Novel thin film composite hollow fiber membranes incorporated with carbon quantum dots for osmotic power generation

Novel thin film composite hollow fiber membranes incorporated with carbon quantum dots for osmotic power generation

Journal of Membrane Science 551 (2018) 94–102 Contents lists available at ScienceDirect Journal of Membrane Science journal homepage: www.elsevier.c...

3MB Sizes 2 Downloads 123 Views

Journal of Membrane Science 551 (2018) 94–102

Contents lists available at ScienceDirect

Journal of Membrane Science journal homepage: www.elsevier.com/locate/memsci

Novel thin film composite hollow fiber membranes incorporated with carbon quantum dots for osmotic power generation Wenxiao Gai, Die Ling Zhao, Tai-Shung Chung

T



Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117585, Singapore

A R T I C L E I N F O

A B S T R A C T

Keywords: Carbon quantum dots (CQDs) Thin film composite (TFC) membrane Pressure retarded osmosis (PRO) Interfacial polymerization Polyamide

By means of carbon quantum dots incorporation, we have developed novel thin film composite (TFC) membranes for osmotic power generation. The newly developed TFC membrane exhibits a peak power density as high as 34.20 W/m2 at 23 bar using 1.0 M NaCl and deionized water as the feed pair. To our best knowledge, this is the highest ever power density reported in the literature. The carbon quantum dots (CQDs) are a new class of carbon nanomaterials with advantages of excellent hydrophilicity, low toxicity, environmental friendliness, easy synthesis and low cost. The CQDs are incorporated into the polyamide selective layers via the conventional interfacial polymerization reaction. The effects of incorporating different CQDs and their loadings on membrane morphology, properties and PRO performance have been examined. It is found that the addition of Na+–functionalized CQDs not only increases the existence of hydrophilic oxygen-containing groups and surface area of the polyamide layer, but also changes the morphology with a looser and thinner polyamide network. The TFC membrane comprising 1 wt% Na–CQD-9 has the optimal performance. Compared with the control, the water flux and power density at 23 bar increase from 44.52 to 53.54 LMH and 28.44 to 34.20 W/m2 respectively, while the reverse salt flux remains unchanged.

1. Introduction Global energy demand has expanded dramatically over the last decades due to the rapid growth of world population and economy [1]. The total primary energy consumption worldwide was about 160,310 million MWh in 2014, and would reach around 240,318 million MWh in 2040 [2]. The search for renewable energy has received substantial attention in recent years because it may not only provide sustainable energy for the future but also mitigate the greenhouse gas emissions from fossil fuels [2,3]. Osmotic energy, also known as salinity gradient energy, is increasingly acknowledged as one of the promising renewable and sustainable energy sources [4–8]. Currently, reverse electrodialysis (RED) and pressure retarded osmosis (PRO) are the two main techniques that aim to harvest the osmotic energy from two solutions with different salinities [6–13]. Different from RED which uses ionic exchange membranes, PRO employs semi-permeable membranes between these two solutions. Since water spontaneously diffuses across the semi-permeable membranes from the low concentration side (i.e., feed solution) to the pressurized high concentration side (i.e., draw solution) due to the chemical potential difference, it results in a higher pressure or higher volume in the draw solution compartment. One can therefore convert



the hydrostatic potential via hydro-turbines or pressure exchangers for power generation [11–15]. Integrations between PRO and seawater reverse osmosis (SWRO) desalination as well as membrane distillation (MD) have been proposed and demonstrated recently from both academia and industries [11,14–18]. The performance of hybrid systems and the amount of energy saving for SWRO are strongly dependent on (1) the performance of PRO membranes, (2) the energy to water price ratio and (3) feed quality and compositions [11,14,19–26]. The heart of PRO process is the semi-permeable membrane, which determines the overall power generation, plant size, capital costs and profitability. According to the experience of Statkraft, who built the first commercial PRO prototype in the world, the employment of high performance PRO membranes is crucial for the commercialization of the PRO technology [24,27]. The energy generated from PRO must be sufficiently higher than the energy consumption for pretreatments of the feed pair and feed pumps in order to have positive economical values. Therefore, intensive efforts have been focused on the development of high performance PRO membranes [8,22,28–32]. Among them, thin film composite (TFC) membranes made from interfacial polymerization have received most attention because they have (1) superior permeation properties compared with traditional phase inversion ones and (2) flexibility to optimize the substrate and the polyamide selective

Corresponding author. E-mail address: [email protected] (T.-S. Chung).

https://doi.org/10.1016/j.memsci.2018.01.034 Received 5 October 2017; Received in revised form 23 December 2017; Accepted 16 January 2018 0376-7388/ © 2018 Elsevier B.V. All rights reserved.

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

O-CQD was dispersed in DI water, and then dialyzed with Slide-A-Lyzer G2 Dialysis Cassettes (2 K MWCO) until there was no significant variation of conductivity in the surrounding DI water. After the dialysis process, a portion of the resultant O-CQD aqueous solution was freeze dried directly to get the O-CQD product, while the other portion was neutralized with a NaOH solution of 5.0 M to pH = 5 and 9. During the neutralization process, the carboxylic acid groups of O-CQD reacted with hydroxyl groups to form neutralized Na-CQD. Afterwards, the NaCQD aqueous solution was also freeze dried to produce two Na-CQD products, which are referred to as Na-CQD-5 and Na-CQD-9 for respective CQDs solutions of pH = 5 and 9. The synthesized O-CQD, Na-CQD-5 and Na-CQD-9 were characterized by high-resolution transmission electron microscopy (HR-TEM, JEOL JEM-2100F), Fourier transform infrared spectroscopy (FT-IR, Vertex 70, Bruker, USA) with a wavenumber range of 4000–500 cm−1 and an X-ray photoelectron spectrometer (XPS, Kratos AXIS UltraDLD spectrometer, Kratos Analytical Ltd) with a mono Al Kα X-ray source.

