1 Introduction

Fuel cell (FC) technology has become increasingly important in recent years as fuel cells provide the highest energy density with low pollution, leading us to a power source revolution [1, 2]. One of the focuses in the development of FC technology is the proton exchange membrane (PEM). In the field of proton exchange membrane fuel cells (PEMFCs), PEM acts as a separator for the reactants, as well as a catalyst support, and at the same time provides ionic pathways for proton transport [3,4,5]. Up to now, it is generally accepted that poly(perfluoroalkyl sulfonic acid) (Nafion®) is the current benchmark material for the PEM due to its outstanding overall properties [6,7,8]. However, it has some disadvantages, including high cost, high methanol permeability, and poor performance at temperatures above 80 °C, which calls for the development of new membrane materials [9, 10].

In this respect, various sulfonated poly(aryl ether)s (SPAEs) have been extensively explored as PEM candidates with low cost and high performance [11,12,13,14]. Nevertheless, most of these SPAEs cannot meet the requirements in durability and proton conductivity. Because of the less pronounced hydrophilic/hydrophobic phase separation in SPAEs compared with that in Nafion®, the proton transport channels in these SPAE membranes are narrower and more distorted. As a result, the proton transport efficiency is lower based on the same ion exchange capacity (IEC) [15]. Sulfonated poly(fluorene ether ketone) (SPFEK), which has been reported as a PEM in FCs, possesses acceptable proton conductivity and stability, but still not comparable with Nafion® [16]. In order to achieve considerable proton conductivity, a high sulfonation degree is required, but unfortunately, it sacrifices durability and dimensional stability. Until now, lots of methods have been introduced to improve the overall performance of these materials, such as hybridization [17, 18], acid-base blending [19, 20], surface coating [21, 22], and cross-linking [23, 24]. One of the most adopted approaches is hybridization, developing composite membranes composed of sulfonated polymeric ionomers and hygroscopic inorganic fillers. SiO2 [25], ZrO2 [26, 27], clays [28], α-zirconium phosphate (α-ZrP) [29, 30], and graphene oxide [31] have been incorporated into neat membranes by either directly dispersing the inorganic fillers in the PEM solution and followed by solution casting or impregnating the inorganic fillers into the porous structure of membranes. Most nanofillers typically improve the mechanical properties of the membranes owing to the nano-reinforcing effect but lower the proton conductivity because such nanofillers are not proton conductive [29, 31,32,33]. Although some proton conductive fillers, such as α-ZrP, have been examined, the composite membrane performance is still far from satisfactory, primarily because of the much lower proton conductivity of the inorganic fillers compared with Nafion® and other PEMs.

In this work, we synthesized sulfonic acid functionalized ZrP (ZrP-SO3H) single-layer nanosheets as robust solid acid fillers for PEMs. The incorporation of such solid acid filler not only improves thermal stability and barrier properties but also offers additional protogenic groups for acid-water clusters, which is beneficial for proton transport [34, 35]. The morphology, water uptake, oxidative property, thermal stability, proton conductivity, and single-cell performance of the composite membranes were thoroughly investigated.

2 Experimental

2.1 Materials

Zirconyl chloride octahydrate (ZrOCl2∙8H2O, 98%, Aldrich), tetra-n-butylammonium hydroxide (TBA, 1.0 M in methanol, Aldrich), (3-mercaptopropyl)trimethoxysilane (MPTMS, 95%, Aldrich), and phosphoric acid (85%, Aldrich) were used as received without further purification. 4,4′-Difluorobenzophenone (DFBP, Aldrich) and 9,9′-bis(4-Hydroxypheyl) fluorene (BHF, Aldrich) were recrystallized from toluene and ethanol, respectively. Reagent-grade sulfuric acid (50%), ethanol, hydrochloric acid, potassium hydroxide, sodium chloride, and hydrogen peroxide (30 wt.%) were obtained from commercial sources and used as received. Dimethylacetamide (DMAc) was dried over 4A molecule sieves, and toluene was dried over sodium wire prior to use. Anhydrous potassium carbonate was dried in a vacuum oven at 150 °C for 10 h.

