Non-planar backbone structure polybenzimidazole membranes with excellent solubility, high proton conductivity, and better anti-oxidative for HT-PEMFCs

Abstract

As an efficient, inexpensive, readily accessible monomer of the polybenzimidazole (PBI) copolymer, the 4,4′-[(4,4′-bipyridine)-2,6-diyl]dibenzoic acid (BPY) containing bipyridine unit was firstly used to synthesize a series of novel non-planar PBI copolymers (BPY–PBI-x) by regulating the ratio with 2,6-pyridinedicarboxylic acid under microwave-assisted conditions. The copolymers exhibit superior solubility in aprotic solvents such as N,N-dimethylacetamide, N,N-dimethylformamide and dimethylsulfoxide than general PBI, and the corresponding BPY–PBI-x serial membranes were prepared by a solution casting method. All the membranes have been characterized by thermal stability, a tension test, scanning electron microscopy, PA-doping ability and swelling ratio, proton conductivity, and a Fenton test. The membranes exhibit excellent comprehensive properties containing high proton conductivity and long lifetime. The conductivity of the doping levels of the BPY–PBI-100% membrane with the highest doping level (342.7%) reaches 78.6 mS cm⁻¹ at 160 °C. The high proton conductivity is attributed to the increase in N atoms of the proton acceptor from the bipyridine monomer and the enhancement of free volume from the existence of a bulky stereostructure BPY unit. The long lifetime is due to the fact that the pyridine unit can quench radicals by the formation of pyridine-N-oxide. The unique anti-oxidation mechanism of the BPY–PBI-x serial membranes, which firstly was pointed out for the application of pyridine heterocyclic rings in HT-PEMs, provides new insights for novel anti-oxidation membrane design. In short, 3-fold objectives, namely, (I) soluble PBI, (II) higher proton conductivity PBI, and (III) anti-oxidative PBI, were achieved in this paper.

---

Introduction

Polybenzimidazole (PBI) has received considerable attention for high-temperature (>100 °C) fuel cell applications.^1,2 In 1995, Wainright et al. first reported the application of PBI doped with phosphoric acid membranes to high temperature polymer electrolyte membrane fuel cells (HT-PEMFCs).³ Since then, PBI doped with phosphoric acid has been extensive researched. Phosphoric acid doped polybenzimidazole (PA-doped PBI) has been proved as an outstanding candidate for HT-PEMFCs, which exhibits relatively high proton conductivity,⁴ good mechanical properties,⁵ and excellent thermal stability⁶ at temperatures of up to 200 °C. The reaction kinetics, CO-tolerance would be improved and water management and cooling system for fuel cell would be simplified under the condition of high temperature.^7,8

Poly[2,2′-(m-phenylene)-5,5′-bibenzimidazole] (mPBI) is the most investigated one among various PA-doped PBI polymer membranes for practical application in PEMFCs. The traditional mPBI from condensation between 3,3′,4,4′-tetraaminobiphenyl and isophthalic acid has poor solubility in DMF solvent, resulting in a difficulty for forming membrane. To overcome this problem, several approaches such as the modification of main chain by ether, sulfone, or aliphatic units were proposed.^9–13 Although the main chain of mPBI containing sulfone units improves the solubility and flexibility, the membranes exhibit inferior proton conductivity.¹¹ Considering the acid-doping levels are closely linked with proton conductivity of the PEM, much more efforts have been done on increase of PA-doping level by increasing the percentage of alkaline group such as pyridine,^14–17 imidazole^18–22 and other alkaline structural unit.^23–25

Durability of membrane is another key parameter for the lifetime of the whole PEMFCs, which is related to the stability of polymer against the attack of oxidative radicals produced during the fuel cell operation. In terms of acid-doped PBI membranes, PBI has aromatic heterocyclic units which contain the reactive sites of electron-rich. Therefore, PBI is easier to be attacked by oxidants such as HO˙ and HO₂˙ radicals. There were some attempts to improve the membrane durability. The passive approaches contain various cross-linking strategies^26–28 and composite membranes with organic or inorganic compounds.^29,30 The active approach is to suppress the free radicals attack, such as avoiding H₂O₂ formation, destroying H₂O₂ ^31,32 or scavenging the free radicals.^33,34 However, the proton conductivity of the anti-oxidative membranes aboved still remains to be improved. Therefore, the PA-doped PBI has no satisfactory results to balance comprehensive properties for HT-PEMs, which motivated us to prepare PBIs of novel structures.

