Electrorheological effect induced quaternized poly(2,6-dimethyl phenylene oxide)-layered double hydroxide composite membranes for anion exchange membrane fuel cells

Abstract

To improve the performance of anion exchange membranes (AEMs), we fabricated quaternized poly(2,6-dimethyl phenylene oxide)-layered double hydroxide composite membranes by combining the advantages of the two components. Also, the electrorheological effect was employed during the casting process to induce the formation of ion conducting channels along the through-plane direction. The membrane with 3% layered double hydroxide (LDH) (QPPO-Im-3% LDH) showed the largest increase in ionic conductivity over that of the pure membrane (QPPO-Im) (15.85 mS cm⁻¹ at 30 °C to 22.52 mS cm⁻¹ at 80 °C vs. 5.93 mS cm⁻¹ at 30 °C to 11.63 mS cm⁻¹ at 80 °C). The ionic conductivity was further improved by applying the electric field treatment, confirming that the addition of LDH and the electric field greatly affect the ionic conductivity of the AEMs. Moreover, the morphology, thermal and mechanical properties, ion-exchange capacity (IEC), water uptake (WU) and swelling ratio (SR) were also studied systematically to determine the effects of LDH and electric field on the membranes.

---

Introduction

Energy shortage and environmental pollution are two big problems in the world, and various power generation applications have been investigated to solve these problems.^1–3 Fuel cells (FCs) are one of the promising technologies to generate electricity from renewable chemical energy with high conversion efficiency, low pollution and noise. Of the various FCs, proton exchange membrane fuel cells (PEMFCs) are the most widely investigated, with Nafion, a typical perfluorosulfonic acid membrane, as a proton conductor.^4–6 However, PEMFCs always suffer from low stability under acidic conditions and high costs of noble-metal-catalysts and membranes. Recently, anion exchange membrane fuel cells (AEMFCs) were developed rapidly as an alternative to PEMFCs to overcome the shortcomings.^7,8

As a key component of AEMFCs, anion exchange membranes (AEMs) have been widely studied; nevertheless, the ion conductivity and chemical stability still need to be improved before AEMs are commercialized. Considerable efforts have been made to meet the AEM commercialization requirements. In addition to promoting phase separation and the formation of network structure, inorganic–organic composite membranes can be used to improve the stability and ion conductivity of AEMs. Shi⁹ and his co-workers recently developed quaternized halloysite nanotubes (QHNTs)/chitosan (CS) hybrid membranes, they found that the QHNTs were well-dispersed within the CS matrix, and the chain mobility of CS was promoted driven by repulsive interactions from QHNTs, in turn affording the increments of water uptake and area swelling. Together with the promoted chain mobility, low-barrier conduction pathways were formed along the QHNTs surface and then significantly enhance the hydroxide conductivity of the hybrid membranes. Liu¹⁰ and his co-workers prepared a series of novel composite anion exchange membranes by incorporating quaternized graphenes (QGs) into chloromethylated polysulfone (CMPSU). These membranes showed good morphologies without phase separation, acceptable thermal properties, alkaline resistances and oxide stabilities, low water uptakes and swelling ratios and enhanced anion conductivity. Li¹¹ and his co-workers prepared a series of quaternized poly(arylene ether sulfone)/nanozirconia composite membranes for AEMFC, and the membrane with 7.5% ZrO₂ showed the highest ion conductivity (41.4 mS cm⁻¹ at 80 °C). Liao¹² and his co-workers fabricated quaternized polysulfone/functionalized montmorillonite nanocomposite membranes, these membranes with functionalized montmorillonite exhibited lower water uptake, higher ultimate stress and larger ionic conductivity than the polysulfone membrane.

The LDH we employed here was a layered functional material with the formula [M_1−x²⁺M_x³⁺(OH)₂](Aⁿ⁻)_x/n·mH₂O, in which the M²⁺ and M³⁺ cations disperse in an ordered and uniform manner in brucite-like layers and anions exist in layers to maintain charge conservation which also can conducted in the interlayer space of LDH^13–15 (Fig. 1). Moreover, tons of hydroxyls on LDHs nanoplates surface would be helpful to trap more water which will be beneficial for hydroxide transportation. In addition, as a basic material, LDH is stable in alkaline solutions. In Kohei¹⁶ and Zhang¹⁷ groups, they also found that with LDHs addition into AEMs would result in the enhancement of the conductivity and the stability to some extent. However, under normal circumstances, LDHs are isotropically dispersed in AEMs, resulting in the isotropy of ionic conductivity which will increase the transmission distance and lower the efficiency of anion conduction. Because the through-plane conductivity of membranes plays a decisive role in the operation of fuel cells, it will be a feasible way to further improve anion conductivity via constructing ion conduct channels by using LDHs to the through-plane direction. It has been reported^18–20 that, nanoparticles affected by the electrorheological effect would arrange to the electric field direction in multiphase system in electric field with sufficiently high strength. Herein, in this study, poly(2,6-dimethyl phenylene oxide) (PPO) was chosen as the backbone of AEMs due to its simple structure and easy modification.^21,22 After bromination and quaternization of PPO, Mg–Al LDH was added to the casting solution to form a multiphase dispersion system. Then, electric field was applied during film casting to induce LDH alignment to the through-plane and then increase the through-plane conductivities. The effects of LDH content and electric field strength on anion conductivity were also systematically studied.

