Graphene oxide incorporated thin-film composite membranes for forward osmosis applications

Author’s Accepted Manuscript Graphene oxide incorporated thin-film composite membranes for forward osmosis applications Liang Shen, Shu Xiong, Yan Wang

www.elsevier.com/locate/ces

PII: DOI: Reference:

S0009-2509(15)00806-4 http://dx.doi.org/10.1016/j.ces.2015.12.029 CES12732

To appear in: Chemical Engineering Science Received date: 17 September 2015 Revised date: 19 December 2015 Accepted date: 29 December 2015 Cite this article as: Liang Shen, Shu Xiong and Yan Wang, Graphene oxide incorporated thin-film composite membranes for forward osmosis applications, Chemical Engineering Science, http://dx.doi.org/10.1016/j.ces.2015.12.029 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 galley proof before it is published in its final citable 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.

Graphene oxide incorporated thin-film composite membranes for forward osmosis applications

Liang Shen, Shu Xiong and Yan Wang*

a

Key Laboratory of Material Chemistry for Energy Conversion and Storage

(Huazhong University of Science and Technology), Ministry of Education, Wuhan, 430074, China b

Hubei Key Laboratory of Material Chemistry and Service Failure, School of

Chemistry and Chemical Engineering, Huazhong University of Science & Technology, Wuhan, 430074, P.R. China

* Corresponding author. Tel.: 86 027-87793436; fax: 86 027-87543632. E-mail address: [email protected] (Yan Wang)

Abstract In this work, graphene oxide (GO) nanosheets are synthesized and incorporated into the polyamide (PA) selective layer to develop a novel thin-film composite (TFC) membrane for forward osmosis (FO) application. The chemical structure and morphology of the synthesized GO and GO-incorporated TFC membranes are studied 1

by various characterization techniques. Compared with the control TFC membrane, GO-incorporated TFC membranes exhibit higher water flux and reasonable draw solute rejection. The effects of the GO loading on the membrane morphology and FO performance of the GO-incorporated TFC membranes are investigated systematically in terms of various characterizations and intrinsic separation performance. The influence of the draw solution concentration is also studied. The GO-incorporated TFC membranes also possess lower fouling propensity in FO test than that without embedded GO.

Keywords: forward osmosis, graphene oxide, polyamide, thin-film composite membrane, anti-biofouling

1.

Introduction

Nowadays, the scarcity of fresh water has terribly constrained the sustainable development of public health, ecosystem, economy, and so on. To solve this global water scarcity problem, extensive researches have been devoted to explore novel technologies for desalination and wastewater reclamation within acceptable cost. Although various techniques including the solvent extraction, distillation and pressure driven membrane processes for water treatment have achieved lots of achievements, high energy consumption of above processes restricts their further development. Alternatively, forward osmosis (FO), an osmotically driven membrane process (ODMP) which utilizes the osmotic pressure difference across the permeable 2

membranes for driving force, has drawn more and more attention in past decades. Since there is low or no hydraulic pressures required, the FO technology exhibits many potential advantages, such as relative low energy consumption (McGinnis and Elimelech, 2007), high water recovery (Martinetti et al., 2009) and low fouling propensity (Achilli et al., 2009; Mi and Elimelech, 2010) compared to these pressure driven membrane process (PDMP) like reverse osmosis, nanofiltration, microfiltration and ultrafiltration. FO membrane is the heart of a successful FO technology with a good separation performance. A desirable FO membrane should own high water permeability, high solute rejection, superior anti-fouling property and good stability. In past years, membrane scientists have developed various FO membranes towards this target, including cellulosic membranes (Ong and Chung, 2012; Su et al., 2010; Zhang et al., 2011), thin film composite (TFC) membranes (Liu et al., 2015a; Sukitpaneenit and Chung, 2012; Wang et al., 2015; Wang and Xu, 2015), layer-by-layer (LBL) self-assembled membranes (Qi et al., 2012; Saren et al., 2011), and so on. Among them, TFC membranes, fabricated by the formation of the polyamide layer on the substrate via interfacial polymerization (IP), have gained much attention for water treatment applications because of the excellent separation performance over a wide operation temperature and pH ranges. However, conventional TFC membranes also face problems of low water flux and fouling propensity ascribed to their relative hydrophobic surfaces formed by TMC and MPD monomers (Rana and Matsuura, 2010). Many efforts have been devoted to 3

solve these issues by appropriate modification approaches, such as using additives or surfactants in the monomer solutions (Cui et al., 2014), immersing the nascent PA membrane into active solvent (Li et al., 2013; Zhang et al., 2013), and grafting the hydrophilic polymer chains (Bernstein et al., 2011; Lu et al., 2013), zwitterions (Azari and Zou, 2012; Mi et al., 2015; Yu et al., 2014) or nanoparticles (Tiraferri et al., 2012b) on the membrane surface of the PA layer. Recently, various nano-structured materials have been widely employed to be incorporated into the PA selective layer of the TFC membrane to improve the membrane performance in membrane separation processes and shown a great performance enhancement with a reasonable particle loading, including inorganic salt (Emadzadeh et al., 2015; Ghanbari et al., 2015), zeolite (Dong et al., 2015; Huang et al., 2013; Liu et al., 2015b), zeolitic imidazolate framework (ZIF) (Duan et al., 2015; Sorribas et al., 2013), carbon nano-tubes (CNT) (Amini et al., 2013; Shen et al., 2013), graphene oxide (Safarpour et al., 2015), etc. Among them, graphene oxide, a novel and promising two-dimensional carbon nanomaterial, has attracted great attention in the field of material research. Its excellent physical properties coupled with flexibility in chemical functionalization ascribed to the abundant oxygen-containing functional groups, make GO a good candidate for various applications (Han et al., 2013; Sun et al., 2011; Wang et al., 2011). Recently, the eccentric water permeability of a graphene oxide membrane (Huang et al., 2014; Joshi et al., 2014; Nair et al., 2012) was reported in the membrane separation field. Later, incorporating GO into the polymer matrix to form mixed matrix membranes (MMMs) has been reported to enhance the membrane 4

hydrophilicity and anti-fouling property (Chang et al., 2014; Xu et al., 2014). But the anti-microbial effect becomes unavailable when GO was buried within the bulk polymer matrix compared to that being incorporated onto the membrane surface (Liu et al., 2011; Tu et al., 2013). Previous researches have incorporated GO in the membrane in the selective layer by layer-by-layer self-assembling (Choi et al., 2013; Kim et al., 2013) or chemical crosslinking (Perreault et al., 2013), and achieved improved separation performance with enhanced water permeability, chlorine resistance, or anti-fouling property. However, those membranes prepared by above methods may suffer the water permeability decline with time since the GO coating layer may interfere with water permeation (Chae et al., 2015). In this study, novel TFC membranes for FO applications are fabricated with GO-incorporated into the PA selective layer by interfacial polymerization, using an aqueous mixture of 1, 3-phenylenediamine (MPD)-GO solutions. Similar works have been demonstrated in previous works for RO (Chae et al., 2015) and NF (Bano et al., 2015a) applications with different GO loadings, but no any work has been reported for FO applications so far. This works studies the structure, morphology and fundamental properties of the synthesized GO and the GO-incorporated PA layer with detailed characterizations. Effects of GO loading on the membrane morphology, intrinsic separation performance, FO performance and antifouling performance of the resultant TFC membranes are also investigated systematically.

