Microstructure and properties of novel SPEEK/PVDF-g-PSSA blends for proton exchange membrane with improved compatibility

Abstract

Novel blend membranes of poly(vinylidene fluoride) grafted poly(styrene sulfonic acid) (PVDF-g-PSSA) with sulfonated poly(ether ether ketone) (SPEEK) were fabricated for direct methanol fuel cell applications. The PVDF-g-PSSA copolymer can be homogeneously dispersed in the SPEEK matrix with a unique hierarchical particle surface. All the SPEEK/PVDF-g-PSSA blend membranes exhibit better dimensional stability, lower methanol permeability, higher selectivity than commercial Nafion® 117 at room temperature. Notably, at the PVDF-g-PSSA content of 5 wt%, the blend membrane exhibits the best overall membrane performance, which is supposed to benefit from the improved compatibility and strong interactions between the PVDF-g-PSSA and SPEEK matrix. The thermal stability and the temperature dependence of water uptake, dimensional change and proton conductivity of the blend membranes are also investigated. This study provides a promising strategy for the design and fabrication of high performance polymer electrolyte membrane by rational control of microstructure and compatibility of the blend membrane.

---

1. Introduction

Polymer electrolyte membrane (PEM) is one of the key components in direct methanol fuel cells (DMFCs), acting as an electrolyte for proton conduction, as well as a barrier for methanol crossover.^1,2 The commercial perfluorinated ionomer membranes such as Nafion® possess high proton conductivity but also high methanol permeability and high cost.³ These shortcomings have stimulated the development of new materials as PEMs for DMFC applications. Sulfonated aromatic polymer membranes have gained considerable attention owing to their excellent thermal and chemical stability, good mechanical properties, low cost and good processability, etc.⁴ The most widely used and intensively studied sulfonated aromatic polymer membrane is SPEEK, which is considered to be the most promising candidate as an alternative to Nafion®.⁵ The properties of SPEEK highly rely on the degree of sulfonation (DS). Increasing the DS of SPEEK is an important approach to achieve high proton conductivity, while high DS will also lead to brittleness, obvious reduction of methanol resistance and dimensional stability,⁶ and inferior hydrolytic stability at elevated temperature, which hinders the use of SPEEK as PEMs in DMFCs.^7,8

Blending SPEEK with another polymer such as polyethersulfone (PES),⁹ polysulfone (PSF),¹⁰ PVDF,¹¹ Nafion®,¹² etc. has been demonstrated to be an effective way to modify the properties of SPEEK for achieving optimum performance of the blend membranes. Among these polymers, PVDF has been widely investigated due to its outstanding mechanical properties, dimensional stability, methanol resistance and commercial availability.¹³ The blend membranes based on SPEEK and PVDF exhibit improved stability and methanol resistance, which is very attractive for DMFCs applications.^14,15 However, these improvements always come at the expense of proton conductivity, and, with increasing PVDF content, may also bring about the instability of membrane properties.^16,17 This phenomenon results from the intrinsic incompatibility of the hydrophilic SPEEK and the hydrophobic PVDF, which leads to non-uniform phase separation of the blend membrane.¹⁸ Therefore, constructing SPEEK/PVDF blend membrane with improved compatibility and uniform phase separation should be a very promising direction for the fabrication of high performance PEM.

In this study, in order to improve the compatibility between SPEEK and PVDF, a novel series of SPEEK/PVDF-g-PSSA blend membranes were designed and fabricated for the first time. The PVDF was slightly grafted by styrene sulfonic acid (SSA) side chains to enhance its hydrophilicity and for the better control of the phase separation with high DS SPEEK matrix. The morphology of the blend membrane and further investigation of the element distribution of an individual PVDF-g-PSSA particle were studied for compatibility analysis. The impact of PVDF-g-PSSA content on the properties of SPEEK/PVDF-g-PSSA blend membrane at room temperature was investigated and compared with that of SPEEK/PVDF blend membrane and Nafion® 117. The thermal stability and the temperature dependence of the water uptake, dimensional stability and proton conductivity of the SPEEK/PVDF-g-PSSA blend membranes were also investigated to evaluate their potential practical applications for DMFCs.

