CHEMSUSCHEM FULL PAPERS DOI: 10.1002/cssc.201402664

A Gemini Quaternary Ammonium Poly (ether ether ketone) Anion-Exchange Membrane for Alkaline Fuel Cell: Design, Synthesis, and Properties Jiangju Si,[a, b] Shanfu Lu,*[a, b] Xin Xu,[a, b] Sikan Peng,[a, b] Ruijie Xiu,[a, b] and Yan Xiang*[a, b] To reconcile the tradeoff between conductivity and dimensional stability in AEMs, a novel Gemini quaternary ammonium poly (ether ether ketone) (GQ-PEEK) membrane was designed and successfully synthesized by a green three-step procedure that included polycondensation, bromination, and quaternization. Gemini quaternary ammonium cation groups attached to the anti-swelling PEEK backbone improved the ionic conductivity of the membranes while undergoing only moderate swelling. The grafting degree (GD) of the GQ-PEEK significantly af-

fected the properties of the membranes, including their ion-exchange capacity, water uptake, swelling, and ionic conductivity. Our GQ-PEEK membranes exhibited less swelling ( 40 % at 25–70 8C, GD 67 %) and greater ionic conductivity (44.8 mS cm1 at 75 8C, GD 67 %) compared with single quaternary ammonium poly (ether ether ketone). Enhanced fuel cell performance was achieved when the GQ-PEEK membranes were incorporated into H2/O2 single cells.

Introduction Anion-exchange membrane fuel cells (AEMFCs)[1] have been investigated as an alternative to proton-exchange membrane fuel cells (PEMFCs) because of their compatibility with nonprecious-metal catalysts[1b, 2] and favorability toward fuel oxidation,[2] where the charge carrier is OH rather than H + . However, the performance of AEMFCs has thus far lagged that of PEMFCs because of the intrinsic low mobility of OH.[3] The improvement of ion-exchange capacity (IEC) by increasing the grafting degree (GD) of cationic functional groups can, to some extent, solve this issue;[4] however, a high IEC is always accompanied by excessive water uptake, swelling, and backbone degradation.[5] Balancing the ionic conductivity and the dimensional stability in AEMs has been a formidable scientific challenge.[6] The following two strategies have been pursued to solve the aforementioned challenge: First, a sufficiently high IEC can be achieved under a relatively lower GD, which is expected to restrict the excessive swelling. Normally, a high IEC requires a large number of grafted quaternary ammonium (QA) cation groups; however, a relatively low GD indicates limited grafting sites on the polymer backbone. Thus, multiple QA groups [a] J. Si, Dr. S. Lu, X. Xu, S. Peng, R. Xiu, Prof. Y. Xiang Beijing Key Laboratory of Bio-inspired Energy Materials and Devices Beihang University Beijing 100191 (P.R. China) E-mail: [email protected] [email protected] [b] J. Si, Dr. S. Lu, X. Xu, S. Peng, R. Xiu, Prof. Y. Xiang Key Laboratory of Bio-Inspired Smart Interfacial Science and Technology of Ministry of Education School of Chemistry and Environment, Beihang University Xueyuan Road No. 37, Haidian District, Beijing 100191 (P.R. China) Supporting Information for this article is available on the WWW under http://dx.doi.org/10.1002/cssc.201402664.

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

must be grafted onto the same site. This proposed structure resembles an arrangement of two ionic head groups linked through a spacer in the structure of Gemini surfactants,[7] possessing strong surface activity and self-aggregation ability. This new grafting pattern, which we refer to as the Gemini grafting pattern, can ensure a sufficiently high IEC at a relatively low GD. This strategy can reconcile the tradeoff between the conductivity and dimensional stability in AEMs and has recently been preliminarily validated by Zhuang[8] and Xu.[9] Furthermore, the presence of a flexible spacer can prevent the repulsion of quaternary ammonium cation groups and accelerate the aggregation of ion clusters,[10] which would facilitate the formation of interconnected and broad ionic channels for the fast diffusion of water and fast conduction of ions through the AEMs.[11] Second, a relatively low swelling can be maintained under a high GD. In general, a high GD always results in excessive water uptake and swelling. For example, for a GD of 62 %, the swelling degree (SD) of quaternary ammonium polysulfone at 25 8C increased to 55 %.[4f] However, Yan et al.,[12] Jasti et al.,[13] and Pan et al.[14] recently developed a new AEM by employing poly(ether ether ketone) (PEEK)[15] as a polymer backbone; this AEM exhibited a considerably lower SD ( 27 %) even at 60 8C and exhibited excellent chemical stability in 10 mol L1 KOH when the GD was increased to 70 %. Its anti-swelling properties and high stability make PEEK an ideal choice as a high-performance AEM backbone material. Here, on the basis of the two aforementioned strategies, a novel Gemini quaternary ammonium poly(ether ether ketone) (GQ-PEEK, Scheme 1) membrane has been designed and successfully synthesized by a green three-step procedure that includes polycondensation, bromination, and quaternization; this approach avoids the traditional sluggish reaction of ChemSusChem 0000, 00, 1 – 8