2.

Materials and Methods 5

2.1. Materials

Natural flake graphite (average particle diameter of 40 mm, 99.95% purity) was purchased from Qingdao Hengsheng graphite Co. Ltd. (China). Polyacrylonitrile (PAN) powder (Mn: 250,000 Da) was purchased from Chushengwei Chemistry Co. Ltd. (Hubei, China), and dried in the vacuum oven at 80 °C for overnight before use. The 1,3-phenylenediamine (MPD, 99.5%) and 1,3,5-benzenetricarbonyl trichloride (TMC, 98%) were obtained from Aladdin and kept in refrigerator before use. Concentrated sulfuric acid (H2SO4, ı 70%), potassium permanganate (KMnO4, 99.5%), sodium nitrate (NaNO3, ı 99%), hydrogen peroxide (H2O2, 30%), concentrated hydrochloric acid (HCl, 16-18%), lithium chloride (LiCl, 97%), N,N-dimethylformamide (DMF, anhydrous, ı 99.5%) and ethanol ( ı 99.7%, anhydrous), sodium hydroxide (NaOH, ı96%), n-hexane (ı97%, anhydrous), sodium chloride (NaCl, ı99.5%), sodium alginate (SA, Mw: 98.11), potassium dihydrogen phosphate (KH2PO4, 99.5%), magnesium sulfate (MgSO4, 99%), sodium bicarbonate (NaHCO3, 99.5%), calcium chloride (CaCl2, 96%) and ammonium chloride (NH4Cl, 99.5%) were all purchased from China National Medicine Corporation.

2.2. Synthesis and characterization of graphene oxide (GO)

Graphene oxide (GO) was synthesized from graphite powder by the modified Hummer’s method (Jiang et al., 2014). The reaction mechanism is shown in Fig. S-1 6

in the Supplementary Information. In brief, 2 g graphite and 2.5 g NaNO3 were added into a 500 ml three-necked flask with 150 ml concentrated sulfuric acid solution, and then stirred under an ice bath. After that, 15 g KMnO4 was slowly added under vigorous stir with the temperature maintained below 20 °C, to allow the mixture to react under stir for overnight at room temperature. Then 180 ml DI water was slowly added, and the reaction was continued for 24 hours with the temperature raised to 98 °C. Next, 80 ml 35 wt% H2O2 aqueous solution was added to reduce the residual KMnO4 and MnO2. The obtained mixture was rinsed repeatedly with 10 wt% HCl aqueous solution to remove metal ions. Subsequently, the mixture was dialyzed by the dialysis bag with MWCO of 3500 Da for one week. Finally, the obtained product was freeze-dried for two days to obtain the sheet-like graphite oxide. The fractionated GO was characterized by Fourier transform infrared (FTIR) (Tensor 27, Bruker, USA), with the wavenumber range of 4000 - 400 cm−1 and a resolution of 2 cm−1, and an average of 16 scans was taken. Its morphology was observed using a scan electron microscopy (SEM, VEGA3, TESCAN, Czech) and an atomic force microscope (SPM9700, Shimadzu, Japan). The SEM sample was prepared by attaching the graphite oxide powder onto the conducting tape, and being coated with gold using a sputter coater (Q150RS, Quorum, England). AFM samples were prepared by dropping the GO/water suspension on the mica surface and allowing water evaporation. Crystal structure of the synthesized GO was also studied using X-ray diffractometer (XRD, 7000S, Shimadzu, Japan) with 2θ ranging from 10 ° to 50 °.

7

2.3. Preparation of hydrolyzed PAN substrates

PAN support layer was prepared by casting a dope solution of 16/5/4/75 wt% PAN/ethanol/LiCl/DMF onto a pre-cleaned glass plate with a casting knife of 100 μm thickness, and then being immersed into a water coagulation bath immediately at room temperature to initiate the phase separation. The water in the coagulation bath was changed every 12 hours for 2 days to remove the residual solvent in the membrane. The as-fabricated membranes were then stored in DI water before the hydrolyzation modification. In order to improve the membrane hydrophilicity, PAN membranes were immersed in 1.5 M NaOH aqueous solution at 45 °C for 1 hour for hydrolyzation modification, and then washed with tap water. The as-prepared substrate is abbreviated as HPAN substrate. The porosity, mean pore size and the molecule weight cut off (MWCO) of the HPAN substrate are 81.1%, 12.7 nm and 352 kDa, respectively, as measured in lab. A water contact angle about 35° and a pure water permeability (PWP) of 54.3 LMHbar-1 of the HPAN substrate were also detected.

2.4. Preparation of TFC membranes

The GO-incorporated TFC membrane was prepared on the HPAN substrate by interfacial polymerization. The mixed MPD/GO solution, as the aqueous monomer solution, was prepared by adding GO into the 1.5 wt% MPD aqueous solution with the concentration of 0-800 ppm, and ultrasonicated for 3 hours to exfoliate the 8

graphite oxide. The HPAN substrate was washed with the ultrapure water, and immersed into the above aqueous solution for 2 minutes. Excess monomer solution was removed by a rubber roller. After that, 0.05 wt% TMC hexane solution was subsequently poured onto the substrate surface to induce the interfacial polymerization, and drained off from the surface after 1-minute contact. The as-formed TFC membranes were dried at ambient condition for 1 minute and then stored in DI water before use. These GO-incorporated TFC membranes were denoted as TFC-0 (control), TFC-50ˈTFC-100ˈTFC-200, TFC-400, TFC-600 and TFC-800, respectively, here the number refers to the GO loading (ppm) in the MPD aqueous solution.

2.5. Characterization of TFC membranes

The water contact angle of the membrane surface was measured using a contact angle goniometer (DSA 25, KRÜSS, Germany) at room temperature by the sessile drop method, and calculated as an average data of at least five points. Atomic force microscope (SPM9700, Shimadzu, Japan) was used to investigate the surface topologies of the selective layer and obtain the mean roughness (Ra) of the TFC membranes under tapping mode in the air. SEM morphologies of the membrane surface and cross-section were observed with a scanning electron microscopy (SEM, VEGA3, TESCAN, Czech) and a field emission scanning electron microscope (FESEM, JSM-7600F, JEOL, Japan). Cross-sectional samples for FESEM observation were prepared by fracturing the freeze-dried membrane strips in liquid nitrogen. All 9

SEM and FESEM samples were coated with gold by a sputter coater (Q150RS, Quorum, England) before observation.

2.6. Pure water flux, water permeability, salt rejection, salt permeability

A lab-scale dead-end filtration setup was employed to evaluate the pure water flux (J, Lm-2h-1, referred to as LMH) and intrinsic water permeability (A, Lm−2h−1/bar, referred to as LMHbar−1), salt rejection rate (Rs, %) and salt permeability (B, L m−2 h−1, referred to as LMH) of the membranes under an applied pressure of 3 bar at room temperature. All membranes were stabilized firstly with the DI water at an applied pressure of 3 bar for 30 min and repeated for at least three times. The pure water flux and water permeability were obtained by using DI water as the feed, which determined by the Eq. (1) and (2). J=

∆

(1)

 ×∆

A = ∆!

(2)

where Am is the effective membrane area (17.5 cm2 in this study), ∆V is the permeate volume change over the test time ∆t, ΔP is the applied hydraulic pressure across the membrane. The salt rejection rate and salt permeability were measured using a 1000 ppm NaCl aqueous solution as the feed and calculated by Eq. (3) and (4). "# = $1 −

&'

*+,-

.