2. Experimental section

2.1. Materials and chemicals

PEEK (M_w = 38 300 g mol⁻¹) and PVDF (M_n = 130 000 g mol⁻¹) were purchased from Victrex and Arkema Company, respectively. Nafion® 117 membrane was purchased from Dupont Company. 4-Styrene sulfonic acid (SSA) and 1,1,4,7,10,10-hexamethyl triethylene tetramine (HMTETA) were purchased from Aladdin Reagent Company (Shanghai, China). Copper(I) chloride (CuCl), 1-methyl-2-pyrrolidinone (NMP), dimethyl sulfoxide (DMSO), H₂SO₄ (98%) were purchased from Beijing Chemical Reagent Factory. All solvent and chemicals were reagent grade, and used without further purification.

2.2. Synthesis of SPEEK and PVDF-g-PSSA graft copolymer

The PVDF-g-PSSA graft copolymer was synthesized following the procedures reported in the literature.¹⁹ Firstly, PVDF (1 g) was dissolved in NMP (10 ml) and SSA (6 g) was dissolved in DMSO (15 ml), respectively. The two solutions were mixed together in a round flask, and then CuCl (0.01 g) and HMTETA (0.05 ml) were added. The reaction flask was sealed and the reaction proceeded at 120 °C for 48 h under N₂ protection. The resultant solution was precipitated into methanol and the precipitate was purified by redissolving in DMSO and reprecipitating in methanol. Finally, the precipitate was dried under vacuum overnight at 80 °C to get PVDF-g-PSSA. The grafting reaction was confirmed by Fourier transform infrared (FTIR) spectroscopy (Fig. S1†).

Typically 5 g of the dried PEEK was dissolved in 125 ml of H₂SO₄ under stirring for 30 min at room temperature (30 °C), followed by vigorous stirring at 50 °C for 3 h. Then, the solution was gradually precipitated into ice water under mechanical agitation overnight. The polymer precipitate was filtered and washed with deionized water until the pH was neutral, then dried under vacuum to get SPEEK. The ion exchange capacity (IEC) of the SPEEK and PVDF-g-PSSA was determined by the classical back-titration method and list in Table S1.†

2.3. Preparation and characterization of the blend membranes

The SPEEK/PVDF-g-PSSA (or SPEEK/PVDF) blend membranes with 0, 5, 10, 15, 20 wt% PVDF-g-PSSA (or PVDF) were prepared by following procedures. Firstly, certain weight of SPEEK and PVDF-g-PSSA (or PVDF) was dissolved in DMSO, respectively. Then the certain proportion of the two solutions was mixed together and magnetic stirred at 80 °C for 4 h to form a homogeneous solution. The solution was cast onto a glass dish and dried at 80 °C for 12 h, then heated at 100 °C for another 12 h under vacuum.

The microstructures of the membranes were analyzed using a FEI F20 transmission electron microscope (TEM) equipped with energy dispersive X-ray spectroscopic analyser (EDS) operated at 200 kV and a FEI Quanta 200F scanning electron microscope (SEM) operated at 30 kV. Differential scanning calorimetry (DSC) measurements were performed on a Netzsch 204 F1 equipped with an intercooler as cooling system under an argon atmosphere. The thermal gravimetric analysis (TGA) was performed with the TA Q5000 thermogravimetric analyzer.

2.4. Evaluation of water uptake, dimensional change, proton conductivity, methanol permeability and single cell performance test

For water uptake and dimensional change measurement, the membrane was previously dried overnight 12 h at 80 °C, and subsequently soaked in deionized water at room temperature for 24 h until fully hydrated. The weight, area and thickness of the dry membrane are W_dry, A_dry and l_dry, respectively. The W_wet, A_wet and l_wet corresponding to the respective parameters in equilibrated swollen states. The water uptake, in-plane swelling ratio and through-plane swelling ratio can be calculated as follows:

Water uptake = (W_wet − W_dry)/W_dry × 100%

In-plane swelling ratio = (A_wet − A_dry)/A_dry × 100%

Through-plane swelling ratio = (l_wet − l_dry)/A_dry × 100%

The in-plane proton conductivity measurements were performed in a two-probe cell using an AC impedance spectroscopic technique over the frequency range of 10 Hz to 1 MHz on a CHI660D electrochemical workstation. The membrane sample in acid form (3 cm × 1 cm) was first soaked in deionized water at room temperature for 48 h, and then put into a cell and heated by water vapor at a certain temperature ranging from 30 to 70 °C. The resistance of the membrane was calculated from the intercept of the real axis in the intermediate frequency domain of the impedance spectrum (Z′′ vs. Z′). The proton conductivity (σ, S cm⁻¹) of the membrane is calculated according to the equation:

σ = l/RA

where l, A, and R are the membrane thickness, membrane area, and membrane resistance, respectively.

The methanol permeability of the membrane was determined using a diaphragm diffusion cell, consisting of two identical compartments (100 ml) separated by the test membrane. One of the compartment A was filled with 1 M methanol and the other compartment B was filled with deionized water. Both compartments were magnetically stirred during the experiments. The increase in methanol concentration with time in compartment B was periodically determined by gas chromatography (GC, BFRL SP-2100A). The methanol permeability P was calculated as follows:

C_B(t) = APC_A(t − t₀)/V_Bl

where A, l and V_B are the effective membrane area, membrane thickness and the volume of compartment, respectively. C_B(t) is the methanol concentration in compartment B at time t, C_A (t − t₀) is the methanol concentration in compartment A between time 0 and t. The selectivity is defined as the ratio of proton conductivity to methanol permeability, which is often applied to evaluate the potential performance of PEMs.

The membrane electrode assembly (MEA) was made by sandwiching the blend membrane between a Pt–Ru catalyst (5 mg cm⁻²) anode and a Pt catalyst (1 mg cm⁻²) cathode to evaluate the single cell performance of the DMFCs. The effective membrane area in the MEA was 4 cm². The single cell performance was measured at room temperature. The anode input methanol solution (2 mol L⁻¹) flow rate was 4 ml min⁻¹, and the cathode input O₂ flow rate was 150 sccm.

3. Results and discussion

3.1. Microstructure of PVDF-g-PSSA particles in SPEEK matrix

The compatibility of polymers in blend membranes is considered to play a critical role on the properties of the membranes.^20–22 The blends of two completely incompatible polymers could lead to a macroscopic phase separation and instability properties, whereas, blending two miscible polymers could form a highly homogeneous phase with single glass transition temperature (T_g), which is unfavorable to exhibit the respective unique characteristics of each component. Based on this consideration, the incompatible highly hydrophobic PVDF and high DS SPEEK were selected in this study, and the former was slightly grafted by hydrophilic styrene sulfonic acid (SSA) side chains in order to enhance the compatibility and control the phase separation of the two polymers. The reaction scheme for the synthesis of PVDF-g-PSSA and SPEEK are presented in Scheme S1.†

Fig. 1a shows the photographs of a typical SPEEK/PVDF-g-PSSA (5 wt%) membrane. It is seen that the blend membrane is freestanding, flexible, and can be easily cut into desired sizes and shapes. Fig. 1b and c show the SEM images of the blend membranes at fluoropolymer content of 5 wt%. It can be seen from the low-magnification SEM image of SPEEK/PVDF-g-PSSA (5 wt%) membrane (Fig. 1b) that the PVDF-g-PSSA particles are homogeneously dispersed in the SPEEK matrix. Moreover, the SEM images of the blend membranes with increasing PVDF-g-PSSA content (Fig. S2a–d†) show that the PVDF-g-PSSA particles can also achieve a homogeneous dispersion with regular shapes and increasing sizes with their content up to 20 wt%. Further analysis of the high-magnification SEM image of the PVDF-g-PSSA particle (inset of Fig. 1b) shows that it exhibits rough outer surface with hierarchical microstructure and therefore attach firmly to the matrix. Besides, as the PVDF-g-PSSA content increase, more and more particles were broken in the cryo-fracture process (Fig. S2j†) instead of being extracted from the matrix, which further indicates the good adhesion at the interfaces. The homogeneous dispersion and good adhesion of PVDF-g-PSSA particles confirm an improved compatibility and strong interfacial interactions between the PVDF-g-PSSA and SPEEK attributing to the grafting reaction. As a comparison, although a roughly similar homogeneous dispersion can be observed from the SEM image of SPEEK/PVDF (5 wt%) membrane (Fig. 1c), the PVDF particles exhibit smooth surfaces (inset of Fig. 1c) and they leave empty cavities when detached from the SPEEK matrix (Fig. S2l†). This reveals the poor compatibility and weak interfacial interaction between the PVDF and SPEEK. In addition, the PVDF particles show obvious heterogeneous dispersion and irregular shapes with a broad distribution of sizes when their contents higher than 10 wt% (Fig. S2e–h†). By considering this tendency, the SPEEK/PVDF (5 wt%) membrane was selected as a reference to evaluate the effect of grafting reaction on the performance of the SPEEK/PVDF-g-PSSA membrane in this study.