Fig. 1 Simulated diagram of Mg–Al LDH.

Experimental

Materials

Poly(2,6-dimethyl phenylene oxide) (PPO, 98%, Beijing Chemical Works, M_n = 53 000 g mol⁻¹, M_w × M_n⁻¹ = 1.2), methanol (99.5%, Beijing Chemical Works), methylene chloride, benzoyl peroxide (BPO), n-bromosuccinimide (NBS) (99%, Aladdin), sodium hydroxide (96%, Beijing Chemical Works), sodium carbonate (99.8%, Beijing Chemical Works), magnesium nitrate (Mg(NO₃)₂·6H₂O, 99%, Beijing Chemical Works), aluminum nitrate (Al(NO₃)₃·9H₂O, 99%, Beijing Chemical Works), were used as received. N-Methyl imidazole (NMI, 99%, Aladdin) and tetrachloromethane (99.5%, Beijing Chemical Works) were distilled over CaH₂ before use.

Synthesis of Mg–Al LDH

Mg–Al LDH was prepared by a co-precipitation method.^23,24 A mixture of 12.80 g of Mg (NO₃)₂·6H₂O and 9.38 g of Al (NO₃)₃·9H₂O (Mg : Al = 2 : 1) was dissolved in 35 mL deionized water to form solution A. A mixture of 7.00 g NaOH and 5.00 g Na₂CO₃ was dissolved in 50 mL of deionized water to form solution B. Solution A was then added dropwise to solution B under vigorous stirring at room temperature. Then, the resulting milky white suspension was aged at 65 °C for 18 h. The resulting solid product was filtered, washed with deionized water and anhydrous ethanol several times, and finally air-dried at 60 °C overnight. Calculated from the molar ratio of Mg and Al and the general molecular formula of LDHs ([M_II1−xM_IIIx(OH)₂](Aⁿ⁻)_x/n), the chemical formula of the synthesized Mg–Al LDH was found to be [Mg_0.68Al_0.32(OH)₂](CO₃²⁻)_0.16. According to the formula of synthesized Mg–Al LDH, the calculated IEC was 4.22 mmol g⁻¹.

Synthesis of brominated PPO (BPPO) and quaternized PPO (QPPO)

4 g PPO and 80 mL distilled tetrachloromethane were added to a 200 mL three-necked round-bottom flask equipped with a magnetic stirrer and matched with a thermometer, a condenser, and a nitrogen inlet–outlet. The mixture was degassed and stirred at 50 °C until the PPO was completely dissolved. Then, 4.64 g NBS and 0.232 g BPO were added to the PPO solution and reaction was initiated at 80 °C. After refluxed for 4 hours, the mixture was cooled to room temperature and precipitated slowly in methanol under vigorous stirring, and then the precipitate was dissolved in methylene chloride. After most of the methylene chloride was steamed out, the product was washed with methanol for another two times and dried at 60 °C for 24 h. A light yellow product (5.3 g, yield 70%) was obtained with a bromomethylation ratio of 30%, as determined by the peak area ratio of H3 and H1 in Fig. 3.

0.3 g of dried BPPO obtained above was dissolved in 15 mL DMF to form a light yellow settled solution. Then, 0.2 g of NMI was added dropwise to the BPPO solution under stirring and maintained for 12 h at room temperature to get a brown QPPO solution. The degree of quaternization (DQ) was 100% according to the disappearance of H3.

Membrane fabrication

The QPPO solution casted on a round glass plate and dried in an oven at 60 °C for 24 h was designated as QPPO-Im. Different fractions of Mg–Al LDH (1 wt%, 3 wt%, 5 wt%, and 7 wt%, in theoretically dried QPPO) were added to the QPPO solution and dispersed by ultrasonic and stirring for each 0.5 h and were denoted as QPPO-Im-f LDH (f is the fraction of LDHs). On the base of QPPO-Im-f LDH, different electric fields were applied during the membrane casting, and the membranes casted under different electric fields were denoted as QPPO-Im-f LDH-g V (g is the electric field strength). Then, all the membranes were immersed in 1 M NaOH solution for 2 days at room temperature to exchange Br⁻ to OH⁻ and the synthesis paths was demonstrated in Scheme 1. The structure of composite membranes before and after applying electric field are shown in Fig. 2, the through-plane direction is also the electric field direction.