,-

&(

) × 100

= (∆!+∆2)

(3) (4)

where Cp and Cf are the salt concentrations in the feed and the permeate solutions, 10

respectively, determined by a conductivity meter (FE30, Mettler Toledo, Switzerland); and Δπ is the osmotic pressure of the feed solution.

2.7. Forward osmosis tests

The schematic diagram of the lab-scale FO test setup was shown in Fig. S-2. The feed and draw solutions employed in this work are DI water and aqueous NaCl solutions (0.5, 1.0 and 2.0 M) respectively, circulated using two variable speed pumps (BT300-2J, Longer, China) with the same flow speeds of 0.3 L min-1 (150 rpm). The solution concentration change was monitored by a conductivity meter (FE30, Mettler Toledo, Switzerland). All TFC membranes were tested in two different operational modes, i.e. FO mode (active layer facing the feed solution (AL–FS)) and pressure retarded osmosis (PRO) mode (active layer facing the draw solution (AL–DS)). Each test was pre-conditioned for 30 minutes to reach the steady state, and the data was collected every 30 min by a digital weight balance (FX3000-GD, AND, Japan) and outputted to a computer. FO tests of all membranes were performed at 25±0.5 °C and repeated for at least three times to yield an average value. The FO performance was evaluated by the water flux (Jv, L m−2 h−1, referred to as LMH) and the reverse salt flux (Js, g m−2 h−1, referred to as gMH), as calculated by Eqs. (5) and (6). 45 = 46 =

∆

 ∆ ∆(&7 7 )  ∆

(5) (6)

where ΔV is the volume change of draw solution over a predetermined test time (Δt), 11

Am is the effective membrane surface area (3.87 cm2 in this study), Ct and Vt are the salt concentration and the feed solution volume in the FO test, respectively. FO water flux could also be expressed by the Eqs. (9) and (10) using the internal concentration polarization (ICP) model. The structural parameter, S, can therefore be calculated by the two equations. FO mode: 8

×2; <.

45 = # 9: ×2

> < ? <.

(9)

PRO mode: 8

45 = # 9:

×2; + ? <. ×2> <.

(10)

where D is the solute diffusion coefficient, B8 and BC are the osmotic pressures of the draw and feed solution, respectively.

2.8. Antifouling test

The fouling test of the TFC membranes was conducted under FO mode at 25±0.5 °C, with sodium alginate (SA) as the organic foulant. Two kinds of synthetic wastewater about 1.5 L were employed as the feed solutions in this study. Feed solution 1 was prepared by dissolving 0.45 mM KH2PO4, 9.20 mM NaCl, 0.61 mM MgSO4, 0.5 mM NaHCO3, 0.5 mM CaCl2, and 0.93 mM NH4Cl in DI water with pH adjusted to 7.4; and feed solution 2 is of the same composition but with 250 mg/L sodium alginate added. Because of the large amount of draw solution (1.5 L) employed in this study, the effect of draw solution dilution is considered negligible. Firstly, a clean membrane sample was stabilized in the system with DI water as both 12

feed and draw solutions for 1 hour. The feed solution and draw solution were then switched to the feed solution 1 and 2 M NaCl aqueous solution in the second stage. This stage was conducted for 1 hour to stabilize the initial water flux before the feed solution 2 was used to start the accelerated fouling test for 18 hours. The flow rates of all feed and draw solutions were maintained at 0.3 L/min (150 rpm) for all above stages. After fouling test, the physical cleaning was performed immediately with a 15 mM NaCl solution circulated through the feed and draw sides for 30 minutes at a crossflow velocity of 0.6 L/min (300 rpm). The water flux of the cleaned membrane was measured again with FO test, using DI water and 2 M NaCl aqueous solution as the feed and draw solutions, respectively. The antifouling property of the membrane was evaluated in terms of the flux reduction ratio (FR%) and flux recovery ratio (FRR%) by Eqs. (7) and (8): D"% =

F+ 7

D""% =

F

G

F

× 100

× 100

(7) (8)

where J0 is the initial flux, Jt is the flux after accelerated fouling test, Jc is the water flux after the physical cleaning stage.

3.

Results and discussion

3.1. Characterization of synthesized GO

The successful synthesis of GO nano-sheets is confirmed by FTIR and XRD characterizations. The presence of O–H groups (3410 cm−1), C=O bonds in carboxyl 13

groups (1728 cm−1), unoxidized sp2 aromatic C=C bonds (1618 cm−1), C–O bonds in epoxy groups (1059 cm−1) can be confirmed in the FTIR spectrum of the synthesized GO (Fig. 1) (Jiang et al., 2014). The phase structure of the as-synthesized GO is also determined by its XRD pattern (Fig. 2). The sharp peak at 2θ = 11.52 ° corresponds to the (001) reflection of GO (Song et al., 2012). There is no obvious peak at about 26 ° which is ascribed to the characteristic peak of un-oxidized graphite (002) (Kuila et al., 2011), indicating that the synthesized GO was oxidized completely. The morphology of the GO nanosheets is also observed by SEM and AFM. The SEM image in Fig. 3 (a) shows the structure of GO agglomerates is multilayered with smooth surface and wrinkled edge. Fig. 3 (b) shows the AFM morphology of synthesized GO dispersed in water. The analyzed results (Fig. 3 (c)) show that the lateral size and thickness of the synthesized GO are 35–90 nm and 0.75–1.25 nm, respectively. According to the XRD result, the interlayer space of the synthesized GO nanosheets can be calculated by the Bragg's law as 0.76 nm. Therefore, the AFM results indicate that the synthesized GO is fully exfoliated to single or double layer.

3.2. Characterization of GO-incorporated TFC membranes

When GO is introduced in the aqueous solution with MPD, it may interact with MPD and TMC molecules during the interfacial polymerization. The possible reaction mechanism is displayed in Fig. 4. As shown, the amine groups of MPD may interact with GO to form new amide bonds and hydrogen bonds during ultrasonication of the aqueous solution in step 1. In step 2, when the TMC solution contacts the membrane, the interfacial polymerization occurs, where acyl chloride groups of TMC would not only react with MPD, but also with GO to form anhydride and ester groups. 14