layer separately. Several strategies have been implemented to improve the mechanical strength and to mitigate the internal concentration polarization (ICP) of the TFC membranes, such as (1) modifying the physicochemical properties of the substrate by pre-compression and polydopamine (PDA) cross-linking [33,34], and (2) molecularly engineering the substrate with different morphology and structures by controlling the phase inversion process [33–36]. Meanwhile, many other approaches have been employed to improve the water permeability of the TFC membranes by appropriate modifications of the polyamide selective layers, such as (1) using additives or surfactants in the monomer solution, and (2) post-treating the nascent polyamide selective layer with chloride, alkaline, alcohol etc. [37–39]. Various nano-materials, such as zeolites, inorganic salts, graphene oxide (GO), carbon nanotubes (CNTs), zeolitic imidazolate frameworks (ZIFs) and metal–organic frameworks (MOFs), have also been employed to tailor the polyamide selective layer with improved separation performance [40–47]. Significant performance enhancements have been reported with a reasonable loading of these nanoscale materials. Carbon quantum dots (CQDs) are a new class of carbon nanomaterials discovered in the last decade. In addition to optical properties, they have unique features such as excellent hydrophilicity, low toxicity, environmental friendliness and low cost [48,49]. They are currently used in chemical sensing, nano-medicine and photo-catalysis [48]. To our best knowledge, there is no exploration to include them in the polyamide layer during interfacial polymerization. Therefore, the first objectives of this study are to (1) explore if CQDs can be incorporated into the polyamide layer during interfacial polymerization and (2) produce novel TFC membranes for PRO applications with a much enhanced power density. Various CQDs would be synthesized and embedded into the polyamide layer using an aqueous mixture of m-phenylenediamine (MPD) and CQDs during interfacial polymerization. The second objectives of this work are to investigate (1) the fundamentals of performance enhancement and (2) the effects of CQDs chemistry and loading on membrane morphology and PRO performance. This study may provide new insights and open up novel strategies to design better PRO membranes for osmotic power generation.

2.3. Fabrication of PES hollow fiber substrates The commercial PES polymer, which had been dried at 60 ℃ under vacuum overnight, was dissolved in a mixture of NMP and PEG (NMP/ PEG = 1:1) by stirring at 60 ℃ first, and then DI water was added dropwise into the polymer solution after it was cooled down to room temperature. Subsequently, the polymer solution was stirred slowly for about 24 h, then the prepared homogenous polymer solution (polymer concentration = 21 wt%) was further degassed under vacuum before spinning. A dry-jet wet spinning process utilizing the co-extrusion technique through a dual layer spinneret was employed to prepare the PES hollow fiber membranes as the substrates for the TFC membranes. DI water, polymer solution and NMP were pumped into the dual layer spinneret through the inner, middle and outer channels, respectively, to fabricate the inner-selective hollow fiber substrate. The details of the spinning process were similar to the previous reports [36,51]. The as-spun PES hollow fiber substrates were immersed in tap water for 2 days before being post-treated by a 50% glycerol aqueous solution for another 2 days followed by air dry for 2 days. Finally, small lab-scale modules were made and each module consisted of three PES hollow fiber substrates.

2. Experimental section 2.1. Materials Veradel® 3100P polyethersulfone (PES, Solvay Specialty Polymers), N-methyl-2-pyrrolidone (NMP, 99.5%, Merck), polyethylene glycol 400 (PEG, Mw = 400 g/mol, Acros Organics) and glycerol (Industrial grade, Aik Moh Pains & Chemicals Pte. Ltd.) were purchased to fabricate and post-treat PES hollow fiber substrates for the TFC membranes. 1,3,5benzenetricarbonyl trichloride (TMC, 98%, Sigma-Aldrich), hexane (99.9%, Fisher Chemicals), m-Phenylenediamine (MPD, 98%, T.C.I.) and sodium dodecyl sulfate (SDS, 99%, Sigma-Aldrich) were employed in the interfacial polymerization process to fabricate the polyamide selective layers of the TFC membranes. Citric acid (99.5%, SigmaAldrich) and sodium hydroxide (98%, Sigma-Aldrich) were utilized to synthesize and functionalize the CQDs. Sodium chloride (NaCl, 99.5%, Merck) was acquired to prepare all the saline solutions in this work. The deionized (DI) water used in this work was produced by a Milli-Q ultrapure water system (Millipore, USA). All chemicals were used as received without further purification.

2.4. Fabrication of PES TFC hollow fiber membranes The polyamide selective layer was synthesized on the inner surface of the PES hollow fiber substrates by interfacial polymerization. First, a 2 wt% MPD aqueous solution containing (1) 0.1 wt% SDS and (2) one of the synthesized O-CQD, Na-CQD-5 and Na-CQD-9 at 0–2 wt% was pumped into the lumen side of the lab-scale module for 3 min at a flow rate of 4.25 mL/min. Then compressed air was purged into the lumen side of the module for 5 min to remove the excessive MPD solution. Subsequently, a 0.15 wt% TMC in hexane solution was pumped into the lumen side of the module at a flow rate of 2.50 mL/min for 5 min to react with the MPD solution saturated on the membrane surface. To remove the excessive TMC solution, the module was purged with air again through the lumen side for 1 min. All prepared lab-scale modules consisting of TFC membranes were kept in air overnight and then soaked in DI water for at least one day before tests or characterizations. The TFC membranes inside the modules were denoted as TFC-0 (control), TFC-(O-CQD)-1, TFC-(Na-CQD-5)-1, TFC-(Na-CQD-9)-0.5, TFC(Na-CQD-9)-1 and TFC-(Na-CQD-9)-2, respectively, where 0.5, 1 and 2 refer to the weight percentages of CQDs in the MPD solutions.

2.2. Syntheses and characterizations of carbon quantum dots (CQDs) The original CQDs (O-CQD) and Na+–functionalized CQDs (NaCQD) were synthesized as illustrated in Fig. 1, following the method reported by Guo et al. with some modifications [50]. Firstly, the grinded citric acid powders were put into a glass container covered with a glass slide and then heated in air at 180 ℃ for 3 h to form O-CQD passivated with carboxyl groups. Subsequently, the product containing

2.5. Membrane characterizations To study the effects of CQDs on the polyamide selective layer, various characterization techniques were employed in this work. To 95

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

Fig. 1. Synthesis schemes of the original CQDs (OCQD) and Na+–functionalized CQDs (Na-CQD).