Fig. 1 (a) Photographs of the SPEEK/PVDF-g-PSSA (5 wt%) membrane, (b) SEM images of the SPEEK/PVDF-g-PSSA (5 wt%) membrane, (c) SEM images of the SPEEK/PVDF (5 wt%) membrane (d) STEM images with EDS elemental mapping of an individual PVDF-g-PSSA particle in the SPEEK/PVDF-g-PSSA (5 wt%) membrane. The scheme shows the cryo-microtome of a 50 nm thick slice of the particle. The red line in the STEM image is the EDS line-scan across a single particle section. In the elemental mapping images, the green region is rich in sulfur and the yellow region is rich in fluorine, and the white curve and blue curve are the element distribution at the red line of sulfur and fluorine, respectively.

To further investigate the compatibility of the PVDF-g-PSSA particle and SPEEK matrix, the SPEEK/PVDF-g-PSSA (5 wt%) membrane was sectioned to yield slice of about 50 nm thick for the element distribution characterization of an individual PVDF-g-PSSA particle (Fig. 1d). According to the chemical structures of the two components (Scheme S1†), sulfur element exists in both SPEEK and PVDF-g-PSSA with higher amounts in the former, whereas fluorine element exists only in PVDF-g-PSSA. The scanning transmission electron microscopy (STEM) and EDS line-scan across a single PVDF-g-PSSA particle section (red line) were performed. The sulfur element (white curve) content is found to be gradually decreased from the SPEEK matrix to the center of the particle, which indicates that the SPEEK chains with high sulfonic acid group concentration diffused into the particle. Moreover, the fluorine element (blue curve) shows an opposite trend, and it can be found not only in the particle but also in the matrix, which suggests that the PVDF-g-PSSA chains can also diffuse into the SPEEK matrix. The inter-diffusion of the two polymers resulted from the entanglement of SPEEK and PVDF-g-PSSA chains owing to the strong interactions between the polymer chains after the grafting reaction. Thus, a co-continuous proton conducting network is believed to be formed for fast water transport and easy proton conduction.

3.2. Water uptake, dimensional change, proton conductivity, methanol permeability, and single cell performance at room temperature

Adequate water uptake in the PEMs is essential to maintaining membrane durability and ensuring efficient proton conduction of the membrane. Proton conduction occurs through water channels via the vehicular mechanism and Grotthuss mechanism.²³ Insufficient water uptake leads to reduction of proton conductivity, whereas excess water uptake results in large dimensional change and deterioration of methanol resistance. Fig. 2a displays the water uptake of the blend membranes at room temperature. As can be seen, the water uptake of the SPEEK/PVDF-g-PSSA blend membranes decreases with increasing PVDF-g-PSSA content as expected. In particular, it is lower than that of SPEEK/PVDF membrane at the same fluoropolymer content, which seems a little unusual as the SPEEK/PVDF-g-PSSA (5 wt%) membrane possesses more sulfonic acid groups content than SPEEK/PVDF (5 wt%). It is understandable because the poor interaction between the PVDF particles and the SPEEK matrix induced numerous voids generating between the two polymers (as shown in the inset of Fig. 1c), which can store water molecules and increase the water uptake. However, these water molecules cannot form water channels for efficient proton conduction due to the lack of sulfonic acid groups, instead, they result in large dimensional change.