2.2 Preparation of ZrP-SO3H Nanosheets

ZrP micro-crystals were synthesized via the one-step hydrothermal method as reported [36,37,38,39,40,41,42]. Briefly, 6.0 g of ZrOCl2∙8H2O was mixed with 60.0 mL 6.0 M H3PO4 in a sealed Teflon-lined pressure vessel. The mixture was reacted at 200 °C for 24 h, and the lateral dimensions of the synthesized ZrP micro-crystals are 600 ~ 800 nm [38, 39].

ZrP micro-crystals were exfoliated into individual single-layer nanosheets using TBA as the exfoliation agent [43,44,45,46,47,48]. A sample of 0.50 g of ZrP was dispersed in 20.0 mL of deionized water with the assistance of ultrasonication (1 h). In an ice bath, 16.7 mL of 0.1 M TBA was added to the mixture, followed by 1 h of magnetic stirring and then another 1 h of ultrasonication. After that, a transparent liquid was obtained as exfoliated ZrP. In order to recover the acidic –OH groups on the ZrP nanosheet surface, a pre-determined amount of HCl solution (50% in excess) was introduced. An exfoliated ZrP gel was then collected by centrifugation and was sequentially rinsed three times with water, acetone, and toluene. A toluene dispersion containing ZrP nanosheets was obtained by treating the final product in toluene via ultrasonication.

As depicted in Fig. 1, the synthesized ZrP nanosheets were functionalized via a grafting reaction [49, 50]. In a 500-mL flask equipped with a condenser, 300 mL of ZrP dispersion (in toluene, containing 0.9 g of ZrP) was mixed with 14.73 g (75 mmol) of MPTMS. The dispersion was refluxed and stirred for 24 h in a N2 atmosphere. The solid product was obtained by centrifuge, rinsed three times with toluene, filtered, and vacuum dried. The MPTMS functionalized ZrP nanosheets are denoted as ZrP-SH. The thiol groups of ZrP-SH were then further oxidized by H2O2 to form sulfuric acid groups according to the following procedures: 0.42 g ZrP-SH and 12.0-mL methanol were added to a flask, and 5.5 mL of H2O2 was introduced dropwise under a N2 atmosphere. The reaction was kept at room temperature for 24 h. The mixture was filtered, and the obtained solid was rinsed with water and ethanol. After being vacuum dried and ground, ZrP-SO3H nanosheets were obtained. The procedures to synthesize ZrP-SO3H are illustrated in Fig. 1a.

Fig. 1
figure 1

a Procedures to functionalize ZrP to prepare ZrP-SO3H nanosheets. b Procedures to prepare SPFEK/ZrP-SO3H nanocomposite membrane

Full size image

2.3 Preparation of SPFEK/ZrP-SO3H composite membranes

Disodium salt of 3,3′-disulfonated-4,4′-difluorobenzophenone (SDFBP) was synthesized according to the procedures described by Wang et al [51]. SPFEK polymer matrix was prepared following the procedures described earlier [16]. The SPFEK/ZrP-SO3H-x (x refers to the weight percentage of ZrP-SO3H, x = 2, 4) composite membranes were prepared by casting from the SPFEK/ZrP-SO3H DMAc uniform dispersion, which was achieved by ultrasonication. The dispersion was spread on a glass plate, left in a dust-free oven at 75 °C for 24 h, and then at 100 °C under vacuum for another 12 h. The synthesized sodium-form membranes were protonated after soaked in 0.50 M H2SO4 aqueous solution at 80 °C for 24 h. Excessive acid was removed by washing the membrane in deionized water at 80 °C for 24 h. The acidified membranes were dried in a vacuum oven at 90 °C for 12 h. The thickness of all membranes was controlled in the range of 80–100 μm.

2.4 Membrane characterization

Fourier-transform infrared (FT-IR) spectra were recorded on a PerkinElmer Spectrum 100 Fourier transform spectrophotometer to confirm the chemical structure of the components in the composite membranes. The membrane micro-morphology was examined using a scanning electron microscopy (SEM, JSM-5600LV system of JEOL) equipped with an energy dispersive X-ray detector (EDX). The accelerating voltage was 15 kV. The cross-sections were obtained via cryo-fracturing in liquid nitrogen. The thermal stability of the composite membranes was analyzed using a PerkinElmer Pyris Diamond thermogravimetric analyzer (TGA) in N2 atmosphere. The temperature was increased from 50 to 150 °C and held for 20 min to eliminate the absorbed water and cooled to 50 °C and then elevated to 500 °C, all at a heating rate of 10 °C min−1.