determined to be 25 bar. The burst pressure was determined by elevating the hydraulic pressure of the draw solution until the water flux across the TFC membranes changed the direction or membrane collapsed. Therefore, all TFC membranes were stabilized at 23 bar before PRO tests. During the stabilization process, the hydraulic pressure applied on the draw solution was increased from 0 bar stepwise up to 23 bar and maintained at that value for 30 min. After the stabilization, the PRO performance was tested at various pressure differences from 23 bar to 0 bar at a decrement of 5 bar. There is no hydraulic pressure applied on the feed solution in all the PRO tests. The determination of burst pressure and the effects of stabilization on PRO membranes had been studied in depth in our previous work [51]. During the PRO tests, the weight loss of the feed solution was recorded as a function of time to calculate the water flux ( Jw , L m−2 h−1, LMH) across the TFC membranes using the following equation:

prepare the samples for characterizations, the as prepared lab-scale modules comprising TFC membranes were opened, the TFC membranes were taken out and then soaked in DI water for at least one day to remove the potential existing glycerol, loosely attached CQDs, unreacted monomers and any other contaminants before being freeze dried. Field emission scanning electronic microscopy (FESEM, JEOL JSM-6700) was employed to examine the morphology of the polyamide layers. Prior to FESEM observation, the freeze-dried TFC membranes were frozen and fractured in liquid nitrogen and coated with platinum using a JOEL JFC-1100E ion sputtering device. Atomic force microscopy (AFM, Nanoscope IIIa, Digital Instrument, USA) was used to probe the surface topology of the polyamide layers and determine the mean roughness (Ra), root mean square roughness (Rq) and surface area under the tapping mode with a scan size of 5 µm × 5 µm in air. An X-ray photoelectron spectrometer (XPS, Kratos AXIS UltraDLD spectrometer, Kratos Analytical Ltd) with a mono Al Kα X-ray source was employed to study the elemental composition of the polyamide layers. In addition, the effects of CQDs incorporation on membrane properties were further investigated by effective surface charge. Since the polyamide layers are on the inner surface of hollow fiber membranes, and the dimension of TFC membranes is very small (outer diameter: 1.12 mm; inner diameter: 0.6 mm), flat sheet TFC membranes were therefore prepared from the same materials and dope compositions for zeta potential measurements with the aid of a SurPASS electrokinetic analyzer (Anton Paar, Austria). To measure the zeta potential, the electrolyte solution (0.01 mol/L NaCl solution, 450 mL) was first autotitrated with a 0.05 mol/L HCl solution from a neutral pH to pH 3, then autotitrated with a 0.05 mol/l NaOH solution from pH 3 to pH 10. These flat sheet TFC membranes were denoted as FSTFC-0 (control), FSTFC-(O-CQD)-1, FSTFC-(Na-CQD-5)-1 and FSTFC-(Na-CQD-9)-1, respectively.

Jw =

∆Vf Am ∆t

(1)

where ∆Vf (L) is the variation of the feed solution volume during a measuring time interval ∆t (h) , and Am (m2) is the effective membrane area of the lab-scale module. The feed conductivity was recorded to estimate the variation of salt concentration in the feed solution and to calculate the reverse salt flux ( Js , g m−2 h−1, gMH) across the TFC membranes using the following equation:

Js =

∆ (Vf Cf ) Am ∆t

(2)

where Vf and Cf are the volume and salt concentration of the feed solution, respectively. The theoretical power density (W, W/m2), which is defined as the energy output per membrane area, was calculated by:

2.6. Pressure retarded osmosis tests

W = Jw ∆P

The experimental setup employed in this work for PRO tests was described elsewhere [35,36]. During the tests, a 1 mol/L NaCl solution of 4 L and DI water of 1 L were employed as draw and feed solutions circulating in the lumen and shell sides of the lab-scale modules, respectively. The flow rates of both streams were controlled at 200 mL/ min. In this work, the burst pressure of the TFC membranes was

(3)

where ∆P (bar) is the pressure difference across the TFC membranes.

Fig. 2. TEM images of the original CQDs (O-CQD) and Na+–functionalized CQDs (i.e., Na-CQD-5 and Na-CQD-9).

96

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

Table 1 Elemental compositions (mass ratio) of the original Na+–functionalized CQDs (i.e., Na-CQD-5 and Na-CQD-9).

C (%) O (%) Na (%)

CQDs

(O-CQD)

surface (Fig. 1). The pH of the MPD solution containing 1 wt% O-CQD is around 5, while the pristine pH of the MPD solution is around 8–9. The acid groups on the surface of the incorporated O-CQD may react with MPD interfering the interfacial polymerization process, which means the lower pH of the amine solution the less amine groups available to react with TMC [45,53,54]. In addition, the ionization of carboxyl groups generates H+, which would inhibit the production of the byproduct HCl during the interfacial polymerization reaction. As a result, the interfacial polymerization reaction would be inhibited in the acid environment. Both factors would contribute to a lower degree of reaction during the interfacial polymerization reaction and form a polyamide network with a nodular-like structure. For the TFC membranes incorporated with Na+–functionalized CQDs, an independent polyamide selective layer reappears. As shown in Fig. 4 (a3) and (a4), both TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)-1 have a polyamide layer on top of their PES substrates with a thickness around 200 and 330 nm, respectively. Their top surfaces also exhibit the typical ridge-and-valley morphology, as shown in Fig. 4 (b3) and (b4). Compared with TFC-0, TFC-(Na-CQD-5)-1 has smaller leaves contributing to a relatively smoother surface, while TFC-(Na-CQD-9)-1 has bigger leaves and more intensive leaves per membrane area contributing to a relatively rougher surface. According to Klaysom et al. [55], the interfacial polymerization reaction could be described into two stages: (1) MPD monomers in the water phase diffuse towards the organic phase and react with TMC monomers, forming the nascent polyamide layer without the obvious ridge-and-valley structure. (2) The surface tension between the water phase and organic phase prompts further migration of MPD monomers to react with TMC monomers, pushing and twisting the nascent polyamide selective layer and resulting in a ridge-and-valley structure. In this work, the Na+–functionalized CQDs dissolved in the MPD solution might not only interfere the crosslinking reaction between the two monomers but also change the surface tension between the water and organic phases. The former factor would lead to a relatively looser polyamide network, enhancing the MPD migration together with the latter factor to form a polyamide network with a high degree of reaction between the two monomers. However, because the MPD solutions have different pH values of around 6.5 and 9 respectively when synthesizing TFC-(NaCQD-5)-1 and TFC-(Na-CQD-9)-1. Due to the relative acidic MPD solution, TFC-(Na-CQD-5)-1 has (1) less amine groups available for reaction and (2) more ionized H+ existing to inhibit the reaction than TFC-(Na-CQD-9)-1. As a result, the former has a thinner and smoother polyamide layer than the latter. To further characterize the control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 at the same loading, the surface topology of polyamide layers was probed by AFM, and the measured data are listed in Table 2. The order of roughness is quite consistent with what observed from the FESEM images. TFC-(O-CQD)-1 and TFC-(Na-CQD-5)-1 have a smaller roughness than TFC-0, while TFC-(Na-CQD-9)-1 has a larger roughness than TFC-0. Compared with TFC-0, the surface areas of TFC-(O-CQD)-1, TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)-1 are increased by 8.83%, 21.62% and 43.03%, respectively, that is consistent with the increasing trend of Na content in CQDs. Interestingly, although TFC-(O-CQD)-1 and TFC-(Na-CQD-5)-1 have a smaller roughness than TFC-0, all the TFC membranes embedded with CQDs have a larger surface area than TFC-0, which may prove the enhanced migration of the MPD monomers during the interfacial polymerization reaction due to the incorporation of CQDs. Table 3 summarizes chemical compositions of polyamide layers of the control and modified TFC membranes analyzed by XPS. As expected, Na is hardly detected by XPS due to its low content in the polyamide selective layers of all modified TFC membranes. Compared with TFC-0, the polyamide layers of the modified TFC membranes have higher O content and O/C ratios due to the incorporation of CQDs, which have high oxygen content as shown in Table 1. The higher O content and O/C ratios in the modified polyamide layers imply the