In this paper, we synthesized the bulky stereostructure of 4,4′-[(4,4′-bipyridine)-2,6-diyl]dibenzoic acid (BPY for short), which was a superior and accessible monomer for the PBI polycondensation. The sulfide-containing PBI copolymers, which were firstly used in the polybenzimidazole for HT-PEMFCs, exhibit satisfactory anti-oxidative capacity by the transformation from sulfide to sulfoxide bonds proved in our previous work.³⁵ In this work, the novel non-planar BPY–PBI-x serial copolymers (x denotes the molar feed percent of BPY in gross diacid monomers) containing bipyridine unit with bulky stereostructure have been prepared by microwave polycondensation technology. On account of the introduction of the bipyridine unit, the anti-oxidative stability of BPY–PBI-x serial membranes will increase because the pyridine unit can quench radicals by the formation of pyridine-N-oxide. Thus the oxidation reactions of imidazole units in PBI will be retarded, and the corresponding durability of PBI membranes will increase as well.

In a word, we planned to obtain a novel non-planar BPY–PBI-x copolymers by polymerizing a bulky stereostructure diacid monomer with the 3-fold objectives, which include (I) good solubility and processability, (II) higher proton conductivity, (III) long lifetime.

Experimental

Materials

3,3′,4,4′-Tetraaminobiphenyl and 2,6-pyridinedicarboxylic acid supplied from J&K Scientific Ltd., Beijing, China. para-Methylacetophenone was obtained from Aladdin Industrial Corporation, Shanghai, China. 4-Pyridinecarboxaldehyde was purchased from Energy Chemical, Shanghai, China. Synthesis of BPY monomer was prepared according to previous researcher work.^36,37 Hydrogen peroxide (H₂O₂) and polyphosphoric acid (PPA, >84% P₂O₅) were purchased from Tianjin Guangfu Technology Development Co., Ltd., Tianjin, China. Ammonium acetate, N,N-dimethylacetamide (DMAc), potassium permanganate (KMnO₄), sodium bicarbonate (NaHCO₃), N,N-dimethylformamide (DMF), pyridine, phosphoric acid (PA, 85%), sulfuric acid (H₂SO₄, 98%) and other general reagents were obtained from Beijing Chemical Reagent Co., Ltd., China. All reagents related above were directly used without further purification. The microwave reactor was purchased from Sineo Microwave Chemistry technology Co., Ltd., Shanghai, China.

Synthesis of copolymer containing bipyridine structure (BPY–PBI-x)

The polybenzimidazole containing bulky bipyridine structure (BPY–PBI-x) was prepared by a microwave-assisted solution polycondensation method according to our previous work.^35,38 A representative synthetic route of BPY–PBI-50%, where 50% represents the percentage of the molar feed percent of BPY in gross diacid monomers, has been handled as below. Firstly, 45 g polyphosphoric acid (PPA) was deaerated under microwave irradiation under nitrogen atmosphere for 30 min. After the process of deaeration was completed, 0.9719 g 3,3′,4,4′-tetraaminobiphenyl (DAB), 0.3790 g 2,6-pyridinedicarboxylic acid (Py-DA) and 0.8990 g BPY were blended together and put in PPA. The mixture would have a vigorously mechanical stirring in nitrogen atmosphere under microwave heating at 200 W under the following temperature conditions: 90 °C for 30 min, 110 °C for 20 min, 140 °C for 60 min, 170 °C for 20 min, 200 °C for 180 min. When the polycondensation reaction was finished, the viscous solution immediately was poured into deionized water, filtered and washed with deionized water several times. Then the polymer was collected and washed thoroughly using NaHCO₃ (10 wt% aqueous) for neutralizing the residual acid, and until the pH value was ca. 7.0. Afterwards, the polymer was also washed with ethyl alcohol to remove low molecular weight organics. In the end, the polymer was dried at 120 °C under vacuum oven for 12 h to afford dry BPY–PBI-50% copolymer powder. Table 1 shows the feed ratio of 3,3′,4,4′-tetraaminobiphenyl, 2,6-pyridinedicarboxylic acid and BPY in the copolymerization of BPY–PBI-x.