2.5 Water uptake

The dry membranes were weighed and immersed in deionized water at 80 °C for 24 h. The water uptake is calculated according to the following equation:

$$ W\ \left(\%\right)=\frac{\left({W}_w-{W}_d\right)}{W_d}\times 100\% $$
(1)

where Wd and Ww are the weight of the membranes before and after water absorption, respectively.

2.6 Ion exchange capacity (IEC)

IEC was measured by titration [52]. The membranes in acid form were immersed in a NaCl solution (2.0 M) for 10 h to replace the protons of sulfonic acid groups with sodium ions. The solutions were titrated using a NaOH solution (0.025 M), with phenolphthalein as an indicator. The moles of the proton denote the moles of the sulfonic groups, and the IEC is calculated from the titration data using the following equation:

$$ IEC=\frac{\Delta {V}_{NaOH}{C}_{NaOH}}{W_s}\kern0.5em \left( mol\ {g}^{-1}\right) $$
(2)

where Ws is the dry membrane weight, ∆VNaOH is the consumed volume of the NaOH solution, and CNaOH is the concentration of the NaOH solution.

2.7 Oxidative stability

A piece of the membrane (1 × 4 cm2) was immersed in Fenton’s reagent (3% H2O2 + 2 ppm FeSO4) in a shaking bath (ZHICHENG ZHWY-110X water bath shaker) at 80 °C and kept for 1 h. The weight and IEC of the membranes before and after being treated in Fenton’s reagent were tested to evaluate the oxidative stability.

2.8 Proton conductivity

The membranes were kept in deionized water overnight prior to each test. The proton conductivity (σ) was measured on a Solartron 1255 B frequency response analyzer coupled with a Solartron 1287 electrochemical interface. The impedance spectrum of a cell with the given membrane sample sandwiched between two gold electrodes was tested. After curve-fitting from the spectrum, the conductivity was calculated using the following equation:

$$ \sigma =\frac{d}{RS} $$
(3)

where S is the area of the electrode, d is the thickness of the membrane, and R is the impedance.

2.9 Methanol permeability

Methanol permeability was tested using a diffusion cell consisting of two chambers (A and B) separated by a membrane. Chamber A is filled with a solution of 5.0 M methanol and 0.40 M H2SO4 in deionized H2O. Chamber B is filled with distilled water. The methanol permeation through the membrane is due to the difference in concentration between chambers A and B. The methanol concentration in chamber B was measured every 15 min for 2 h by gas chromatography (GC). The methanol permeability can be calculated according to the following equation.

$$ {C}_{\mathrm{B}}(t)=\frac{A}{V_{\mathrm{B}}}\frac{P}{L}{C}_{\mathrm{A}}\left(t-{t}_0\right) $$
(4)

where P is the methanol permeability, CB(t) is the concentration of methanol in permeate (chamber B), CA is the concentration of methanol in the feed (chamber A), A and L are the membrane area and thickness, and VB is the volume of permeate in chamber B. The methanol permeability (P) in Eq. (4) is directly proportional to the slope of the linear expression for the concentration of methanol in chamber B as the Y ordinate versus time (t − t0) as the X-axis.

$$ P=\frac{d{C}_{\mathrm{B}}(t)}{dt}\frac{V_{\mathrm{B}}}{A}\frac{L}{C_{\mathrm{A}}} $$
(5)

where \( \frac{d{C}_{\mathrm{B}}(t)}{dt} \) is the slope of the linear function of CB versus t in Eq. (4).

