicator to predict interaction as well as membrane fouling caused by gelling foulant adhesion.

Fig. 4. Effects of AB surface tension of membrane on the total specific energy in contact for

the gelling foulants based on sensitivity analysis.

3.3. Effects on interfacial interactions regarding sludge floc foulant adhesion

Fig. 5a shows the energy profiles of interaction between a sludge floc and the modeled rough membrane surface. AB and LW interactions are attractive in the whole separation distance, and AB interaction is attractive and acts as a chief constitution of the total 14

interaction. The profiles are similar to those in Fig. 3. Whereas, as compared with Fig. 3, LW and EL interactions become flatten and long-ranged. Fig. 5b services a reference which shows the energy profiles of interaction between a sludge floc and the smooth planar membrane surface. As compared with Fig. 5b, the AB interaction range becomes shorten, and moreover, the interaction strength at least weakens for 100 times for rough membrane surface. In addition, the energy barrier shown in Fig. 5b disappears in Fig. 5a. Due to lack of the energy barrier, sludge flocs will more easily adhere to rough membrane surface. However, thereafter, the attached sludge flocs will more easily detach from rough membrane surface due to the significantly reduced strength of interaction in contact. Fig. 5. The energy profiles of interactions between a sludge floc and (a) the modeled rough

membrane surface, (b) the smooth planar membrane surface.

Fig. 6a illustrates the profile of total interaction energy variation with γ m− component of membrane. When membrane γ m− component increases, absolute value of total energy decays gradually. This is especially the case for the total energy in contact. For example, the total interaction energy in contact decreases from -517 kT to -115 kT as γ m− component of membrane increases from 17.43 mJ·m-2 to 34.86 mJ·m-2. Fig. 6b shows the change of the total interaction energy in contact with AB surface tensions of membrane at an increase interval of 20%. Similar to Fig.4, γ m− has a strong effect, while γ m+ exerts a marginal effect on the total interaction in contact. Starting from the zero point, 100% increase in absolute − + value of γ m and γ m induces about 78% and 0.8% decrease in absolute value of total

energy in contact, respectively. The apparent monotonic decrease trend of total interaction energy with γ m− suggests the potential application of γ m− in membrane fouling prediction.

Fig. 6. (a) Variation of interaction energy between a sludge floc and rough membrane surface 15

with membrane γ − component, and (b) effects of AB surface tension of membrane on the total interaction energy in contact for the sludge floc foulants based on sensitivity analysis.

3.4. Implications for membrane fouling control

The experimental results give important implications for membrane fouling control in MBRs. van Oss (2003) introduced conception “hydrophobic attraction” in adhesion process study, which was defined as the strongest non-electrostatic and non-covalent binding force existing between polymer molecules immersed in water. Hydrophobic attraction straightforwardly determines the adhesion process caused by biopolymers such as SMPs and EPSs. As shown in Fig. 3 and Fig. 5a, LW interaction does not much contribute to the hydrophobic attraction, and the main contributor of hydrophobic attraction is virtually the AB interaction. Therefore, foulant adhesion process can be controlled by adjusting hydrophobic attraction effect, which can be further achieved by control of AB interaction. It has been frequently reported the failure to predict membrane fouling by using surface hydrophilicity (Choo and Lee, 1996; Chen et al., 2012; Subhi et al., 2012). Zhang et al. (2015) reported that there was no definite relationship between membrane surface hydrophilicity (defined as △Gsws) and the total interaction energy. These studies suggested the inapplicability of surface hydrophilicity for membrane fouling prediction and mitigation. Therefore, there is a need to develop a more reliable fouling indicator and control strategy. The monotonic variation tendency of interaction energy in contact with membrane γ m− component for adhesion of gelling foulants and sludge flocs (Fig. 4 and Fig. 6), not only suggests that γ m− is a reliable indicator to predict membrane fouling, but also implies that adhesion of both gelling foulants and sludge flocs can be reduced by improving the γ m− component of membrane. This actually provides a novel membrane fouling control strategy. 16