Fig. 2 (a) Water uptake and proton conductivity, (b) in-plane and through-plane swelling ratio, (c) methanol permeability and selectivity, and (d) polarization and power density curves of membranes at room temperature.

Fig. 2b is the dimensional change of the blend membranes. It is seen that the in-plane swelling ratio of the SPEEK/PVDF blend membranes decreases with the addition of pristine PVDF, while a sharp increase in through-plane swelling ratio is also accompanied. In comparison, both the in-plane and through-plane swelling ratio of the SPEEK/PVDF-g-PSSA blend membranes dramatically decreases with increasing PVDF-g-PSSA content at room temperature, lower than that of Nafion® 117 membrane (in-plane swelling ratio = 31.79%, through-plane swelling ratio = 14.75%) measured under the same condition. These results indicate that the SPEEK/PVDF-g-PSSA blend membranes exhibit remarkably enhanced dimensional stability with isotropic swelling behavior after the grafting reaction, which is suggested to be benefit from the improved compatibility.

Proton conductivity is a key factor for the performance of PEMs in DMFCs, which is closely related to the sulfonic acid group concentration and the microstructure of the membrane.^24,25 The proton conductivity of the SPEEK/PVDF-g-PSSA membranes at room temperature ranges from 6.54 × 10⁻² to 1.19 × 10⁻¹ S cm⁻¹ (Fig. 2a), which is comparable to that of Nafion® 117 membrane (9.56 × 10⁻² S cm⁻¹) measured under the same condition. Notably, unlike other blend membranes which usually exhibit decreased proton conductivity by the incorporation of low IEC component, the SPEEK/PVDF-g-PSSA membrane exhibits enhanced proton conductivity at the PVDF-g-PSSA content of 5 wt%. This intriguing phenomenon could be due to at least two factors: (i) the low water uptake leading to the high sulfonic acid group concentration; (ii) the microstructure inducing the formation of the co-continuous proton conducting network. It is well accepted that the Grotthuss mechanism involves the proton hopping from one proton carrier site to the next through hydrogen bonds.²⁶ The introduction of 5 wt% PVDF-g-PSSA decreases the water uptake of pristine SPEEK by 12.1% of its original value, which indicates a higher sulfonic acid group concentration in the SPEEK/PVDF-g-PSSA (5 wt%) membrane. As a result, the distance between the adjacent sulfonic acid groups becomes much shorter, which is easier for proton hopping via Grotthuss mechanism. On the other hand, as discussed in Section 3.1, the SPEEK/PVDF-g-PSSA (5 wt%) membrane exhibits hierarchical microstructure and the polymer chains of the two components inter-diffuse into each other, which can induce the formation of large amounts of interfaces and the co-continuous proton conducting network for fast proton hopping. However, when further increase the PVDF-g-PSSA content, the sulfonic acid group concentration of the blend membrane decrease, which becomes the determining factor for proton conduction and consequently results in the decrease of proton conductivity.

Fig. 2c shows the methanol permeability and selectivity of the blend membranes. It can be seen that the SPEEK/PVDF-g-PSSA blend membranes exhibit improved methanol resistance than pristine SPEEK and SPEEK/PVDF blend membrane. It is suggested that the enhanced interaction between SPEEK and PVDF-g-PSSA can suppress the membrane swelling in methanol solution, and thus prevents the hydrophilic ionic clusters from aggregating into larger ionic clusters and consequently reduces the methanol crossover. Moreover, the hierarchical surfaces of PVDF-g-PSSA particles can increase the tortuosity of the methanol pathway, and retard the methanol crossover through the membranes. Besides, among all the membranes investigated in this paper, the SPEEK/PVDF-g-PSSA (5 wt%) membrane possesses the highest selectivity value, which is about 1.3-fold greater than pristine SPEEK and 2.1-fold greater than Nafion® 117 at room temperature.