Scheme 1 Synthesis pathway of composite membranes.

Fig. 2 Schematic structural of composite membranes before and after applying electric field.

Characterization and measurement

¹H NMR spectra were obtained at 400.13 MHz with a Bruker Avance III 400 MHz NMR spectrometer using CDCl₃ or DMSO-d₆ as solution. The morphologies and arrangement of LDHs in different membranes were performed on a transmission electron microscope (JEOL JEM-3010HR, Japan). The membranes were cut in small pieces, then immersed in epoxy resin until totally solidification. Diamond knife together with the ultramicrotome was used to cut the fixed membranes into thin sections. The front and section morphologies of the membranes were observed under a scanning electron microscope (Zeiss Supra 55). In addition, the membranes were immersed in liquid nitrogen and bent to break to stick to the sample stage. The XPS measurements were performed by using an ESCALAB 250 instrument (Thermo Electron) with Al Kα radiation. The thermal stability of the membranes was determined by thermogravimetric analysis (TGA) apparatus (Mettler Toledo TGA/DSC 1/1100 SF) from 25 to 800 °C at a heating rate of 10 °C min⁻¹ under a nitrogen flow.

Water uptake (WU)

The water uptake of the membranes was calculated by the following equations after the membranes were saturated with deionized water:

WU = (m_wet − m_dry)/m_dry × 100% (1)

where m_wet and m_dry are the weights of the wet and dry membranes, respectively.

Swelling ratio (SR)

The swelling ratio of the membranes was characterized by the expansion of dry membranes after soaked in deionized water for 24 h (1 cm in length and 1 cm in width). The SR was calculated by the following equation:

SR = (l_wet − l_dry)/l_dry × 100% (2)

where l_wet and l_dry are the perimeter of the wet and the dry membranes, respectively.

Ion-exchange capacity (IEC)

Being alkaline, LDH will react with hydrochloric acid, and the IECs of the membranes were determined by Mohr titration. About 0.2 g of membrane was ion exchanged in 0.01 M NaCl for 24 h and then washed with DI water completely. The membrane in chloride form was immersed in 50 mL of 0.01 M NaNO₃ for 24 h. All the NaNO₃ solution was collected and titrated with 0.01 M AgNO₃, using K₂CrO₄ as a colorimetric indicator. After titration the membrane was dried at 60 °C in vacuum for 24 h and weighed. The IEC was calculated from the membrane dry mass and the amount of AgNO₃ consumed in the titration as follows:

IEC (mmol) = (V₁ − V₂)C_NaCl/m_dry (3)

where V₁ and V₂ are the original and the end volume of the AgNO₃ solution, respectively, C_NaCl is the concentration of NaCl, and m_dry is the weight of the dry membranes.

Hydroxide conductivity

A two-probe AC impedance spectroscopy (Zahner IM6ex (Germany) electrochemical workstation, Germany) was applied to measure the through-plane ionic conductivities of the membranes in the frequency from 1 Hz to 100 kHz. All the membranes were washed with deionized water before test and then put in a chamber with DI water from 30 °C to 80 °C. The ionic conductivity was calculated as follows:

σ = l/AR_m (4)

where l is the distance between two electrodes, A (cm²) is the contact area of the membranes and electrodes, R_m is resistance of the membrane.

Alkaline stability

To evaluate the stability of the membranes in alkaline solution, the membranes were immersed into a 1 M KOH solution at room temperature, and the change of the ionic conductivity of the membranes was observed every 48 h.

Before testing, the membranes were washed with deionized water until the pH of the deionized water was constant. Then, the conductivities of the membranes at 60 °C were recorded.

Single cell tests

The single cell was fabricated by using 40 wt% Pt/C (Hesen, Shanghai) with a metal loading of 0.5 mg cm² as both anode and cathode catalysts. Quaternized PPO was used as the ionomer with a 20% loading at both electrodes. The catalyst ink was obtained by mixture of water, methanol, catalyst and ionomer, where the catalyst: ionomer ratio was 3 : 1. Afterwards, the catalyst ink was coated on a carbon paper (Hesen, Shanghai) to obtain anode/cathode catalyst layers with an effective electrode area of 2 cm × 2.5 cm. The membrane electrode assemblies (MEAs) were prepared by hot-pressing the membranes between the anode and cathode electrodes at 20 MPa and 105 °C for 6 min. Fuel cell tests were carried out with H₂ and O₂ at 50 °C with 100% relative humidity (RH). The flow rates of H₂ and O₂ were 200 mL min⁻¹ with a 0.2 MPa backpressure during test.