Membrane surface γ − component can be altered by some measures such as membrane blending (Zhang et al., 2014), surface coating and grafting (Kaur et al., 2013). It was reported that adsorptive fouling was significantly reduced when γ m− component was increased from 13.91 to 27.58 mJ·m-2 by membrane blending modification (Zhang et al., 2014). Zuo and Wang (2013) found that the increase of γ m− component from 0 to 54.05 mJ·m-2 via plasma induced membrane grafting of polyethylene glycol (PEG) and subsequent TiO2 particles deposition, corresponded well to the enhanced anti-oil fouling performance of membrane. The consistency of reduced membrane fouling with enhanced γ m− indicates the correctness of the findings obtained in this study. According to XDLVO theory, surface hydrophilicity is quantitatively defined by the free energy of interaction between two identical surfaces immersed in water (van Oss, 1995), and γ m− denotes the ability of membrane surface to donate electrons to another entity. There should be no definite relationship between membrane surface hydrophilicity and γ m− . In summary, this study demonstrated that membrane modification for fouling mitigation should orient to improving membrane surface γ m− component rather than hydrophilicity.

4. Conclusions

This study investigated the effects of AB property of membrane on the total interaction energy in two different interaction scenarios. The samples of PVDF membrane, gelling foulants and sludge flocs simultaneously possessed relatively high γ- and low γ+ component. AB interaction accounted for the largest proportion of the total interaction in both of interaction scenarios. The total interaction monotonically decreased with membrane surface γ-, indicating that γ- could be a reliable indicator for membrane fouling prediction. It was

suggested that fouling mitigation based on membrane modification should orient to 17

improving membrane surface γ- rather than hydrophilicity.

Acknowledgements

Financial support of National Natural Science Foundation of China (No. 51578509), Public Welfare Project of the Science and Technology Department of Zhejiang Province (2016C31106), Special Supporting Project for Zhejiang Provincial Research Institutes (2015F50013), and Natural Science Foundation of Zhejiang Province (LQ16E080003) is highly appreciated.

References

1.

Chakraborty, P., Zachariah, M.R., 2007. "Effective" negative surface tension: a property of coated nanoaerosols relevant to the atmosphere. J. Phys. Chem. A 111 (25), 5459-64.

2.

Chen, L., Tian, Y., Cao, C.-q., Zhang, J., Li, Z.-n., 2012. Interaction energy evaluation of soluble microbial products (SMP) on different membrane surfaces: Role of the reconstructed membrane topology. Water Res. 46 (8), 2693-2704.

3.

Choo, K.H., Lee, C.H., 1996. Effect of anaerobic digestion broth composition on membrane permeability. Water Sci. Technol. 34 (9), 173-179.

4.

Feng, L., Li, X., Du, G., Chen, J., 2009. Adsorption and fouling characterization of Klebsiella oxytoca to microfiltration membranes. Process Biochem. 44 (11), 1289-1292.

5.

Hermansson, M., 1999. The DLVO theory in microbial adhesion. Colloids Surf. B 14 (1–4), 105-119.

6.

Hoek, E.M.V., Agarwal, G.K., 2006. Extended DLVO interactions between spherical particles and rough surfaces. J Colloid Interf. Sci. 298 (1), 50-58.

7.

Hong, H., Peng, W., Zhang, M., Chen, J., He, Y., Wang, F., Weng, X., Yu, H., Lin, H., 18

2013. Thermodynamic analysis of membrane fouling in a submerged membrane bioreactor and its implications. Bioresour. Technol. 146, 7-14. 8.

Hwang, G., Lee, C.-H., Ahn, I.-S., Mhin, B.J., 2011. Determination of reliable Lewis acid–base surface tension components of a solid in LW–AB approach. J. Ind. Eng. Chem. 17 (1), 125-129.

9.

Kaur, I., Kumari, V., Sharma, B., Gupta, N., 2013. Characterization and applications of PVF film grafted with binary mixture of methacrylic acid and 4-vinyl pyridine by gamma radiations: Effect of swift heavy ions. Appl. Radiat. Isotopes 79, 118-130.