&1&

These are not the final page numbers! ÞÞ

CHEMSUSCHEM FULL PAPERS

www.chemsuschem.org 1273 cm1, indicated the successful polycondensation of the monomers. In addition, the single peak at 2.2 ppm (Ha of  CH3) in the 1H NMR spectrum (Figure 1) also confirmed that the polymer backbone methyl PEEK (mPEEK) was successfully synthesized.

Scheme 1. Simplified illustration of Gemini quaternary ammonium poly (ether ether ketone) (GQ-PEEK) with the Gemini grafting pattern.

chloromethylation and its highly toxic reagents. A detailed investigation of the SD and ion conductivity of GQ-PEEK membranes with different GDs was performed to validate the two aforementioned strategies. The AEMFC H2/O2 fuel cell performance using GQ-PEEK membranes as the electrolyte membrane are also demonstrated.

Results and Discussion Synthesis and characterization of GQ-PEEK polymers

Figure 1. 1HNMR spectra of methyl PEEK (mPEEK), bromomethylated PEEK (Br-mPEEK), and GQ-PEEK membranes. When the substituent was changed from CH3 to CH2Br and CH2N + , the chemical shift of the corresponding protons changed from 2.2 to 4.5 and 4.8 ppm.

Successful monobromination was proven by the change in the chemical shift of the methyl group protons from 2.2 to 4.5 ppm (Hb). Therefore, the degree of bromination (DB) of bromomethylated PEEK (Br-mPEEK) could be determined by the following Equation (1)

The traditional procedure for synthesizing quaternary ammonium PEEK (QPEEK) includes sulfonation, chloromethylation, and quaternization of commercially available PEEK; however, the chloromethylation reaction is usually sluggish and involves DB ¼ 3Ab =ð3Ab þ 2Aa Þ ð1Þ toxic compounds. Furthermore, a general degradation of the PEEK backbone is unavoidable because concentrated sulfuric acid is used as the solvent for sulfonation. Here, we report a much more green and simplified three-step route that includes polycondensation, brominationand quaternization (Scheme 2). The obtained GQPEEK membranes with GDs of 53, 67, and 85 % were labeled as GQ-PEEK 53 %, GQ-PEEK 67 %, and GQ-PEEK 85 %, respectively. Their chemical structures were further confirmed using FTIR and 1 HNMR spectroscopy. In the FTIR spectra (Figure S1), the characteristic peaks, such as the benzene skeleton vibrational peaks at 1586 and 1453 cm1, the CH stretching vibrational peak of methyl groups at 2920 and 2875 cm1, and the COC antisymmetric stretching vibra- Scheme 2. Synthesis route to the GQ-PEEK membranes: Gemini quaternary ammonium (QA) cation groups, linked tional peaks of aromatic ether at through a flexible alkyl chain containing hydroxy and three carbons, were attached to a PEEK backbone.  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ChemSusChem 0000, 00, 1 – 8