Results and discussion

Synthesis and characterization of Mg–Al-LDH

The XRD and TEM were measured and emerged in Fig. 3. Fig. 3(a) shows the XRD patterns of the Mg–Al LDH, with the most distinct characteristic peaks at 2θ = 11° and 24°, and other characteristic peaks are also the same as those reported in the literatures.^25,26 The sharp peaks demonstrate that the Mg–Al LDH has good crystal shape, which together with the morphology of Mg–Al LDH (Fig. 3(b)) demonstrates the successful synthesis of Mg–Al-LDH. And the small particle size (68 nm) of the Mg–Al-LDH (Fig. 3(c)) ensures its good distribution in the casting solution.

Fig. 3 (a) XRD of Mg–Al LDH, (b) TEM of Mg–Al LDH and (c) particle size distribution of Mg–Al LDH.

Synthesis and characterization of BPPO

Fig. 4 shows the ¹H NMR spectra of the PPO, BPPO, and QPPO membranes. For PPO, the peak at 2.1 ppm is associated with the methyl on the benzene (H1), and the peak at 6.5 ppm is assignable to the proton of the benzene ring (H2). A new peak at around 4.3 ppm, which is ascribed to methylene connected with benzene and bromine (H3), appears after bromination. Meanwhile, the resonance change at 6.3–6.6 ppm is also a powerful evidence to bromination. All the membranes used were from the same batch of QPPO, and the only difference is the content of LDH, so the ¹H NMR spectra of the QPPO membranes are quite similar. After quaternization, the characteristic proton bands at 9.6–9.8 ppm, 7.5–7.8 ppm, and 3.6–3.8 ppm are assigned to NCH₂N (H7), NCHCHN (H5, H6) and N–CH₃ (H4), respectively, indicating the successful incorporation of NMI groups into the AEMs. This conclusion is further supported by the H3′ signal change from 4.3 ppm to 5.25 ppm assigned to Ar–CH₂–Br and Ar–CH₂–N (H3′).

Fig. 4 ¹H NMR spectra of PPO, BPPO and QPPO.

Hydroxide conductivity

Hydroxide conductivity is one of the key properties of AEMs, and the basic requirement for hydroxide conductivity is a value above 10 mS cm⁻¹.²⁷ Fig. 5 shows the through-plane conductivities of all the synthesized membranes. The through-plane conductivities of the composite membranes were all higher than the QPPO-Im's, especially for the QPPO-Im-3% LDH membrane (5.93 mS cm⁻¹ at 30 °C and 11.63 mS cm⁻¹ at 80 °C for QPPO-Im, 15.85 mS cm⁻¹ at 30 °C and 22.52 mS cm⁻¹ at 80 °C for QPPO-Im-3% LDH). However, the through-plane conductivities do not increase with increasing content of LDH, because the LDH nanoplates were mixed and disorderly dispersed in the membranes without an external force and the number of nanoplates perpendicular to the ionic conduction direction increased with the increase of LDH content and blocked the ionic conduction paths and reduced the through-plane conduction efficiency, which is an evidence to prove the importance of arrangement of LDH in the membrane.

Fig. 5 Ion conductivities of composite membranes with different fractions of LDH.

In order to avoid the negative effects of the mixed and disorderly arrangement of LDH nanoplates, different strengths of electric field were applied to control the arrangement direction. The QPPO-Im-3% LDH membrane, which has the highest through-plane conductivity, was chosen as the reference. Fig. 6 shows that the ionic conductivity can be improved only when the electric field strength is above 4000 V cm⁻¹. However, the through-plane conductivities decrease with the electric field decrease when the electric field strength below 4000 V cm⁻¹. This is because when the electric field strength below 4000 V cm⁻¹, although the electric field force is not enough to align the LDH to the through-plane direction, however the lower electric field will arrange LDHs more deviates from the direction of the electric field and then decline the ions transport. In order to verify the electrorheological effect was initiated at electric field strength above 4000 V cm⁻¹, theoretical calculation was carried out as follows:

where E_p is the lowest electric field strength that can initiated electrorheological effect, ε_d is the dielectric constant of the dispersion medium, ε_m is dielectric constant of dispersion particles, k is Boltzmann constant, T is temperature; ε₀ is vacuum permittivity, and r is particle size (Fig. 3(c)). With ε_DMF = 38 F m⁻¹ and ε_{Mg–Al LDH} = 3 F m⁻¹, the result of E_p is 3870 V cm⁻¹, consistent with our experiment results. Because of equipment limitation, 5000 V cm⁻¹ was the highest electric strength obtained, at which the highest ionic conductivity was observed (17.23 mS cm⁻¹ at 30 °C to 25.45 mS cm⁻¹ at 80 °C) under this circumstance.

Fig. 6 Ion conductivities of composite membranes at different electric field strengths.