This reaction mechanism could be confirmed by the FTIR characterization of the control and GO-embedded TFC membranes in the Fig. 5. Firstly, distinct characteristic peaks of the formed amide groups by interfacial polymerization can be observed in all spectra at 1658 cm-1 and 1543 cm-1 (Belfer et al., 1998; Fulmer and Wynne, 2011). The former could be ascribed to the carbonyl stretching vibration of the amide I group, while the later could due to the coupling of the in-plane N-H bending and C-N stretching vibrations of the amide II group. What’s more, an slight enhancement in the peak intensity at 1658 cm-1 should be ascribed to the formation of the new amide linkages by the reaction of the –COOH groups in GO with –NH2 groups in MPD (Bano et al., 2015a). Besides that, the gradual increment in the peak intensity is observed at 2925 cm-1 and 2856 cm-1 with the GO incorporation, which are due to the asymmetric and symmetric stretching vibrations of more C–H bonds. Additionally, compared to the control TFC membrane, the GO-incorporated TFC membranes show an intense and wider absorption band at 3341 cm-1, which could be attributed to the stretching vibration of hydroxyl groups with GO incorporation. All above changes indicate the successful incorporation of GO nanosheets into the polyamide structure during the interfacial polymerization TGA result in Fig. 6 further confirms the reaction mechanism. The PA curve shows the first sharp weight loss at about 100 °C due to the remaining solvent/adsorbed water. When the temperature increases above 300 °C, the rate of weight loss slows down till the main decomposition stage at about 600 °C. The curve of GO shows the weight loss at the first and second stage around 180 °C and 450 °C ascribed to the removal of oxygen functional groups (-COOH, -OH and epoxide) in the GO sheets and carbon oxidation, respectively (Fang et al., 2010). Relatively, the weight loss of 15

the first decomposition stage of PA@GO composite is much less than that of GO, which may be ascribed to the fact that a large amount of oxygen-containing groups in GO have reacted with the TMC and MPD molecules. In addition, the first and the second decomposition stage in the PA@GO are also observed to shift to higher temperatures, possibly because of the successful bonding of the GO nanosheets to the PA polymer chains. Similarly, in comparison to of the pristine PA, the main decomposition of the PA@GO due to main chain pyrolysis also shifts to a higher temperature around 650 °C and shows to be of less weight loss, possibly because of the covalent bond between GO nanosheets and PA polymer chains. The surface and cross-sectional morphology of the control TFC and GO-incorporated TFC membranes are also investigated and shown in Fig. 7. Typical ridge-and-valley surface structures are observed in all membranes, but the addition of GO into the selective layer can greatly affect the membrane surface morphology. Compared to the rough surface of the control TFC membrane, the GO-incorporated TFC membranes exhibit denser and smoother surfaces. This could be further proved by their AFM images in Fig. 9. Clearly, the GO-incorporated TFC membranes exhibit smoother surfaces with lower surface roughness than that of the control TFC membrane. Meanwhile, the ridge height of the control TFC membranes is also much higher than those of GO-incorporated TFC membranes. According to previous works, the smoother surfaces of GO-incorporated membranes may be contributed by the following three factors. Firstly, when the HPAN substrate is taken out from the aqueous solution vertically, GO nanosheets tend to orient along the membrane surface 16

horizontally because of the Langmuir–Blodgett film deposition (Chae et al., 2015; Yang et al., 2014). The MPD diffusion usually leads to the ridge formation, while the horizontally oriented GO nanosheets would retard the diffusion of MPD into the organic phase, leading to a smoother surface. Secondly, oxygen-containing functional groups of GO could react with the MPD and TMC, thus the competition would affect the reaction rate between MPD and TMC (Xia et al., 2015). Thirdly, hydrogen bonds present in the hydroxyl groups of GO can contribute to a more compact chain structure (Bano et al., 2015). The average thickness of PA selective layer is calculated by measuring the average height of the PA layer in the control and GO-incorporated membranes according to their SEM cross-sectional morphology (Fig. 7), and the results are presented in Fig. 8. It shows that the PA layer thickness decreases with the increase in the GO loading.

3.3. Intrinsic separation properties of control and GO-incorporated TFC membranes

Table 1 summarizes the fundamental parameters of the control and GO-incorporated TFC membranes, which are generally used to represent the intrinsic performance of the FO membrane regardless of the external operation conditions. It can be found that, the water permeability (A) of the GO-incorporated TFC membranes (2.20±0.03 LMHbar−1) is approximately two times that of the control TFC membranes (1.12±0.02 LMHbar−1), which may be attributed to the following several factors. Firstly, the incorporation of GO in the thin selective layer increases the surface hydrophilicity of 17

the membranes, which can attract water molecules into the membrane matrix and facilitates their transport through the membrane (Zhao et al., 2013). Secondly, the incorporation of GO reduces the thickness of the selective layer, contributing to a lower transport resistance and an enhanced water flux. Thirdly, it may provide additional transport passages (i.e. the nano-channels) for water molecules by the interfacial gap between GO nanosheets and polyamide (Hu and Mi, 2013; Mahmoud et al., 2015). The fractional free volume may also increase in the membrane matrix due to the disrupted polymer chain packing by the addition of GO (Bano et al., 2015b). What’s more, graphene pores functionalized with hydroxylated groups were also reported to provide a surface with less friction, resulting in an overall faster flow of water molecules (Wang and Karnik, 2012). The enhanced hydrophilicity of GO-incorporated TFC membranes could be confirmed by the water contact angle result in the Fig. 10. As shown, the water contact angle of the TFC membrane gradually decreases with the increase in the GO loading in the selective layer, indicating that the incorporated GO with oxygen-containing functional groups could enhance the hydrophilicity of the membrane significantly, and contribute to a greater water flux. Compared to the control TFC membrane (82.23±0.43), the GO-incorporated TFC membranes also exhibit higher salt rejections (82.63±0.67, 93.83±0.71, 94.60±0.76, 91.93±0.53, 86.43±0.75) because of their denser skin layers. While the TFC-800 membrane (81.53±0.46) has slight lower salt rejection (81.53±0.46), possibly due to the defects formed in the selective layer by GO agglomeration during the interfacial 18

polymerization (Amini et al., 2013). That is consistent with the B/A ratio which directly indicates the membrane selectivity. Accordingly, the salt permeability (B) decreases initially up to 200 pm GO loading and then increases with higher GO loading. Another important index of the FO membrane, the structure parameter (S), is determined by the classical ICP model developed by Loeb et al (Eqs. (9) and (10)), and also listed in the Table 1. Generally, the GO-incorporated membranes with higher water permeability have lower S values, ascribed to their higher hydrophilicity and less ICP effect. In summary, above results prove that the GO-incorporated TFC membranes exhibit higher water permeability and reasonable salt rejection as compared to the control TFC membrane, and are expected to have a superior FO performance.

3.4. Effect of GO loading on the FO performance

Fig. 11 shows the FO performance of TFC membranes with different GO loadings. It shows that the water fluxes of all GO-incorporated TFC membranes are higher than that of the control TFC membranes in both operation modes. And the water flux increases with the increase in the GO loading, showing a consistent trend with that of the water permeability exhibited in Table 1. Similarly, this enhancement should be attributed to the combined effect of the enhanced hydrophilicity, thinner selective layer, additional passages formed, etc., as mentioned in section 3.3. And, the water fluxes in FO mode vary from 21.6±0.4 to 35.4±0.7 LMH, lower than those in PRO 19

mode (31.1±0.9 to 56.6±0.8 LMH), due to the severer ICP effect in the former. The reverse salt fluxes of the TFC membranes are also shown in Fig. 11 (b). Basically, except for the TFC-800 membrane, the reverse salt fluxes in the GO-incorporated TFC membranes are all lower than that of the control TFC membranes. However, the reverse salt flux decreases initially up to 200 ppm GO loading and then increases with higher GO loading from 50 to 800 ppm in the selective layer, which is in accordance with the trend of the salt rejection change. Therefore, a GO loading of 400-600 ppm in the selective layer could be the optimal loading to produce promising FO performance of TFC membranes. Meanwhile, the effect of the draw solution concentration on the FO performance of TFC-400 and TFC-600 TFC membranes is also studied. From Fig. 12 (a) and (b), it is noted that both water flux and reverse salt flux of the two membranes exhibit similar increasing trends with the increase in the draw solution concentration in both PRO and FO modes. The water flux increases steadily as the draw solution concentration increases, attributed to the larger osmotic driving force supplied by the draw solution of higher concentration.