The single cell performance of the SPEEK/PVDF-g-PSSA (5 wt%) membrane was measured because of its best overall membrane properties. Fig. 2d shows the polarization and power density curves of the DMFCs with SPEEK/PVDF-g-PSSA (5 wt%), SPEEK/PVDF (5 wt%), and Nafion® 117 as PEMs operated at room temperature. The DMFC with the SPEEK/PVDF-g-PSSA (5 wt%) membrane exhibits improved single cell performance compared to that of the SPEEK/PVDF (5 wt%) and even Nafion® 117 membrane. This improvement mainly attributed to the good dimensional stability, high proton conductivity and low methanol permeability of the SPEEK/PVDF-g-PSSA (5 wt%) membrane as demonstrated above. In comparison, the SPEEK/PVDF (5 wt%) membrane displays the lowest power density due to its poor dimensional stability and inferior proton conductivity. The above results indicate that the SPEEK/PVDF-g-PSSA membranes hold great promise as PEMs for DMFC applications.

3.3. Thermal analysis of the SPEEK/PVDF-g-PSSA blend membranes

The thermal property of the PEM is an important parameter for the long-term durability of DMFCs operation at high temperature. DSC analysis was used to characterize the thermal transition of the SPEEK/PVDF-g-PSSA blend membranes with results shown in Fig. 3a. The pristine SPEEK (black curve) shows a broad endothermic peak around 90 °C owing to the melting of PEEK semi-crystalline nanodomains and the evaporation of bound water in SPEEK.²⁷ The melting point (T_m) of pristine PVDF is 162 °C (blue curve), which increases to 166 °C in PVDF-g-PSSA (red curve), resulting from the suppressing of chain mobility by the grafted PSSA side chain. The T_m of PVDF-g-PSSA in the blend membranes was depressed (red curves), which indicates that the sulfonic acid groups of PVDF-g-PSSA take part in the interactions with SPEEK chains, hence suppressing the formation of crystal of PVDF-g-PSSA in the blend membrane. With increasing PVDF-g-PSSA content, the melting peak of PVDF-g-PSSA shifts to high temperature (black arrow) and becomes close to that of the pristine PVDF-g-PSSA, which indicates a greater phase separation of PVDF-g-PSSA and SPEEK.

Fig. 3 (a) DSC and (b) TGA curves of the SPEEK/PVDF-g-PSSA blend membranes.

The thermal stability of the SPEEK/PVDF-g-PSSA blend membranes is shown in Fig. 3b. Three weight losses are observed in the TGA curves. The first weight loss from 50 to 200 °C is the evaporation of water and residual solvent. The second weight loss that occurred between 200 °C and 350 °C can be attributed to the decomposition of sulfonic acid groups in the PVDF-g-PSSA and SPEEK. The third weight loss at 450–900 °C is caused by the degradation of the PVDF and SPEEK backbone. As can be seen from the second weight loss stage, the thermal stability of the blend membranes increase with increasing PVDF-g-PSSA content. This is because the strong interactions between the SPEEK and PVDF-g-PSSA inhibits the polymer chain motion at high temperature and thus enhances the thermal stability of the blend membranes. All the SPEEK/PVDF-g-PSSA blend membranes show sufficiently high thermal stability for the application as PEMs in DMFCs and their thermal stability increase with increasing PVDF-g-PSSA content.

3.4. Water uptake, dimensional change and proton conductivity as a function of temperature

Water uptake, dimensional change and proton conductivity of sulfonic acid polymer electrolyte membranes are known to be significantly dependent on temperature. Thus investigating the temperature dependence of these properties is important for the practical applications of the PEMs. Fig. 4a displays the water uptake of the blend membranes as a function of temperature. It can be seen that the water uptake of SPEEK/PVDF-g-PSSA (5 wt%) and SPEEK/PVDF (5 wt%) membranes are similar at 20 °C and both increase with increasing temperature, while the latter exhibits a larger temperature dependence and even exceeds that of the pristine SPEEK. This is because the increasing temperature results in bigger membrane swelling, thus more water molecules were absorbed into the voids between the PVDF and SPEEK (as discussed in Fig. 2a). In contrast, all the SPEEK/PVDF-g-PSSA membranes displays decreased water uptake comparing with pristine SPEEK membrane at various temperature because there is no such voids for the storage of water molecules between the two polymers due to the improved compatibility.

