Highly sulfonated poly(ether ether ketone) grafted on graphene oxide as nanohybrid proton exchange membrane applied in fuel cells

Electrochimica Acta 283 (2018) 428e437

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Highly sulfonated poly(ether ether ketone) grafted on graphene oxide as nanohybrid proton exchange membrane applied in fuel cells Shuitao Gao a, Hulin Xu b, Zhou Fang a, Amina Ouadah a, Huan Chen a, Xin Chen c, Lubin Shi a, Bing Ma a, Chaojun Jing a, Changjin Zhu a, * a b c

School of Chemistry and Chemical Engineering, Beijing Institute of Technology, Beijing 100081, China Beijing Qintian Science & Technology Development Co., Ltd., China Department of Chemical Engineering, Columbia University, NY 10027, United States

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 April 2018 Received in revised form 14 June 2018 Accepted 27 June 2018 Available online 28 June 2018

Highly sulfonated poly(ether ether ketone) (PEEK) polymer is hydrogenated and then readily grafted on GO to provide nanohybrid material GO-g-SPEEK. The material shows good enough properties for the preparation of proton exchange membranes. First, the GO-g-SPEEK membrane has much more water uptake but less water swelling compared with Nafion®, and an excellent proton conductivity of 0.219 Scm
1 at 90  C. Then, it demonstrates a peak power density of 112 mWcm
2 when utilized in a H2/air fuel cell at circumstance temperature. Further, blending of the materials with Nafion leads to composite membranes of GO-g-SPEEK/Nafion, which has a close proton conductivity to that of GO-g-SPEEK but far higher cell performances. Of the composite membranes, GO-g-SPEEK/Nafion-33 reveals cell performance of 182 mWcm
2 at 25  C and 213 mWcm
2 at 60  C. These results indicate that Nafion might serve only as an enhancer of compatibility between membrane and Nafion-supported catalyst phase in the membrane electrode assembly, and therefore suggest that the GO-g-SPEEK may be a promising membrane material for the application in proton exchange membrane fuel cells. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Sulfonated poly(ether ether ketone) Graphene oxide Grafting-to Proton exchange membrane Fuel cell

1. Introduction To reduce reliance on fossil fuels and increase demands for clean energy technology worldwide, there is currently a growing interest in the use of fuel cells as energy-efficient and environmentally friendly power generators [1]. Among all major types of fuel cell, proton exchange membrane fuel cells (PEMFCs) are in the forefront stage and have gained substantial attention for vehicle and portable applications, owing to a compact structure, room-temperature start-up capacity, high power density features and etc. Proton exchange membrane (PEM) fuel cells, which are composed of a cathode, an anode, and a PEM, convert chemical energy directly into electrical energy cleanly via electrochemical reactions of hydrogen and oxygen, to yield water and heat as the only byproducts [2]. In PEMFCs, one of few pivotal components that underpin the overall performance of fuel cells is the PEM, which serves as the separator of an anode and a cathode along with the

* Corresponding author. E-mail address: [email protected] (C. Zhu). https://doi.org/10.1016/j.electacta.2018.06.180 0013-4686/© 2018 Elsevier Ltd. All rights reserved.

proton conducting channel. Thus, an ideal PEM should possess not only fast proton transport but also superior mechanical properties in the presence of water [3]. Although poly(perfluorosulfonic acid) (PFSA) membranes (for example, Nafion®) have been superior to others in overall performance, they still have several disadvantages such as high cost, low thermal stability, and poor function at elevated temperatures (above 80  C) or low humidity, hindering their worldwide commercialization. In this context, great efforts have been dedicated to developing cost-effective and alternate membranes based on sulfonate aromatic polymers like the sulfonated poly(ether ether ketone) (SPEEK). That is, the SPEEK not only is low at cost, but also holds sufficient mechanical qualities and appropriate thermal strength [4]. Proton conductivity of SPEEK is largely related to the degree of sulfonation. Increasing the degree of sulfonation values helps obtain the ostensible result of “higher conductivity,” while it causes undesirable and excessive dimensional swelling under high humidity conditions. Moreover, the excessive water swelling drawback of the membranes leads to fuel crossover and even deterioration in mechanical stability [5], and it is observed in our work that the membranes with the sulfonation degree of 77% were

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

ready to melt down at 30  C in hydrate environment and of 84% even at room temperature. This greatly halts the effective application of SPEEK in PEM. Thus, it is of importance to overcome the trade-off effect between the proton conductivity and dimensional stability and comprehensively to promote the performance of SPEEK-based PEM. Graphene oxide (GO) is an attractive material due to its great promise in a number of application areas. One area where GO has opened up exciting ways as a reinforcement agent is in solid electrolytes owing to its mechanical, thermal, and electronical insulating properties. In particular, GO is a proton conductive material. In-plane and through-plane conductivities of spray-painted GO membrane were reported to be 49.9 and 0.3 mScm
1, respectively, reported by Bayer, Lyth and coworker [6]. They also obtained a maximum power density of 33.8 mWcm
2 for GO membrane at 30  C, the highest reported so far for pure GO [7], and further conducted a systematic impedance spectroscopy investigation of GO membranes to present information about the nature of the conductivity and permittivity of GO paper [8]. On the other hand, the proton conductivity for GO nearly 10 mScm
1, higher than a graphene oxide/proton hybrid, was reported by the different research group [9]. Moreover, bulk GO film exhibits mixed conductivity with different oxidation degrees and its electrochemical behavior may also be influenced by the water content of the film [10,11]. Also, a freestanding sulfonated GO paper showed the inplane and through-plane conductivity values of 40 and 12 mScm
1 at 30  C, respectively [12]. Incorporation of GO into polymer can moderately adjust proton conductivity as a result of rendering the membrane with enhanced hydrophilic property and managing the state of water confined in ionic channels in the polymer matrix, as the presence of oxygen-containing group, such as carboxyl groups and hydroxyl groups [13]. Moreover, the involvement of these functional groups would improve the solubility of GO and afford reaction sites for further use. Apart from that, the existence of substantial band gap due to sp3-hybridized carbon atoms, and large surface area in GO also make it a smart nanomaterial with the possibility of facile surface modification, thus providing the opportunity of tailoring the comprehensive properties of GO-based membranes in PEMFC [14,15]. GO-based membranes have been considered to be a very promising alternative to conventional filled polymers or polymer blends of the fuel-cell community, because of their impermeable nature to H2 and O2 gases while exhibiting decent Hþ transport, satisfying the important prerequisites for a successful fuel-cell membrane [16,17]. The covalent modification of graphitic materials by functional molecules and various polymers can be obtained generally by “grafting from” and “grafting to” strategies. As for the “grafting from” strategy, the initiator used for polymerization is normally anchored onto the body and in the edges of graphitic materials. In contrast, via “grafting to” strategy, end-grouptransformed polymer chains react with the functional groups of substrates [18e21]. In fact, exceptional efforts have been undertaken to show that graphene and graphene oxide (GO) are potentially effective reinforcements [21e25]. Specially, previously reported works about SPEEK/functionalized GO have confirmed the advantage of GO-based materials as PEM [26,27]. In our previous work, we paid great attention to the reinforcement of polymer composites by GO-based additives. Herein, a novel procedure was presented to make a new type of composite PEMs using GO nanosheets as the mechanically-strong barrier for highly sulfonated PEEK. Firstly, the partially hydroxyl-functionalized SPEEK polymers with high sulfonation degree were prepared by the reduction of some benzophenone moieties of SPEEK. Separately, graphite oxides were brominated for further chemical modification using highly reactive CeBr bonds. Then, “graft-to”