2.10 Membrane electrode assembly (MEA) fabrication and fuel cell test

The catalyst ink was prepared from Pt/C catalyst (40 wt.% Pt, Johnson Mattey Corp.), 5 wt.% Nafion® solution, isopropyl alcohol, and de-ionized water. Water was added dropwise to moisturize the catalyst, and then 5 wt.% Nafion® with isopropyl alcohol was added dropwise to form the ink. The mixture was dispersed by an ultrasonic stirrer bath for 20 min to form the catalyst slurry. The electrodes were prepared by directly spraying the catalyst slurry on 2 cm × 2 cm carbon papers (EC-TP1–060, Electrochem. Inc.) and subsequently drying at 60 °C under vacuum. The Pt loading was 0.4 mg/cm2, and the mass ratio of dry Nafion® to Pt/C is 1:4. A membrane was sandwiched between two pieces of electrodes and then hot pressed.

The performance of the membrane-electrode assembly (MEA) was tested using a 4 cm2 single cell in humidified H2/O2 (for PEMFC), and 1.0 M methanol/humidified O2 (for direct methanol fuel cell (DMFC)), respectively. For PEMFC, the single cell was operated at 85 °C under 100% relative humidity (RH) with 0 MPa gas back-pressure. The gas feeding rate of oxygen was fixed at twice of hydrogen to provide adequate oxidant for the fuel. For DMFC, the single cell was operated at 80 °C at ambient pressure. The polarization curves were measured by applying a constant current for 1 min at each point using a fuel cell test station (Arbin Instruments, 160,269). The power densities were calculated from the steady-state voltages and applied currents.

3 Results and discussion

3.1 Compositional and morphological analysis

The representative FT-IR spectra of ZrP, ZrP-SH, ZrP-SO3H, SPFEK, and SPFEK/ZrP-SO3H composite membrane are shown in Fig. 2. In the spectra of ZrP and ZrP-SH, the samples exhibit a broad band peaked at ca. 1057 cm−1 due to the overlap of P-O and Zr-O peaks. There are two absorption bands at ~3570 and 1636 cm−1, assigned to the stretching vibration and bending vibrations of –OH groups, respectively. The additional peaks at 472, 1120, 2558, and 2932 cm−1 in the spectrum of ZrP-SH are ascribed to the Si-O-Si bending vibration, Si-O-Si symmetric stretching vibration, S-H stretching vibration of thiol groups, and C-H stretching vibration of methylene groups, respectively, suggesting the successful graft reaction of MTPMS with ZrP. In the spectra of ZrP-SO3H, the absence of a peak at 2558 cm−1 indicates that all thiol groups attached to the ZrP-SH have been oxidized to sulfonic acid groups, indicating the successful synthesis of ZrP-SO3H. Furthermore, the bands ranging from 1000 to 1126 cm−1 are complicated due to the overlap of S-O symmetric and asymmetric stretching vibration of sulfonic acid groups (1084 and 1029 cm−1) with Si-O-Si symmetric stretching vibration, as well as P-O stretching vibration.

Fig. 2
figure 2

FT-IR spectra of ZrP, ZrP-SH, ZrP-SO3H, SPFEK, and SPFEK/ZrP-SO3H composite membrane

Full size image

In order to investigate the distribution of the functionalized ZrP nanosheets in the composite membranes, SEM-EDX was applied to the cross-section of the composite membranes, as depicted in Fig. 3. The ZrP-SO3H fillers were well dispersed throughout the SPFEK-ZrP-2% composite membrane (Fig. 3a). When the filler loading was increased to 4%, aggregation of ZrP-SO3H fillers within some regions were observed (Fig. 3b). Similarly, as shown in Fig. 3c and d, based on the cross-sectional images with linear analysis of Zr atom content along the transverse direction of the composite membranes by EDX, SPFEK/ZrP-SO3H nanosheets were virtually uniformly dispersed in SPFEK/ZrP-SO3H-2%, but a lower level of dispersion was achieved in SPFEK/ZrP-SO3H-4%. Nevertheless, no aggregation was observed at the bottom of the membranes.