3.5 Dynamic bio-antifouling performance of GO incorporated TFC membranes

Fig. 13 shows the FO fouling test result of the control TFC and GO-incorporated TFC membranes. It can be found that the water flux decline of GO-incorporated TFC membranes with time is much slower than that of the control TFC membrane, suggesting the lower fouling propensity. It was reported that a cross-linked alginate 20

gel layer will form on the membrane surface with the existence of Ca2+ ions in the synthetic wastewater which acts as the “bridges” between alginate molecules (Mi and Elimelech, 2008; Tiraferri et al., 2012a). This gel layer will lead to an increased transport resistance, and significantly decrease the water flux. This phenomenon suggests that a thinner and/or discontinuous SA gel layer may form on the surface of the GO-incorporated TFC membrane because of its hydrophilicity, negative charge, and surface smoothness. The hydrophobic foulants, such as SA and BSA, can easily adsorb on the hydrophobic surfaces to minimize the interfacial energy, but doesn’t show any significant thermodynamic advantage for the high hydrophilic surface. What’s more, strongly bounded water molecules can easily adhere to the hydrophilic surface via hydrogen bonding, and bring out a thin water boundary layer as a barrier to restrain adsorption of hydrophobic foulants. In addition, the foulant SA is negatively charged, while the membrane surface of the GO-incorporated TFC membranes possess many negatively charged functional groups, such as carboxyl, hydroxyl and epoxide groups. The electrostatic repulsion of the membrane surface may therefore exist against the extracellular polymeric substance which is in close association with biofilm formation, such as SA. Lastly, the GO-incorporated TFC membranes with smoother surface can potentially avoid the accumulation of SA molecules on the polyamide due to less adhesion sites on the membrane surface, thus contributing to the improved antifouling property. The final water flux after the fouling test and cleaning are further normalized by the initial water flux and the results are presented in Fig. 14. The flux recovery ratio (FRR) 21

is generally employed to assess the reversibility of SA fouling. We can see that, the water flux of the control TFC membrane exhibits a sharp drop to about 50±3.7% after fouling and a low FRR value of 60.95±2.16% after cleaning. Alternatively, the water flux drops of the GO-incorporated TFC membranes are much less, and all FRRs exceed 90% with a highest FRR value of 97.5±0.65% of the TFC-800 membrane. The results demonstrate that the fouling of the GO-incorporated TFC membranes is mainly reversible, but that of the control TFC membranes is typically partial reversible.

4.

Conclusion

In this study, a novel GO-incorporated TFC membranes is developed by interfacial polymerization process, using an MPD/GO mixed solution as the aqueous phase. The synthesized GO is characterized by the FTIR, XRD and SEM to confirm its successful synthesis and morphology. The GO-incorporated TFC membranes exhibit enhanced hydrophilicity and surface smoothness, as well as thinner selective layer, because of the affected reaction rate between TMC and MPD molecules with the addition of GO. Various characterizations were employed to elucidate the changes in the chemical structure and membrane morphology of the GO-incorporated TFC membranes. As a result, the GO-incorporated TFC membranes basically possess higher water permeability, lower salt permeability, higher salt rejection, lower structural parameter as compared to the control TFC membrane. But the salt permeability decreases initially up to 200 ppm GO loading and then increases with higher GO loading from 22

50 to 800 ppm, due to the severer GO agglomeration and defect formation. With the GO loading increase, the water flux of the GO-incorporated TFC membranes is much higher than that of the control TFC membranes, while the reverse salt flux is lower except for the TFC-800 membrane. The fouling test also demonstrates that the embedment of GO in the PA layer can sufficiently suppress the undesired fouling phenomenon and the fouling of the GO incorporated TFC membranes was almost reversible. The GO-incorporated TFC membranes with desirable performance may hold a good prospect for FO applications in the future. However, when GO loading reaches a certain amount, GO agglomeration may occur, leading to the defect formation on the membrane surface. It is necessary to improve the process of synthesized GO and the preparation of GO-incorporated PA TFC membranes. These possible solutions could include: (1) higher oxidation degree of the synthesized GO to obtain a smaller size; (2) chemical modification of GO to improve its dispersibility in the aqueous solution; (3) exfoliation of GO by ultrasonication with higher frequency and power.

Acknowledgement

We thank the financial support from National Natural Science Foundation of China (no. 21306058), Huazhong University of Science and Technology, China (nos. 0124013041 and 2014YQ012), and “Thousand Youth Talent Plan”. Special thanks are also due to the Analysis and Testing Center and the State Key Laboratory of Materials Processing and Die & Mould Technology, Huazhong University of Science and 23

Technology for their help with material characterizations.