and

O-CQD

Na-CQD-5

Na-CQD-9

52.96 46.97 0.06

45.93 43.23 10.84

50.35 36.08 13.57

3. Results and discussion 3.1. Characterizations of the synthesized CQDs TEM images shown in Fig. 2 confirm the successful syntheses of the original CQDs (O-CQD) and Na+–functionalized CQDs (i.e., Na-CQD-5 and Na-CQD-9). They have a nearly spherical structure with sizes about 3–9 nm. The presence of Na in Na+–functionalized CQDs was quantitatively confirmed by XPS and Table 1 summarizes the results. A negligible Na content could be detected in O-CQD, while the Na content in Na-CQD-5 and Na-CQD-9 are 10.84% and 13.57%, respectively. The Na content in Na+–functionalized CQDs increases with an increase in the NaOH dose during the neutralization process. Besides, all three kinds of CQDs have high oxygen content, indicating that a large number of oxygen-containing groups exist in these CQDs. Both the ultra-fine size of CQDs and the existence of sodium and oxygen-containing groups contribute to the excellent hydrophilicity of CQDs, which allows good dispersion of CQDs in aqueous solutions. Fig. 3 depicts the FT-IR spectra of the original CQDs and Na+–functionalized CQDs. Consistent with the report by Pan et al. [52], the wide absorption of stretching vibrations of C-OH around 3400 cm−1 and the absorption of stretching vibration of C=O at 1711.82 cm−1 and 1578.73 cm−1 confirm the existence of carboxyl groups in CQDs. Therefore, three kinds of CQDs with different Na content and acid-base properties have been synthesized. 3.2. Characterizations of TFC membranes incorporated with different CQDs Fig. 4 presents the (a) cross-section and (b) surface morphology of the polyamide selective layers of the control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 at the same loading of 1 wt%. Fig. 4 (a1) shows that the polyamide layer of the control TFC membrane (i.e., TFC-0) is an independent layer on top of the PES substrate with a thickness around 420 nm, while Fig. 4 (b1) displays that this polyamide layer has a typical ridge-and-valley surface morphology [31–36]. When the MPD solution comprises 1 wt% O-CQD, the resultant polyamide layer is hardly observed for the modified TFC membrane (i.e., TFC-(O-CQD)-1), as shown in Fig. 4 (a2). Moreover, as illustrated in Fig. 4 (b2), this polyamide layer possesses a nodular-like surface structure instead of the typical ridge-and-valley morphology and is much smoother than TFC-0. This may be attributed to the acid nature of O-CQD, which has a great number of carboxyl groups on its

Fig. 3. FT-IR spectra of the original CQDs (O-CQD) and Na+–functionalized CQDs (i.e., Na-CQD-5 and Na-CQD-9).

97

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

TFC-(O-CQD)-1

TFC-0 (a1)

TFC-(Na-CQD-5)-1

(a2)

100 nm

(b1)

(a4)

(a3)

100 nm

100 nm

(b2)

1 μm

TFC-(Na-CQD-9)-1

100 nm

(b4)

(b3)

1 μm

1 μm

1 μm

Fig. 4. FESEM images of (a) cross-section and (b) surface of the polyamide layers of (1) TFC-0 (control), (2) TFC-(O-CQD)-1, (3) TFC-(Na-CQD-5)-1 and (4) TFC-(Na-CQD-9)-1.

increased existence of hydrophilic oxygen-containing groups in TFC-(OCQD)-1, TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)-1, which would potentially lead to higher membrane hydrophilicity. It should be noticed that the O content in the polyamide layers of TFC-(O-CQD)-1, TFC-(NaCQD-5)-1 and TFC-(Na-CQD-9)-1 show an increasing trend, which is contrary to the decreasing trend of oxygen content in their respective incorporated CQDs. This phenomenon may be attributed to the increased degree of reaction during the interfacial polymerization process, as witnessed by the increased N content in their corresponding polyamide layers. In other words, the higher the degree of reaction between MPD and TMC monomers, the more CQDs are incorporated into the polyamide layer. It should be noticed that the TFC-(O-CQD)-1 has a lower degree of reaction between MPD and TMC monomers as suggested by the lower N contents compared with TFC-0, which is consistent with the Fig. 4 (a2) and (b2). The effects of CQDs incorporation on membrane properties were further investigated by effective surface charge. The zeta potential of flat sheets, including the control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 at the same loading of 1 wt%, are summarized in Fig. 5. In general, all the TFC membranes have a negative zeta potential, and the zeta potential of all the TFC membranes decreases with an increase in pH value. However, the modified TFC membranes with CQDs show a much higher negative charge than the control due to the increased existence of hydrophilic oxygen-containing groups, i.e., carboxyl groups. The difference in zeta potential between the control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 is small at low pH values, while the difference becomes larger at high pH values. This is because the carboxyl groups of the incorporated CQDs remain as the form of COOH at low pH due to the existence of lots of H+ in the surroundings. These carboxyl groups would be deprotonated to form COO- with an increase in pH, which contribute greatly to the dramatically increased negative charge.

Table 3 Surface elemental compositions (mass ratio) of polyamide layers of the control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 at the same loading of 1 wt%.

Na (%) O (%) C (%) N (%) O/C

TFC-0 (control)

TFC-(O-CQD)1

TFC-(Na-CQD5)-1

TFC-(Na-CQD9)-1

0.09 12.74 82.78 4.40 0.15

0.10 14.79 81.22 3.89 0.18

0.06 15.88 78.69 5.36 0.20

0.06 16.91 75.37 7.66 0.22

Fig. 5. Zeta potential of polyamide layers of the flat sheet control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 at the same loading of 1 wt%.