429

reaction between the brominated GO and hydroxyl-functionalized SPEEK polymers yielded final nanohybrid material GO-g-SPEEK. The modified SPEEK chains were found to covalently anchor on graphene surfaces and this afforded reinforcement effects of dimensional stability for GO-g-SPEEK membrane compared to SPEEK. Therefore, it is suggested the improved compatibility and strong interfacial interaction between the graphene and modified SPEEK polymer, and in turn a restriction for their relaxation or segmental motion in the membrane even at higher temperatures. As a result, prevention of excessive water swelling and superior proton conductivity were achieved for the membrane by the combination of materials. Furthermore, polybenzimidazole (s-PBI) and Nafion as blends were combined with the GO-g-SPEEK material, respectively, in order to reinforce the membrane mechanical strength and investigate the electrode assembly behaviour of the GO-g-SPEEK composite membrane. The performances of the resulting membranes were fully tested to illustrate the potential application of GO-composited membranes in PEMs for fuel cells. Our results indicated an effective way to prepare the highly sulfonated SPEEK-based PEMs, which could have excellent membrane and fuel cell performances. 2. Experimental 2.1. Materials Polybenzimidazole (PBI) powder was supplied from HEOWNS (Tianjin, China). Concentrated sulfuric acid (95e98%), poly(ether ether ketone)(PEEK) 450G Victrex, N-Methyl-2-pyrrolidinone (NMP), dimethylsulfoxide (DMSO), 2-Propanol, potassium carbonate (K2CO3) and hydrobromic acid were obtained from TCI Chemical Co. And used without further purification. DMSO was freshly distilled prior to use. Graphene oxide (GO) and Nafion (Du Pont) solution (5 wt%) D520 were produced by Nanjing XFNANO Materials Tech Co., Ltd. 2.2. Synthesis of SPEEK PEEK (4.0 g) was gradually added to 100 ml of concentrated sulfuric acid (95e98%) at ambient temperature under argon atmosphere. After complete dissolution of PEEK, solution was stirred at 70  C under vigorous mechanical stirring for 3 h. Then, polymer solution was then cooled on ice water bath to terminate the reaction and poured into excess cold water to give a fibrous type SPEEK polymer. Precipitated polymer was filtered and washed several times with distilled water until the pH was neutral and then dried under vacuum at 80  C for 24 h. 2.3. Preparation of hydroxylation of SPEEK (SPEEK-OH) To a solution of SPEEK (0.48 g) in purified DMSO (30 ml), sodium borohydride (0.12 g) was slowly added and the mixture was stirred for 12 h at 120  C under nitrogen atmosphere. The mixture was then cooled to room temperature and purified by three cycles of centrifugation with 2-propanol. The precipitate was collected and dried in vacuum at 70  C for 24 h, resulting in white powders (Scheme S1 in the supplementary data). 2.4. Synthesis of brominated graphene oxide (GO-Br) Graphite oxide (135 mg) and hydrobromic acid (20 ml) were added into a 100 ml reaction flask with a stirrer. After stirring and reflux for 5 h, the reaction product was filtrated and purified by three cycles of centrifugation with deionized water. Finally, the product was dried in vacuum at 80  C for 24 h. These samples were

430

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

termed ‘GO-Br’ (Scheme 1 and Scheme S2 in the supplementary data). 2.5. Synthesis of s-PBI PBI (1 g, 3.3 mmol) and purified DMSO (50 ml) were added into a 100 ml two-necked flask with a stirrer. After completely dissolved, pulverized NaH (0.4 g, 16.5 mmol) was slowly added and stirred at 85  C for 3 h, followed by adding 1,4-butanesultone (0.5 g, 0.34 ml). After being stirred for another 24 h at 85  C, the resulting brown mixture was poured into a large excess of 2-propanol to precipitate the polymer, and then centrifuged. The isolated polymer was washed with a copious amount of 2-propanol and distilled water, and dried in a vacuum oven at 80  C overnight (Scheme S3 in the supplementary data). 2.6. Synthesis of polymer GO-g-SPEEK First, the SPEEK-OH (0.18 g) was first completely dissolved in NMP. K2CO3 (100 mg) was dissolved in NMP at 90  C for 1 h separately. After sonication for 0.5 h in NMP, brown dispersion of GO-Br (0.2 g) was formed. Then, the above solutions were mixed together. After being stirred at 135  C under a nitrogen atmosphere for 3 days, the resultant black solution was then poured into 2-propanol to precipitate the polymer, and purified by centrifugation. Afterwards, the polymer was collected and washed thoroughly using a copious amount of 2-propanol and distilled water, and dried under vacuum overnight to provide the polymer powder and designated as GO-g-SPEEK. 2.7. Membrane fabrication Taking mass ratio of raw Nafion: GO-g-SPEEK ¼ 1:2 (w/w) as an example (GO-g-SPEEK/Nafion-33), the detailed preparation procedures were described as follows: first, the Nafion membrane was prepared individually by casting Nafion (Du Pont) solution (5 wt%) D520 onto Teflon plates, being dried in a vacuum oven at 80  C for next use. And then, Nafion was dissolved in NMP at a concentration of 10% (w/v). A certain amount of GO-g-SPEEK was added to NMP under stirring to form a uniform dispersion. To produce the composite membrane, the obtained transparent Nafion solution was then uniformly blended with the GO-g-SPEEK dispersion. Afterward, the obtained mixture was stirred for another 2 h, and directly cast onto Teflon plate. Heating the solution at 80  C overnight gave a 60e80 mm thick and bendable composite membrane. The membrane was treated with 2 M H2SO4 for 24 h at room temperature, washed with deionized water several times. All composite membranes were denoted as GO-g-SPEEK/Nafion-x, GO-g-SPEEK/s-PBIx where x stands for the mass content of Nafion or s-PBI in the composite membranes. 2.8. Measurements Ion Exchange Capacity (IEC) for the membranes was determined using a classical titration method. For this, approximately a piece of the membrane (50 mg) was treated with 2.0 M NaCl solution for 1 day. After that, the HCl formed as a result of the ion exchange was

titrated with a standard NaOH (0.01 M) solution using phenolphthalein as the indicator. Each sample was tested three times. The measurement of in-plane proton conductivity of each membrane was carried out using a conductivity cell equipped with an electrochemical impedance spectroscopy technique (CHI660D). A detailed procedure was reported previously [3]. Water uptake for the membrane was determined after soaking in water at a certain temperature for 24 h. Then the membrane was taken out, removed excess water by wiping with a tissue paper, and quickly weighed on a microbalance until a constant mass was obtained. The determination of water sorption (l-value) and swelling ratio can be conducted according to the testing method described in our previous work [3]. Both the water uptake and swelling ratio of each membrane were measured three times and the results were reported as the average. Typically, the oxidative stability of the membranes was assessed by immersing the samples (Hþ form) in Fenton's reagent (3% H2O2 containing 2 ppm FeSO4) at 80  C for 1 h, and the residual weight was recorded. TGA experiments were performed on a TGA-Q500 thermogravimetric analyzer with heating from 25 to 700  C (10  C min
1) in N2 flow. Membrane electrode assembly (MEA) was fabricated by hot pressing technique, and conditions were described in our previously reported method [3]. Herein, the single fuel cell test was measured at 25  C and 60  C under 50% relative humidity (RH). Hydrogen (100 ml min
1) and air (150 ml min
1) were supplied to the anode and the cathode, respectively (Supplementary data). 2.9. Characterizations [3] 1 H NMR was used to determine the chemical structures with a Varian mercury-plus 400 spectrometer using deuterated dimethyl sulfoxide (DMSO‑d6) as solvent to dissolve materials. Fourier transform infrared spectroscopy (FTIR) of the polymers was measured on a Horiba FT-720 Fourier transform spectrometer. For the SEM, TEM and AFM observation, the overall morphology of the membrane samples was obtained using a SHIMADZU SSX-550 SEM, a JEOL JEM-2010 TEM and a Digital Instruments Nanoscope IIIa, respectively. Electrical impedance spectroscopy (EIS) experiments of MEAs were carried out on a Parstat 2273 advanced electrochemical systems. The frequency range was between 2 MHz and 0.1 Hz with the AC signal amplitude of 10 mV.