Morphology

The QPPO-Im, QPPO-Im-3% LDH, QPPO-Im-3% LDH-5000 V cm⁻¹ membranes were chosen to further study the effect of LDH and electrorheological effect on the membranes. SEM and TEM were applied to study the arrangement of LDH nanoplates in the membranes affected by the electrorheological effect. As shown in Fig. 7(a), the QPPO-Im membrane shows a smooth surface, because it could hardly to observe the phase separate of PPO-Im membranes even after bromination and quaternization by SEM.^28,29 With the addition of 3% of LDH, the surface of the membrane becomes rough (Fig. 7(b)) because the addition of LDH disorganizes the homogeneous system. However, the electrorheological effect limits the disorganization (Fig. 7(c)). EDX-mapping was used to observe the elements distribution. Nitrogen was distributed uniform in the QPPO-Im membrane (Fig. 7(d)), even after the addition of LDH and electric field, the distribution of nitrogen was also uniform (Fig. 7(e) and (f)). To the QPPO-Im-3% LDH membrane, magnesium (Fig. 7(g)) and aluminum (Fig. 7(i)) were homogeneous distributed due to the super scattered system of Mg–Al-LDH and DMF. However, after the application of electric field, magnesium (Fig. 7(h)) and aluminum (Fig. 7(j)) were significantly distributed to specific direction, this phenomenon was induced by electrorheological effect caused by the enough electric field strength.

Fig. 7 SEM of (a) QPPO-Im (b) QPPO-Im-3% LDH (c) QPPO-Im-3% LDH-5000 V cm⁻¹; EDX-mapping of N (d) of QPPO-Im; EDX-mapping of N (e), Mg (g), Al (i) of QPPO-Im-3% LDH; EDX-mapping of N (f), Mg (h), Al (j) of QPPO-Im-3% LDH-5000 V cm⁻¹.

The QPPO-Im, QPPO-Im-3% LDH and QPPO-Im-3% LDH-5000 V cm⁻¹ membranes were observed under a transmission scanning microscope to determine the LDH arrangement in the membranes. The scratches in the micrographs are coming from the diamond knife. The internal structure of the QPPO-Im membrane (Fig. 8(a)) is quite smooth. To the QPPO-Im-3% LDH membrane, the LDH nanoplates were just treated by ultrasonic before casting to avoid aggregation, but there is still aggregation in Fig. 8(b). However, for membrane with electric field treatment, the casting solution were treated with ultrasonic before the casting process as well as the electric field in the casting process. Under the AC electric field, the LDH nanoplates tend to swing which will make the LDH nanoplates hard to aggregate. Especially with the strength higher than 4000 V cm⁻¹ the driven force is high enough to align LDH nanoplates to the electric field direction which the aligned and connected LDH nanoplates will improve the ion conduction (Fig. 8(c)). But if with the strength lower than 4000 V cm⁻¹, the LDH nanoplates tend to overturn and hard to connect with each other edge to edge which will block the ion transport. However, even with the imperfect aligned LDH in the membrane, the conductivity of QPPO-Im-3% LDH-1000 V is also higher than that of QPPO-Im which due to the higher IEC and water uptake after the addition of LDH. The simulated diagram were shown in Fig. 2, before applying the electric field, the positive and negative charges of the electrical double layer uniformly distributed around the LDH, but, after applying the electric field, the positive and negative charges are driven to the direction of electric field by the electrorheological effect and the attractive force induced the LDH arrange to the direction of electric field.

Fig. 8 TEM images of (a) QPPO-Im (b) QPPO-Im-3% LDH, and (c) QPPO-Im-3% LDH-5000 V cm⁻¹. The red arrows indicate the through-plane direction of the membranes.

Thermal and mechanical properties

The normal working temperature of AEMFCs is 80 °C, so the thermal stability is an important parameter to test the availability of AEM. The LDH decomposition temperature shown in Fig. 10 is the same as that reported in the literature.^30–32 Fig. 9 shows that the thermal decomposition of all the membranes can be divided into two steps. The 6–10% weight loss before 100 °C is attributed to the evaporation of water and residual solvent. The main decomposition temperatures are at 280 °C and 450 °C, demonstrating that the AEMs are stable below 300 °C and thus can be used in AEMFC.

Fig. 9 Thermograms of LDH, QPPO-Im, QPPO-Im-3% LDH, and QPPO-Im-3% LDH-5000 V cm⁻¹.

The membranes used in fuel cells must have sufficient mechanical strength to satisfy operation requirements. The mechanical properties of the membranes are shown in Table 1. With the addition of LDH, the tensile strength and Young's modulus increased from 25.32 MPa and 700.01 MPa to 29.55 MPa and 878.77 MPa after added LDHs, but the elongation at break decreased from 9.59% to 7.64%. The application of an electric field hardly changes the mechanical properties, because only the arrangement of the LDH is changed.