10. Khayet, M., 2004. Membrane surface modification and characterization by X-ray photoelectron spectroscopy, atomic force microscopy and contact angle measurements. Appl. Surf. Sci. 238 (1–4), 269-272. 11. Liao, B.Q., Allen, D.G., Droppo, I.G., Leppard, G.G., Liss, S.N., 2001. Surface properties of sludge and their role in bioflocculation and settleability. Water Res. 35 (2), 339-350. 12. Lin, H., Zhang, M., Wang, F., Meng, F., Liao, B.-Q., Hong, H., Chen, J., Gao, W., 2014. A critical review of extracellular polymeric substances (EPSs) in membrane bioreactors: Characteristics, roles in membrane fouling and control strategies. J. Membr. Sci. 460 (0), 110-125. 13. Mahendran, B., Lin, H., Liao, B., Liss, S.N., 2011. Surface Properties of Biofouled Membranes from a Submerged Anaerobic Membrane Bioreactor after Cleaning. Journal of Environmental Engineering 137 (6), 504-513. 14. Meinders, J.M., van der Mei, H.C., Busscher, H.J., 1995. Deposition Efficiency and Reversibility of Bacterial Adhesion under Flow. J. Colloid Interface Sci. 176 (2), 329-341. 15. Meng, F., Chae, S.-R., Drews, A., Kraume, M., Shin, H.-S., Yang, F., 2009. Recent 19

advances in membrane bioreactors (MBRs): Membrane fouling and membrane material. Water Res. 43 (6), 1489-1512. 16. Meng, F., Zhang, H., Yang, F., Zhang, S., Li, Y., Zhang, X., 2006. Identification of activated sludge properties affecting membrane fouling in submerged membrane bioreactors. Sep. Purif. Technol. 51 (1), 95-103. 17. Nguyen, V.T., Chia, T.W.R., Turner, M.S., Fegan, N., Dykes, G.A., 2011. Quantification of acid–base interactions based on contact angle measurement allows XDLVO predictions to attachment of Campylobacter jejuni but not Salmonella. J. Microbiol. Meth. 86 (1), 89-96. 18. Santos, A., Judd, S., 2010. The commercial status of membrane bioreactors for municipal wastewater. Sep. Sci. Technol. 45 (7), 850 - 857. 19. Subhi, N., Verliefde, A.R.D., Chen, V., Le-Clech, P., 2012. Assessment of physicochemical interactions in hollow fibre ultrafiltration membrane by contact angle analysis. J. Membr. Sci. 403–404, 32-40. 20. Tian, Y., Li, Z., Ding, Y., Lu, Y., 2013. Identification of the change in fouling potential of soluble microbial products (SMP) in membrane bioreactor coupled with worm reactor. Water Res. 47 (6), 2015-2024. 21. van den Brink, P., Zwijnenburg, A., Smith, G., Temmink, H., van Loosdrecht, M., 2009. Effect of free calcium concentration and ionic strength on alginate fouling in cross-flow membrane filtration. J. Membr. Sci. 345 (1-2), 207-216. 22. van Oss, C.J., 1993. Acid-base interfacial interactions in aqueous media. Colloids Surf. A 78 (0), 1-49. 23. van Oss, C.J., 1995. Hydrophobicity of biosurfaces - Origin, quantitative determination and interaction energies. Colloids Surf. B 5 (3–4), 91-110. 20