3.3. PRO performance of TFC membranes incorporated with different CQDs Fig. 6 summarizes the water flux, reverse salt flux and power density as a function of pressure difference (ΔP) across the membrane for the control and modified TFC membranes. For TFC-0, the water flux

Table 2 Characteristics of the polyamide selective layers of the control TFC membrane and those modified with O-CQD, Na-CQD-5 and Na-CQD-9 at the same loading of 1 wt%.

Rq (nm) Ra (nm) Surface area (nm2)

TFC-0 (control)

TFC-(O-CQD)-1

TFC-(Na-CQD-5)-1

TFC-(Na-CQD-9)-1

90.39 ± 6.98 71.51 ± 5.28 33.30 ± 2.10

38.03 ± 1.74 29.53 ± 1.22 36.24 ± 0.76

64.57 ± 5.17 48.52 ± 5.99 40.50 ± 2.38

100.40 ± 8.31 80.40 ± 6.16 47.63 ± 1.24

98

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

Fig. 6. (a) Water flux, (b) reverse salt flux and (c) power density of TFC-0 (control) (○), TFC-(O-CQD)-1 (■), TFC-(Na-CQD-5)-1 (▲) and TFC-(Na-CQD-9)-1 (◇) as a function of pressure difference.

reverse salt flux. The deteriorated water flux and power density of TFC(O-CQD)-1 are resulted from the lower degree of reaction during the interfacial polymerization process that forms a nodular-like polyamide layer with a lower roughness because of the acid nature of O-CQD. Since the TFC-(O-CQD)-1 has the lowest roughness, the performance of TFC-(O-CQD)-1 is the worst. Ren et al. reported a similar phenomenon that the roughness of TFC membranes influences the membrane performance in osmotic tests, claiming that the membrane with the highest roughness shows the highest water flux [56]. While the incorporation of Na+–functionalized CQDs could significantly enhance the water flux and power density, especially for the TFC membranes containing NaCQD-9. This should be attributed to four factors: (1) the decreased polyamide layer thickness in TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)1; (2) the increased surface area due to the increased degree of reaction between MPD and TMC monomers; (3) a looser polyamide network due to the interference of crosslinking reaction between the two monomers with the presence of Na-CQDs; and (4) the increased existence of hydrophilic oxygen-containing groups in the polyamide layers.

decreases from 66.59 to 44.52 LMH when ΔP is increased from 0 to 23 bar. Similarly, the water fluxes of the CQDs modified TFC membranes monotonically decrease with an increase in ΔP. In general, compared with TFC-0, TFC-(O-CQD)-1 has a lower water flux, while both TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)-1 have higher water fluxes, and TFC-(Na-CQD-9)-1 shows the highest water flux. The reverse salt fluxes of all TFC membranes rise monotonically with an increase in ΔP. They are overlapped one another within the pressure range from 0 to 15 bar. However, the CQDs modified TFC membranes have slightly higher reverse salt fluxes than TFC-0 at high ΔP (i.e., 20 and 23 bar) due to their relatively looser polyamide layers because of the CQDs incorporation. However, it should be noticed that the highest reverse salt flux of the modified TFC membranes is only 81.34 gMH for TFC-(NaCQD-5)-1, which implies no significant defects are formed in the polyamide layers after the CQDs incorporation at this loading. The power density of all TFC membranes is almost proportional to ΔP within the pressure range of 0–23 bar. Thus, the peak power density is achieved at 23 bar. The respective power densities for TFC-0, TFC-(OCQD)-1, TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)-1 at 23 bar are 28.44, 24.98, 31.03 and 34.20 W/m2. Compared with TFC-0, the peak power densities of TFC-(O-CQD)-1, TFC-(Na-CQD-5)-1 and TFC-(Na-CQD-9)-1 increase at − 12.18%, 9.12% and 20.27%, respectively. In summary, the dissolution of CQDs into the MPD aqueous solution indeed influences the interfacial polymerization reaction between the amine and acyl chloride monomers, resulting in polyamide layers with different morphology and properties. However, the incorporation of nanoscale CQDs into the polyamide selective layers at 1 wt% loading does not induce significant defects. All TFC membranes show a low

TFC-0 (control)

3.4. Characterizations of TFC membranes incorporated with Na-CQD-9 at different loadings Fig. 7 shows the micrographs of (a) cross-section and (b) surface of the polyamide layers of the control and modified TFC membranes as a function of Na-CQD-9 loading. For all modified TFC membranes, an independent polyamide layer can be observed on top of PES substrates with a thickness of around 300 nm. A comparison between Fig. 7 (b2) and (b3) indicates that TFC-(Na-CQD-9)-0.5 has a quite similar

TFC-(Na-CQD-9)-1

TFC-(Na-CQD-9)-0.5

(a1)

(a2)

100 nm

(b1)

(a4)

(a3)

100 nm

100 nm

(b2)

1 μm

TFC-(Na-CQD-9)-2

(b4)

(b3)

1 μm

100 nm

1 μm

1 μm

Fig. 7. FESEM images of (a) cross-section and (b) surface of the polyamide layers of (1) TFC-0 (control), (2) TFC-(Na-CQD-9)-0.5, (3) TFC-(Na-CQD-9)-1 and (4) TFC-(Na-CQD-9)-2.

99

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

polyamide layers. The O content and O/C ratio of the polyamide layer show an increasing trend from TFC-0, TFC-(Na-CQD-9)-0.5, TFC-(NaCQD-9)-1 to TFC-(Na-CQD-9)-2 due to the increased Na-CQD-9 loading, indicating the increasing existence of hydrophilic oxygen-containing groups. The N content in the polyamide layers of TFC-(Na-CQD-9)-0.5, TFC-(Na-CQD-9)-1 and TFC-(Na-CQD-9)-2 are all higher than that of TFC-0, suggesting the formers have a higher degree of reaction between MPD and TMC monomers than the latter.

Table 4 Characteristics of the polyamide layers of the control TFC membrane and those modified with Na-CQD-9 at different loadings.