&2&

These are not the final page numbers! ÞÞ

CHEMSUSCHEM FULL PAPERS where Ab and Aa are the integrated areas of the Hb peak (CH2Br) and Ha peak (CH3), respectively. The DB of the BrmPEEK is an important parameter; it determines the IEC of the targeted AEMs. In our study, the DB of Br-mPEEK was efficiently controlled by varying the amount of bromination agent (Nbromosuccinimide, NBS). Notably, no fractions with an extremely high molecular weight due to crosslinking, as reported by Wang and Roovers,[15c] were observed in the bromination products. In addition, the bromination process of mPEEK is much more green and safer than chloromethylation of chloromethylmethyl ether. In general, the introduction of substituted groups would destroy the regularity of polymer chains, and thus improve the solubility of polymers. Although the pristine PEEK is insoluble in regular solvents, Br-mPEEK, as expected, exhibits excellent solubility in chloroform, tetrahydrofuran (THF), N,N-dimethylformamide (DMF), etc. The excellent solubility of the Br-mPEEK samples enables their molecular weights to be measured by gel permeation chromatography (GPC). In the case of Br-PEEK with DB 53 %, the measured Mn was 16 993 and the Mw was 37 560, resulting in a Mw/Mn ratio of 2.2. The Br-mPEEK was further functionalized with 3-(dimethylamino)-2-hydroxy-N,N,N-trimethylpropan-1-aminium chloride (DHTC) to obtain polymer-attached Gemini QA groups (GQPEEK) with high efficiency under moderate reaction conditions. In this process, chloroform, in which Br-PEEK could completely dissolve and the final GQ-PEEK was insoluble, was first used as the best solvent for the quaternization, resulting in a sharp decrease in the wrapping of impurities in the final polymer. The GQ-PEEK membranes exhibited lower solubility in organic solvents than those of the parent Br-mPEEK membranes. The structure of GQ-PEEK was confirmed by 1H NMR spectroscopy ; the methylene protons shifted completely from 4.50 to 4.77 ppm (Hc), indicating that all bromomethyl groups were completely converted to aminomethyl groups during the quaternization. Thus, the GD value of the final GQ-PEEK should be the same as the DB of the corresponding Br-PEEK. In particular, the methyl protons (Hg and Hh) of the QA split into two peaks (Figure 1, g and h), which clearly indicates that GQ-PEEK contains two types of N-connected methyl groups in slightly different chemical environments; these different environments correspond to the Gemini functional group structure in GQPEEK. The anion exchange process clearly illustrates one of the fundamental problems with many previously developed AEMs. Within 48 h of the initial immersion of the GQ-PEEK membranes in 1 mol L1 KOH solution, the membrane changed from a brown material to a pale yellow, transparent, and flexible membrane that could no longer dissolve in DMF and DMSO. Thus, the 1HNMR spectrum of GQ-PEEK membranes in OH form were not available. The FTIR spectra of GQ-PEEK in Br and Cl forms and those of GQ-PEEK membranes in OH form were similar, except that the water peak appearing at 3300 cm1 broadened because of the increased water uptake. These results indicate that the Hofmann degradation reaction never occurred during the anion-exchange process, even though Gemini QA groups have beta hydrogens. This phenom 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org enon confirmed the high durability of the GQ-PEEK membranes under alkaline conditions.

Swelling degree and ion conductivity of GQ-PEEK membranes with different GDs under fully hydrated conditions Water in the membrane solvates the ion-exchange groups and reduces the electrostatic interactions between the anions and cations, which is critical for ion-pair separation and ion conductivity in AEMs. The ionic conductivity and dimensional stability are two important, but usually contradictory, properties of AEM membranes. To enhance ionic conductivity, the AEM membranes should possess sufficient IEC, whereas a high IEC leads to excessive water uptake and membrane swelling. As previously mentioned, the purpose of designing the Gemini cation grafting pattern was to achieve a sufficiently high IEC. The results shown in Table 1 are consistent with our

Table 1. Comparison of the properties of single quaternary ammonium PEEK (SQ-PEEK and Gemini quaternary ammonium PEEK (GQ-PEEK) membranes with grafting degrees (GDs) of 53, 67, and 85 % at 25 8C. GD[a]

53 % 67 % 85 %

IECi[b] [mmol g1]

SD (%)

s [mS cm1]

SQ

GQ

SQ

GQ

GQ/SQ

SQ

GQ

GQ/SQ

28 33 127

23 30 40

1.42 1.69 2.02

2.39 2.80 3.23

1.68 1.66 1.60

6.6 7.5 8.7

8.6 10 10

1.30 1.33 1.15

[a] Calculated from 1H NMR spectra. [b] The ideal IEC.

expectations: the IECi (ideal IEC) of the GQ-PEEK significantly increased to 1.68–1.60 times the IEC of the single quaternary ammonium PEEK (SQ-PEEK) under the same GD. The IEC of GQ-PEEK membranes is controlled by varying the GD; thus, its properties such as its water uptake (WU), number of absorbed water molecules per QA group coordinated with OH (designated as l), SD, and ion conductivity are related to the GD (Table S1 in the Supporting Information). In particular, for membrane with a GD of 85 % (denoted GQPEEK 85 %, which represents GQ-PEEK membrane with 85% grafting degree), the values of WU, l, and SD at 25 8C increased several-fold compared with the corresponding values for the GQ-PEEK 53 % membrane because of the overlap of the hydration regions near the ions and the decreased volume fraction of the unfunctionalized polymer, which prevents significant swelling and ion transport at higher GDs. Furthermore, the SD of three GQ-PEEK membranes at higher temperatures was investigated; the results are shown in Figure 2 A. The temperature more strongly affected the SD of the GQ-PEEK membranes with higher GDs. When the temperature was less than 50 8C, the SDs of the GQ-PEEK 53 % and GQ-PEEK 67 % membranes remained almost unchanged. In contrast, the SD of the GQ-PEEK 85 % membranes increased substantially with increasing temperature. When the temperature was elevated continuously, the SD of the GQ-PEEK 53 % membranes began to climb within a small range, but still remained less than 30 %. However, the GQ-PEEK 85 % membranes become completely fragChemSusChem 0000, 00, 1 – 8

