Comb-shaped phenolphthalein-based poly(ether sulfone)s as anion exchange membranes for alkaline fuel cells

Abstract

A series of novel comb-shaped phenolphthalein-based poly(ether sulfone)s was synthesized for preparing anion exchange membranes (AEMs). Hexadecyldimethylamine with a long alkyl chain was used as the quaternization reagent to form a comb-shaped architecture of the copolymers. Due to the presence of a long alkyl side chain with hydrophobicity, the as-synthesized comb-shaped AEMs possess a self-anti-swelling property resulting in a low water uptake and swelling ratio. The PES-B100-C16 membrane exhibits excellent alkaline stability due to the presence of large volumetric β-alkyl chains linking to the cationic group that resist the attack of OH⁻, and retain available ionic conductivity in a 2 M KOH solution at 60 °C for 360 h. An open circuit voltage of the single cell reached 0.67 V, and the maximum power density was 43 mW cm² at a current density of 125 mA cm⁻² without optimization in a single H₂/O₂ alkaline fuel cell at 50 °C.

---

Introduction

Fuel cells are considered to be promising energy conversion devices for mobile applications which generate electric power directly from the chemical energy of fuels and oxidants that are highly efficient, have high energy density and low emission.^1,2 Ion exchange membranes as a critical component of fuel cells have a great effect on cell performance and lifetime. Proton exchange membrane fuel cells (PEMFCs) based on commercialized perfluorosulfonic acid (such as Nafion) membranes have been well developed during the past decades.^3–5 Although Nafion membranes exhibit high ionic conductivity, excellent thermal and chemical stability, the commercialization of PEMFCs is hindered for the use of high-cost precious metal electro-catalysts and its slow oxygen reduction kinetics in acidic condition.^6,7

Recently, alkaline fuel cells (AFCs) based on anion exchange membranes (AEMs) have attracted extensive attention due to their improved oxygen reduction kinetics and fuel oxidation kinetics in alkaline condition, which allow for the use of non-precious metal electro-catalysts (such as Ag, Ni and Co), greatly reducing the cost of the fuel cell devices.^8–10 As a crucial component of AFCs, various types of AEMs based on poly(ether sulfone) (PES),^11,12 poly(ether ketone) (PEK),^13,14 poly(phenylene oxide) (PPO),^15–17 polyethylene,¹⁸ polystyrene,¹⁹ polymerizable ionic liquid,^20,21 etc. have been studied over the past years. Phenolphthalein-based polymers were regarded as an promising material for AEMs considering their excellent chemical resistance as well as robust mechanical properties, thermal stability and good solubility in membrane-forming organic solvents.^22,23 In previous studies, phenolphthalein-based AEMs were prepared via chloromethylation of the active sites on aromatic ring to introduce chloromethyl groups into the backbone, followed by a Menshutkin reaction with trimethylamine to form quaternary ammonium groups.^13,24,25 However, the commonly used chloromethylation reagents such as chloromethyl methyl ether were undesirable due to their toxicity and carcinogenicity. Furthermore, conventional cationic group such as tetramethylammonium attached to the backbone could be easily degraded, especially at elevated temperature and high pH.²⁶

Benzyl bromide groups on the polymers are allowed for nearly quantitative conversion to quaternary ammonium moieties when they react with a tertiary amine under suitable condition. Compared to chloromethylation, benzylic bromination using N-bromosuccinimide is a faster and more efficient approach for preparing AEMs.²⁷ Thus, in this work, dimethyl phenolphthalein-based PES was synthesized for preparing AEMs and the benzyl groups were brominated into benzyl bromide groups with different degrees of bromomethylation by controlling the amount of bromination reagent.

According to the recent research on AEMs, comb-shaped architecture is a promising approach for designing AEMs with high ionic conductivity as well as excellent dimensional stability.²⁸ N,N-Dimethylhexadecylamine (DMHDA) as a kind of tertiary amine has a long alkyl chain of sixteen carbon atoms that can not only present a comb-shaped architecture when it is attached to the backbone, but also form cationic groups via Menshutkin reaction with benzyl bromide groups. Moreover, DMHDA had been verified as a good quaternization reagent for its good chemical stability in alkaline environment.²⁹ To the best of our knowledge, DMHDA has not been introduced into phenolphthalein-based poly(ether sulfone)s to form AEMs for alkaline fuel cells. The objective of this work was thus to choose DMHDA as the quaternization reagent to introduce functional groups into the brominated phenolphthalein-based PES and to form a comb-shaped architecture. In addition, the as-synthesized copolymers were characterized by ¹H NMR spectroscopy, Fourier transform infrared spectroscopy (FT-IR) and gel permeation chromatography (GPC). Furthermore, water uptake, swelling ratio, ionic exchange capacity, ionic conductivity, mechanical properties as well as thermal and alkaline stability of the as-synthesized AEMs were comprehensively investigated.

Experimental

Materials

Bis(4-fluorophenyl)sulfone (FPS) (99.0%, TCI, Japan), o-cresolphthalein (OCP) (indicator, TCI, Japan), N-bromosuccinimide (NBS) (99%, Aladdin, China), trimethylamine (TMA) (CP, Sinopharm, China), N,N-dimethylhexadecylamine (95%, Aldrich, U.S.), N,N-dimethylacetamide (DMAc) (99.8%, Aladdin, China), 1,1,2,2-tetrachloroethane (TCE) (AR, Sinopharm, China) and dimethyl formamide (DMF) (AR, Sinopharm, China) were used as received. Toluene (AR, Sinopharm, China) was stirred over CaH₂ for 24 h, then refined by vacuum distillation and stored over 4 Å molecular sieves. Benzoyl peroxide (BPO) (AR, Aladdin, China) was recrystallized twice from chloroform. All other chemicals were obtained from Shanghai Sinopharm Chemical Reagent Co. Ltd (China) and used without further purification.