Table 1 Mechanical properties of various membranes at room temperature

Membrane	Tensile strength (MPa)	Young's modulus (MPa)	Elongation at break (%)
QPPO-Im	25.32 ± 2.5	700.01 ± 24.5	9.59 ± 1.4
QPPO-Im-3% LDH	29.55 ± 1.3	878.77 ± 20.8	7.64 ± 2.7
QPPO-Im-3% LDH-5000 V cm^−1	29.70 ± 2.3	890.72 ± 23.7	7.54 ± 2.5

---

IEC, water uptake (WU) and swelling ratio (SR)

The ion-exchange capacity, water uptake and swelling ratio of the QPPO-Im, QPPO-Im-3% LDH, and QPPO-Im-3% LDH-5000 V cm⁻¹ membranes are listed in Table 2. The IEC of the QPPO-Im-3% LDH membrane is increased compared to the QPPO-Im membrane (1.94 mmol g⁻¹ vs. 1.83 mmol g⁻¹) because of the addition of LDH and high theoretical IEC (4.22 mmol g⁻¹) of LDH, and there is no significant increase in IEC after the application of electric field (1.94 mmol g⁻¹ without electric field vs. 1.93 mmol g⁻¹ with electric field). Similarly, the WU increases after the addition of LDH because of the water imbibition property of LDHs (90.92% at 30 °C and 100.32% at 60 °C versus 112.35% at 30 °C and 118.35% at 60 °C), and the application of electric field has no significant effect on the WU. As we know, the relationship of IEC and conductivity is not a simple linear relationship, little change of IEC may induce great changes of conductivity.³³ And the water uptake also has a big influence on the conductivity.³⁴ So with a big increased WU (20% higher than that of QPPO-Im membrane) and introduction of LDH which increased the IEC of the composites membrane lead to about 2.5 times increase of conductivity of QPPO-Im-3% LDH, compared with QPPO-Im membrane in Fig. 5. Under normal circumstances, the SR increases with increasing WU, it doesn't work in our research. On the other hand, the LDH nanoplates worked as a barrier of SR in the membranes, so the SR of AEMS was hampered with the addition of LDH and electric field (34.03%, 32.56% and 32.47%, at 30 °C respectively), and this phenomenon is more obvious at higher temperature. IEC on behalf of the number of ion exchangeable groups in membranes and the available ion exchangeable groups determine the anion transfer ability, and a high IEC usually means a high ion conductivity. Compared with those of QPPO-Im-3% LDH, the IEC and WU of QPPO-Im-3% LDH-5000 V cm⁻¹ remain almost the same, but the ionic conductivity is higher (Fig. 6). This phenomenon reconfirms that the electrorheological effect rearranges the LDH nanoplates to form anion conduction channels along the through-plane direction. We also compared the IEC values and ionic conductivities of our composite membranes with those of other composite membranes reported in the literature. Our composite membranes, especially the QPPO-Im-3% LDH-5000 V membrane, have conductivities as high as those of other composite membranes and thus are promising in AEMFC applications (Table 3).

Table 2 IEC, WU and SR of membranes

Membrane	IEC (mmol g^−1)	WU (%)		SR (%)
		30 °C	60 °C	30 °C	60 °C
QPPO-Im	1.83 ± 0.2	90.92 ± 5	100.32 ± 4	34.03 ± 3	42.34 ± 3
QPPO-Im-3% LDH	1.94 ± 0.2	112.35 ± 3	118.35 ± 3	32.56 ± 2	38.67 ± 3
QPPO-Im-3% LDH-5000 V cm^−1	1.93 ± 0.3	110.46 ± 3	118.46 ± 3	32.47 ± 2	37.28 ± 3

---

Table 3 IEC values and ionic conductivities of various membranes in this work and the literature

Membrane	IEC (mmol g^−1)	Temperature (°C)	σ (mS cm^−1)	Reference
QPVA/5 wt% GA	0.41	30	1.2	35
QPSF/5% MMT-1	—	25	10	12
PVA/DGBE-15/SiO2-1	1.01	25	2.9	36
QPPO/SiO2-2.8%	2.25	30	12	37
PVA/TiO2-2%	—	30	5.6	38
QPVA/Al2O3-10%	—	30	35	39
QPPO	0.70	30	4.3	40
PPO–MIm	1.94	20	12.9	41
QPPO-Im	1.83	30	5.93	This work
QPPO-Im-3% LDH	1.94	30	15.85	This work
QPPO-Im-3% LDH-5000 V cm^−1	1.93	30	17.23	This work