Fig. 3
figure 3

SEM images of the cross-section of (a) SPFEK/ZrP-SO3H-2% and (b) SPFEK/ZrP-SO3H-4% composite membranes; SEM images with zirconium concentration profile of (c) SPFEK/ZrP-SO3H-2% and (d) SPFEK/ZrP-SO3H-4% composite membranes

Full size image

3.2 Water uptake and proton conductivity

The water within the membrane plays a vital role in the transportation of protons (vehicle mechanism or the Grotthuss mechanism) [53,54,55], which affects the overall FC performance significantly. The water uptake of the membranes is dependent on several parameters, including content and distribution of sulfonic acid groups, temperature, and relative humidity. Generally, the water uptake increases with the increase of membranes’ IEC and temperature. Although a larger water uptake facilitates proton transportation to a greater extent, excessive water uptake could result in an apparent decrease in mechanical strength of the membrane and a dimensional mismatch between the membrane and the electrodes. The IEC value of ZrP-SO3H was tested to be 2.81 meq/g, which is higher than the value of SPFEK (1.75 meq/g). As shown in Table 1, the IEC values of the SPFEK/ZrP-SO3H composite membranes increased with an increasing doping degree. On the other hand, the water uptake of SPFEK barely changed after the incorporation of SPFEK/ZrP-SO3H nanosheets. As is known, the water absorption of polymers usually depends on the concentration of sulfonic acid groups. In our investigation, the water uptake of the composite membranes showed an almost constant value. These results suggest that the addition of ZrP-SO3H nanosheets is effective in elevating IEC values of the resultant composite membranes without increasing water absorption. The introduction of ZrP-SO3H nanosheets leads to a high concentration of sulfonic acid groups within the membrane, which should enhance water uptake. However, the ZrP-SO3H nanosheets locate at the interspace of the macromolecular chains and occupy some space of water molecules. The combined effects balance the water uptake to a virtually constant value. A similar phenomenon was reported in the literature [56].

Table 1 Water uptake and proton conductivity of pure SPFEK, SPFEK/ZrP-SO3H-2% and SPFEK/ZrP-SO3H-4% composite membranes, and Nafion® 117
Full size table

Proton conductivity was measured under the fully hydrated condition at 80 °C. Table 1 clearly reveals that the incorporation of ZrP-SO3H nanosheets into the SPFEK matrix can significantly improve its proton conductivity. These results can be ascribed to the increased concentration of sulfonic acid groups, which enhances the proton transfer.

3.3 Oxidative stability

Chemical stability of proton exchange membranes is one of the most important factors that affect their durability. It is well known that the formation of HO· and HOO· radicals from the catalytic process during FC operation causes membrane degradation [57]. Fenton’s reagent is often used for the determination of a membrane’s stability, for an accelerated testing. The highly reactive HO· and HOO· radicals generated by Fenton’s reagent can attack the polymer main chains and result in considerable degradation of composite membranes. The oxidative stability of the SPFEK/ZrP-SO3H composite membranes was examined by the weight loss and IEC loss of the membranes after being treated in Fenton’s reagent for 1 h. As shown in Table 2, both SPFEK/ZrP-SO3H composite membranes show significantly enhanced oxidative stability than the neat SPFEK membrane. The weight losses of both composite membranes are much lower than that of the neat SPFEK membrane. In particular, SPFEK/ZrP-SO3H-2% membrane demonstrates that a small amount of ZrP-SO3H is effective in enhancing oxidative stability. It should be noted that, after the treatment by Fenton’s reagent, the IEC value of the composite membrane increased with the ZrP-SO3H content. This may be caused by the acidic hydroxyl groups in ZrP-SO3H. Before the test of oxidative stability, due to the interactions between the –OH and –SO3H groups of ZrP-SO3H, the proton-donating ability of –OH is low and protons are more difficult to exchange via Na+ during the titration. After the treatment by Fenton’s reagent, the membrane became less dense, and the protons on the –OH groups in ZrP-SO3H can be easily exchanged by Na+ cations. Thus, the IEC of SPFEK/ ZrP-SO3H composite membranes became higher. The phenomenon that IEC rose after an oxidative stability test was also reported by other researchers [56]. Meanwhile, the IEC loss and weight loss of Nafion® 117 are much lower than that of SPFEK, mainly because Nafion® 117 is a fluoropolymer, where fluorine atoms protect the polymer chains from being attacked by water molecules containing highly oxidizing radical species [57].