Rq (nm) Ra (nm) Surface area (nm2 )

TFC-0 (control)

TFC-(NaCQD-9)-0.5

TFC-(Na-CQD-9)-1

TFC-(NaCQD-9)-2

90.39 ± 6.98 71.51 ± 5.28 33.30 ± 2.10

73.81 ± 3.08 58.66 ± 2.47 44.11 ± 0.62

100.40 ± 8.31 80.40 ± 6.16 47.63 ± 1.24

21.80 ± 1.33 14.78 ± 0.83 29.20 ± 0.45

3.5. PRO performance of TFC membranes incorporated with Na-CQD-9 at different loadings Fig. 8 shows the water flux, reverse salt flux and power density of the control and Na-CQD-9 incorporated TFC membranes as a function of ΔP. All TFC membranes comprising Na-CQD-9 have higher water fluxes and power densities than TFC-0 due to the decreased thickness, increased existence of hydrophilic oxygen-containing groups, increased surface area and looser polyamide network of the polyamide selective layers. TFC-(Na-CQD-9)-0.5 has a relatively lower water flux and power density than TFC-(Na-CQD-9)-1 possibly because of the smaller surface area, denser polyamide network and less oxygen-containing groups in the polyamide layer. While TFC-(Na-CQD-9)-2 shows a relatively lower water flux and power density than TFC-(Na-CQD-9)-1 because of the smaller surface area of the polyamide selective layer and the increased ICP effect caused by the increased reverse salt flux. It should be noticed that TFC-(Na-CQD-9)-2 has a reverse salt flux of 103.8 gMH, which almost doubles the value of TFC-(Na-CQD-9)-1. The remarkable increment in reverse salt flux may be caused by the defects formed in the polyamide layer when the Na-CQD-9 loading is too high. As a conclusion, the addition of 1 wt% Na-CQD-9 into the MPD aqueous solution is the optimal condition, which can not only induce a desirable water flux and power density but also maintain a low salt reverse flux.

Table 5 Surface elemental compositions (mass ratio) of the polyamide layers of the control TFC membrane and those modified with Na-CQD-9 at different loadings.

Na (%) O (%) C (%) N (%) O/C

TFC-0 (control)

TFC-(Na-CQD9)-0.5

TFC-(Na-CQD9)-1

TFC-(Na-CQD9)-2

0.09 12.74 82.78 4.40 0.15

0.06 15.09 77.65 7.20 0.19

0.06 16.91 75.37 7.66 0.22

0.03 21.51 71.00 7.47 0.30

polyamide surface morphology with TFC-(Na-CQD-9)-1 but the former has smaller leaves and a smoother surface than the latter. Interestingly, TFC-(Na-CQD-9)-2 has a more nodular-like surface structure, as illustrated in Fig. 7 (b4). As a result, it has a smoother surface than TFC-0, TFC-(Na-CQD-9)-0.5 and TFC-(Na-CQD-9)-1. As discussed in Section 3.2, the presence of Na-CQD-9 might not only interferes the crosslinking reaction between the two monomers leading to a relatively looser polyamide network, but also changes the surface tension between the water and organic phases. Both factors would enhance the MPD migration and form a polyamide network with a high degree of reaction between MPD and TMC monomers. Compared with TFC-(NaCQD-9)-1, TFC-(Na-CQD-9)-0.5 has a relatively denser polyamide selective layer and less MPD migration to the organic phase due to the lower concentration of Na-CQD-9, while TFC-(Na-CQD-9)-2 has a relatively looser polyamide selective layer and more MPD migration to the organic phase due to the higher concentration of Na-CQD-9. As a result, the leaves of TFC-(Na-CQD-9)-2 are connected together to form a smoother surface, as shown in Fig. 7 (a4) and (b4). Table 4 summarizes the surface topology of the polyamide layers on the control TFC membrane and those containing Na-CQD-9 at different loadings probed by AFM. The results are quite consistent with the observation from FESEM images. Both TFC-(Na-CQD-9)-0.5 and TFC-(NaCQD-9)-2 have a smaller roughness than TFC-(Na-CQD-9)-1. Compared with TFC-0, TFC-(Na-CQD-9)-0.5 and TFC-(Na-CQD-9)-1 have an enlarged surface area, while TFC-(Na-CQD-9)-2 has a decreased surface area. Among the modified TFC membranes, TFC-(Na-CQD-9)-1 has the highest surface area. Table 5 tabulates their surface chemical compositions as a function of Na-CQD-9 loading analyzed by XPS. As expected, all modified TFC membranes have negligible Na content in the

4. Conclusions In this work, carbon quantum dots (CQDs), including the original CQDs (O-CQD) and Na+–functionalized CQDs (Na-CQD), have been synthesized and incorporated into the polyamide selective layers to develop novel TFC membranes for PRO applications. According to the membrane morphology and PRO performance of the modified TFC membranes with different CQDs, it could be concluded that the pH of the MPD solution significantly influences the interfacial polymerization process. Compared with an acidic MPD solution, a basic MPD solution is more preferable to form a polyamide layer with a ridge-and-valley structure and a larger surface area for water transport. The incorporation of Na+–functionalized CQDs, especially Na-CQD-9, could significantly enhance the water flux and power density by decreasing the thickness of the polyamide layer, increasing the oxygen-containing groups and surface area of the polyamide layer, and forming a looser polyamide network on TFC membranes. The newly developed TFC membranes comprising 1 wt% Na+–functionalized CQDs exhibit a peak power density as high as 34.20 W/m2 at 23 bar using 1.0 M NaCl solution and deionized water as the feed pair for osmotic power

Fig. 8. (a) Water flux, (b) reverse salt flux and (c) power density of TFC-0(control) (○), TFC-(Na-CQD-9)-0.5 (■), TFC-(Na-CQD-9)-1 (◇)and TFC-(Na-CQD-9)-2 (▲) as a function of pressure difference.

100

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

generation.