Preparation of anion exchange membranes

Synthesis of poly(ether sulfone)s (PESs). A typical procedure for preparing PES copolymers is described below (as shown in Scheme 1). FPS (3.9912 g, 15.7 mmol), OCP (5.4382 g, 15.7 mmol), K₂CO₃ (4.3398 g, 31.4 mmol), DMAc (60 mL), and toluene (20 mL) were put into a three-necked flask equipped with a Dean–Stark trap, a mechanical stirrer, a condenser, and a nitrogen gas inlet. Under the protection of nitrogen, the mixture was heated to 145 °C for 4 h, subsequently increased to 165 °C and continued to react for 20 h. Afterwards, the mixture was cooled to room temperature and coagulated into 1000 mL of an aqueous methanol solution (deionized water/methanol = 1 : 1, v/v) with vigorous stirring. After washing the fibrous copolymer with methanol or deionized water for several times and vacuum drying at 80 °C for 24 h, a grey solid product was finally obtained.

Scheme 1 Synthesis of the comb-shaped copolymer AEMs and their schematic structure.

Bromination of PES (PES-Bx). A typical procedure for the synthesis of bromo-substitution copolymer PES-Bx, where x denotes the degree of bromomethylation (DB), is described as follows (Scheme 1). Taking PES-B100 as an example, 1.5000 g of PES (5.3512 mmol –CH₃) was dissolved in 20 mL of TCE to form a solution, 0.9524 g of NBS (5.3512 mmol) and 0.0864 g of BPO (0.3567 mmol) were then added into the solution and constantly stirred at 85 °C for 5 h under reflux. After cooling to room temperature, the solution was precipitated in methanol and the solid was collected by filtration. The obtained bromide polymers were washed with methanol several times and dried in a vacuum oven at 80 °C for 24 h to give brominated PES-B100 copolymers (a pale yellow solid). The degree of bromomethylation is 82.4%.

Synthesis of comb-shaped poly(ether sulfone)s (PES-Bx-Cy) with functional groups. Herein, N,N-dimethylhexadecylamine was used as the quaternization reagent to form a comb-shaped architecture of the copolymers. PES-Bx-Cy was synthesized via nucleophilic substitution reaction of PES-Bx with DMHDA or TMA (Scheme 1), where y denotes the carbon chain length attached to cationic groups. A typical procedure, illustrated by the preparation of PES-B100-C16, is described as follows. 1.0000 g of PES-B100 (2.3862 mmol of –CH₂Br) and 20 mL of DMF were added into a three-neck flask with a magnetic stirrer. After the copolymer being dissolved completely, 1.9293 g of DMHDA (7.1586 mmol) was charged into the reaction flask at room temperature, and the mixture was stirred vigorously for 48 h. The amount of quaternization reagents used in the reaction was 300% molar excess to ensure that benzyl bromide groups were sufficiently converted into quaternary ammonium groups. The resulting solution was poured into anhydrous ether to yield product. The obtained copolymers were filtered and washed with excess anhydrous ether for several times, and then dried under vacuum at 80 °C for 24 h to obtain PES-B100-C16 (brown).

Membrane formation. PES-B100-C16 (1.0000 g) was completely dissolved in 20 mL of DMF followed by filtrating through a 0.45 μm PTFE filter. The obtained solution was then cast onto a custom-built flat glass dish with an area of 5 cm × 5 cm and heated under vacuum at 60 °C for 24 h to obtain the membranes in bromine form, followed by immersing them into a 1 M NaOH solution at room temperature for 24 h to obtain hydroxide form membranes. The membranes were then washed thoroughly and immersed in deionized water for more than 24 h to remove residual NaOH.

Characterization and measurements

¹H NMR and FT-IR spectroscopy. ¹H NMR spectra were recorded on an Avance III 500 MHz spectrometer (Bruker, Switzerland) using CDCl₃, DMSO-d₆ or DMF-d₇ as the solvent, and tetramethylsilane (TMS) as the internal standard. FT-IR spectra of the samples were determined using a FT-IR spectrophotometer (Nicolet Avatar 330, Thermo Electron Corporation, USA) in the range 4000–500 cm⁻¹ with a resolution of 4 cm⁻¹.

Gel permeation chromatography (GPC) measurement. The apparent molecular weight of the polymers was measured on a GPC system (Waters, USA) which was equipped with a Waters 1515 HPLC pump, and three Styragel columns (Waters HT4, HT5E, and HT6) at 30 °C using tetrahydrofuran as the eluent at a flow rate of 1.0 mL min⁻¹. Standard polystyrene samples were used to calibrate the curve of the molecular weight.

Scanning electron microscopy (SEM). The morphology of the as-synthesized membranes was observed using a scanning electron microscope (Sigma, Zeiss, Germany) operating at 15 kV. Before observation, the membrane sample was fractured with liquid nitrogen and coated with a platinum layer using a JFC-1600 autofine coater to prevent charging and improve the membrane conductivity.

Atomic force microscopy (AFM). An atomic force microscope (5500, Agilent Technologies, USA) was used to study the surface microstructure of the membranes. The samples were observed in the tapping mode under room temperature. Before observation, the membranes were kept in 60% RH for 24 h.

Transmission electron microscopy (TEM). Before TEM imaging, the membrane was stained with tungstate ions by ion exchange of the quaternary ammonium groups in a 1 M aqueous Na₂WO₄ solution at room temperature, and dried in a vacuum oven at 60 °C. Then, the dry membrane sample was embedded in epoxy resin, followed by slicing into a thickness of approximately 60 nm, and collected on the copper grids for observation. TEM images were recorded on a JEM-1400 electron microscope (JEOL, Japan) operating at 100 kV.

Small angle X-ray scattering (SAXS). SAXS patterns of the dry AEMs were recorded using a SAXSess-MC2 X-ray diffractometer (Anton paar, Austria) at room temperature. The dry AEMs were placed under vacuum at 80 °C for 24 h before testing.

Ionic exchange capacity (IEC). The IEC of the membranes was determined by the classical back-titration method and ¹H NMR spectroscopy. The hydroxide form membranes were dried to a constant weight and then equilibrated in a large excess of a 0.1 M aqueous HCl solution for 48 h, followed by back titrating with a standard 0.05 M NaOH solution. The IEC value can be calculated bywhere M_1,HCl and M_2,HCl (meq.) are the milliequivalents of HCl required before and after equilibrium, respectively. m_dry (g) is the weight of the dry membrane.