3. Results and discussion 3.1. Synthesis and structural characterization of the membranes 3.1.1. 1H NMR SPEEK was prepared from the previous work by the postsulfonation of PEEK with a corresponding degree of sulfonation of 84%, which was estimated by 1H NMR spectra from the ratio of the peak area of the distinct He signal (AHe) to the integrated peak area of all the other aromatic hydrogens (AHa,a’,b,b’,c,d). ((Fig. S1 and Fig. 1a) [3]. As shown in Fig. 1b, the product s-PBI showed that the NeH protons at 13.0 ppm disappeared incompletely, while new signals appeared at 4.37, 2.48, 1.85 and 1.59 ppm, corresponding to the sulfobutyl protons, which confirmed the attachment of

Scheme 1. Synthesis of brominated graphene oxide.

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

431

Fig. 2. FT-IR spectra of GO, GO-Br, SPEEK, SPEEK-OH and GO-g-SPEEK.

Fig. 1. 1H NMR spectrum of SPEEK, SPEEK-OH and GO-g-SPEEK (a) 1HNMR spectra of sPBI (b).

sulfobutyl groups onto parts of imidazole amino sites in PBI. Besides, strong characteristic peaks at 1051 and 1229 cm
1 for s-PBI and the bands at 1072 and 1218 cm
1 for SPEEK were observed. Consequently, all data indicated that sulfonic acid groups were added to PEEK and PBI polymers, respectively (Fig. S2). The reduction of SPEEK was carried out by treatment with sodium borohydride (NaBH4) in anhydrous DMSO solution at 120  C for 12 h to convert carbonyls to hydroxyl groups of the main chain. The composition and structure of resulting SPEEK-OH and GO-g-SPEEK samples were confirmed by 1H NMR spectroscopy (Fig. 1a). A resonance peak in the 1H NMR spectra at around 5.7 ppm, corresponding to the methine proton, was identified for SPEEK-OH but not for the pristine SPEEK sample, consistent with the reported results [28e30]. After combination with GO, it can be seen that proton signals became a little wider and much smaller, similar to the previous work [31]. Therefore, these spectral data could support the structures of the expected nanohybrid product. 3.1.2. FTIR FTIR spectroscopy was used to monitor synthetic steps for the

production of GO-g-SPEEK. As depicted in Fig. 2, a reduction in the peak intensity at 1650 cm
1 (carbonyl band) and a broad absorption band at 3400 cm
1 appeared for SPEEK-OH [30], which corresponds to stretching of hydrogen bonded OH groups. This further confirms the hydroxylation of PEEK and they were absent in the SPEEK spectra. On the other hand, GO-Br clearly showed the CeBr band at 470 cm
1 compared to GO [32]. An interesting blue shift of the characteristics band at 1625 cm
1 attributed to eC]Ce stretching of GO could be observed. This peak shift could be attributed to changes of the chemical surroundings which might also suggest the reaction of hydrobromic acid with unsaturated CeC bonds under the simultaneous formation of CeBr and CeH bonds [32]. After grafting reaction, the peak assigned to v(CeBr) disappeared and the band at 3400 cm
1 (OeH stretching) observed in SPEEK-OH largely reduced, suggesting the successful attachment of SPEEK-OH onto the GO surface. Moreover, GO-g-SPEEK showed the bands at 1072 and 1218 cm
1, assigned for S]O bond of sulphonic acid group. 3.1.3. Raman spectroscopy and SEM-EDS Raman spectroscopy was employed to evaluate the bromination of GO forming GO-Br, defects and ordering degree of graphene materials, and the state of carbon hybridization [33]. Two obvious characteristic peaks at 1328 and 1591 cm
1 of both GO and GO-Br, commonly corresponding to the diamondoid (D) and graphitic (G) bands, were clearly observed (Fig. S3). Moreover, the I(D)/I(G) values of GO and GO-Br were calculated as 1.02 and 1.15, respectively, which indicated an increase in defects upon the rigorous bromination of GO [34,35]. Also, SEM-EDS data confirmed the presence of bromine in the GO-Br material and the bromine content in GO-Br was estimated to be about 3.08 wt % (Fig. S4). In addition, the results for element distribution mapping were shown in Fig. S5, and showed uniform bromine distribution. 3.1.4. TEM The morphology of the GO-Br and GO-g-SPEEK was characterized by TEM. The samples were prepared by placing a drop of ethanol/Br-GO or GO-g-SPEEK dispersion onto holey carbon copper grids and then drying. As displayed in Fig. 3A, the monolayer or multilayer GO-Br platelets showed a transparent clean surface and had a typical shape resembling crumpled silk veil waves.

432

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

Fig. 3. TEM images of (A) GO-Br and (B) GO-g-SPEEK. The surface SEM image of GO-g-SPEEK (C), and cross-sectional SEM image GO-g-SPEEK (D).

Apparently, the morphology of GO-g-SPEEK (Fig. 3B) was rather different from that of GO-Br, with dark blocks from the grafted SPEEK-OH particles clearly visible on the folded nanostructures surface, which is similar with other polymer-functionalized GO or grapheme [20,23,36]. Therefore, reactive sites for the attachment of SPEEK-OH by the GO bromination were readily produced and this led to successful SPEEK bonding to the surface of GO sheets. 3.1.5. SEM The performance of the nanocomposites is largely related to the dispersion state of the nanofillers in the polymer matrix. Thus, to evaluate the dispersion of GO and further to get more information about the interfacial interaction between GO and the SPEEK polymer matrix, SEM was employed to investigate the morphology of the GO-g-SPEEK membrane. As shown in Fig. 3C, the surface of the membrane was smooth and rarely crumpled, indicating that the graphene nano-sheets were dispersed homogeneously in the SPEEK matrix. This result may arise from the stronger covalent interfacial bonding with the polymer matrix [37]. From the crosssection SEM images, the membrane displayed a laminated and relatively dense morphology structure (Fig. 3D). Also, the graphite sheets dispersed as few-layer stacks in the polymers could be attributed to the interfacial interactions [38e40]. 3.1.6. XRD Fig. 4 showed X-ray diffraction profiles of the GO, GO-Br, SPEEKOH and GO-g-SPEEK nanocomposites. On the GO profile, the characteristic peak of 001 reflection appeared clearly at 11.3 , corresponding to the layer-to-layer distance of 0.78 nm [41]. The presence of this reflection also indicated that the GO sheets stacked themselves on top of each other [42]. The XRD pattern of GO-Br showed a peak shift at 26.3 , implying a slight decrement occurred in the interlayer distance, likely attributed to the partial restacking upon bromination [43]. In the nanocomposites GO-gSPEEK, diffraction peak of GO or GO-Br could not be detected, revealing the disordering and loose stacking of GO or GO-Br [20]. Moreover, the intensity of the main peaks of the SPEEK-OH moderately decreased and became broader, implying a decline in

Fig. 4. XRD spectra of GO, SPEEK-OH, GO-g-SPEEK and GO-Br.

crystallinity ascribed to the compatibility effect of amphiphilic GO sheets onto the two domains of SPEEK polymer [44], and also indicating that some interactions between polymer chains and the fillers might take place [45], thus making the nanocomposite membrane more flexible which would be benefit to ionic conductivity [46].