Membr. Sci. 479 (2015) 148–158. [23] H. Sakai, T. Ueyama, M. Irie, K. Matsuyama, A. Tanioka, K. Saito, A. Kumano, Energy recovery by PRO in sea water desalination plant, Desalination 389 (2016) 52–57. [24] T.S. Chung, L. Luo, C.F. Wan, Y. Cui, G. Amy, What is next for forward osmosis (FO) and pressure retarded osmosis (PRO), Sep. Purif. Technol. 156 (2015) 856–860. [25] A. Altaee, P. Palenzuela, G. Zaragoza, A.A. AlAnezi, Single and dual stage closedloop pressure retarded osmosis for power generation: feasibility and performance, Appl. Energy 191 (2017) 328–345. [26] M.H. Sharqawy, S.M. Zubair, J.H. Lienhard V, Second law analysis of reverse osmosis desalination plants: an alternative design using pressure retarded osmosis, Energy 36 (2011) 6617–6626. [27] F. Helfer, C. Lemckert, Y.G. Anissimov, Osmotic power with pressure retarded osmosis: theory, performance and trends-a review, J. Membr. Sci. 453 (2014) 337–358. [28] N.N. Bui, J.R. McCutcheon, Nanofiber supported thin-film composite membrane for pressure-retarded osmosis, Environ. Sci. Technol. 48 (2014) 4129–4136. [29] X. Song, Z. Liu, D.D. Sun, Energy recovery from concentrated seawater brine by thin-film nanofiber composite pressure retarded osmosis membranes with high power density, Environ. Sci. Technol. 6 (2013) 1199–1210. [30] S. Chou, R. Wang, L. Shi, Q. She, C. Tang, A.G. Fane, Thin-film composite hollow fiber membranes for pressure retarded osmosis (PRO) process with high power density, J. Membr. Sci. 389 (2012) 25–33. [31] S. Zhang, T.S. Chung, Minimizing the instant and accumulative effects of salt permeability to sustain ultrahigh osmotic power density, Environ. Sci. Technol. 47 (2013) 10085–10092. [32] G. Han, T.S. Chung, Robust and high performance pressure retarded osmosis hollow fiber membranes for osmotic power generation, AIChE J. 60 (2014) 1107–1119. [33] S. Zhang, F. Fu, T.S. Chung, Substrate modifications and alcohol treatment on thin film composite membranes for osmotic power, Chem. Eng. Sci. 87 (2013) 40–50. [34] X. Li, S. Zhang, F. Fu, T.S. Chung, Deformation and reinforcement of thin-film composite (TFC) polyamide-imide (PAI) membranes for osmotic power generation, J. Membr. Sci. 434 (2013) 204–217. [35] X. Li, T.S. Chung, Thin-film composite P84 co-polyimide hollow fiber membranes for osmotic power generation, Appl. Energy 114 (2014) 600–610. [36] S. Zhang, P. Sukitpaneenit, T.S. Chung, Design of robust hollow fiber membranes with high power density for osmotic energy production, Chem. Eng. J. 241 (2014) 457–465. [37] S.P. Sun, T.S. Chung, Outer-selective pressure retarded osmosis hollow fiber membranes from vacuum-assisted interfacial polymerization for osmotic power generation, Environ. Sci. Technol. 47 (2013) 13167–13174. [38] Y. Cui, X.Y. Liu, T.S. Chung, Enhanced osmotic energy generation from salinity gradients by modifying thin film composite membranes, Chem. Eng. J. 242 (2014) 195–203. [39] N.Y. Yip, A. Tiraferri, W.A. Phillip, J.D. Schiffman, L.A. Hoover, Y.C. Kim, M. Elimelech, Thin-film composite pressure retarded osmosis membranes for sustainable power generation from salinity gradients, Environ. Sci. Technol. 45 (2011) 4360–4369. [40] M.M. Pendergast, E.M.V. Hoek, A review of water treatment membrane nanotechnologies, Environ. Sci. Technol. 4 (2011) 1946–1971. [41] M.L. Lind, B.H. Jeong, A. Subramani, X. Huang, E.M.V. Hoek, Effect of mobile cation on zeolite-polyamide thin film nanocomposite membranes, J. Mater. Res. 24 (2009) 1624–1631. [42] S.H. Kim, S.Y. Kwak, B.H. Sohn, T.H. Park, Design of TiO2 nanoparticle self-assembled aromatic polyamide thin-film-composite (TFC) membrane as an approach to solve biofouling problem, J. Membr. Sci. 211 (2003) 157–165. [43] M. Ghanbari, D. Emadzadeh, W.J. Lau, T. Matsuura, M. Davoody, A.F. Ismail, Super hydrophilic TiO2/HNT nanocomposites as a new approach for fabrication of high performance thin film nanocomposite membranes for FO application, Desalination 371 (2015) 104–114. [44] J. Duan, Y. Pan, F. Pacheco, E. Litwiller, Z. Lai, I. Pinnau, High-performance polyamide thin-film-nanocomposite reverse osmosis membranes containing hydrophobic zeolitic imidazolate framework-8, J. Membr. Sci. 476 (2015) 303–310. [45] L. Shen, S. Xiong, Y. Wang, Graphene oxide incorporated thin-film composite membranes for forward osmosis applications, Chem. Eng. Sci. 143 (2016) 194–205. [46] J.N. Shen, C.C. Yu, H.M. Ruan, C.J. Gao, B. Van der Bruggen, Preparation and characterization of thin-film nanocomposite membranes embedded with poly(methyl methacrylate) hydrophobic modified multiwalled carbon nanotubes by interfacial polymerization, J. Membr. Sci. 442 (2013) 18–26. [47] D. Ma, S.B. Peh, G. Han, S.B. Chen, Thin-film nanocomposite (TFN) membranes incorporated with super-hydrophilic metal-organic framework (MOF) UiO-66: toward enhancement of water flux and salt rejection, ACS Appl. Mater. Interfaces 9 (2017) 7523–7534. [48] S.Y. Lim, W. Shen, Z. Gao, Carbon quantum dots and their applications, Chem. Soc. Rev. 44 (2015) 362–381. [49] Y. Wang, A. Hu, Carbon quantum dots: synthesis, properties and applications, J. Mater. Chem. C 2 (2014) 6921–6939. [50] C.X. Guo, D. Zhao, Q. Zhao, P. Wang, X. Lu, Na+-functionalized carbon quantum dots: a new draw solute in forward osmosis for seawater desalination, Chem. Commun. 50 (2014) 7318–7321. [51] W. Gai, X. Li, J.Y. Xiong, C.F. Wan, T.S. Chung, Evolution of micro-deformation in inner-selective thin film composite hollow fiber membranes and its implications for osmotic power generation, J. Membr. Sci. 516 (2016) 104–112. [52] J. Pan, Z. Zheng, J. Yang, Y. Wu, F. Lu, Y. Chen, W. Gao, A novel and sensitive fluorescence sensor for glutathione detection by controlling the surface passivation degree of carbon quantum dots, Talanta 166 (2017) 1–7.