Water uptake and swelling ratio. A piece of membrane sample (50 mm × 10 mm) was immersed in deionized water at a certain temperature for 24 h. After taking the wet membrane out of water, the excess surface water was wiped with tissue and the mass and length were quickly measured. The water uptake (WU) of the membranes is calculated fromwhere m_wet and m_dry are the weight of the membrane sample in the wet and dry states, respectively.

The swelling ratio (SR) of both the in-plane and through-plane directions of the membranes is calculated by:

where L_w and L_d are the length of the wet and dry membranes, T_w and T_d are the thickness of the wet and dry membranes.

Ionic conductivity. For in-plane conductivity measurements, the membrane was sandwiched between two pairs of copper-plate electrodes. For through-plane conductivity measurements, the membrane was sandwiched between two square copper-plate electrodes. The measurement was carried out under a fully hydrated state or equilibrated at a certain humidity controlled by a fuel cell system (GFEC-10, Guangdong Electron. Tech. Res. Inst.). The ionic conductivity (σ, mS cm⁻¹) of each membrane sample can be calculated by:where L is the distance (cm) between the reference electrodes for in-plane measurement and the thickness of the membrane for through-plane measurement. A is the cross-sectional area (cm²) of the membrane for in-plane measurement and the area of the electrode for through-plane measurement. The resistance R (kΩ) of the membrane was determined over the frequency range 0.1 Hz to 100 kHz by two-probe alternating current (AC) impedance spectroscopy using an electrochemical workstation (Parstat 263, Princeton Advanced Technology, USA).

Mechanical property. A dumbbell-shaped sample (5.0 mm × 2.0 mm) was punched from membranes for tensile test. Measurements were carried out using an Instron 3343 machine with a tensile speed of 0.2 mm s⁻¹ at room temperature. Before test, the AEMs were dried under vacuum at 60 °C for 24 h.

Thermal stability. The thermal stability of the membranes was analyzed using a thermogravimetric analyzer (SDT-Q600, TA instruments, USA). A membrane sample (3 mg) was heated from 30–700 °C with a heating rate of 10 °C min⁻¹ under a nitrogen atmosphere (nitrogen flow rate: 100 mL min⁻¹). The membrane samples were dried under vacuum at 80 °C for 24 h to remove the moisture before test.

Alkaline stability. To evaluate the alkaline stability, the membranes were immersed into a 2 M KOH solution at 60 °C, the change in the ionic conductivity and chemical structure of the membrane was observed. Before testing, the OH⁻ form membrane was equilibrated in deionized water for more than 24 h.

Gas permeation measurements. The permeability of pure hydrogen and oxygen through the membrane was measured using the constant pressure volume method.³⁰ The permeation experiments were carried out at 50 °C. The gas permeability (P, cm³ cm cm⁻² s⁻¹ cmHg⁻¹) is calculated according to the following equation:where T is the operating temperature (K), and dV/dt is the flow rate on the permeate side (cm³ s⁻¹), which is measured using a bubble flow meter. l and A are the thickness (cm) and effective area (cm²) of the membrane, respectively. Δp is the pressure difference (cmHg) between the feed and permeate side. The feed pressure was kept at 0.1 MPa, and the permeate side was kept at atmospheric pressure.

Membrane electrode assembly (MEA) fabrication and single-cell performance. The MEA is made up of an AEM and anode/cathode electrodes. The PES-B100-C16 membrane in hydroxide form was chosen as the AEM for the construction of MEA considering its overall properties. The catalyst Pt/C (40 wt% Pt, Johnson Matthey) and glycol were mixed with ionomer solution (5 wt% OH⁻ form membrane in DMF) at room temperature under ultrasonic dispersion for 30 min and vigorously stirred for 24 h to obtain a homogeneous catalyst ink. The weight ratio of ionomer to catalyst in the catalyst ink was controlled to be 3 : 7. Afterwards, the catalyst ink was coated on a carbon–carbon composite paper (Toray TGP-H-060, Japan) to obtain anode/cathode catalyst layers with an effective electrode area of 2 cm × 2 cm and a catalyst loading of 1 mg cm⁻². The membrane was sandwiched between the anode and cathode electrodes and followed by hot-pressing at 0.5 MPa and 50 °C for 5 min to obtain MEA for fuel cell test. The single cell test was carried out at 50 °C with a humidity of 100% on an electronic load (ZY8714, ZHONGYING Electronic Co., Ltd.). The flow rate of H₂ and O₂ supplied to the anode and cathode channels was controlled at 100 mL min⁻¹ during the test.

Results and discussion

Synthesis and characterization of PES

The PES copolymers were synthesized via polycondensation between FPS and OCP using K₂CO₃ as the acid binding agent in anhydrous DMAc, as shown in Scheme 1. The high molecular weight of PES (M_n = 70 kg mol⁻¹, M_w = 100 kg mol⁻¹) was obtained as white fibers which were soluble in commonly used organic solvents such as chloroform, TCE, DMAC, DMF, and dimethyl sulfoxide (DMSO). The chemical structure was confirmed by ¹H NMR using CDCl₃ as the solvent. As shown in Fig. S1 (ESI†), all of the aromatic protons are belong to the proposed polymer structure. The peaks around 6.82, 7.18, 7.62, 7.77, 7.86 and 8.00 ppm are assigned to the chemical shifts of protons on the benzene ring (from H2 to H10). The peak around 2.14 ppm is ascribed to the chemical shifts of the proton on –CH₃ (H1).