3.1.7. TGA TGA measurement was performed to assess the thermal stability and the amount of SPEEK chains grafted onto GO as shown in Fig. 5. Two sharp weight loss process approximately in the ranges of 100e250  C and 500e650  C were observed for GO, corresponding to the pyrolysis of unstable oxygen-containing groups and carbon backbone, respectively [20,22]. For GO-Br, rapid mass loss started at the lower temperature (150  C, versus 200  C for GO) reflecting

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

433

3.2. Electrical performance and dimensional stability of the membranes

Fig. 5. TGA thermographs for GO, GO-Br, SPEEK-OH and GO-g-SPEEK.

higher defect density in the brominated GO, because the higher defect density graphite and/or graphene sheets have, the lower their thermal stability is [24,47]. This is consisting with the Raman results. Further for GO-g-SPEEK, 10% weight loss temperature (Td10%) was recorded at 308  C, and at 200  C the weight loss was 3.3%. Therefore, the composite GO-g-SPEEK was sufficiently stable below 200  C, which might be adequate to meet the requirement for long-term application. Moreover, the TGA curve of GO-g-SPEEK can be regarded as the linear addition of GO-Br and SPEEK-OH, and thus the content of SPEEK polymer in GO-g-SPEEK was calculated to be around 73 wt % based on the overall weight loss values of GOBr (90%), SPEEK-OH (50%), and GO-g-SPEEK (61%) [25,48].

3.1.8. DSC Fig. S6 showed the glass transition behaviours of SPEEK-OH alone and in the nanocomposite. SPEEK-OH showed a glass transition temperature (Tg) around 112  C and the GO-g-SPEEK nanocomposite at about 125  C. The covalent attachment of the polymer into the graphitic layers prevented polymer chains from the segmental motions and then enhanced the Tg [24]. In the fabrication process, GO could not be well dispersed in the SPEEK solution due to the less interfacial compatibility between the compositions. However, in this system we observed that the GO-g-SPEEK composite was more readily homogeneous in organic solvents such as NMP or DMSO to cast film than simply GO-doped polymer. The ability of GO to disperse well and interact intimately with the polymer matrixes could maximize the fillers' effect on reducing mobility of polymer chains, consistent with the results from XRD [25]. The similar results were also reported [31,49].

3.2.1. IEC, water uptake and dimensional swelling Water uptake (WU), swelling ratio, and IEC (Table 1) are the important parameters in determining the hydrophilic nature of the membranes. For most of proton conductive polymers, water acts as the carrier which transports the proton ions through membranes, while excessive water uptake may lead to dimensional instability [50]. The pristine SPEEK with a high degree of sulfonation of 84% showed too high dimensional swelling to form a mechanically stable proton-exchange membrane. In the present study, constraint of the swelling as a consequence of the reinforcement with nanofibers was successfully achieved for GO-based membranes. GO-gSPEEK had WU more than Nafion by as much as 3.0 times at room temperature but only slightly increased water swelling, and even at 90  C the swelling ratio of 25.2% was observed for GO-gSPEEK (Table 1). This suggests that the stronger covalent interfacial bonding between GO and SPEEK worked well to restrain the mobility of polymer chains in the hybrid membrane [35,39]. At this stage, blends of sulfonated polybenzimidazole (s-PBI) and Nafion, respectively, into the GO-g-SPEEK material forming composite membranes were also examined. The former blend was used to intensify mechanical properties of the resulting GO-gSPEEK membrane, and the latter mainly to investigate the electrode assembly behavior. As shown in Table 1, s-PBI-incorporated composite membranes exhibited relatively lower IEC values, which might result from a reduced density of sulfonic acid groups in membranes. Also, the formation of an acid-base complex between benzimidazole and sulfonic acid groups stabilized part of ionic protons in sulfonic groups, not being available for the exchange of protons for Naþ ions in the titration [51]. WU and water swelling of the membranes were successively decreased, and generally followed the trend of IEC values. In comparison, GO-g-SPEEK/Nafion composite membranes had an accelerating effect on IEC and in turn on WU and water swelling because the raising of hydrophilic functional groups from Nafion.

3.2.2. Oxidative stability and mechanical property of membranes The oxidative stability was evaluated by the weight loss of the aforesaid membranes treated in Fenton's reagent. Residual weight (RW) of GO-g-SPEEK membrane was above 92% after exposure (Table S1). This indicated a very low impact of radicals on the GO-gSPEEK composite membrane compared with that of SPEEK, and that GO nanosheets within the SPEEK matrix protected the polar groups of the SPEEK polymer from attack by radicals [16]. Also, the combinations of s-PBI and Nafion with GO-g-SPEEK increased the weight residues of resulting membranes, respectively. The GO-g-SPEEK membrane exhibited an acceptable mechanical properties including tensile strength of 5.1 MPa, Young's modulus of 259 MPa, and elongation at break of 17.1% in the wet state (Table S1), whereas, the pristine membrane of highly sulfonated

Table 1 Water-related properties of membranes. Membranes

IEC (mequiv g
1)a

Water uptake (%)

Swelling ratio (%)

l

GO-g-SPEEK GO-g-SPEEK/s-PBI-3 GO-g-SPEEK/s-PBI-8 GO-g-SPEEK/Nafion-10 GO-g-SPEEK/Nafion-33 Nafion® 212

1.27 1.14 0.71 1.34 1.45 0.98

85.3a/88.7b 38.2a/42.3b 18.6a/19.7b 98.5a/111.6b 117.1a/136.3b 28a/-b

20.6a/25.2b 14.3a/15.8b 9.7a/10.9b 27.8a/33.5b 32.7a/41.5b 17a/-b

37 18 14 41 45 16

a b

Measured at room temperature; an average of three measurements. Measured at 90  C; an average of three measurements.