Fig. 4 Temperature dependence of (a) water uptake, (b) in-plane swelling ratio and (c) through-plane swelling ratio of the Nafion® 117, pristine SPEEK, SPEEK/PVDF (5 wt%) membrane and SPEEK/PVDF-g-PSSA blend membranes.

Similar to water uptake, the in-plane (Fig. 4b) and through-plane swelling ratio (Fig. 4c) of all the membranes increase with increasing temperature. Consistent with the swelling behavior at room temperature, the SPEEK/PVDF (5 wt%) membrane exhibits anisotropic swelling behavior, while all the SPEEK/PVDF-g-PSSA blend membranes show isotropic swelling behavior at various temperature. The increase of both the in-plane and through-plane swelling ratio of SPEEK/PVDF-g-PSSA blend membranes is much more slowly with temperature compared with that of SPEEK/PVDF membrane, and it trends to level off with increasing PVDF-g-PSSA content. All the SPEEK/PVDF-g-PSSA blend membranes exhibit comparable dimensional stability to Nafion® 117. These results indicate that the SPEEK/PVDF-g-PSSA blend membrane exhibit superior dimensional stability at elevated temperature attributing to the improved compatibility.

The proton conductivity of the blend membranes as function of temperature obtained from fitting the equivalent circuit to the AC impedance spectra (Fig. S3†) are shown in Fig. 5a. It can be seen that the proton conductivity of both SPEEK/PVDF and SPEEK/PVDF-g-PSSA blend membranes increase with increasing temperature. The SPEEK/PVDF-g-PSSA (5 wt%) membrane exhibits the highest proton conductivity at each temperature, surpassing that of the pristine SPEEK and Nafion® 117. It can be concluded that proper control of PVDF-g-PSSA contents of the blend membrane can facilitate proton conduction. The activation energy for proton conductivity reveals the minimum energy required for the proton conduction across the membrane. The activation energy of pristine SPEEK, SPEEK/PVDF (5 wt%) and SPEEK/PVDF-g-PSSA (5 wt%) membranes were obtained by linear fitting using the Arrhenius equation (results shown in Fig. 5b),²⁸ ln(σT) = ln σ₀ − E_a/RT, where σ₀ is the pre-exponential factor, E_a is the proton conducting activation energy and R is the ideal gas constant. The activation energy of the pristine SPEEK membrane is 11.53 kJ mol⁻¹, which is in good agreement with previous reported values.²⁹ The activation energies of SPEEK/PVDF-g-PSSA (5 wt%) and PEEK/PVDF (5 wt%) membranes are 10.37 and 15.95 kJ mol⁻¹, respectively. The SPEEK/PVDF-g-PSSA (5 wt%) membrane exhibits the lowest activation energy, which verifies that the hierarchical microstructure of the PVDF-g-PSSA particles induces the generation of a co-continuous proton conducting network for fast proton conduction with a low energy barrier via Grotthuss mechanism.

Fig. 5 (a) Proton conductivity of the membranes as function of temperature, and (b) Arrhenius plots for the proton conductivity of pristine SPEEK, SPEEK/PVDF (5 wt%) and SPEEK/PVDF-g-PSSA (5 wt%) membrane.

4. Conclusions

In summary, we designed and fabricated a new series of SPEEK/PVDF-g-PSSA blend membranes for DMFC applications. The grafted PSSA side chains on the PVDF backbone improve the compatibility between the hydrophilic SPEEK and hydrophobic PVDF, and contribute to the formation of the hierarchical microstructure of the PVDF-g-PSSA particles, which can induce the generation a co-continuous proton conducting network for fast proton conduction and increase the tortuosity of the methanol pathway to retard the methanol crossover. All the SPEEK/PVDF-g-PSSA blend membranes exhibit superior dimensional stability, methanol resistance and selectivity than commercial Nafion® 117 at room temperature. In particular, the SPEEK/PVDF-g-PSSA (5 wt%) membrane displays the best overall membrane performance, which processes the highest selectivity and superior single cell performance compared with the SPEEK/PVDF membrane and Nafion® 117. In addition, the SPEEK/PVDF-g-PSSA blend membranes exhibit excellent thermal stability and small temperature dependence of water uptake and dimensional change benefiting from the improved compatibility. Summarily, the high overall performance SPEEK/PVDF-g-PSSA membranes were produced at low cost and with superior properties, offering great potential as PEMs for DMFC applications.