List of abbreviations and nomenclatures

Abbreviations

CaCl2

: calcium chloride

CNT

: carbon nano-tube

DMF

: N,N-dimethylformamide

DI

: deionized

FESEM

: field emission scanning electron microscope

FTIR

: fourier transform infrared

FO

: forward osmosis

GO

: graphene oxide

IP

: interfacial polymerization

HCl

: hydrochloric acid

H2O2

: hydrogen

HPAN

: hydrolyzed polyacrylonitrile

H2SO4

: sulfuric acid

KH2PO4

: potassium dihydrogen phosphate

KMnO4

: potassium

LBL

: layer-by-layer

LiCl

: lithium chloride

MgSO4

: magnesium sulfate

MMM

: mixed matrix membrane

peroxide

permanganate

24

MnO2

: manganese dioxide

MPD

: m-phenylenediamine

NaCl

: sodium chloride

NaHCO3

: sodium bicarbonate

NaNO3

: sodium nitrate

NaOH

: sodium hydroxide

NF

: nanofiltration

NH4Cl

: ammonium chloride

ODMP

: osmotic driven membrane process

PA

: polyamide

PAN

: polyacrylonitrile

PDMP

: pressure driven membrane process

RO

: reverse osmosis

SA

: sodium alginate

SEM

: scan electron microscopy

TEM

: transmission electron microscopy

TFC

: thin-film composite

TMC

: 1,3,5-trimesoyl chloride

XRD

: X-ray diffractometer

ZIF

: zeolitic imidazolate framework

Nomenclatures

A

: water permeability

AL-DS

: active layer facing draw solution

AL-FS

: active layer facing feed solution 25

Am

: effective membrane area

B

: salt permeability

Cf

: feed concentration

Cp

: permeate

Ct

: salt concentration

D

: diffusion coefficient

FR%

: flux reduction ratio

FRR%

: flux recovery ratio

ICP

: internal concentration polarization

J

: pure water flux

Js

: reverse salt flux

Jv

: water flux

MWCO

: molecular weight cut off

PRO

: pressure retarded osmosis

PWP

: pure water permeability

Rs

: salt rejection

ΔP

: hydraulic pressure

∆t

: test time

∆V

: volume change

Δπ

: osmotic pressure

B8

: draw solution osmotic pressure

BC

: feed solution osmotic pressure

concentration

26

References

Achilli, A., Cath, T.Y., Marchand, E.A., Childress, A.E., 2009. The forward osmosis membrane bioreactor: a low fouling alternative to MBR processes. Desalination 239, 10-21. Amini, M., Jahanshahi, M., Rahimpour, A., 2013. Synthesis of novel thin film nanocomposite (TFN) forward osmosis membranes using functionalized multi-walled carbon nanotubes. Journal of Membrane Science 435, 233-241. Azari, S., Zou, L., 2012. Using zwitterionic amino acid l-DOPA to modify the surface of thin film composite polyamide reverse osmosis membranes to increase their fouling resistance. Journal of Membrane Science 401, 68-75. Bano, S., Mahmood, A., Kim, S.-J., Lee, K.-H., 2015a. Graphene oxide modified polyamide nanofiltration membrane with improved flux and antifouling properties. Journal of Materials Chemistry A 3, 2065-2071. Bano, S., Mahmood, A., Kim, S.-J., Lee, K.-H., 2015b. Graphene oxide modified polyamide nanofiltration membrane with improved flux and antifouling properties. J. Mater. Chem. A 3, 2065-2071. Belfer, S., Purinson, Y., Kedem, O., 1998. Surface modification of commercial polyamide reverse osmosis membranes by radical grafting: An ATRϋFTIR study. Acta polymerica 49, 574-582. Bernstein, R., Belfer, S., Freger, V., 2011. Bacterial attachment to RO membranes surface-modified by concentration-polarization-enhanced graft polymerization. Environ Sci Technol 45, 5973-5980. Chae, H.-R., Lee, J., Lee, C.-H., Kim, I.-C., Park, P.-K., 2015. Graphene oxide-embedded thin-film composite reverse osmosis membrane with high flux, anti-biofouling, and chlorine resistance. Journal of Membrane Science 483, 27

128-135. Chang, X., Wang, Z., Quan, S., Xu, Y., Jiang, Z., Shao, L., 2014. Exploring the synergetic effects of graphene oxide (GO) and polyvinylpyrrodione (PVP) on poly(vinylylidenefluoride)

(PVDF)

ultrafiltration

membrane

performance.

Applied Surface Science 316, 537-548. Choi, W., Choi, J., Bang, J., Lee, J.-H., 2013. Layer-by-layer assembly of graphene oxide nanosheets on polyamide membranes for durable reverse-osmosis applications. ACS Appl Mater Interfaces 5, 12510-12519. Cui, Y., Liu, X.-Y., Chung, T.-S., 2014. Enhanced osmotic energy generation from salinity gradients by modifying thin film composite membranes. Chemical Engineering Journal 242, 195-203. Dong, H., Zhao, L., Zhang, L., Chen, H., Gao, C., Winston Ho, W.S., 2015. High-flux reverse osmosis membranes incorporated with NaY zeolite nanoparticles for brackish water desalination. Journal of Membrane Science 476, 373-383. Duan, J., Pan, Y., Pacheco, F., Litwiller, E., Lai, Z., Pinnau, I., 2015. High-performance

polyamide

thin-film-nanocomposite

reverse

osmosis

membranes containing hydrophobic zeolitic imidazolate framework-8. Journal of Membrane Science 476, 303-310. Emadzadeh, D., Lau, W.J., Rahbari-Sisakht, M., Daneshfar, A., Ghanbari, M., Mayahi, A., Matsuura, T., Ismail, A.F., 2015. A novel thin film nanocomposite reverse osmosis membrane with superior anti-organic fouling affinity for water desalination. Desalination 368, 106-113. Fang, M., Wang, K., Lu, H., Yang, Y., Nutt, S., 2010. Single-layer graphene nanosheets with controlled grafting of polymer chains. Journal of Materials Chemistry 20, 1982-1992. 28

Fulmer, P.A., Wynne, J.H., 2011. Development of broad-spectrum antimicrobial latex paint surfaces employing active amphiphilic compounds. ACS Appl Mater Interfaces 3, 2878-2884. Ghanbari, M., Emadzadeh, D., Lau, W.J., Matsuura, T., Davoody, M., Ismail, A.F., 2015. Super hydrophilic TiO2/HNT nanocomposites as a new approach for fabrication of high performance thin film nanocomposite membranes for FO application. Desalination 371, 104-114. Han, Y., Xu, Z., Gao, C., 2013. Ultrathin Graphene Nanofiltration Membrane for Water Purification. Advanced Functional Materials 23, 3693-3700. Hu, M., Mi, B., 2013. Enabling graphene oxide nanosheets as water separation membranes. Environ Sci Technol 47, 3715-3723. Huang, H., Qu, X., Ji, X., Gao, X., Zhang, L., Chen, H., Hou, L., 2013. Acid and multivalent ion resistance of thin film nanocomposite RO membranes loaded with silicalite-1 nanozeolites. Journal of Materials Chemistry A 1, 11343. Huang, K., Liu, G., Lou, Y., Dong, Z., Shen, J., Jin, W., 2014. A graphene oxide membrane with highly selective molecular separation of aqueous organic solution. Angew Chem Int Ed Engl 53, 6929-6932. Jiang, Z., Jiang, Z.-j., Tian, X., Chen, W., 2014. Amine-functionalized holey graphene as a highly active metal-free catalyst for the oxygen reduction reaction. Journal of Materials Chemistry A 2, 441-450. Joshi, R., Carbone, P., Wang, F., Kravets, V., Su, Y., Grigorieva, I., Wu, H., Geim, A., Nair, R., 2014. Precise and ultrafast molecular sieving through graphene oxide membranes. Science 343, 752-754. Kim, S.G., Hyeon, D.H., Chun, J.H., Chun, B.-H., Kim, S.H., 2013. Novel thin nanocomposite RO membranes for chlorine resistance. Desalination and Water 29

Treatment 51, 6338-6345. Kuila, T., Khanra, P., Bose, S., Kim, N.H., Ku, B.-C., Moon, B., Lee, J.H., 2011. Preparation of water-dispersible graphene by facile surface modification of graphite oxide. Nanotechnology 22, 305710. Li, X., Zhang, S., Fu, F., Chung, T.-S., 2013. Deformation and reinforcement of thin-film composite (TFC) polyamide-imide (PAI) membranes for osmotic power generation. Journal of Membrane Science 434, 204-217. Liu, S., Zeng, T.H., Hofmann, M., Burcombe, E., Wei, J., Jiang, R., Kong, J., Chen, Y., 2011. Antibacterial activity of graphite, graphite oxide, graphene oxide, and reduced graphene oxide: membrane and oxidative stress. ACS Nano 5, 6971-6980. Liu, T.-Y., Bian, L.-X., Yuan, H.-G., Pang, B., Lin, Y.-K., Tong, Y., Van der Bruggen, B., Wang, X.-L., 2015a. Fabrication of a high-flux thin film composite hollow fiber nanofiltration membrane for wastewater treatment. Journal of Membrane Science 478, 25-36. Liu, T.-Y., Liu, Z.-H., Zhang, R.-X., Wang, Y., Bruggen, B.V.d., Wang, X.-L., 2015b. Fabrication of a thin film nanocomposite hollow fiber nanofiltration membrane for wastewater treatment. Journal of Membrane Science 488, 92-102. Lu, X., Romero-Vargas Castrillon, S., Shaffer, D.L., Ma, J., Elimelech, M., 2013. In situ surface chemical modification of thin-film composite forward osmosis membranes for enhanced organic fouling resistance. Environ Sci Technol 47, 12219-12228. Mahmoud, K.A., Mansoor, B., Mansour, A., Khraisheh, M., 2015. Functional graphene nanosheets: The next generation membranes for water desalination. Desalination 356, 208-225. 30