&3&

These are not the final page numbers! ÞÞ

CHEMSUSCHEM FULL PAPERS

Figure 2. Temperature-dependent SD (a) and ionic conductivities (b) of GQPEEK 53, 67, and 85 % and single quaternary SQ-PEEK 67 % membranes at 25–80 8C. GQ-PEEK 67 % membranes exhibited a much lower SD and a greater ionic conductivity compared with SQ-PEEK 67 % with the same GD because of its unique Gemini grafting pattern.

mented because their high GD resulted in excessive water uptake and swelling. By this token, the GQ-PEEK 53 % and GQPEEK 67 % membranes exhibited much greater resistance to swelling and greater dimensional stability. This advantage makes the GQ-PEEK 53 % and GQ-PEEK 67 % membranes suitable for fuel cell applications. Surprisingly, compared with the SQ-PEEK membranes with the same GD, the GQ-PEEK membranes exhibited much lower swelling at 25 8C (Table 1), even though they have a much higher IECi. The SD of the GQ-PEEK 85 % membrane, in particular, was only one third the SD of the SQ-PEEK 85 % membrane. When the temperature was increased from 25 to 80 8C, the GQ-PEEK 67 % membrane still exhibited a lower SD than that of SQ-PEEK 67 % (Figure 2 A), which is consistent with the results of Pan et al.[8] This phenomenon indicates that the Gemini grafting pattern also decreases swelling in addition to resulting in a sufficiently high IEC. A possible reason for the reduced swelling is the presence of flexible spacers in the Gemini grafting pattern, which can alleviate the repulsion of grafting QA cation groups, thereby resulting in a limited extension of the backbone and accelerating the enlargement and aggregation of hydrophobic domains. Notably, in the case of the three reported AEMs based on quaternary ammonium PEEK (GQ-PEEK, QAPEEKOH,[12] and xQAPEEK[14]), their SD exhibited a low dependence on temperature compared with other AEMs with different polymer backbones, such as polysulfone (PSF)[4f] because of the inherent low water imbibition and antiswelling of PEEK. This result also confirmed our original hypothesis that employing PEEK as the backbone of an AEM  2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org could reconcile the tradeoff between conductivity and dimensional stability. The IEC-dependent ionic conductivity represents a very important parameter that influences the behavior of AEMFCs. As previously mentioned, under the same GD, the IECis of the GQPEEK membranes are approximately 1.60–1.68 times greater than those of SQ-PEEK membranes (Table 1). Accordingly, the ionic conductivity of GQ-PEEK membranes at 25 8C increases by approximately 1.15–1.3-fold. This may due to the fact that the ionic conductivity relies on not only the IEC but also on the water uptake content of the membranes.[8] With respect to fuel cell performance, good ionic conductivity over a wide temperature range (25–80 8C) is also required (Figure 2 B). As expected, the ionic conductivities for the GQPEEK membranes are temperature dependent. At 25 8C, no significant difference in the ionic conductivity of GQ-PEEK membranes with different GDs was observed, likely because of the membranes’ similar IECm values (Table S1). When the temperature was greater than 30 8C, the ionic conductivity of the GQPEEK membranes with three different GDs were all greater than 10 mS cm1. When the temperature was increased continuously, the ionic conductivity of GQ-PEEK membranes tended to increase with increased GD and temperature. However, higher GDs made greater contributions to the rate of increase of ionic conductivity. Of particular note, the GQ-PEEK 67 % membrane exhibited a high ionic conductivity (44.8 mS cm1) at 75 8C, which was 4.48 times its conductivity at 25 8C. The ionic conductivity of the GQ-PEEK 85 % could not be determined because of its tendency to swell excessively and even dissolve in water at temperatures greater than 50 8C. However, the GQ-PEEK 53 % and GQ-PEEK 67 % membranes exhibited remarkably stable ionic conductivities even at 75 8C, where water evaporation dramatically affects the hydration of the ion membrane; thus, the GQ-PEEK 53 % and GQ-PEEK 67 % membranes were less susceptible to dehydration. Compared with SQ-PEEK under the same GD, the GQPEEK 67 % membrane exhibited enhanced ionic conductivity at 25–75 8C, as expected, and exhibited a substantially greater growth rate with increased temperature. When the temperature was higher than 40 8C, the ionic conductivity of GQ-PEEK was 1.66 times (the ratio value of IECi between GQ-PEEK and SQ-PEEK) greater than that of SQ-PEEK, suggesting that, in addition to the IEC, another factors existing that enhanced the ionic conductivity of the GQ-PEEK membranes. As previously mentioned, the presence of flexible spacers in the Gemini grafting pattern can alleviate the repulsion of grafting QA cation groups and accelerate the aggregation of ion clusters, which would facilitate the formation of ionic channels for the fast diffusion of water and conduction of ions. The presence of ion clusters in the case of the GQ-PEEK 67 % membrane was further confirmed by transmission electron microscopy (TEM) observations (Figure 3); the clusters were 3–5 nm in size. In addition, the conductivities of the GQ-PEEK membranes with three different GDs followed an approximate Arrheniustype temperature dependence, which was promoted by the thermal activation of water motion (Figure S2),[16] indicating that ion migration in the GQ-PEEK membranes occurs via hopChemSusChem 0000, 00, 1 – 8