Synthesis and characterization of brominated poly(ether sulfone) (PES-Bx)

So far, most of the phenolphthalein-containing AEMs are based on precursors generated via chloromethylation. However, the use of toxic chloromethyl methyl ether or chloromethyl octyl ether as the chloromethylation reagents limits its improvement. Besides, the chloromethylation reaction is time-consuming (several days) and often requires large excess of reagents.²⁷ Compared to chloromethylation, radical substitution of active hydrogen in benzylmethyl groups using NBS as the bromination reagent and BPO as the radical initiator is an effective approach for preparing copolymers containing high reactive bromomethyl groups.^31,32

We explored the effect of feed ratio on the degree of bromomethylation at 85 °C in a short time (5 h) and obtained PES with various degrees of bromomethylation under a similar reaction condition, as shown in Table S1 (ESI†). The degree of bromomethylation ranging from 18.6% to 82.4% depended on the molar ratio of NBS to –CH₃ and increased with increasing the NBS content. The yield of product in the range of 82–91% was calculated from the mole of PES used in the reaction and the amount of obtained PES-Bx.

Fig. S2 (ESI†) shows the ¹H NMR spectrum of PES-Bx, the peaks around 2.05 ppm and 4.29 ppm were assigned to the methyl protons and bromomethyl protons, respectively. Herein, the degree of bromomethylation could be determined by comparison of the integration of methylene with methyl. The peaks above 6.5 ppm were attributed to the aryl hydrogen atoms except for the peak around 7.25 ppm (CDCl₃). All the spectra agreed well with the proposed molecular structure, demonstrating the successful synthesis of target brominated copolymers.

Quaternization, membrane formation and FT-IR spectra

PES-Bx-C16 (x = 60, 80 and 100) copolymers were synthesized via the Menshutkin reaction of PES-Bx copolymers with DMHDA. Taking PES-B100-C16 as an example, the membrane was characterized by ¹H NMR, as shown in Fig. S3 (ESI†). The peak of bromomethyl protons at 4.29 ppm disappeared in the ¹H NMR spectra while a broad hump appeared from 4.53–4.74 ppm, which could be attributed to the two benzylic protons H11. The characteristic peaks of the methylene of long alkyl chain of DMHDA salts at around 1.22, 1.66 and 3.00 ppm and methyl protons at around 0.83 and 3.00 ppm indicate that quaternary ammonium groups were introduced into the copolymers successfully.

The chemical structure of the copolymers and AEMs were further evidenced by FT-IR spectra. As shown in Fig. S4 (ESI†), the peak at 1248 cm⁻¹ was attributed to the Ar–O bond, suggesting the successful synthesis of PES. For PES-B100, the characteristic peak at 628 cm⁻¹ which is disappeared in the spectra of PES-B100-C16 was assigned to the C–Br stretching vibration. The new characteristic peaks at 3407 and 1692 cm⁻¹ were attributed to the O–H and C–N stretching vibration, respectively. All the spectra agreed well with the proposed molecular structure, indicating the successful synthesis of the target copolymers.

Morphology of the membranes

The appearance, SEM morphologies of surface and cross-section of PES-B100-C16 membrane are shown in Fig. S5.† A uniform and smooth structure with approximately 41 μm in thickness can be observed. This indicates a homogeneous and dense membrane being synthesized successfully. As shown in Fig. S5(a),† the membrane can be bended in any way, indicating that the as-synthesized membranes are flexible enough for fuel cell applications.

AFM phase images (tapping mode) of the PES-B60-C1 and PES-B100-C16 membranes were obtained, as shown in Fig. 1(a) and (b). The dark region was attributed to the hydrophilic domain which was comprised of quaternary ammonium groups containing small amounts of water, while the brighter one was assigned to the hydrophobic domain which was probably made up of aliphatic side chains. In addition, PES-B100-C16 has more dark domains than PES-B60-C1 with the aggregation of quaternary ammonium groups resulting in the formation of more ion conducting channels. Fig. 1(c) and (d) shows the TEM image of PES-B60-C1 and PES-B100-C16. The dark and bright regions are attributed to the hydrophilic and hydrophobic domains, respectively. The TEM images clearly show hydrophilic/hydrophobic phase separation regions for PES-B100-C16, which is in favour of providing a nano-channel for efficient ion conducting. However, no obvious phase separation can be observed for PES-B60-C1. Similar results were found in the AFM phase images. This suggests that comb-shaped architecture is effective at forming hydrophilic/hydrophobic phase separation, as noted by Li.²⁸

Fig. 1 (a and b) AFM phase images and (c and d) TEM images of PES-B60-C1 (left) and PES-B100-C16 (right) membranes.

The phase separation structure of the AEMs was also characterized using small angle X-ray scattering (SAXS). A non-comb-shaped PES-B60-C1 AEM based on PES-B60 and trimethylamine was synthesized for comparison. Fig. 2 shows large width scattering peak at 3.05 nm⁻¹ for all the comb-shaped PES-Bx-C16 membranes, indicating the generation of micro-phase segregation in the membranes. However, no obvious scattering peak was found for the non-comb-shaped PES-B60-C1 membrane. Since the peak position (q_max) is relative to hydrophilic aggregates, the value of interdomain spacing (d) can be calculated from the Bragg equation, d = 2π/q_max.^33,34 The average interdomain spacing is 2.06 nm for the PES-Bx-C16 membranes, roughly equivalent to the length of the aliphatic side chains. Micro-phase segregation in the membranes are beneficial to the enhancement of localized ion concentration, which could increase the efficiency of ion hopping and thus improve ion conduction in the AEMs.³⁵ AFM, TEM and SAXS results suggest that a micro-phase segregation structure was formed for the comb-shaped AEMs. As shown below, this would have a great influence on ionic conductivity of the AEMs.

Fig. 2 SAXS profiles of PES-Bx-C16 and PES-B60-C1 membranes in the dry state.

Ionic exchange capacity, water uptake, swelling ratio of the PES-Bx-C16 membranes

Ionic exchange capacity (IEC) is a critical factor that dominates the properties such as water uptake and swelling ratio of the AEMs. As listed in Table 2, the experimental IEC values of the comb-shaped PES-Bx-C16 membranes determined by back titration ranging from 0.94 to 1.35 meq. g⁻¹ agreed well with the values from the ¹H NMR spectra of quaternized copolymers. The IEC values of the AEMs can be improved by increasing the amount of ion exchange groups in the membrane. However, this is often accompanied with excessive water uptake which may result in a loss of mechanical properties. Therefore, the IEC values of the AEMs should be controlled within an appropriate range to get the optimized performance.