24. van Oss, C.J., 2003. Long-range and short-range mechanisms of hydrophobic attraction and hydrophilic repulsion in specific and aspecific interactions. J. Mol. Recognit. 16 (4), 177-190. 25. Van Oss, C.J., Good, R.J., Chaudhury, M.K., 1988. Additive and nonadditive surface tension components and the interpretation of contact angles. Langmuir 4 (4), 884-891. 26. Wang, Q., Wang, Z., Zhu, C., Mei, X., Wu, Z., 2013. Assessment of SMP fouling by foulant–membrane interaction energy analysis. J. Membr. Sci. 446, 154-163. 27. Wang, X.M., Waite, T.D., 2009. Role of gelling soluble and colloidal microbial products in membrane fouling. Environ. Sci. Technol. 43 (24), 9341-7. 28. Wang, Z., Ma, J., Tang, C.Y., Kimura, K., Wang, Q., Han, X., 2014. Membrane cleaning in membrane bioreactors: A review. J. Membr. Sci. 468 (0), 276-307. 29. Wang, Z., Wu, Z., Yin, X., Tian, L., 2008. Membrane fouling in a submerged membrane bioreactor (MBR) under sub-critical flux operation: Membrane foulant and gel layer characterization. J. Membr. Sci. 325 (1), 238-244. 30. Wu, J., He, C., Zhang, Y., 2012. Modeling membrane fouling in a submerged membrane bioreactor by considering the role of solid, colloidal and soluble components. J. Membr. Sci. 397–398 (0), 102-111. 31. Yan, X., Wu, Q., Sun, J., Liang, P., Zhang, X., Xiao, K., Huang, X., 2016. Hydrodynamic optimization of membrane bioreactor by horizontal geometry modification using computational fluid dynamics. Bioresour. Technol. 200, 328-334. 32. Zhang, J., Wang, Q., Wang, Z., Zhu, C., Wu, Z., 2014. Modification of poly(vinylidene fluoride)/polyethersulfone blend membrane with polyvinyl alcohol for improving antifouling ability. J. Membr. Sci. 466, 293-301. 33. Zhang, M., Liao, B.-q., Zhou, X., He, Y., Hong, H., Lin, H., Chen, J., 2015. Effects of 21

hydrophilicity/hydrophobicity of membrane on membrane fouling in a submerged membrane bioreactor. Bioresour. Technol. 175, 59-67. 34. Zhao, L., Zhang, M., He, Y., Chen, J., Hong, H., Liao, B.-Q., Lin, H., 2016. A new method for modeling rough membrane surface and calculation of interfacial interactions. Bioresour. Technol. 200, 451-457. 35. Zuo, G., Wang, R., 2013. Novel membrane surface modification to enhance anti-oil fouling property for membrane distillation application. J. Membr. Sci. 447, 26-35.

22

Figure captions:

Fig. 1.

Schematics of a particle on top of rough surface: (a) outline view, (b) side view, (c) modeled rough membrane surface, and (d) top view.

Fig. 2.

Schematics of the interaction scenarios: (a) associated with gelling foulants and rough membrane surface, and (b) associated with sludge floc foulants and rough membrane surface.

Fig. 3.

Profiles of the specific energies with separation distance.

Fig. 4.

Effects of AB surface tension of membrane on the total specific energy in contact for the gelling foulants based on sensitivity analysis.

Fig. 5.

The energy profiles of interactions between a sludge floc and (a) the modeled rough membrane surface, (b) the smooth planar membrane surface.

Fig. 6.

(a) Variation of interaction energy between a sludge floc and rough membrane surface with membrane γ − component, and (b) effects of AB surface tension of membrane on the total interaction energy in contact for the sludge floc foulants based on sensitivity analysis.

23

Fig. 1. Schematics of a particle on top of rough surface: (a) outline view, (b) side view, (c)

modeled rough membrane surface, and (d) top view.

24

(a)

(b) sludge floc

solutes colloids

membrane

membrane

Fig. 2. Schematics of the interaction scenarios: (a) associated with gelling foulants and rough

membrane surface, and (b) associated with sludge floc foulants and rough membrane surface.

-2

Energy per unit area (mJ•m )

3

EL interaction

0 -3

LW interaction

-6 -9 -12

AB interaction

-15 total interaction

-18 -21

0.158 nm

-24 0

1

2 3 4 Separation distance (nm)

5

Fig. 3. Profiles of the specific energies with separation distance.