---

Alkaline stability

As we know, the QPPO-Im membrane is not stable in high concentration aqueous KOH solution and at high temperature. We chose the QPPO-Im membrane because the structure of it is simple and be much better for studying the electrorheological effect to improve the through-plane conductivities of QPPO AEMs. For this paper, it is proved even though the composite membranes degraded, the through-plane conductivities are still higher than that of QPPO-Im membrane. So the QPPO-Im, QPPO-Im-3% LDH and QPPO-Im-3% LDH-5000 V cm⁻¹ membranes were chosen and immersed in 1 M NaOH solution at room temperature. Fig. 10 shows that the conductivities of QPPO-Im membrane at 60 °C decrease by 35% compared to its original conductivity. For QPPO-Im-3% LDH membrane, the conductivities of this membrane at 60 °C for different time decrease more drastically than that of QPPO-Im membrane, this is due to the loss of LDHs along with the degradation of QPPO-Im. But the conductivities of QPPO-Im-3% LDH are still higher than that of QPPO-Im. Taking QPPO-Im-3% LDH and QPPO-Im-3% LDH-5000 V cm⁻¹ membranes for contrast, the conductivities of these two membranes tend to be close after degradation, which may also be due to the degradation of QPPO-Im, resulting in the collapse of orderly arranged LDH. The point of this paper is applying the electrorheological effect to improve the though-plane conductivities of QPPO AEMs. The alkaline stability of composite membranes could be drastically improved by using stable polymer matrix.

Fig. 10 The alkaline stability of QPPO-Im, QPPO-Im-3% LDH and QPPO-Im-3% LDH-5000 V cm⁻¹.

Performance of single cell

Considering the preparation of MEA in our group is imperfect, the dispersion of catalyst ink and the spraying of catalyst ink exist considerable error. Therefore, in order to avoid unnecessary disputes, we chose the Im-3% LDH-5000 V cm⁻¹ membrane, which with the highest conductivity to measure the single cell test. Fig. 11 shows the polarization and power density curves for Im-3% LDH-5000 V cm⁻¹ under fully humidified inlet gas conditions at 50 °C. The open circuit voltage (OCV) of 0.94 V and a maximum power density of 63.9 mW cm⁻² can be achieved at a current density of 170 mA cm⁻². It's worth noting that the single cell test was carried out without any optimization and the performance of single cell has the potential to increase by optimizing the electrode structure, water management and double plate.

Fig. 11 The fuel cell performance of QPPO-Im-3% LDH-5000 V cm⁻¹ membrane.

Conclusions

We have synthesized poly(2,6-dimethyl phenylene oxide)-layered double hydroxide composite membranes with electrorheological effect for anion exchange membrane fuel cells. It is concluded that adding LDHs to AEMs can dramatically improve the ion conductivity, and the highest ion conductivity was achieved by adding 3% of LDH. The ion conductivity of composite membranes can be further improved by applying an electric field with a strength higher than 4000 V cm⁻¹. The results show that a sufficiently high electric field strength can induce the rearrangement of LDHs to the conducting direction of OH⁻, facilitating the conduction of anions.

Acknowledgements

We gratefully appreciate the financial support from the National Natural Science Foundation of China (No. 21176022, 21176023 and 21376022), the International S&T Cooperation Program of China (No. 2013DFA51860), the Fundamental Research Funds for the Central Universities (No. JC1504).