&4&

These are not the final page numbers! ÞÞ

CHEMSUSCHEM FULL PAPERS

www.chemsuschem.org

Figure 3. TEM image of a GQ-PEEK 67 % membrane impregnated with I . The dark spots in the image represent the ion clusters formed by the aggregation of the hydrophilic domains.

ping processes between coordination sites.[17] On the basis of the Arrhenius relationship between ionic conductivity and temperature, the apparent activation energies (Ea) of ionic conductivity were calculated to be 15.4, 24.8, and 38.0 kJ mol1 for the GQ-PEEK 53, 67, and 85 % membranes, respectively; these Ea values are higher than those of other quaternary ammonium aromatic polymers.[12] It has been suggested that that the high Ea of such membranes is associated with incomplete exchange from Br and Cl to OH , as demonstrated by the IECm (measured IEC) values (Table S1) obtained by titration being substantially lower than the IECi values. Compared to OH , Br has much stronger electrostatic interactions with QA groups, making it difficult to be exchanged with more mobile OH ions. Future work will focus on the improvement of the ion-exchange efficiency. H2/O2 fuel cell performance The performance of a serial H2/O2 AEMFC assembled with GQPEEK membranes as the electrolyte membrane was investigated. Figure 4 A clearly shows that the fuel cell performance of the GQ-PEEK membranes at 40 8C improved with increasing GD, consistent with the trend observed for the ion conductivity data. The open-circuit voltage (OCV) and maximum power density were 1.02 V and 72 mW cm2 when the GD was increased to 85 %. In addition, the fuel cell constructed with a GQ-PEEK 67 % membrane exhibited higher power density compared with that constructed with a SQ-PEEK 67 % membrane (Figure 4 B), which also confirmed the enhanced performance of the Gemini QA grafting pattern. The data further suggest that the GQ-PEEK membranes have strong potential for use as AEMs in alkaline fuel cells.

Conclusions A novel AEM (GQ-PEEK) with Gemini QA grafting pattern and anti-swelling PEEK backbone was designed and successfully synthesized. The resulting GQ-PEEK membranes exhibited enhanced ionic conductivity and H2/O2 single cell performance while resisting excessive swelling, thereby reconciling the con 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Figure 4. Polarization curves and power density curves of a) GQ-PEEK 53, 67, 85 % and b) SQ-PEEK 67 % membranes at 40 8C. Test conditions: membrane thickness = 50 mm; catalyst loading = 0.5 mg Pt cm2 (Pt/C) for both anode and cathode; gas flow rate = 0.1 L min1 for both H2 and O2.

ductivity/stability dilemma in AEMs. These results suggest that the GQ-PEEK membranes have strong potential for use as AEMs in alkaline fuel cells. The strategy for designing and synthesizing GQ-PEEK membranes may lead to the emergence of new AEMs materials that meet the demanding challenges of alkaline fuel cells.

Experimental Section Materials Methyl hydroquinone, 4,4-difluorobenzophenone, K2CO3, toluene, 1-methyl-2-pyrrolidinone (NMP), tetrahydrofuran (THF), N-bromosuccinimide (NBS), 2,2’-azobis-isobutyronitrile (AIBN), glycidyltrimethylammonium chloride, 33 wt % aqueous solution of dimethylamine were purchased from commercial sources and used as received without purification.