434

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

SPEEK had excessive water uptake and lost its mechanical integrity. Such great improvement in the mechanical properties for GO-gSPEEK could be attributed to dispersive uniformity of the filler with the SPEEK matrix as well as the strong interface interaction between the two components [46]. All of GO-g-SPEEK/s-PBI and GO-g-SPEEK/Nafion composites had more excellent tensile strength and tensile modulus than GO-g-SPEEK, making the further enhancement of mechanical properties by the fillers. Overall, these results indicated that the as-prepared membranes were strong and ductile in mechanical properties for use in fuel cells. 3.2.3. Proton conductivity The proton conduction properties of membranes were measured as a function of temperature (Fig. 6). Acidic functional groups (eSO3H from SPEEK and epoxide groups from GO) of the GO-g-SPEEK membrane dissociate under hydration conditions and allow transport of hydrated proton (H3Oþ). The conductivity for the membrane was observed at 0.089 Scm
1 at 25  C and 0.219 Scm
1 at 90  C, and better than that of Nafion® 212 (Table S2). Remarkably, the conductivity of the GO-g-SPEEK membrane increased even above 80  C, while that of Nafion decreased under the temperature conditions because of dehydration, resulting in shrinkage of the hydrated ionic clusters [52]. The greater proton conduction of the GO-g-SPEEK membrane at high temperature might be interpreted here in terms of water confined in the reorganized nanochannels as a consequence of strong interactions between GO and SPEEK [53]. Also, their sheet-like structure could provide outstanding connectivity of proton paths, especially the narrow or dead-end ion channels in SPEEK matrix, thus revealing excellent proton conductivity over a temperature range of 25e90  C [13,54]. Unexpectedly, the blends of Nafion with GO-g-SPEEK could almost not accelerate the proton conduction of the composite membranes and even moderately declined it as seen for the GO-gSPEEK/Nafion-10 membrane (Fig. 6), although positive contribution of Nafion on the conductivity of the SPEEK-Nafion blended membranes was observed by previous reports [55,56]. Only the GO-gSPEEK/Nafion-33 membrane at temperatures of 80  C and 90  C had a slight conductivity increment. However, both the composite membranes were significantly superior to Nafion itself revealing an obvious synergistic effect of the present material GO-g-SPEEK with Nafion. These findings suggested an outstanding achievement of SPEEK polymer grafted on GO in the proton conduction.

Fig. 6. Proton conductivity of membranes at different temperatures.

The incorporation of s-PBI into the matrix of GO-g-SPEEK has significantly decreased the proton conductivity, particularly with 8 wt% loading of s-PBI due to the reduced water uptake caused by the low IEC value of s-PBI as well as the cross-linking mechanism, where some of sulfonic acid groups were occupied by amine base of s-PBI to form the acid-base electrostatic bonds in the blend membranes. The activation energy (Ea), which is the minimum energy required for proton transfer from one free-site to another, was obtained for selected membranes from the gradient of the Arrhenius plot with proton conductivity and temperature depicted in Fig. S7. A good proton conductor is expected to show low Ea value. The activation energies calculated for GO-g-SPEEK, GO-g-SPEEK/ Nafion-33, GO-g-SPEEK/Nafion-10, GO-g-SPEEK/s-PBI-3 and GO-gSPEEK/s-PBI-8 were 0.159, 0.156, 0.169, 0.181 and 0.193 eV, respectively, which gave a support to the proton conductivity improvement of corresponding membranes [57,58]. 3.2.4. Microstructure of the membrane The morphology of the membranes was investigated by using TEM and AFM, respectively. It is well known that proton conductivity and dimensional stability of the membranes are in close relationship with their morphology. The bright and dark regions are accounted for hard segments corresponding to hydrophobic domains and soft segments corresponding to hydrophilic units, which provided mechanical strength as well as dimensional stability in the membranes, and facilitated the transport of water and protons, respectively [59,60]. As shown in the picture Fig. 7, the membrane GO-g-SPEEK exhibited a clear hydrophobic/hydrophilic phase-separated morphology and the interconnectivity of hydrophilic domains of the membrane GO-g-SPEEK/Nafion-33 appeared to be pronounced, which is beneficial for proton transport. However, with respect to the membrane GO-g-SPEEK/s-PBI-3, the morphology tended to be more hydrophobic, which could be due to the presence of s-PBI polymer backbone, thus causing an increase of mechanical property of the corresponding membrane (Fig. S8). 3.3. Electrochemical performances of the single cells with the membranes The single cell performances of all the membranes were assessed at 25  C and 60  C under 50% relative humidity (RH) and the polarization curves were shown in Fig. 8. All the as-prepared membranes exhibited open circuit potential of about 0.93 V. This result indicated that there was no great gas permeability of the membranes. GO-g-SPEEK membrane achieved a peak power density of 112 mWcm
2 at 25  C and elevated performance to 139 mWcm
2 at 60  C, consistent with proton conductivities recorded at high temperatures. For the membranes of GO-g-SPEEK/ s-PBI-3 and GO-g-SPEEK/s-PBI-8, the peak power density was largely dropped to 80.7 mWcm
2 and 67 mWcm
2, respectively, revealing a s-PBI-dependent decline in the PEMFC performance. It also reflected their lowered proton conduction by the s-PBI in the performance. However, by the presence of Nafion in the GO-gSPEEK matrix, cell performance was clearly improved in comparison with the pristine GO-g-SPEEK membrane with a maximum value of 163 mWcm
2 for GO-g-SPEEK/Nafion-10. Interestingly, as both GO-g-SPEEK and GO-g-SPEEK/Nafion exhibited close proton conductivities (Fig. 6), the relatively poor cell performance recorded on GO-g-SPEEK could be ascribed to its different behavior from that of Nafion in the membrane electrode assembly (MEA) such as interfacial interactions between the phases of electrode and Nafion-supported catalyst [61,62]. Three phases boundary including the reactant gases, solid catalysts, and electrolyte is found in a catalyst layer (CL), and their contact is a very important factor

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

435

Fig. 7. TEM image of membrane GO-g-SPEEK (a) and AFM phase image of membrane GO-g-SPEEK/Nafion-33 (b).

Fig. 8. Polarization curves and power densities of H2/air fuel cells based on all the prepared membranes operated under 50% RH at 25  C (a) and at 60  C (b). The flux rates of H2 and air are 100 and 150 ml min
1, respectively.

for electrochemical performance in fuel cell systems [63]. In the MEA preparation of the present work, as Nafion ionomer was employed as the binder in the electrocatalyst layer and GO-g-SPEEK membrane was made of different type of polymer from Nafion, it was inferred that the poor compatibility between the two polymers would compromise the connectivity of the membrane-electrode [64,65]. For this reason, we blended Nafion solution with the GOg-SPEEK material in the membrane preparation to enhance the interfacial compatibility of the membrane with optimum threephase reaction structure, and to obtain a better performance of cells [66,67]. Therefore, we simply checked the adhesion by immersing the MEA into hot water, which leads to delamination of catalyst layer from hydrocarbon membrane [68,69]. In the hot water test, the GO-g-SPEEK MEA interface was delaminated in 12 min of the immersion whereas the Nafion-containing membranes showed a retarded delamination after 28 min immersion, indicating an improved adhesion. As a result, the improved cell performance at as high as 182 mWcm
2 was obtained for GO-gSPEEK/Nafion-33 at 25  C, and it further showed a peak power density of 213 mWcm
2 at a current density of 650 mAcm
2 at 60  C. Given that the Nafion blending could not promote significantly the proton conductivity of membranes (Fig. 6), it is suggested that the Nafion blended in the membrane served nothing but an enhancer providing compatibility with Nafion-supported catalyst phase in MEA. The difference in the interfacial contact was also

reflected in the impedance spectroscopy results. The bulk resistance (ROhm) including the contributions from proton transport through the membrane as well as the membrane/CL interface, is derived from inflection point, where a semicircle converts into a straight line [68,70]. As shown in Fig. S9, the resistance at electrode/ electrolyte interfaces for GO-g-SPEEK/Nafion-33 MEA is lower than that of GO-g-SPEEK MEA. This preliminarily proves that the improved fuel cell performance is achieved by the better electrode/ electrolyte interfaces alone since the MEAs were fabricated in the same manner and other parameters such as membrane thicknesses and ionomer binder loading were identical except for the membrane intrinsic properties in this experiment. To evaluate the durability of the MEA with GO-g-SPEEK and GO-g-SPEEK/Nafion-33 membranes, a long-term performance test of 50 h at 30  C was carried out at the constant current density of 297 mAcm
2 and 398 mAcm
2, respectively. After 50 h of operation, the cell with GOg-SPEEK and GO-g-SPEEK/Nafion-33 membranes showed reasonable stability with peak power density losses of 11 mWcm
2 and 8 mWcm
2, respectively, which could be attributed to the presence of GO with the slight degradation point. 4. Conclusions Highly sulfonated PEEK was bonded to GO by the “graft-to” reaction producing nanohybrid materials GO-g-SPEEK. The thus