Table 2 Oxidative stability of pure SPFEK, SPFEK/ZrP-SO3H-2% and SPFEK/ZrP-SO3H-4% composite membranes, and Nafion® 117
Full size table

3.4 Thermal properties

Thermal properties of SPFEK/ZrP-SO3H composite membranes in their acid form were characterized by TGA, and the results are shown in Fig. 4. The degradation of pure SPFEK has two major steps. The first weight loss is assigned to the elimination of sulfonic acid groups, which occurred during a temperature range of 220–380 °C. The second weight loss at about 450–610 °C is due to the decomposition of the polymer main chains. The composite membranes exhibit a three-step degradation. The additional weight loss before 200 °C (step I) is corresponding to the removal of hydration water of ZrP-SO3H [34, 58, 59]. Overall, the TGA thermograms of SPFEK/ZrP-SO3H composite membranes are similar to that of the neat SPFEK membrane. Note that the decomposition temperature of the polymer main chains (step III in Fig. 4) rises to 530 °C for SPFEK/ZrP-SO3H-2% and 535 °C for SPFEK/ZrP-SO3H-4%, which are much higher than that of pure SPFEK (450 °C). The increase in decomposition temperature implies that the existence of thermally stable ZrP-SO3H nanosheets can help improve the thermal stability of the resultant composite membranes without changing the decomposition mechanism. In addition, SPFEK/ZrP-SO3H-2% with a lower ZrP-SO3H weight content shows less weight loss of hydration water in the first step but more weight loss overall, compared with SPFEK/ZrP-SO3H-4%. The hydration water in the membrane samples was mainly retained by ZrP-SO3H; thus a lower weight content of ZrP-SO3H corresponded to less hydration water loss. Meanwhile, ZrP-SO3H nanosheets are more thermally stable than SPFEK, and their presence helped further improve the thermal stability of SPFEK. As a result, SPFEK/ZrP-SO3H-4% exhibited less weight loss overall.

Fig. 4
figure 4

TGA thermograms of pure SPFEK, SPFEK/ZrP-SO3H-2%, and SPFEK/ZrP-SO3H-4% composite membranes

Full size image

3.5 Single PEMFC performance

The polarization and power density curves of the H2/O2 fuel cells assembled with the composite membranes SPFEK/ZrP-SO3H-2% and SPFEK/ZrP-SO3H-4% are shown in Fig. 5a. Both SPFEK and Nafion® 117 were used as controls. All single PEMFCs were performed at a cell temperature of 85 °C, 100% relative humidity, with a back-gas pressure of 0 MPa. The open-circuit voltages (OCV) of the fuel cells composed of SPFEK/ZrP-SO3H-2%, SPFEK/ZrP-SO3H-4%, SPFEK, and Nafion® 117 were 0.99, 0,98, 0.98, and 0.99 V, respectively. This indicates that all membranes exhibit very close gas permeability. The highest current densities of the fuel cells using SPFEK/ZrP-SO3H-2%, SPFEK/ZrP-SO3H-4%, SPFEK, and Nafion® 117 were 1.29, 0.93, 1.09, and 1.15 A/cm2, respectively, with a cut-off voltage of 0.1 V; the highest power densities of the four corresponding fuel cells were 0.268, 0.198, 0.255, and 0.271 W/cm2, respectively. It is obvious that the fuel cell using composite membrane SPFEK/ZrP-SO3H-2% exhibited better performance than the fuel cell using neat SPFEK. However, similar to many other composites [34, 35, 60], a higher filler loading does not guarantee better results. The fuel cell using SPFEK/ZrP-SO3H-4% exhibited reduced performance. This is probably because excessive ZrP-SO3H nanosheets resulted in a non-uniform distribution, which was observed from the SEM results. This could explain why SPFEK/ZrP-SO3H-4% had a high proton conductivity but gave inferior fuel cell performance.