Martinetti, C.R., Childress, A.E., Cath, T.Y., 2009. High recovery of concentrated RO brines using forward osmosis and membrane distillation. Journal of Membrane Science 331, 31-39. McGinnis, R.L., Elimelech, M., 2007. Energy requirements of ammonia–carbon dioxide forward osmosis desalination. Desalination 207, 370-382. Mi, B., Elimelech, M., 2008. Chemical and physical aspects of organic fouling of forward osmosis membranes. Journal of Membrane Science 320, 292-302. Mi, B., Elimelech, M., 2010. Organic fouling of forward osmosis membranes: fouling reversibility and cleaning without chemical reagents. Journal of Membrane Science 348, 337-345. Mi, Y.-F., Zhao, Q., Ji, Y.-L., An, Q.-F., Gao, C.-J., 2015. A novel route for surface zwitterionic functionalization of polyamide nanofiltration membranes with improved performance. Journal of Membrane Science 490, 311-320. Nair, R., Wu, H., Jayaram, P., Grigorieva, I., Geim, A., 2012. Unimpeded permeation of water through helium-leak–tight graphene-based membranes. Science 335, 442-444. Ong, R.C., Chung, T.-S., 2012. Fabrication and positron annihilation spectroscopy (PAS) characterization of cellulose triacetate membranes for forward osmosis. Journal of Membrane Science 394-395, 230-240. Perreault, F.o., Tousley, M.E., Elimelech, M., 2013. Thin-film composite polyamide membranes

functionalized

with

biocidal

graphene

oxide

nanosheets.

Environmental Science & Technology Letters 1, 71-76. Qi, S., Li, W., Zhao, Y., Ma, N., Wei, J., Chin, T.W., Tang, C.Y., 2012. Influence of the properties of layer-by-layer active layers on forward osmosis performance. Journal of Membrane Science 423-424, 536-542. 31

Rana, D., Matsuura, T., 2010. Surface modifications for antifouling membranes. Chem Rev 110, 2448-2471. Safarpour, M., Khataee, A., Vatanpour, V., 2015. Thin film nanocomposite reverse osmosis membrane modified by reduced graphene oxide/TiO2 with improved desalination performance. Journal of Membrane Science 489, 43-54. Saren, Q., Qiu, C.Q., Tang, C.Y., 2011. Synthesis and characterization of novel forward osmosis membranes based on layer-by-layer assembly. Environ Sci Technol 45, 5201-5208. Shen, J.n., Yu, C.c., Ruan, H.m., Gao, C.j., Van der Bruggen, B., 2013. Preparation and characterization of thin-film nanocomposite membranes embedded with poly(methyl methacrylate) hydrophobic modified multiwalled carbon nanotubes by interfacial polymerization. Journal of Membrane Science 442, 18-26. Song, P., Zhang, X., Sun, M., Cui, X., Lin, Y., 2012. Synthesis of graphene nanosheets via oxalic acid-induced chemical reduction of exfoliated graphite oxide. RSC Advances 2, 1168-1173. Sorribas, S., Gorgojo, P., Tellez, C., Coronas, J., Livingston, A.G., 2013. High flux thin film nanocomposite membranes based on metal-organic frameworks for organic solvent nanofiltration. J Am Chem Soc 135, 15201-15208. Su, J., Zhang, S., Chen, H., Chen, H., Jean, Y.C., Chung, T.-S., 2010. Effects of annealing on the microstructure and performance of cellulose acetate membranes for pressure-retarded osmosis processes. Journal of Membrane Science 364, 344-353. Sukitpaneenit, P., Chung, T.S., 2012. High performance thin-film composite forward osmosis hollow fiber membranes with macrovoid-free and highly porous structure for sustainable water production. Environ Sci Technol 46, 7358-7365. 32

Sun, Y., Wu, Q., Shi, G., 2011. Graphene based new energy materials. Energy & Environmental Science 4, 1113-1132. Tiraferri, A., Kang, Y., Giannelis, E.P., Elimelech, M., 2012a. Superhydrophilic thin-film composite forward osmosis membranes for organic fouling control: fouling behavior and antifouling mechanisms. Environ Sci Technol 46, 11135-11144. Tiraferri, A., Kang, Y., Giannelis, E.P., Elimelech, M., 2012b. Superhydrophilic thin-film composite forward osmosis membranes for organic fouling control: fouling behavior and antifouling mechanisms. Environ Sci Technol 46, 11135-11144. Tu, Y., Lv, M., Xiu, P., Huynh, T., Zhang, M., Castelli, M., Liu, Z., Huang, Q., Fan, C., Fang, H., 2013. Destructive extraction of phospholipids from Escherichia coli membranes by graphene nanosheets. Nat Nanotechnol 8, 594-601. Wang, E.N., Karnik, R., 2012. Water desalination: Graphene cleans up water. Nat Nanotechnol 7, 552-554. Wang, L., Pu, K.Y., Li, J., Qi, X., Li, H., Zhang, H., Fan, C., Liu, B., 2011. A Graphene–Conjugated Oligomer Hybrid Probe for LightϋUp Sensing of Lectin and Escherichia Coli. Adv Mater 23, 4386-4391. Wang, Y., Ou, R., Wang, H., Xu, T., 2015. Graphene oxide modified graphitic carbon nitride as a modifier for thin film composite forward osmosis membrane. Journal of Membrane Science 475, 281-289. Wang, Y., Xu, T., 2015. Anchoring hydrophilic polymer in substrate: An easy approach for improving the performance of TFC FO membrane. Journal of Membrane Science 476, 330-339. Xu, Z., Zhang, J., Shan, M., Li, Y., Li, B., Niu, J., Zhou, B., Qian, X., 2014. 33

Organosilane-functionalized graphene oxide for enhanced antifouling and mechanical properties of polyvinylidene fluoride ultrafiltration membranes. Journal of Membrane Science 458, 1-13. Yu, H.-Y., Kang, Y., Liu, Y., Mi, B., 2014. Grafting polyzwitterions onto polyamide by click chemistry and nucleophilic substitution on nitrogen: A novel approach to enhance membrane fouling resistance. Journal of Membrane Science 449, 50-57. Zhang, S., Fu, F., Chung, T.-S., 2013. Substrate modifications and alcohol treatment on thin film composite membranes for osmotic power. Chemical Engineering Science 87, 40-50. Zhang, S., Wang, K.Y., Chung, T.-S., Jean, Y.C., Chen, H., 2011. Molecular design of the cellulose ester-based forward osmosis membranes for desalination. Chemical Engineering Science 66, 2008-2018. Zhao, C., Xu, X., Chen, J., Yang, F., 2013. Effect of graphene oxide concentration on the morphologies and antifouling properties of PVDF ultrafiltration membranes. Journal of Environmental Chemical Engineering 1, 349-354. Zhao, H., Qiu, S., Wu, L., Zhang, L., Chen, H., Gao, C., 2014. Improving the performance of polyamide reverse osmosis membrane by incorporation of modified multi-walled carbon nanotubes. Journal of Membrane Science 450, 249-256.

34

List of Tables

Table 1.

Transport properties of the control TFC and GO incorporated TFC membranes.

List of Figures

Fig. 1.

FTIR spectrum of synthesized GO.

Fig. 2.

X-ray diffraction pattern of the synthesized GO.

Fig. 3.