436

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

obtained GO-g-SPEEK membrane was found to have much more water uptake but less rise in the water swelling compared with Nafion®, and excellent proton conductivity (0.219 Scm
1 at 90  C). This indicates that SPEEK polymers were effectively bounded to GO graphitic layers by covalent bonds, and therefore prevented the segmental motions of their hydrophilic polymer chains even coated by water molecules. In addition, about 112 mWcm
2 of peak power density at current density of 297 mAcm
2 in H2/air were obtained under fuel cell testing conditions for the membrane. Further blending of the materials with polymers of s-PBI and Nafion, respectively, aiming at the improvement of dimensional stability of the membrane and the insight into the behavior of the material in MEA, yielded membranes of GO-g-SPEEK/s-PBI and GO-g-SPEEK/ Nafion. Compared with SPEEK with a high degree of sulfonation of 84%, the GO-g-SPEEK membrane was found to have balanced ionic conductivity and lower swelling ratio (25.2% at 90  C), ascribed to the effective covalent attachment of the polymer to the layers, including improved compatibility and strong interfacial interaction that prevented the segmental motions of the polymer chains. GO-g-SPEEK/s-PBI membranes declined in the proton conduction and significantly in cell performance although they had the improvement in mechanical properties. However, it is noteworthy that GO-g-SPEEK/Nafion membranes were close to GO-g-SPEEK in the proton conductivity but far higher in the cell performances than GO-g-SPEEK and comparable to Nafion®. These findings indicate that Nafion composited in the membranes gave a little impact on the proton conduction of the membranes but very big on the cell performance, and therefore the interfacial compatibility of the GOg-SPEEK material with the Nafion-supported catalyst phase in the electrode assembly was impacted by Nafion. As a result, we suggest that the GO-g-SPEEK membranes may have excellent cell performance if a compatible catalyst phase could be developed instead of the Nafion-supported catalyst phase for the electrode assembly, and the modification with GO may provide a significant strategy for the highly sulfonated PEEK in the PEMFC applications. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.electacta.2018.06.180. References [1] X. Ge, Y. He, M.D. Guiver, L. Wu, J. Ran, Z. Yang, T. Xu, Alkaline anion-exchange membranes containing mobile ion shuttles, Adv. Mater. 28 (2016) 3467e3472. [2] J. Han, Q. Liu, X. Li, J. Pan, L. Wei, Y. Wu, H. Peng, Y. Wang, G. Li, C. Chen, L. Xiao, J. Lu, L. Zhuang, An effective approach for alleviating cation-induced backbone degradation in aromatic ether-based alkaline polymer electrolytes, ACS Appl. Mater. Interfaces 7 (2015) 2809e2816. [3] S. Gao, H. Xu, T. Luo, Y. Guo, Z. Li, A. Ouadah, Y. Zhang, Z. Zhang, C. Zhu, Novel proton conducting membranes based on cross-linked sulfonated polyphosphazenes and poly(ether ether ketone), J. Membr. Sci. 536 (2017) 1e10. [4] S. Qu, Y. Sun, J. Li, Sulfonate poly(ether ether ketone) incorporated with ammonium ionic liquids for proton exchange membrane fuel cell, Ionics 23 (2017) 1607e1611. [5] W.H. Lee, K.H. Lee, D.W. Shin, D.S. Hwang, N.R. Kang, D.H. Cho, J.H. Kim, Y.M. Lee, Dually cross-linked polymer electrolyte membranes for direct methanol fuel cells, J. Power Sources 282 (2015) 211e222. [6] T. Bayer, R. Selyanchyn, S. Fujikawa, K. Sasaki, S.M. Lyth, Spray-painted graphene oxide membrane fuel cells, J. Membr. Sci. 541 (2017) 347e357. [7] T. Bayer, S.R. Bishop, M. Nishihara, K. Sasaki, S.M. Lyth, Characterization of a graphene oxide membrane fuel cell, J. Power Sources 272 (2014) 239e247. [8] T. Bayer, S.R. Bishop, N.H. Perry, K. Sasaki, S.M. Lyth, Tunable mixed ionic/ electronic conductivity and permittivity of graphene oxide paper for electrochemical energy conversion, ACS Appl. Mater. Interfaces 8 (2016) 11466e11475. [9] M.R. Karim, K. Hatakeyama, T. Matsui, H. Takehira, T. Taniguchi, M. Koinuma, Y. Matsumoto, T. Akutagawa, T. Nakamura, S. Noro, T. Yamada, H. Kitagawa, S. Hayami, Graphene oxide nanosheet with high proton conductivity, J. Am. Chem. Soc. 135 (2013) 8097e8100.