Water uptake (WU) and swelling ratio (SR) are great important properties for AEMs due to their enormous effect on the ionic conductivity and mechanical properties of AEMs. A membrane containing appropriate water can dissociate the alkali functionality and provide extra active transport channels for hydroxide and thus improve ionic conductivity. However, excess water content will result in over-swelling and thus decline mechanical properties. Therefore, an ideal strategy for designing AEMs would be expected to maintain water at a suitable level that can facilitate efficient ion transport channels and control swelling ratio at a low value.

As shown in Table 1, the water uptake of the comb-shaped PES-Bx-C16 membranes increased with increasing IEC and temperature. For example, the water uptake of the comb-shaped PES-Bx-C16 membranes increased from 5.6% to 10.4% at 30 °C and from 6.7% to 12.6% at 60 °C. This may be attributed to the increased mobility of the molecular chain at elevated temperature, which offers more free volume for holding water. Fig. 3 shows the swelling ratio of PES-Bx-C16 and PES-B60-C1 in the in-plane and through-plane direction at 30 °C. The comb-shaped PES-Bx-C16 showed much lower water uptake and swelling ratio than non-comb-shaped PES-B60-C1 even at a similar IEC. This indicates that long alkyl side chain with hydrophobicity in PES-Bx-C16 is effective at constraining water uptake and suppressing the swelling of the copolymer membranes resulting in a good dimensional stability.^28,36 Besides, the comb-shaped PES-Bx-C16 membranes exhibited slight anisotropic swelling behavior, in which the dimensional change in the through-plane direction was a little larger than that in the in-plane direction.

Table 1 IEC, water uptake (WU) and mechanical properties of the membranes

Membrane	IEC (meq. g^−1)		WU (%)		Tensile strength (MPa)	Elongation at break (%)
	Cal.^a	Exp.^b	30 °C	60 °C
a Calculated from ^1H NMR spectra.b Measured by back titration.
PES-B60-C16	0.97	0.94 ± 0.08	5.6 ± 0.3	6.7 ± 0.2	48 ± 5	9.4 ± 0.5
PES-B80-C16	1.18	1.11 ± 0.11	7.2 ± 0.3	8.2 ± 0.3	36 ± 6	8.7 ± 0.4
PES-B100-C16	1.43	1.35 ± 0.09	10.4 ± 0.4	12.6 ± 0.3	27 ± 4	7.4 ± 0.5
PES-B60-C1	1.49	1.40 ± 0.06	20.8 ± 0.3	32.1 ± 0.5	33 ± 5	5.8 ± 0.5

---

Fig. 3 Dimensional swelling of the AEMs at 30 °C.

Ionic conductivity

Ionic conductivity of AEMs is a critical property that can determine the output performance of a fuel cell system. The in-plane and through-plane ionic conductivity of the as-synthesized AEMs was measured under hydrated conditions as a function of temperature using the two-electrode AC impedance method. As shown in Fig. 4(a) and (b), the ionic conductivity of the as-prepared AEMs increased with increasing temperature because water mobility was enhanced at elevated temperature. It is also observed that the ionic conductivity increased with increasing IEC. This can be explained by that the more concentration of functional groups, the more OH⁻ charge carriers are. As apparent from Fig. 4(a), the in-plane conductivity of the comb-shaped PES-Bx-C16 membranes was in the range of 9.8–18.9 mS cm⁻¹ at 30 °C and 25.7–44.0 mS cm⁻¹ at 80 °C, and was much higher than that of the non-comb-shaped PES-B60-C1 membrane (5.8–22 mS cm⁻¹) under the same testing conditions. Similar results were obtained in the through-plane conductivity of the AEMs versus temperature plots. Besides, the in-plane conductivity of the comb-shaped PES-Bx-C16 membranes was found to be a little higher than the through-plane conductivity indicating slight anisotropy. The high ionic conductivity of the comb-shaped PES-Bx-C16 membranes at a very low swelling ratio was ascribed to the presence of hydrophilic/hydrophobic phase segregation that fabricated ion channels inside the membrane (as shown in Fig. 1(b) and (d)), which is effective for conducting hydroxide.

Fig. 4 (a) In-plane and (b) through-plane conductivity of the AEMs as a function of temperature.

Fig. S6(a) and (b)† shows the relationship between ln σ and 1000/T, which is conformed to Arrhenius behavior. The ionic transport activation energies (E_a) of the membranes could be calculated from: E_a = −bR, where b is the slope of the regressed linear ln σ–1000/T plots and R is the gas constant of 8.314 J mol⁻¹ K⁻¹. The apparent activation energies (E_a) were calculated to be 14.7, 14.0, 17.4 kJ mol⁻¹ for the in-plane conductivity of PES-B100-C16, PES-B80-C16 and PES-B60-C16, which were lower than that of PES-B60-C1 (24.02 kJ mol⁻¹). Similar results were obtained in the Arrhenius plots for the through-plane conductivity of the AEMs. This suggests that the hydroxyl ions could transfer more easily in comb-shaped PES-Bx-C16 than in non-comb-shaped PES-B60-C1. It may be attributed to the presence of micro-phase separation in PES-Bx-C16 with good continuity of the hydrophilic regions for conducting ions.

Herein, we introduce in-plane conductivity against swelling ratio (σ/SR) to assess the feature of the as-synthesized comb-shaped membranes, as listed in Table 2. A higher σ/SR value means that a membrane obtained a higher conductivity at the same swelling ratio. Although the PES-Bx-C16 membranes exhibited a relative lower ionic conductivity, the σ/SR value was higher than those of block type,³⁷ side-chain type and main-chain type membranes.^38,39 It was ascribed to the phase segregation structure inside membrane which enhances conductivity and the presence of long hydrophobic alkyl side chain that constrains the water uptake and thus restrains the swelling.