Polycondensation of monomers Methyl hydroquinone (0.1 mol), 4,4-difluorobenzophenone (0.1 mol), K2CO3 (0.11 mol), 150 mL of 1-methyl-2-pyrrolidinone (NMP), and toluene (60 mL) were charged into a four-necked 500 mL flask equipped with N2 inlet, mechanical stirrer, thermometer and Dean–Stark trap. The mixture was heated in oil bath. The temperature was raised to 145 8C, when the toluene began to reflux, and then to 175 8C over a 2.5 h period to remove all water. The polymerization was continued for another 4–5 h while the temperature was raised to 190 8C. Then the mixture was cooled to room temperature, diluted with 400 mL THF, and filtered to ChemSusChem 0000, 00, 1 – 8

&5&

These are not the final page numbers! ÞÞ

CHEMSUSCHEM FULL PAPERS remove inorganic salts. 10 mL of acetic acid was added to the filtrate and the polymer was precipitated in methanol. The white polymer was washed twice with boiling methanol and dried in a vacuum oven at 70 8C.

Bromination The obtained polymers mPEEK (1 mol) from polymerization of monomers, NBS (0.5–0.9 mol), 2,2’-azobis(2-methylpropionitrile) (0.01 mol), and chloroform were charged into a four-necked flask and refluxed 24 h with stirring. The mixture was cooled to room temperature and then precipitated into methanol. The target product was collected by filtration and then washed twice with boiling methanol and dried in a vacuum oven at 70 8C for overnight to get Br-mPEEK with a yield of 95 %.

Quaternization and membrane casting 3-(Dimethylamino)-2-hydroxy-N,N,N-trimethylpropan-1-aminium chloride (DHTC): Glycidyltrimethylammonium chloride (20 g) was dissolved in water (100 mL) to form a solution and added into a 33 wt % aqueous solution of dimethylamine (60 mL) at room temperature. Then the reaction mixture was stirred overnight at 60 8C. After the reaction, the solvent and excess dimethylamine was removed on a rotary vacuum evaporator at 50 8C to yield the DHTC (87 %). Gemini quaternary ammonium mPEEK (GQ-PEEK): Dried Br-mPEEK powders were dissolved in chloroform to form a solution of 10 wt %, into which DHTC was added and stirred for 5 h at 40 8C to produce GQ-PEEK. The resultant brown participation was filtered, washed by chloroform, and dried on vacuum. The GQ-PEEK membranes in Br and Cl form were cast from DMF solution (5 wt %) in a clean, flat glass plate and dried in oven at 48 8C for 48 h. To replace the Br and Cl anion in GQ-PEEK for OH , the GQ-PEEK membrane was immersed in 1 mol L1 KOH solution for 48 h. This process was repeated for four times to ensure a complete displacement. Finally, GQ-PEEK membrane with OH anion was repeatedly rinsed with deionized water until the pH of residual water was neutral. Single quaternary ammonium poly (ether ether ketone) (SQ-PEEK) was synthesized according to the similar procedure, only trimethylamine substituted for DHTC was used as the quaternization agent.

www.chemsuschem.org The ionic conductivity was calculated from Eq. (2): s ¼ L=ðZ 0  AÞ

ð2Þ

Where L is the length between sense electrodes (1 cm), Z’ is the real component of the impedance response at high frequency, and A is membrane area available for ion conduction. The water uptake of the membranes was evaluated according to Eq. (3):   WUð%Þ ¼ Wwet  Wdry  100 Wdry

ð3Þ:

Where Wdry is the dry mass of the membranes determined after drying in a desiccator and Wwet is the wet mass of the membranes without excess surface water after soaking for 24 h. TEM observation: 2 mL of GQ-PEEK in Br and Cl form in DMF solution (0.5 wt %) was dropped on copper grids and then dried at 40 8C for 12 h. In order to obtain TEM images, the obtained membrane placed on copper grids was dyed with I by immersing in a 1 mol L1 KI solution for two days, and then rinsed with excessive water and finally dried at room temperature for 12 h. The images were taken on an ultrahigh-resolution transmission electron microscope (JEOL JEM-2010FEF) using an accelerating voltage of 200 kV. Mechanical strength: A tensile tester (CMT6503, Shengzhen SANS Test Machine Co. Ltd, China) was employed to analyze the tensile stress–strain behavior of fully hydrated membranes in OH- form at room temperature. A constant crosshead speed of 5 mm min1 was used for samples of 1 cm in width and 3 cm in length.

Fuel cell tests Pt/C (60 %, Johnson Matthey Co.) was mixed with GQ-PEEK solution and sprayed on each side of the GQ-PEEK membrane (50  2 mm in thickness) to produce the catalyst-coated membrane (CCM). The Pt loading in both anode and cathode was 0.5 mg cm2, and the area of the electrodes was 4 cm2. The percentage of GQ-PEEK in both the anode and the cathode was 15 wt %. The resulted CCM was pressed between two pieces of Teflon-treated carbon paper (Toray250) to make a membrane electrode assembly (MEA). H2/O2 fuel cell tests were conducted by a fuel cell testing system G20 (Greenlight, Canada) using fully humidified (RH = 100 %) H2 and O2 flowing at 100 mL min1.