[10] Q.-L. Meng, H.-C. Liu, Z. Huang, S. Kong, X. Lu, P. Tomkins, P. Jiang, X. Bao, Mixed conduction properties of pristine bulk graphene oxide, Carbon 101 (2016) 338e344. [11] V.A. Smirnov, V.P. Vasil’ev, N.N. Denisov, Y.V. Baskakova, V.A. Dubovitskii, Electric behavior of interlayer water in graphene oxide films, Chem. Phys. Lett. 648 (2016) 87e90. [12] K. Ravikumar, Scott, Freestanding sulfonated graphene oxide paper: a new polymer electrolyte for polymer electrolyte fuel cells, Chem. Commun. 48 (2012) 5584. [13] B. Zhang, Y. Cao, S. Jiang, Z. Li, G. He, H. Wu, Enhanced proton conductivity of Nafion nanohybrid membrane incorporated with phosphonic acid functionalized graphene oxide at elevated temperature and low humidity, J. Membr. Sci. 518 (2016) 243e253. [14] A. Kundu, S. Nandi, P. Das, A.K. Nandi, Fluorescent graphene oxide via polymer grafting: an efficient nanocarrier for both hydrophilic and hydrophobic drugs, ACS Appl. Mater. Interfaces 7 (2015) 3512e3523, https://doi.org/10.1021/ am507110r. [15] K. Nawaz, U. Khan, N. Ul-Haq, P. May, A. O'Neill, J.N. Coleman, Observation of mechanical percolation in functionalized graphene oxide/elastomer composites, Carbon 50 (2012) 4489e4494. [16] L. Sha Wang, A. Nan Lai, C. Xiao Lin, Q. Gen Zhang, A. Mei Zhu, Q. Lin Liu, Orderly sandwich-shaped graphene oxide/Nafion composite membranes for direct methanol fuel cells, J. Membr. Sci. 492 (2015) 58e66. [17] S. Stankovich, D.A. Dikin, G.H.B. Dommett, K.M. Kohlhaas, E.J. Zimney, E.A. Stach, R.D. Piner, S.T. Nguyen, R.S. Ruoff, Graphene-based composite materials, Nature 442 (2006) 282e286. [18] B. Wang, D. Yang, J.Z. Zhang, C. Xi, J. Hu, Stimuli-responsive polymer covalent functionalization of graphene oxide by Ce(IV)-Induced redox polymerization, J. Phys. Chem. C 115 (2011) 24636e24641. [19] T. Mondal, A.K. Bhowmick, R. Krishnamoorti, Conducting instant adhesives by grafting of silane polymer onto expanded graphite, ACS Appl. Mater. Interfaces 6 (2014) 16097e16105. [20] Z. Tang, H. Kang, Z. Shen, B. Guo, L. Zhang, D. Jia, Grafting of polyester onto graphene for electrically and thermally conductive composites, Macromolecules 45 (2012) 3444e3451. [21] K.-J. Peng, J.-Y. Lai, Y.-L. Liu, Nanohybrids of graphene oxide chemicallybonded with Nafion: preparation and application for proton exchange membrane fuel cells, J. Membr. Sci. 514 (2016) 86e94. [22] J. Lim, H. Yeo, M. Goh, B.-C. Ku, S.G. Kim, H.S. Lee, B. Park, N.-H. You, Grafting of polyimide onto chemically-functionalized graphene nanosheets for mechanically-strong barrier membranes, Chem. Mater. 27 (2015) 2040e2047. [23] G. Gonçalves, P.A.A.P. Marques, A. Barros-Timmons, I. Bdkin, M.K. Singh, N. Emami, J. Gr acio, Graphene oxide modified with PMMA via ATRP as a reinforcement filler, J. Mater. Chem. 20 (2010) 9927. [24] M. Fang, K. Wang, H. Lu, Y. Yang, S. Nutt, Covalent polymer functionalization of graphene nanosheets and mechanical properties of composites, J. Mater. Chem. 19 (2009) 7098. [25] H.K.F. Cheng, N.G. Sahoo, Y.P. Tan, Y. Pan, H. Bao, L. Li, S.H. Chan, J. Zhao, Poly(vinyl alcohol) nanocomposites filled with poly(vinyl alcohol)-grafted graphene oxide, ACS Appl. Mater. Interfaces 4 (2012) 2387e2394. [26] N. Cao, C. Zhou, Y. Wang, H. Ju, D. Tan, J. Li, Synthesis and characterization of sulfonated graphene oxide reinforced sulfonated poly (ether ether ketone) (SPEEK) composites for proton exchange membrane materials, Materials 11 (2018) 516. [27] D. Liu, J. Peng, Z. Li, B. Liu, L. Wang, Improvement in the mechanical properties, proton conductivity, and methanol resistance of highly branched sulfonated poly(arylene ether)/graphene oxide grafted with flexible alkylsulfonated side chains nanocomposite membranes, J. Power Sources 378 (2018) 451e459. [28] S. Feng, Y. Shang, Y. Wang, G. Liu, X. Xie, W. Dong, J. Xu, V.K. Mathur, Synthesis and crosslinking of hydroxyl-functionalized sulfonated poly(ether ether ketone) copolymer as candidates for proton exchange membranes, J. Membr. Sci. 352 (2010) 14e21. [29] D. An, B. Wu, G. Zhang, W. Zhang, Y. Wang, Gradiently crosslinked polymer electrolyte membranes in fuel cells, J. Power Sources 301 (2016) 204e209. [30] S. Feng, Y. Shang, G. Liu, W. Dong, X. Xie, J. Xu, V.K. Mathur, Novel modification method to prepare crosslinked sulfonated poly(ether ether ketone)/silica hybrid membranes for fuel cells, J. Power Sources 195 (2010) 6450e6458. mez, G. Martínez, Polymeric modification of gra[31] H.J. Salavagione, M.n.A. Go phene through esterification of graphite oxide and poly(vinyl alcohol), Macromolecules 42 (2009) 6331e6334.  jkova , M. Pumera, [32] O. Jankovský, P. Simek, K. Klimov a, D. Sedmidubský, S. Mate Z. Sofer, Towards graphene bromide: bromination of graphite oxide, Nanoscale 6 (2014) 6065e6074. [33] A.K. Sahu, K. Ketpang, S. Shanmugam, O. Kwon, S. Lee, H. Kim, Sulfonated grapheneenafion composite membranes for polymer electrolyte fuel cells operating under reduced relative humidity, J. Phys. Chem. C 120 (2016) 15855e15866. [34] A.K. Mishra, N.H. Kim, D. Jung, J.H. Lee, Enhanced mechanical properties and proton conductivity of NafioneSPEEKeGO composite membranes for fuel cell applications, J. Membr. Sci. 458 (2014) 128e135. [35] X. Qiu, T. Dong, M. Ueda, X. Zhang, L. Wang, Sulfonated reduced graphene oxide as a conductive layer in sulfonated poly(ether ether ketone) nanocomposite membranes, J. Membr. Sci. 524 (2017) 663e672. [36] M.-C. Hsiao, S.-H. Liao, M.-Y. Yen, P.-I. Liu, N.-W. Pu, C.-A. Wang, C.-C.M. Ma, Preparation of covalently functionalized graphene using residual oxygen-

S. Gao et al. / Electrochimica Acta 283 (2018) 428e437

[37]