Table 2 Conductivity (σ), in-plane swelling ratio (SR) and σ/SR of the AEMs

Membrane	Temp. (°C)	σ (mS cm^−1)	SR (%)	σ/SR	Reference
PES-B60-C16	30	9.8	2.9	3.4	This work
	60	18.3	3.3	5.5
PES-B80-C16	30	13.7	3.6	3.8	This work
	60	22.9	4.0	5.7
PES-B100-C16	30	18.9	4.8	3.9	This work
	60	29.2	5.7	5.1
ImPESN-9-22	30	38.5	18.3	2.1	37
	60	60.0	23.2	2.6
PAES-Q-100	25	28.2	15.6	1.8	38
	80	62.9	23.4	2.7
QPPO-0.36	25	12.9	9.1	1.4	39
BQAPPO-0.15	25	36.3	11.1	3.3	39

---

Furthermore, the humidity dependence of in-plane conductivity was measured for comb-shaped PES-Bx-C16 and non-comb-shaped PES-B60-C1 at 50 °C, as shown in Fig. 5. It is also observed that the comb-shaped architecture greatly improved the ionic conductivity of the membranes. PES-B100-C16 (IEC = 1.35 meq. g⁻¹) showed higher ionic conductivity than PES-B60-C1 with comparable IEC value (1.40 meq. g⁻¹) over a relative humidity ranging from 30–90%. The difference in the ionic conductivity is more significant under low humidity conditions. Combined with morphology analysis of the membranes, all these results suggest that the comb-shaped PES-Bx-C16 membranes is effective at forming ion conducting channels resulting in improved conductivity.

Fig. 5 In-plane conductivity of the AEMs as a function of RH at 50 °C.

Mechanical property and thermal stability

The mechanical properties of the AEMs are of great importance for fabricating the membrane electrode assembly for alkaline fuel cells. The mechanical properties of the as-synthesized membranes in the hydroxide form were determined at room temperature. As shown in Table 1, the tensile strength of the comb-shaped PES-Bx-C16 membranes ranged from 27 to 48 MPa and the elongation at break varied from 7.4% to 9.4%, indicating that the comb-shaped membranes are strong enough for potential application in fuel cells. In general, the tensile strength of the comb-shaped membranes decreased with increasing IEC values because a higher IEC value results in higher water content and swelling. The similar behavior was observed in the literature.^37,40

The thermal stability of the copolymers and AEMs in the hydroxide form was investigated by TGA, as shown in Fig. 6. Although the samples were dried before thermal test, there is still a slight weight loss (<5%) of the PES-B100-C16 membrane below 140 °C because of the evaporation of absorbed water and residual solvent (DMF). The weight loss above 330 °C is assigned to the degradation of the copolymer main chain, which is similar to the curves of PES and PES-B100. In summary, the comb-shaped PES-B100-C16 membrane exhibited a good thermal stability below 140 °C, which can meet the demand of alkaline fuel cells (operating at 23–70 °C).⁴¹

Fig. 6 TGA curves of PES, PES-B100 and PES-B100-C16 membranes.

Solubility

The solubility of the membranes in commonly used membrane-forming solvents was tested, as listed in Table S2 (ESI†). PES and PES-Bx were soluble in strong polar aprotic solvents with high-boiling-point, such as DMF, DMAc, DMSO and NMP, but insoluble in lower alcohols. After quaternization, the PES-Bx-C16 was still soluble in most polar aprotic solvents, allowing easy formation of high-quality membranes by solvent casting method. It might be ascribed to the presence of phenolphthalein groups on the synthesized copolymers, which could reduce the crystallinity of the copolymers so that they were soluble in polar aprotic solvents.²³ Therefore, the good solubility of the comb-shaped PES-Bx-C16 membranes in organic solvents made them possible to use as an ionomer in the catalyst layer for membrane electrode assemble. In addition, the PES-Bx-C16 membranes were insoluble in lower alcohols with low boiling point in spite of their good solubility in organic solvents, indicating that they may also be used for direct alcohol fuel cells without loss arising from solubility. For the application in alkaline fuel cells, AEMs should possess long term resistant ability in a strong alkaline environment.

Alkaline stability

To evaluate the tolerance of the membranes in alkaline condition, the alkaline stability of PES-B100-C16 and PES-B60-C1 membranes was investigated in a 2 M aqueous KOH solution at 60 °C, and the change of ionic conductivity with soaking time was observed. Fig. 7 shows that the ionic conductivity of the PES-B100-C16 membrane decreased sharply within 96 h, after which the ionic conductivity tended to be constant. The comb-shaped PES-B100-C16 membrane still maintained about 70% of the initial ionic conductivity after 360 h while the ionic conductivity of the contrast non-comb-shaped PES-B60-C1 membrane decreased even more in the initial 72 h and broke into small pieces after 3 days. The excellent alkaline stability of PES-B100-C16 membrane is mainly ascribed to the introduction of large β-alkyl side chains with high electron density, which may impede Hofmann elimination and prevent the quaternary ammonium groups from OH⁻ attack.^28,35

Fig. 7 The alkaline stability of the PES-B100-C16 and PES-B60-C1 membranes in a 2 M aqueous KOH solution at 60 °C.

The ¹H NMR (Fig. 8) and FT-IR spectra (Fig. S7†) of the comb-shaped PES-B100-C16 membrane before and after alkaline stability test were further recorded to examine the chemical structure changes. Compared with the ¹H NMR spectra before test (Fig. 8(a)), a slight decrease of the peak area around 3.20, 1.86, 1.28 and 0.88 ppm assigned to the alkyl chain close to cationic group was found after the alkaline stability test (Fig. 8(b)). This indicates a small number of quaternary ammonium groups being degraded during the test. A small peak around 3.5 ppm (Fig. 8(a)) was ascribed to the proton resonances from impurities, and it doesn't belong to the PES-B100-C16 membrane. During alkaline stability test, the impurities may be decomposed and dissolved in the KOH solution, and thus disappeared in the ¹H NMR spectrum. As shown in Fig. S7,† the peak at 1248 cm⁻¹ and 1692 cm⁻¹ assigned to the Ar–O and C–N bond without significant change before and after alkaline test suggest the integrity of the groups. All the results indicate that the comb-shaped PES-B100-C16 membrane was stable enough for fuel cell testing.

Fig. 8 The ¹H NMR spectra of PES-B100-C16 (a) before and (b) after alkaline stability test.