Acknowledgement This research study is supported by the National Research Foundation (1102-IRIS-11-02), Prime Minister's Office, Republic of Singapore, under its Environmental & Water Technologies Strategic Research Programme and administered by the Environment & Water Industry Programme Office (EWI) of the PUB. This research work was funded by the project entitled “Membrane Development for Osmotic Power Generation, Part 2, Module Fabrication and System Integration” (1102-IRIS-11-02) and NUS Grant no. of R-279-000-382-279. The authors also wish to sincerely thank Ms. Yu Zhang, Dr. Yue Cui and Mr. Japip Susilo for their valuable advices and kind support. References [1] M.I. Hoffert, K. Caldeira, G. Benford, D.R. Criswell, C. Green, H. Herzog, A.K. Jain, H.S. Kheshgi, K.S. Lackner, J.S. Lewis, H.D. Lightfoot, W. Manheimer, J.C. Mankins, M.E. Mauel, L.J. Perkins, M.E. Schlesinger, T. Volk, T.M.L. Wigley, Advanced technology paths to global climate stability: energy for a greenhouse planet, Science 298 (2002) 981–987. [2] T.T.D. Tran, A.D. Smith, Evaluation of renewable energy technologies and their potential for technical integration and cost-effective use within the U.S. energy sector, Renew. Sustain. Energy Rev. 80 (2017) 1372–1388. [3] M. Bilgili, A. Ozbek, B. Sahin, A. Kahraman, An overview of renewable electric power capacity and progress in new technologies in the world, Renew. Sustain. Energy Rev. 49 (2015) 323–334. [4] T. Thorsen, T. Holt, The potential for power production from salinity gradients by pressure retarded osmosis, J. Membr. Sci. 335 (2009) 103–110. [5] K. Gerstandt, K.V. Peinemann, S.E. Skilhagen, T. Thorsen, T. Holt, Membrane processes in energy supply for an osmotic power plant, Desalination 224 (2008) 64–70. [6] A. Achilli, T.Y. Cath, A.E. Childress, Power generation with pressure retarded osmosis: an experimental and theoretical investigation, J. Membr. Sci. 343 (2009) 42–52. [7] N.Y. Yip, M. Elimelech, Performance limiting effects in power generation from salinity gradients by pressure retarded osmosis, Environ. Sci. Technol. 45 (2011) 10273–10282. [8] G. Han, S. Zhang, X. Li, T.S. Chung, Progress in pressure retarded osmosis (PRO) membranes for osmotic power generation, Prog. Polym. Sci. 51 (2015) 1–27. [9] J.G. Hong, B. Zhang, S. Glabman, N. Uzal, X. Dou, H. Zhang, X. Wei, Y. Chen, Potential ion exchange membranes and system performance in reverse electrodialysis for power generation: a review, J. Membr. Sci. 486 (2015) 71–88. [10] J.W. Post, J. Veerman, H.V.M. Hamelers, G.J.W. Euverink, S.J. Metz, K. Nymeijer, C.J.N. Buisman, Salinity-gradient power: evaluation of pressure-retarded osmosis and reverse electrodialysis, J. Membr. Sci. 288 (2007) 218–230. [11] M. Kurihara, M. Hanakawa, Mega-ton water system: Japanese national research and development project on seawater desalination and wastewater reclamation, Desalination 308 (2013) 131–137. [12] E. Sivertsen, T. Holt, W. Thelin, G. Brekke, Pressure retarded osmosis efficiency for different hollow fibre membrane module flow configurations, Desalination 312 (2013) 107–123. [13] N.Y. Yip, M. Elimelech, Comparison of energy efficiency and power density in pressure retarded osmosis and reverse electrodialysis, Environ. Sci. Technol. 48 (2014) 11002–11012. [14] C.F. Wan, T.S. Chung, Energy recovery by pressure retarded osmosis (PRO) in SWRO-PRO integrated processes, Appl. Energy 162 (2016) 687–698. [15] A. Achilli, J.L. Prante, N.T. Hancock, E.B. Maxwell, A.E. Childress, Experimental results from RO-PRO: a next generation system for low-energy desalination, Environ. Sci. Technol. 48 (2014) 6437–6443. [16] K. Saito, M. Irie, S. Zaitsu, H. Sakai, H. Hayashi, A. Tanioka, Power generation with salinity gradient by pressure retarded osmosis using concentrated brine from SWRO system and treated sewage as pure water, Desalin. Water Treat. 41 (2012) 114–121. [17] G. Han, J. Zuo, C.F. Wan, T.S. Chung, Hybrid pressure retarded osmosis-membrane distillation (PRO-MD) process for osmotic power and clean water generation, Environ. Sci.: Water Res. Technol. 1 (2015) 507–515. [18] A. Altaee, A. Sharif, Pressure retarded osmosis: advancement in the process applications for power generation and desalination, Desalination 356 (2015) 31–46. [19] V.S.T. Sim, Q. She, T.H. Chong, C.Y. Tang, A.G. Fane, W.B. Krantz, Strategic colocation in a hybrid process involving desalination and pressure retarded osmosis (PRO), Membranes 3 (2013) 98–125. [20] H. Gong, D.D. Anastasio, K. Wang, J.R. McCutcheon, Finding better draw solutes for osmotic heat engines: understanding transport of ions during pressure retarded osmosis, Desalination (2016), http://dx.doi.org/10.1016/j.desal.2017.03.030. [21] J.L. Prante, J.A. Ruskowitz, A.E. Childress, A. Achilli, RO-PRO desalination: an integrated low-energy approach to seawater desalination, Appl. Energy 120 (2014) 104–114. [22] C.F. Wan, T.S. Chung, Osmotic power generation by pressure retarded osmosis using seawater brine as the draw solution and wastewater brine as the feed, J.

101

Journal of Membrane Science 551 (2018) 94–102

W. Gai et al.

[55] C. Klaysom, S. Hermans, A. Gahlaut, S.V. Craenenbroeck, I.F.J. Vankelecom, Polyamide/Polyacrylonitrile (PA/PAN) thin film composite osmosis membranes: film optimization, characterization and performance evaluation, J. Membr. Sci. 445 (2013) 25–33. [56] J. Ren, M.R. Chowdhury, J. Qi, L. Xia, B.D. Huey, J.R. McCutcheon, Relating osmotic performance of thin film composite hollow fiber membranes to support layer surface pore size, J. Membr. Sci. 540 (2017) 344–353.

[53] M. Fathizadeh, H.N. Tien, K. Khivantsev, Z. Song, F. Zhou, M. Yu, Polyamide/nitrogen-doped graphene oxide quantum dots (N-GOQD) thin film nanocomposite reverse osmosis membranes for high flux desalination, Desalination (2017), http:// dx.doi.org/10.1016/j.desal.2017.07.014. [54] L. Shen, J. Zuo, Y. Wang, Tris(2-aminoethyl)amine in-situ modified thin-film composite membranes for forward osmosis applications, J. Membr. Sci. 537 (2017) 186–201.

102