[38] [39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

containing functional groups, ACS Appl. Mater. Interfaces 2 (2010) 3092e3099. Y. Zhan, X. Yang, H. Guo, J. Yang, F. Meng, X. Liu, Cross-linkable nitrile functionalized graphene oxide/poly(arylene ether nitrile) nanocomposite films with high mechanical strength and thermal stability, J. Mater. Chem. 22 (2012) 5602. C. Wan, B. Chen, Reinforcement and interphase of polymer/graphene oxide nanocomposites, J. Mater. Chem. 22 (2012) 3637. X. Wang, W. Xing, P. Zhang, L. Song, H. Yang, Y. Hu, Covalent functionalization of graphene with organosilane and its use as a reinforcement in epoxy composites, Compos. Sci. Technol. 72 (2012) 737e743. Y. Wang, Z. Shi, J. Yu, L. Chen, J. Zhu, Z. Hu, Tailoring the characteristics of graphite oxide nanosheets for the production of high-performance poly(vinyl alcohol) composites, Carbon 50 (2012) 5525e5536. X. Zhao, Q. Zhang, D. Chen, P. Lu, Enhanced mechanical properties of graphene-based poly(vinyl alcohol) composites, Macromolecules 43 (2010) 2357e2363. S. Morimune, T. Nishino, T. Goto, Ecological approach to graphene oxide reinforced poly (methyl methacrylate) nanocomposites, ACS Appl. Mater. Interfaces 4 (2012) 3596e3601. W. Wu, Y. Li, P. Chen, J. Liu, J. Wang, H. Zhang, Constructing ionic liquid-filled proton transfer channels within nanocomposite membrane by using functionalized graphene oxide, ACS Appl. Mater. Interfaces 8 (2016) 588e599. B.G. Choi, Y.S. Huh, Y.C. Park, D.H. Jung, W.H. Hong, H. Park, Enhanced transport properties in polymer electrolyte composite membranes with graphene oxide sheets, Carbon 50 (2012) 5395e5402.  mez, Synthesis of poly(vinyl alcohol)/ H.J. Salavagione, G. Martínez, M.A. Go reduced graphite oxide nanocomposites with improved thermal and electrical properties, J. Mater. Chem. 19 (2009) 5027. P. Salarizadeh, M. Javanbakht, S. Pourmahdian, Fabrication and physicochemical properties of iron titanate nanoparticles based sulfonated poly (ether ether ketone) membrane for proton exchange membrane fuel cell application, Solid State Ionics 281 (2015) 12e20. S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S.T. Nguyen, R.S. Ruoff, Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide, Carbon 45 (2007) 1558e1565. D. Wang, Y. Bao, J.-W. Zha, J. Zhao, Z.-M. Dang, G.-H. Hu, Improved dielectric properties of nanocomposites based on poly(vinylidene fluoride) and poly(vinyl alcohol)-functionalized graphene, ACS Appl. Mater. Interfaces 4 (2012) 6273e6279. T. Huang, R. Lu, C. Su, H. Wang, Z. Guo, P. Liu, Z. Huang, H. Chen, T. Li, Chemically modified graphene/polyimide composite films based on utilization of covalent bonding and oriented distribution, ACS Appl. Mater. Interfaces 4 (2012) 2699e2708. J. Ren, S. Zhang, Y. Liu, Y. Wang, J. Pang, Q. Wang, G. Wang, A novel crosslinking organiceinorganic hybrid proton exchange membrane based on sulfonated poly(arylene ether sulfone) with 4-amino-phenyl pendant group for fuel cell application, J. Membr. Sci. 434 (2013) 161e170. H. Li, G. Zhang, W. Ma, C. Zhao, Y. Zhang, M. Han, J. Zhu, Z. Liu, J. Wu, H. Na, Composite membranes based on a novel benzimidazole grafted PEEK and SPEEK for fuel cells, Int. J. Hydrogen Energy 35 (2010) 11172e11179. C.-Y. Tseng, Y.-S. Ye, M.-Y. Cheng, K.-Y. Kao, W.-C. Shen, J. Rick, J.-C. Chen, B.J. Hwang, Sulfonated polyimide proton exchange membranes with graphene oxide show improved proton conductivity, methanol crossover impedance, and mechanical properties, Adv. Energy Mater. 1 (2011) 1220e1224. B.G. Choi, J. Hong, Y.C. Park, D.H. Jung, W.H. Hong, P.T. Hammond, H. Park, Innovative polymer nanocomposite electrolytes: nanoscale manipulation of ion channels by functionalized graphenes, ACS Nano 5 (2011) 5167e5174. K. Kim, J. Bae, M.-Y. Lim, P. Heo, S.-W. Choi, H.-H. Kwon, J.-C. Lee, Enhanced

[55] [56]

[57]

[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

437

physical stability and chemical durability of sulfonated poly(arylene ether sulfone) composite membranes having antioxidant grafted graphene oxide for polymer electrolyte membrane fuel cell applications, J. Membr. Sci. 525 (2017) 125e134. I.T. Kim, J. Choi, S.C. Kim, Blend membranes of Nafion/sulfonated poly(aryl ether ketone) for direct methanol fuel cell, J. Membr. Sci. 300 (2007) 28e35. J.-C. Tsai, J.-F. Kuo, C.-Y. Chen, Nafion®/nitrated sulfonated poly(ether ether ketone) membranes for direct methanol fuel cells, J. Power Sources 194 (2009) 226e233. X. He, G. He, A. Zhao, F. Wang, X. Mao, Y. Yin, L. Cao, B. Zhang, H. Wu, Z. Jiang, Facilitating proton transport in nafion-based membranes at low humidity by incorporating multifunctional graphene oxide nanosheets, ACS Appl. Mater. Interfaces 9 (2017) 27676e27687. T. Luo, Y. Zhang, H. Xu, Z. Zhang, F. Fu, S. Gao, A. Ouadah, Y. Dong, S. Wang, C. Zhu, Highly conductive proton exchange membranes from sulfonated polyphosphazene-graft-copolystyrenes doped with sulfonated single-walled carbon nanotubes, J. Membr. Sci. 514 (2016) 527e536. C. Wang, N. Li, D.W. Shin, S.Y. Lee, N.R. Kang, Y.M. Lee, M.D. Guiver, Fluorenebased poly(arylene ether sulfone)s containing clustered flexible pendant sulfonic acids as proton exchange membranes, Macromolecules 44 (2011) 7296e7306. G. Wang, K.H. Lee, W.H. Lee, D.W. Shin, N.R. Kang, D.H. Cho, D.S. Hwang, Y. Zhuang, Y.M. Lee, M.D. Guiver, Durable sulfonated poly(benzothiazole-cobenzimidazole) proton exchange membranes, Macromolecules 47 (2014) 6355e6364. B. Wu, B. Li, W. Liu, J. Liu, M. Zhao, Y. Yao, J. Gu, Z. Zou, The performance improvement of membrane and electrode assembly in open-cathode proton exchange membrane fuel cell, Int. J. Hydrogen Energy 38 (2013) 10978e10984. W. Liu, L. Wan, J. Liu, M. Zhao, Z. Zou, Performance improvement of the opencathode proton exchange membrane fuel cell by optimizing membrane electrode assemblies, Int. J. Hydrogen Energy 40 (2015) 7159e7167. S.-Y. Oh, T. Yoshida, G. Kawamura, H. Muto, M. Sakai, A. Matsuda, Inorganiceorganic composite electrolytes consisting of polybenzimidazole and Cssubstituted heteropoly acids and their application for medium temperature fuel cells, J. Mater. Chem. 20 (2010) 6359. T.A. Sherazi, M.D. Guiver, D. Kingston, S. Ahmad, M.A. Kashmiri, X. Xue, Radiation-grafted membranes based on polyethylene for direct methanol fuel cells, J. Power Sources 195 (2010) 21e29. S. Ren, C. Li, X. Zhao, Z. Wu, S. Wang, G. Sun, Q. Xin, X. Yang, Surface modification of sulfonated poly(ether ether ketone) membranes using Nafion solution for direct methanol fuel cells, J. Membr. Sci. 247 (2005) 59e63. K.-Y. Cho, J.-Y. Eom, H.-Y. Jung, N.-S. Choi, Y.M. Lee, J.-K. Park, J.-H. Choi, K.W. Park, Y.-E. Sung, Characteristics of PVdF copolymer/Nafion blend membrane for direct methanol fuel cell (DMFC), Electrochim. Acta 50 (2004) 583e588. R. Wycisk, J. Chisholm, J. Lee, J. Lin, P.N. Pintauro, Direct methanol fuel cell membranes from Nafionepolybenzimidazole blends, J. Power Sources 163 (2006) 9e17. K.H. Oh, H.S. Kang, M.J. Choo, D.H. Jang, D. Lee, D.G. Lee, T.H. Kim, Y.T. Hong, J.K. Park, H.T. Kim, Interlocking membrane/catalyst layer interface for high mechanical robustness of hydrocarbon-membrane-based polymer electrolyte membrane fuel cells, Adv. Mater. 27 (2015) 2974e2980. S. Yuk, M.J. Choo, D. Lee, H. Guim, T.H. Kim, D.G. Lee, S. Choi, D.H. Lee, G. Doo, Y.T. Hong, H.T. Kim, Three-dimensional interlocking interface: mechanical nanofastener for high interfacial robustness of polymer electrolyte membrane fuel cells, Adv. Mater. 29 (2017). S. Mikhailenko, M. Guiver, S. Kaliaguine, Measurements of PEM conductivity by impedance spectroscopy, Solid State Ionics 179 (2008) 619e624.