Single cell performance

Although the comb-shaped PES-Bx-C16 AEMs exhibit high ionic conductivity and excellent alkaline stability, the most practical evaluation of the AEMs is their performance in single cells. Considering the overall properties of the AEMs, the PES-B100-C16 membrane in OH⁻ form was chosen as the ionomer in anode/cathode catalyst layer for the single fuel cell performance testing. The test was running at 50 °C with H₂/O₂ flow rate of 100 mL min⁻¹ under 100% relative humidity and the corresponding polarization curve and power density curve were obtained, as shown in Fig. 9. The open circuit voltage (OCV) of the single cell is 0.67 V, and is lower than the theoretical value (1.23 V). This may be attributed to the permeation of H₂/O₂ through the PES-B100-C16 membrane resulting in a decrease of fuel cell performance.⁴² As shown in Table S3,† the permeability of hydrogen and oxygen through the PES-B100-C16 membrane at 50 °C is 4.2 × 10⁻⁹ and 3.2 × 10⁻¹⁰ cm³ cm cm⁻² s⁻¹ cmHg⁻¹, respectively, and is close to Nafion membrane.⁴³ This may be ascribed to the introduction of bulky and rigid phenolphthalein groups in the main chain that provided free volume in membranes which will result in gas molecules crossing through the membrane.^44,45

Fig. 9 Polarization curve and power density curve of H₂/O₂ fuel cell using PES-B100-C16 membrane at 50 °C.

In addition, the ohmic resistance recorded during the test was close to a constant value of 0.7 Ω cm² (Fig. S8†), which may contain the membrane resistance and inter-layer contact resistance of catalyst. The maximum power density for the fuel cell was 43 mW cm² at a current density of 125 mA cm⁻² without optimization, which is higher than that of membranes GPPO-0.54 (16 mW cm²) and QPPT-35 (6 mW cm²) at 50 °C,^46,47 suggesting that the comb-shaped phenolphthalein-containing membranes have the potential application in the alkaline fuel cell. It's also worth pointing out that the single cell performance is also affected by the architecture of catalyst layer, MEA fabrication and operating conditions.^48,49 Hence, lots of work should be focused on these issues for a better single cell performance.

Conclusions

In this work, we have synthesized and characterized a series of comb-shaped phenolphthalein-containing AEMs via condensation polymerization, bromination, quaternization and alkalization. Compared with non-comb-shaped AEMs, the comb-shaped AEMs not only have an improved dimensional stability but also exhibit high ionic conductivity. The highest ionic conductivity of the AEMs with an IEC value of 1.35 mmol g⁻¹ is reach up to 44.0 mS cm⁻¹ at 80 °C. In addition, the comb-shaped membranes exhibit good solubility in high-boiling-point solvents as well as robust mechanical strength, excellent alkaline and thermal stability. The PES-B100-C16 membrane was tested in a H₂/O₂ alkaline fuel cell system and showed a good single cell performance. These results show that the comb-shaped phenolphthalein-containing membranes have the potential application for the AEM materials in alkaline fuel cells. Further work will focus on optimizing MEA fabrication and fuel cell testing condition.

Acknowledgements

Financial support from the National Nature Science Foundation of China (grant no. 21376194 & 21576226), the Nature Science Foundation of Fujian Province of China (grant no. 2014H0043), and the research fund for the Priority Areas of Development in Doctoral Program of Higher Education (no. 20130121130006) is gratefully acknowledged.