(a) SEM and (b) TEM images of synthesized GO.

Fig. 4.

Schematic illustration of interactions of GO with TMC and MPD.

Fig. 5.

FTIR spectra of TFC-0 and TFC-400 membranes.

Fig. 6.

TGA curves of GO, PA, PA@GO-400.

Fig. 7.

SEM images of surface and cross-sectional morphologies of control and GO-incorporated TFC membranes.

Fig. 8.

AFM images of surface morphologies of (a) TFC-0, (b) TFC-200, (c) TFC-400, (d) TFC-600, and (e) TFC-800 membranes.

Fig. 9.

Water contact angles of the control TFC and GO-incorporated TFC membranes.

Fig. 10.

(a) Water flux and (b) reverse salt flux of the control and GO-incorporated TFC membranes.

Fig. 11.

Water flux and reverse salt flux of the TFC-400 and TFC-600 membranes against different concentration of draw solution under (a) FO mode and (b) PRO mode. 35

Fig. 12.

Forward osmosis fouling test of the control and GO-incorporated TFC membranes. (Jw/Jw,0 ratio were taken with a 10-minutue interval during the fouling test.)

Fig. 13.

Forward osmosis fouling test results of the control TFC and GO incorporated TFC membranes.

Table 1. Intrinsic transport properties of the control and GO-incorporated TFC membranes. Membrane Rejection B/A, Aa, LMH/Bar Bb, LMH Sc, mm ID Rs, % Bar

TFC-0

1.12±0.02

0.67±0.02

82.23±0.43

0.56

0.109

TFC-50

1.23±0.08

0.66±0.01

82.63±0.67

0.54

0.108

TFC-100

1.58±0.07

0.27±0.02

93.83±0.71

0.17

0.106

TFC-200

1.66±0.03

0.24±0.03

94.60±0.76

0.15

0.104

TFC-400

1.93±0.04

0.44±0.04

91.93±0.53

0.23

0.099

TFC-600

2.04±0.05

0.83±0.07

86.43±0.75

0.40

0.085

TFC-800

2.20±0.03

1.29±0.05

81.53±0.46

0.58

0.083

a

DI water is used as the feed solution in the RO test with an applied pressure of 3 bar (2.5 rpm); b 1000 ppm NaCl solution is used as the feed solution in the RO test with an applied pressure of 3 bar (2.5 rpm); c Calculated based on the FO test under FO mode using 0.5 M NaCl as the draw solution and DI water as the feed solution.

36

Research Highlights for the manuscript “Graphene oxide incorporated thin-film composite membranes for forward osmosis applications” by Liang Shen, Shu Xiong, Yan Wang

·

TFC FO membranes are fabricated with GO incorporated PA layer.

·

GO nanosheets are covalently bonded to the polyamide chains.

·

GO-incorporated TFC membranes exhibit enhanced hydrophilicity and water flux.

·

GO-incorporated TFC membranes also exhibit low fouling propensity.

·

The optimized GO loading in the PA layer is about 400-600 ppm.

37

Transmittance(%)

1059 1728 3410 1618

4500 4000 3500 3000 2500 2000 1500 1000 500

Wavenumbers(cm-1) Fig. 1 FTIR spectrum of synthesized GO.

Intensity 10

15

20

25 30 35 2 Theta(°)

40

45

Fig. 2 X-ray diffraction pattern of the synthesized GO.

50

a

b

c

Fig. 3 (a) SEM image, (b) AFM image and (c) lateral size and thickness of synthesized GO

(1)

(2-1)

(2-2)

(2-3)

Fig. 4 Schematic illustration of interactions of GO with TMC and MPD.

Transmittance(%)

TFC-0

TFC-GO

2856 2925

3341

1658

1543

4000 3500 3000 2500 2000 1500 1000 500

Wavenumbers(cm-1) Fig. 5. FTIR spectra of TFC-0 and TFC-GO membranes.

Weight(%)

100 90 80 70 60

PA@GO GO PA

50 0

100 200 300 400 500 600 700 800 900

Temperature(℃) Fig. 6 TGA curves of GO, PA, PA@GO.

(A) TFC-0

TFC-200

TFC-50

TFC-400

TFC-100

TFC-600

TFC-800

(B) TFC-0

TFC-200

TFC-50

TFC-400

TFC-100

TFC-600

TFC-800

Fig. 7 SEM images of surface (A) and cross-sectional (B) morphologies of control and GO-incorporated TFC membranes.

280

Thickness (nm)

240 200 160 120 80 0

200

400

600

800

Conc.GO (ppm) Fig. 8 Average thickness of PA selective layer of control and GO-incorporated membranes.

a, Ra=60.5 nm

e, Ra=27.4 nm

b, Ra=56.3 nm

c, Ra=48.8 nm

f Ra=20.1 nm

d, Ra=46.2 nm

g, Ra=14.0 nm

Fig. 9 AFM images of surface morphologies of (a) TFC-0, (b) TFC-50, (c) TFC-100, (d) TFC-200, (e) TFC-400, (f) TFC-600, and (g) TFC-800 membranes.

75

WCA (°)

70 65 60 55 50 0

200

400

600

800

Conc.GO (ppm) Fig. 10 Water contact angles of the control TFC and GO-incorporated TFC membranes.

(a)

FO PRO

60

Jv (LMH)

50 40 30 20 10 0 0

200

400

600

800

Conc.GO (ppm) (b)

16 14

FO PRO

Js (gMH)

12 10 8 6 4 2 0 0

200

400

600

800

Conc.GO (ppm) Fig. 11 (a) Water flux and (b) reverse salt flux of the control and GO-incorporated TFC membranes. (2 M NaCl solution and DI water were used as draw solution and feed solution, respectively.)

(a)

30

Jv(LMH)

25

8

Jv-400 Jv-600 Js-400 Js-600

7 6 5

20

4

15

3

10

2

5

1

0

0

-5

Js(gMH)

35

-1 0.0

0.5

1.0

1.5

2.0

Conc.DS (M) (b)

Jv(LMH)

40

12 10 8

30

6

20

4 10

Js(gMH)

50

14

Jv-400 Jv-600 Js-400 Js-600

2

0

0 0.0

0.5

1.0

1.5

2.0

Conc.DS (M) Fig. 12 Water flux and reverse salt flux of the TFC-400 and TFC-600 membranes against different concentration of draw solution under (a) FO mode and (b) PRO mode. (NaCl solution and DI water were used as draw solution and feed solution, respectively.)

Normalized water flux Jw/Jw,0(%)

100 90 80 70 TFC-800 TFC-600 TFC-400 TFC-200 TFC-0

60 50 0

200

400

600

800

1000 1200

Time (min) Fig. 13 Forward osmosis fouling test of the control and GO-incorporated TFC membranes. (2 M NaCl solution and synthetic wastewater were used as draw solution and feed solution, respectively; the Jw/Jw,0 ratio was taken with a 10-minutue interval during the fouling test.)

Normalized water flux Jw/Jw,0 (%)

100

100 After fouling After cleaning

95

95

90

90

85

85

80

80

75

75

70

70

65

65

60

60

55

55

50

50

45

0

200

400

600

800

45

Conc.GO (ppm) Fig. 14 Forward osmosis fouling test results of the control TFC and GO incorporated TFC membranes. (2 M NaCl solution and synthetic wastewater were used as draw solution and feed solution, respectively.)