25

6

Specific energy in contact (%)

200

γ m− varation rsγ m+ variation rs+

150 100 50 0 -100

-80

-60

-40

-20 0 -50

20

40

60

80

100

-100 -150 -200 AB surface r s variation(%) tension variation (%)

Fig. 4. Effects of AB surface tension of membrane on the total specific energy in contact for

the gelling foulants based on sensitivity analysis.

26

(a) 100

EL interaction AB interaction

Interaction energy (kT)

0

total interaction LW interaction

-100

-200

-300

-400

0.158 nm -500 0

2

4

6 8 10 S eparation distance (nm)

12

14

12

14

(b)5000 4000

EL interaction

Interaction energy (kT)

3000 2000

total interaction

1000 0

LW interaction

-1000

AB interaction

-2000 -3000

0.158 nm

-4000 -5000 0

2

4

6 8 10 Separation distance (nm)

Fig. 5. The energy profiles of interactions between a sludge floc and (a) the modeled rough

membrane surface, (b) the smooth planar membrane surface.

27

(b)

0

energy (kT)

-100 -200 -300

Interaction

-400 -500 -600

100

40

rm- variatio

20 0

-20

0

0 -70 0 -80 Sepa r

n (%)

80 60

Interaction energy in contact (%)

(a)

200

γ m− varation rsγ m+ variation rs+

150 100 50 0 -100 -80 -60

-40 -20 0 -50

20

40

60

80

-100 -150

3

2

1

-40

5

4

ation

-60

-80

7

6

dista

10

9

8

nce ( nm)

-200

AB surface tension variation (%) r - variation(%)

-100

Fig. 6. (a) Variation of interaction energy between a sludge floc and rough membrane surface

with membrane γ − component, and (b) effects of AB surface tension of membrane on the total interaction energy in contact for the sludge floc foulants based on sensitivity analysis.

28

100

Table list

Table 1

Contact angle and zeta potential data for membrane, gelling foulants and sludge flocs

Table 2

Measured surface tension parameters (mJ·m-2) for membrane, gelling foulants and sludge flocs

29

Table 1 Contact angle and zeta potential data for membrane, gelling foulants and sludge

flocs Materials

Pure water

Contact angle ( º ) Glycerol Diiodomethane

Zeta potential (mV)

virgin PVDF membrane

59.28 ± 2.41

53.63 ± 1.39

18.66 ± 0.57

-32.6 ± 1.8

gelling foulants

61.98 ± 0.71

61.29 ± 0.37

23.84 ± 0.75

-17.15 ± 0.21

sludge flocs

71.33 ± 0.52

67.70 ± 0.53

30.97 ± 0.56

-15.75 ± 0.35

Table 2 Measured surface tension parameters (mJ·m-2) for membrane, gelling foulants and

sludge flocs Materials

γ LW

γ+

γ−

γ AB

γ Tot

Tot ∆Gsws

virgin PVDF membrane

48.16

0.089

17.43

2.50

50.66

-21.78

gelling foulants

46.83

0.009

18.64

-0.81

46.02

-19.78

sludge flocs

43.81

0.042

12.97

-1.49

42.32

-34.61

30

Graphical abstract

γ

− m

γ

electron donor

+ m

solutes colloids

electron acceptor

acid base (AB) property C

N

C

O

O

H

O

H

H

N

H

N

C

C

N

C

thermodynamic assessment

150 100

monotonic50decrease tre 0 -10 -80 -60 -40 -20 -50 0 0 -100

20

-150

-200 AB surface tension variatio

interaction scenario 1# C

200

sludge floc

Interaction energy in contact (%)

O

membrane

Specific energy in contact (%)

200

thermodynamic assessment

rough membrane surface membrane interaction scenario 2#

31

150 100

monotonic decrease tre 50 0 -10 -80 -60 -40 -50 -20 0 20 0 -100 -150

-200 AB surface tension variatio

Research highlights ► total interaction energy monotonically decreases with membrane electron donor ► total interaction energy is marginally affected by membrane electron acceptor ► membrane electron donor is a reliable indicator predicting membrane fouling ► fouling control shall improve membrane electron donor rather than hydrophilicity

32