Acknowledgements Measurements 1

HNMR (400 MHz) analysis was performed on a Bruker spectrometer at 400 MHz by using [D6] DMSO or CDCl3 as solvent. FTIR spectra of membranes were obtained on a Nicolet 6700 ATRIR spectrometer with a wave number resolution of 4 cm1 and range of 400–4000 cm1. The molecular weights of the polymers were determined by using GPC with a Waters 515 HPLC pump coupled with a Waters 410 differential refractometer detector and a Waters 996 photodiode array detector. THF was used as the eluant and the m-Styragel columns were calibrated by polystyrene standards. The in-plane ionic conductivity measurements under fully hydrated conditions were performed in a four-probe alternating current (ac) impedance spectroscopy by use of a PAR 2273 potentiostat. The membrane strips (1  3 cm) were mounted in a conductivity cell immersed in water and stabilized at a specific temperature under nitrogen. The frequency region from 1 Hz to 1 MHz was scanned, where the impedance had a constant value.

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

This work was financially supported by grants from the National Natural Science Foundation of China (No. 51422301, U1137602), the National High Technology Research and Development Program of China (2013AA031902), National Program on Key Basic Research Project (No. 2011CB935700), the National Science Foundation of Beijing (No. 2132051), Beijing Higher Education Young Elite Teacher Project (No. 29201493) and the Fundamental Research Funds for the Central Universities. Keywords: conducting materials · electrochemistry · fuel cell · Gemini quaternary ammonium grafting pattern · membranes · poly (ether ether ketone) [1] a) K. Asazawa, K. Yamada, H. Tanaka, A. Oka, M. Taniguchi, T. Kobayashi, Angew. Chem. Int. Ed. 2007, 46, 8024 – 8027; Angew. Chem. 2007, 119, 8170 – 8173; b) W. Sheng, A. P. Bivens, M. Myint, Z. Zhuang, R. V. Forest,

ChemSusChem 0000, 00, 1 – 8

&6&

These are not the final page numbers! ÞÞ

CHEMSUSCHEM FULL PAPERS Q. Fang, J. G. Chen, Y. Yan, Energy Environ. Sci. 2014, 7, 1719 – 1724; c) S. Gu, R. Cai, T. Luo, K. Jensen, C. Contreras, Y. Yan, ChemSusChem 2010, 3, 555 – 558; d) S. Gu, R. Cai, T. Luo, Z. Chen, M. Sun, Y. Liu, G. He, Y. Yan, Angew. Chem. Int. Ed. 2009, 48, 6499 – 6502; Angew. Chem. 2009, 121, 6621 – 6624; e) J. Wang, S. Gu, R. B. Kaspar, B. Zhang, Y. Yan, ChemSusChem 2013, 6, 2079 – 2082; f) S. D. Poynton, R. C. T. Slade, T. J. Omasta, W. E. Mustain, R. Escudero-Cid, P. Ocon, J. R. Varcoe, J. Mater. Chem. A 2014, 2, 5124 – 5130; g) X. Lin, J. R. Varcoe, S. D. Poynton, X. Liang, A. L. Ong, J. Ran, Y. Li, T. Xu, J. Mater. Chem. A 2013, 1, 7262 – 7269; h) X. Lin, Y. Liu, S. D. Poynton, A. L. Ong, J. R. Varcoe, L. Wu, Y. Li, X. Liang, Q. Li, T. Xu, J. Power Sources 2013, 233, 259 – 268; i) O. I. Deavin, S. Murphy, A. L. Ong, S. D. Poynton, R. Zeng, H. Herman, J. R. Varcoe, Energy Environ. Sci. 2012, 5, 8584 – 8597; j) N. Li, M. D. Guiver, W. H. Binder, ChemSusChem 2013, 6, 1376 – 1383. [2] S. Lu, J. Pan, A. Huang, L. Zhuang, J. Lu, Proc. Natl. Acad. Sci. USA 2008, 105, 20611 – 20614. [3] M. R. Hibbs, M. A. Hickner, T. M. Alam, S. K. McIntyre, C. H. Fujimoto, C. J. Cornelius, Chem. Mater. 2008, 20, 2566 – 2573. [4] a) M. Tanaka, K. Fukasawa, E. Nishino, S. Yamaguchi, K. Yamada, H. Tanaka, B. Bae, K. Miyatake, M. Watanabe, J. Am. Chem. Soc. 2011, 133, 10646 – 10654; b) N. J. Robertson, H. A. Kostalik, T. J. Clark, P. F. Mutolo, H. D. AbruÇa, G. W. Coates, J. Am. Chem. Soc. 2010, 132, 3400 – 3404; c) M. Tanaka, M. Koike, K. Miyatake, M. Watanabe, Macromolecules 2010, 43, 2657 – 2659; d) M. R. Hibbs, C. H. Fujimoto, C. J. Cornelius, Macromolecules 2009, 42, 8316 – 8321; e) T. J. Clark, N. J. Robertson, H. A. Kostalik Iv, E. B. Lobkovsky, P. F. Mutolo, H. D. AbruÇa, G. W. Coates, J. Am. Chem. Soc. 2009, 131, 12888 – 12889; f) J. Pan, S. Lu, Y. Li, A. Huang, L. Zhuang, J. Lu, Adv. Funct. Mater. 2010, 20, 312 – 319.