Fig. 5
figure 5

(a) Polarization and power density curves of the PEMFCs using SPFEK, Nafion® 117, SPFEK/ZrP-SO3H-2%, and SPFEK/ZrP-SO3H-4% as PEMs operating at 85 °C; (b) polarization and power density curves of a PEMFC using SPFEK/ZrP-SO3H-2% at different temperatures

Full size image

It was very encouraging that membrane SPFEK/ZrP-SO3H-2% exhibited a comparable cell performance with Nafion® 117 in H2/O2 fuel cells. The polarization and power density curves of a H2/O2 fuel cell using SPFEK/ZrP-SO3H-2% at temperatures of 75, 80, 85, 90 and 95 °C at 100% RH and ambient pressure were recorded and are shown in Fig. 5b. In the temperature range of 75 ~ 85 °C, the fuel cell performance significantly increased with the operating temperature. When the temperature reached 90 °C, the cell showed the highest current density of 1.29 A/cm2 and a power density of 0.26 W/cm2, which was the same as a fuel cell operated at 85 °C. However, when the temperature was further elevated to 95 °C, cell performance dropped. This is probably because at such a high temperature, the water retentivity of the membrane decreased, and hence the proton conductivity decreased. In brief, the fuel cell using composite membrane SPFEK/ZrP-SO3H-2% displayed an optimal working temperature of 85 ~ 90 °C.

SPFEK/ZrP-SO3H-2% composite membrane was further investigated in a DMFC. DMFCs using SPFEK/ZrP-SO3H-2%, SPFEK, and Nafion® 117 as PEMs were tested at 80 °C with 1.0 M methanol and humidified oxygen gas at ambient pressure. As expected, the methanol permeability of the composite membranes decreased with an increasing ZrP-SO3H content in the composite membrane (Table 3), indicating the composite membranes have better methanol resistances. The methanol permeability of SPFEK/ZrP-SO3H-2% is 2.25 × 10−7 cm2 s−1, which is only ca. 15.4% of the value of Nafion@ 117 (1.46 × 10−6 cm2 s−1). These results are consistent with another report of a composite membrane using SiO2 as fillers [61]. As shown in Fig. 6, there is a marked improvement in DMFC performance for the cell composed of SPFEK/ZrP-SO3H-2%. The DMFC using SPFEK/ZrP-SO3H-2% shows a peak power density of 61 mW/cm2, which is much higher than 38 and 41 mW/cm2 for the DMFCs using SPFEK and Nafion® 117, respectively. Because the methanol permeability and membrane conductivity serve as two key parameters in DMFC performances, the composite membrane SPFEK/ZrP-SO3H-2% possessing low methanol permeability and high membrane conductivity exhibited the best DMFC performance. Overall, the composite membrane with conductive ZrP-SO3H nanosheets displayed significant improvement in DMFC performance. These results demonstrate that the ZrP-SO3H-doped SPFEK membrane is a promising candidate for polymer electrolyte membranes for high-performance DMFCs.

Table 3 Methanol permeability (cm2 s−1) of Nafion® 117, pure SPFEK, SPFEK/ZrP-SO3H-2% and SPFEK/ZrP-SO3H-4% composite membranes at 80 °C
Full size table
Fig. 6
figure 6

Polarization and power density curves of DMFCs using SPFEK/ZrP-SO3H-2%, SPFEK, and Nafion® 117 as PEMs, using 1.0 M methanol at 80 °C

Full size image

4 Conclusion

A functionalized proton conductive filler ZrP-SO3H was synthesized and used to prepare SPFEK/ZrP-SO3H composite membranes with 2 wt.% and 4 wt.% filler loadings. When the loading was 2 wt.%, a uniform dispersion of ZrP-SO3H nanosheets was achieved in the composite membrane SPFEK/ZrP-SO3H-2%. The composite membranes showed improved proton conductivity, oxidative stability, and methanol barrier property. SPFEK/ZrP-SO3H-2% exhibited better fuel cell performance than neat SPFEK in the H2/O2 fuel cell test. Moreover, in the DMFC test, SPFEK/ZrP-SO3H-2% showed a peak power density of 61 mW/cm2, which is much higher than 38 and 41 mW/cm2 for the DMFCs using SPFEK and Nafion® 117, respectively. These results show that the incorporation of ZrP-SO3H to SPFEK is effective in improving fuel cell performances in all respects. The SPFEK/ZrP-SO3H-2% membrane is promising for high-performance H2/O2 fuel cells and DMFCs as a polymer electrolyte membrane.