Acknowledgements

We thank the North China Electric Power University for the support in the methanol permeability measurements. This study was supported financially by the National Natural Science Foundation of China (Grant no. 51303211).

Notes and references

1. S. J. Peighambardoust, S. Rowshanzamir and M. Amjadi, Int. J. Hydrogen Energy, 2010, 35, 9349–9384.
2. Y. L. Liu, Polym. Chem., 2012, 3, 1373–1383.
3. K. A. Mauritz and R. B. Moore, Chem. Rev., 2004, 104, 4535–4586.
4. H. Zhang and P. K. Shen, Chem. Rev., 2012, 112, 2780–2832.
5. C. H. Park, C. H. Lee, M. D. Guiver and Y. M. Lee, Prog. Polym. Sci., 2011, 36, 1443–1498.
6. W. Dai, Y. Shen, Z. Li, L. Yu, J. Xi and X. Qiu, J. Mater. Chem. A, 2014, 2, 12423–12432.
7. M. X. Reinholdt and S. Kaliaguine, Langmuir, 2010, 26, 11184–11195.
8. Z. Jiang, X. Zhao and A. Manthiram, Int. J. Hydrogen Energy, 2013, 38, 5875–5884.
9. C. Manea and M. Mulder, J. Membr. Sci., 2002, 206, 443–453.
10. X. Li, S. de Feyter and I. F. J. Vankelecom, J. Membr. Sci., 2008, 324, 67–75.
11. Z. Li, J. Xi, H. Zhou, L. Liu, Z. Wu, X. Qiu and L. Chen, J. Power Sources, 2013, 237, 132–140.
12. J. C. Tsai, H. P. Cheng, J. F. Kuo, Y. H. Huang and C. Y. Chen, J. Power Sources, 2009, 189, 958–965.
13. E. M. Tsang, Z. Zhang, Z. Shi, T. Soboleva and S. Holdcroft, J. Am. Chem. Soc., 2007, 129, 15106–15107.
14. J. Wootthikanokkhan and N. Seeponkai, J. Appl. Polym. Sci., 2006, 102, 5941–5947.
15. S. Ren, G. Sun, C. Li, Z. Wu, W. Jin, W. Chen, Q. Xin and X. Yang, Mater. Lett., 2006, 60, 44–47.
16. S. Xue and G. Yin, Polymer, 2006, 47, 5044–5049.
17. T. Y. Inan, H. Doğan, E. E. Unveren and E. Eker, Int. J. Hydrogen Energy, 2010, 35, 12038–12053.
18. H. Y. Jung and J. K. Park, Int. J. Hydrogen Energy, 2009, 34, 3915–3921.
19. Y. W. Kim, D. K. Lee, K. J. Lee and J. H. Kim, Eur. Polym. J., 2008, 44, 932–939.
20. J. Hou, J. Li, D. Mountz, M. Hull and L. A. Madsen, J. Membr. Sci., 2013, 448, 292–299.
21. A. Mokrini, M. A. Huneault and P. Gerard, J. Membr. Sci., 2006, 283, 74–83.
22. H. Y. Li and Y. L. Liu, J. Mater. Chem. A, 2014, 2, 3783–3793.
23. T. J. Peckham and S. Holdcroft, Adv. Mater., 2010, 22, 4667–4690.
24. G. Gebel, Macromolecules, 2013, 46, 6057–6066.
25. N. W. Li and M. D. Guiver, Macromolecules, 2014, 47, 2175–2198.
26. R. Kumar, M. Mamlouk and K. Scott, RSC Adv., 2014, 4, 617–623.
27. V. Di Noto, M. Piga, G. A. Giffin and G. Pace, J. Membr. Sci., 2012, 390–391, 58–67.
28. W. Zhengbang, H. Tang and P. Mu, J. Membr. Sci., 2011, 369, 250–257.
29. H. Zhang, T. Zhang, J. Wang, F. Pei, Y. He and J. Liu, Fuel Cells, 2013, 13, 1155–1165.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra11894h

---