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemsuschem.org [5] a) J. Pan, Y. Li, L. Zhuang, J. Lu, Chem. Commun. 2010, 46, 8597 – 8599; b) C. G. Arges, V. Ramani, Proc. Natl. Acad. Sci. USA 2013, 110, 2490 – 2495. [6] J. Pan, C. Chen, L. Zhuang, J. Lu, Acc. Chem. Res. 2012, 45, 473 – 481. [7] F. M. Menger, J. S. Keiper, Angew. Chem. Int. Ed. 2000, 39, 1906 – 1920; Angew. Chem. 2000, 112, 1980 – 1996. [8] J. Pan, Y. Li, J. Han, G. Li, L. Tan, C. Chen, J. Lu, L. Zhuang, Energy Environ. Sci. 2013, 6, 2912 – 2915. [9] J. Ran, L. Wu, T. Xu, Polym. Chem. 2013, 4, 4612. [10] L. D. Song, M. J. Rosen, Langmuir 1996, 12, 1149 – 1153. [11] a) J. Pan, C. Chen, Y. Li, L. Wang, L. Tan, G. Li, X. Tang, L. Xiao, J. Lu, L. Zhuang, Energy Environ. Sci. 2014, 7, 354 – 360; b) T. J. Peckham, S. Holdcroft, Adv. Mater. 2010, 22, 4667 – 4690. [12] X. Yan, G. He, S. Gu, X. Wu, L. Du, H. Zhang, J. Membr. Sci. 2011, 375, 204 – 211. [13] A. Jasti, S. Prakash, V. K. Shahi, J. Membr. Sci. 2013, 428, 470 – 479. [14] J. Han, H. Peng, J. Pan, L. Wei, G. Li, C. Chen, L. Xiao, J. Lu, L. Zhuang, ACS Appl. Mater. Interfaces 2013, 5, 13405 – 13411. [15] a) D. P. Jones, D. C. Leach, D. R. Moore, Polymer 1985, 26, 1385 – 1393; b) A. Jonas, R. Legras, Polymer 1991, 32, 2691 – 2706; c) F. Wang, J. Roovers, J. Polym. Sci. Part A 1994, 32, 2413 – 2424. [16] A. A. Kornyshev, A. M. Kuznetsov, E. Spohr, J. Ulstrup, J. Phys. Chem. B 2003, 107, 3351 – 3366. [17] G. A. Giffin, S. Lavina, G. Pace, V. Di Noto, J. Phys. Chem. C 2012, 116, 23965 – 23973. Received: July 11, 2014 Revised: September 1, 2014 Published online on && &&, 0000

ChemSusChem 0000, 00, 1 – 8

&7&

These are not the final page numbers! ÞÞ

FULL PAPERS J. Si, S. Lu,* X. Xu, S. Peng, R. Xiu, Y. Xiang* && – && A Gemini Quaternary Ammonium Poly (ether ether ketone) Anion-Exchange Membrane for Alkaline Fuel Cell: Design, Synthesis, and Properties

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

PEEK and ye shall find: To reconcile the tradeoff between conductivity and dimensional stability in anion exchange membranes, a novel Gemini quaternary ammonium poly (ether ether ketone) (GQ-PEEK) membrane was successfully synthesized. The GQ-PEEK membranes exhibited enhanced ionic conductivity and fuel cell performance while undergoing only moderate swelling.

ChemSusChem 0000, 00, 1 – 8

&8&

These are not the final page numbers! ÞÞ

A gemini quaternary ammonium poly (ether ether ketone) anion-exchange membrane for alkaline fuel cell: design, synthesis, and properties.

To reconcile the tradeoff between conductivity and dimensional stability in AEMs, a novel Gemini quaternary ammonium poly (ether ether ketone) (GQ-PEE...
1MB Sizes 0 Downloads 8 Views