Notes and references

1. B. C. Steele and A. Heinzel, Nature, 2001, 414, 345–352.
2. M. Jacobson, W. Colella and D. Golden, Science, 2005, 308, 1901–1905.
3. W. Y. Hsu and T. D. Gierke, J. Membr. Sci., 1983, 13, 307–326.
4. K. A. Mauritz and R. B. Moore, Chem. Rev., 2004, 104, 4535–4586.
5. T. Mochizuki, K. Kakinuma, M. Uchida, S. Deki, M. Watanabe and K. Miyatake, ChemSusChem, 2014, 7, 729–733.
6. Y. J. Wang, J. Qiao, R. Baker and J. Zhang, Chem. Soc. Rev., 2013, 42, 5768–5787.
7. O. Z. Sharaf and M. F. Orhan, Renewable Sustainable Energy Rev., 2014, 32, 810–853.
8. G. McLean, T. Niet, S. P. Richard and N. Djilali, Int. J. Hydrogen Energy, 2002, 27, 507–526.
9. N. Wagner, M. Schulze and E. Gülzow, J. Power Sources, 2004, 127, 264–272.
10. J. R. Varcoe and R. C. Slade, Fuel Cells, 2005, 5, 187–200.
11. W. Lu, Z.-G. Shao, G. Zhang, Y. Zhao, J. Li and B. Yi, Int. J. Hydrogen Energy, 2013, 38, 9285–9296.
12. H. Sun, G. Zhang, Z. Liu, N. Zhang, L. Zhang, W. Ma, C. Zhao, D. Qi, G. Li and H. Na, Int. J. Hydrogen Energy, 2012, 37, 9873–9881.
13. Y. Xiong, Q. L. Liu and Q. H. Zeng, J. Power Sources, 2009, 193, 541–546.
14. J. Fang and P. K. Shen, J. Membr. Sci., 2006, 285, 317–322.
15. L. Wu, Q. Pan, J. R. Varcoe, D. Zhou, J. Ran, Z. Yang and T. Xu, J. Membr. Sci., 2015, 490, 1–8.
16. J. Ran, L. Wu, Y. Ru, M. Hu, L. Din and T. Xu, Polym. Chem., 2015, 6, 5809–5826.
17. J. Ran, L. Wu, B. Wei, Y. Chen and T. Xu, Sci. Rep., 2014, 4, 1–5.
18. T. A. Sherazi, S. Zahoor, R. Raza, A. J. Shaikh, S. A. R. Naqvi, G. Abbas, Y. Khan and S. Li, Int. J. Hydrogen Energy, 2015, 40, 786–796.
19. Q. H. Zeng, Q. L. Liu, I. Broadwell, A. M. Zhu, Y. Xiong and X. P. Tu, J. Membr. Sci., 2010, 349, 237–243.
20. J. Fang, M. Lyu, X. Wang, Y. Wu and J. Zhao, J. Power Sources, 2015, 284, 517–523.
21. Y. Ye, S. Sharick, E. M. Davis, K. I. Winey and Y. A. Elabd, ACS Macro Lett., 2013, 2, 575–580.
22. J. H. Chen, Q. L. Liu, Y. Xiong, Q. G. Zhang and A. M. Zhu, J. Membr. Sci., 2008, 325, 184–191.
23. G. L. Han, Q. G. Zhang, A. M. Zhu and Q. L. Liu, Sep. Purif. Technol., 2013, 107, 211–218.
24. L. Li and Y. Wang, J. Membr. Sci., 2005, 262, 1–4.
25. A. K. Singh, R. P. Pandey and V. K. Shahi, RSC Adv., 2014, 4, 22186.
26. S. Chempath, B. R. Einsla, L. R. Pratt, C. S. Macomber, J. M. Boncella, J. A. Rau and B. S. Pivovar, J. Phys. Chem. C, 2008, 112, 3179–3182.
27. J. Yan and M. A. Hickner, Macromolecules, 2010, 43, 2349–2356.
28. N. W. Li, T. Z. Yan, Z. Li, T. Thurn-Albrecht and W. H. Binder, Energy Environ. Sci., 2012, 5, 7888–7892.
29. N. Li, Y. Leng, M. A. Hickner and C. Y. Wang, J. Am. Chem. Soc., 2013, 135, 10124–10133.
30. X. M. Wu, Q. G. Zhang, P. J. Lin, Y. Qu, A. M. Zhu and Q. L. Liu, J. Membr. Sci., 2015, 493, 147–155.
31. X. Li, S. Cheng, L. Wang, Q. Long, J. Tao, G. Nie and S. Liao, RSC Adv., 2014, 4, 29682–29693.
32. P. Y. Xu, K. Zhou, G. L. Han, Q. G. Zhang, A. M. Zhu and Q. L. Liu, J. Membr. Sci., 2014, 457, 29–38.
33. A. N. Lai, K. Zhou, Y. Z. Zhuo, Q. G. Zhang, A. M. Zhu, M. L. Ye and Q. L. Liu, J. Membr. Sci., 2016, 497, 99–107.
34. G. L. Han, P. Y. Xu, K. Zhou, Q. G. Zhang, A. M. Zhu and Q. L. Liu, J. Membr. Sci., 2014, 464, 72–79.
35. J. R. Varcoe, P. Atanassov, D. R. Dekel, A. M. Herring, M. A. Hickner, P. A. Kohl, A. R. Kucernak, W. E. Mustain, K. Nijmeijer and K. Scott, Energy Environ. Sci., 2014, 7, 3135–3191.
36. A. H. Rao, S. Nam and T.-H. Kim, J. Mater. Chem. A, 2015, 3, 8571–8580.
37. A. N. Lai, L. S. Wang, C. X. Lin, Y. Z. Zhuo, Q. G. Zhang, A. M. Zhu and Q. L. Liu, ACS Appl. Mater. Interfaces, 2015, 7, 8284–8292.
38. G. Nie, X. Li, J. Tao, W. Wu and S. Liao, J. Membr. Sci., 2015, 474, 187–195.
39. Y. He, J. Pan, L. Wu, Y. Zhu, X. Ge, J. Ran, Z. Yang and T. Xu, Sci. Rep., 2015, 5, 1–7.
40. Y. Z. Zhuo, A. N. Lai, Q. G. Zhang, A. M. Zhu, M. L. Ye and Q. L. Liu, J. Membr. Sci., 2015, 491, 138–148.
41. G. Merle, M. Wessling and K. Nijmeijer, J. Membr. Sci., 2011, 377, 1–35.
42. N. Asano, M. Aoki, S. Suzuki, K. Miyatake, H. Uchida and M. Watanabe, J. Am. Chem. Soc., 2006, 128, 1762–1769.
43. H. Ono, J. Miyake, S. Shimada, M. Uchida and K. Miyatake, J. Mater. Chem. A, 2015, 3, 21779–21788.
44. J. F. Zheng, J. Wang, S. B. Zhang, T. Yuan and H. Yang, J. Power Sources, 2014, 245, 1005–1013.
45. A. Y. Houde, S. S. Kulkarni and M. G. Kulkarni, J. Membr. Sci., 1994, 95, 147–160.
46. X. Lin, L. Wu, Y. Liu, A. L. Ong, S. D. Poynton, J. R. Varcoe and T. Xu, J. Power Sources, 2012, 217, 373–380.
47. X. Lin, Y. Liu, S. D. Poynton, A. L. Ong, J. R. Varcoe, L. Wu, Y. Li, X. Liang, Q. Li and T. Xu, J. Power Sources, 2013, 233, 259–268.
48. S. D. Poynton, R. C. T. Slade, T. J. Omasta, W. E. Mustain, R. Escudero-Cid, P. Ocón and J. R. Varcoe, J. Mater. Chem. A, 2014, 2, 5124–5130.
49. Y. Zhao, H. Yu, F. Xie, Y. Liu, Z. Shao and B. Yi, J. Power Sources, 2014, 269, 1–6.

---

Footnote

† Electronic supplementary information (ESI) available: ¹H NMR spectra of PES in CDCl₃; results of the bromination of PES using different amounts of NBS; ¹H NMR spectra of PES-Bx in CDCl₃; ¹H NMR of PES-B100-C16 in bromine form in DMSO-d₆; FT-IR spectra of PES, PES-B100, and PES-B100-C16; digital photo and SEM images: cross-section, surface of the PES-B100-C16 membrane; solubility of the comb-shaped PES-Bx-C16 membranes in commonly used solvents; FT-IR spectra of PES-B100-C16 (a) before and (b) after alkaline stability test; H₂ and O₂ permeability of the PES-B100-C16 membrane in hydroxide form at 50 °C; ohmic resistance of MEA using the PES-B100-C16 membrane. See DOI: 10.1039/c5ra22774g

---