Notes and references

1. K. Han, J. M. Shen, S. Q. Hao, H. Q. Ye, C. Wolverton and M. C. Kung, ChemSusChem, 2014, 9, 2545–2553.
2. M. Liu, F. Dai, Z. Ma, M. Ruthkosky and Y. Li, J. Power Sources, 2014, 268, 37–44.
3. B. Li, M. Gu, Z. Nie, X. Wei, C. Wang, V. Sprenkle and W. Wang, Nano Lett., 2014, 14, 158–165.
4. Z. Chai, C. Wang, H. Zhang, M. D. Cara, P. L. Bradley, J. H. Anita and H. Wang, Adv. Funct. Mater., 2010, 20, 4394–4399.
5. H. Wang, X. Zhuang, J. Tong, X. Li, W. Wang, B. Cheng and Z. Cai, J. Appl. Polym. Sci., 2015, 132, 42843.
6. Y. Wei, S. Matar, L. Shen, X. Zhang, Z. Guo, H. Zhu and H. Liu, Int. J. Hydrogen Energy, 2012, 37, 1857–1864.
7. Z. Yang, J. Zhou, S. Wang, J. Hou, L. Wu and T. Xu, J. Mater. Chem. A, 2015, 3, 15015–15019.
8. W.-H. Lee, Y. S. Kim and C. Bae, ACS Macro Lett., 2015, 4, 814–818.
9. B. Shi, Y. Li, H. Zhang, W. Wu, R. Ding, J. Dang and J. Wang, J. Membr. Sci., 2016, 498, 242–253.
10. L. Liu, C. Tong, Y. He, Y. Zhao and C. Lü, J. Membr. Sci., 2015, 487, 99–108.
11. X. Li, Y. Yu and Y. Meng, ACS Appl. Mater. Interfaces, 2013, 5, 1414–1422.
12. X. Liao, L. Ren, D. Chen, X. Liu and H. Zhang, J. Power Sources, 2015, 286, 258–263.
13. C. Li, J. Zhou, W. Gao, J. Zhao, J. Liu, Y. Zhao, M. Wei, D. G. Evans and X. Duan, J. Mater. Chem. A, 2013, 1, 5370–5376.
14. D. Sachdev, M. A. Naik, A. Dubey and B. G. Mishra, Catal. Commun., 2010, 11, 684–688.
15. C. Li, Y. Chen, S. Zhang, J. Zhou, F. Wang, S. He, M. Wei, D. G. Evans and X. Duan, ChemCatChem, 2014, 6, 824–831.
16. K. Miyazaki, T. Abe, K. Nishio, H. Nakanishi and Z. Ogumi, J. Power Sources, 2010, 195, 6500–6503.
17. P. Zhang, T. Yamaguchi, B. N. Nair, K. Miyajima and G. M. Anilkumar, RSC Adv., 2014, 4, 41051–41058.
18. Y. Ko, H. J. Lee, Y. J. Chun, U. S. Choi and K. P. Yoo, ACS Appl. Mater. Interfaces, 2013, 5, 1122–1130.
19. X. Guo, Y. Chen, M. Su, D. Li, G. Li, C. Li, Y. Tian, C. Hao and Q. Lei, ACS Appl. Mater. Interfaces, 2015, 48, 26624–26632.
20. C. F. d. G. Ferreira and P. C. Camargo, J. Phys. Chem. B, 2015, 119, 12561–12567.
21. H. S. Dang and P. Jannasch, Macromolecules, 2015, 48, 5742–5751.
22. H. S. Dang, E. A. Weiber and P. Jannasch, J. Mater. Chem. A, 2015, 3, 5280–5284.
23. L. Yang, Z. Jiang, G. Fan and F. Li, Catal. Sci. Technol., 2014, 4, 1123–1131.
24. Z. Tian, X. Xiang, L. Xie and F. Li, Ind. Eng. Chem. Res., 2013, 52, 288–296.
25. J. Y. Uan, J. K. Lin and Y. S. Tung, J. Mater. Chem., 2010, 20, 761–766.
26. T. Ishizaki, S. Chiba, K. Watanabe and H. Suzuki, J. Mater. Chem. A, 2013, 1, 8968–8977.
27. Y. Lu, A. A. Armentrout, J. Li, H. L. Tekinalp, J. Nanda and S. Ozcan, J. Mater. Chem. A, 2015, 3, 13350–13356.
28. Y. Yang, N. Sun, P. P. Sun and L. Q. Zheng, RSC Adv., 2016, 6, 25311–25318.
29. T. Y. Feng, B. C. Lin, S. Zhang, N. Y. Yuan, F. Q. Chu, M. A. Hickner, C. Y. Wang, L. Zhu and J. N. Ding, J. Membr. Sci., 2016, 508, 7–14.
30. Z. Q. Zhang, M. C. Liao, H. Y. Zeng, S. Xu, L. H. Xu, X. J. Liu and J. Z. Du, Fuel Process. Technol., 2014, 128, 519–524.
31. P. Zhang, S. Sago, T. Yamaguchi and G. M. Anilkumar, J. Power Sources, 2013, 230, 225–229.
32. V. Radha, P. V. Kamath, N. Ravishankar and C. Shivakumara, Langmuir, 2007, 23, 7700–7706.
33. Y. Yang, N. Sun, P. P. Sun and L. Q. Zheng, RSC Adv., 2016, 6, 25311–25318.
34. T. Y. Feng, B. C. Lin, S. Zhang, N. Y. Yuan, F. Q. Chu, M. A. Hickner, C. Y. Wang, L. Zhu and J. N. Ding, J. Membr. Sci., 2016, 508, 7–14.
35. C. C. Yang, S. Chiu, S. Kuo and T. Liou, J. Power Sources, 2012, 199, 37–45.
36. G. Das, B. K. Deka, S. H. Lee, Y. B. Park and Y. S. Yoon, Macromol. Res., 2015, 23, 256–264.
37. Y. Wu, C. Wu, J. R. Varcoe, S. D. Poynton, T. Xu and Y. Fu, J. Power Sources, 2010, 195, 3069–3076.
38. C. C. Yang, J. Membr. Sci., 2007, 288, 51–60.
39. C. C. Yang, S. J. Chiu, W. C. Chien and S. S. Chiu, J. Power Sources, 2010, 195, 2212–2219.
40. K. HariGopi, S. G. Peera, S. D. Bhat, P. Sridhar and S. Pitchumani, Int. J. Hydrogen Energy, 2014, 39, 2659–2668.
41. Q. Li, L. Liu, S. E. Liang, Q. B. Dong, B. K. Jin and R. K. Bai, RSC Adv., 2013, 3, 13477–13485.

---