Table 1 Feed ratio of the monomers for the copolymerization of BPY–PBI-x

BPY–PBI-x	DAB/mmol	Py-DA/mmol	BPY/mmol
BPY–PBI-0%	10	10	0
BPY–PBI-10%	10	9	1
BPY–PBI-30%	10	7	3
BPY–PBI-50%	10	5	5
BPY–PBI-70%	10	3	7
BPY–PBI-100%	10	0	10

---

Preparation of polymer membranes and acid doping

The BPY–PBI-x serial membranes were prepared by the solution casting technique using DMAc as solvent. A typical procedure for the membrane fabrication was as follows: the polymer powder was dissolved in DMAc at 140 °C to make a (1.5 wt%) homogeneous solution, then filtered and poured into a glass culture dish and dried at 50 °C for 4 h, at 80 °C for 4 h, and eventually at 120 °C for 3 h to remove the solvent in an oven. Afterwards, the membranes obtained were cut into 2 cm × 2 cm, weighed and immersed in phosphoric acid (80 wt%) at room temperature for 3–4 days to obtain acid-doped PBI membrane. After that, the residual acid was wiped with filter paper, and dried in vacuum oven at 80 °C overnight until no weight loss.

Structural characterization

Fourier transform infrared (FT-IR) spectra of polymer samples were recorded using a Thermo Scientific Nicolet 6700 FT-IR spectrometer and the wavenumber range was 4000–400 cm⁻¹. The molecular weight distribution was determined by gel permeation chromatography (GPC), which was performed with an Agilent Technologies 1200 Series LC system. DMF with 2 wt% LiCl was used as the eluent at a flow rate of 1.0 mL min⁻¹ under nitrogen atmosphere. The molecular weight was calibrated with a polystyrene standard. The morphologies of the BPY–PBI-x serial membranes were observed under a scanning electron microscope (Zeiss Supra 55).

Tension test

The mechanical properties of BPY–PBI-x serial membranes were measured at room temperature with an Instron Model 1185 instrument at ambient conditions (25 °C and 40 ± 5% relative humidity) according to the ASTM D882-02 with a strain speed of 10 mm min⁻¹. The membranes samples required the thickness to retain around 80 μm were cut into dumbbell-shaped specimens with area of 50 mm × 4 mm according to ASTM D882-02. According to the test of mechanical properties, we have recorded the tensile strength and elongation at break (%).

Acid leaching

The PA-doped membranes were kept hanging over boiling water for a period of 5 h and the weight of the acid leached membranes noted every hour. Acid leaching percentages of the membranes were calculated according to the following eqn (1):

Acid loss percentage = (W₀ − W_i)/W_k × 100% (1)

where W₀ is the original weight of PA doped membrane, and the W_i the weight of the PA doped membrane measured at different time. W_k is the original weight of PA present in the membranes calculated from the PA doping level of the membranes.

Thermal stability

The thermal stability of the membranes was carried out using thermo-gravimetric analysis (TGA) on a Beijing Henven Thermo Gravimetric Analyzer (HCT-1) from 50 to 800 °C with a heating rate of 15 °C min⁻¹ under nitrogen atmosphere.

PA-doping ability and swelling ratio

The BPY–PBI-x serial membranes were immersed into 80 wt% H₃PO₄ solution at room temperature for 2 days to be fully saturated. The dimensions of membranes were 1 cm × 1 cm and measured before immersing. Then, the wet membranes were wiped dry and quickly washed with deionized water several times to remove free H₃PO₄ of membranes surface. After that, the wet membranes need be dried at 120 °C for 12 h to remove the bound water until no weight loss.

The acid doping weight ratio (r_a) was determined by eqn (2),

r_a = [(m₁ − m₀)/m₀] × 100% (2)

where m₀ refers to the pristine weight of dry membrane before acid treatment, and m₁ represent the weight of the membrane after soaking and drying.

The swelling ratio (r_s) were determined by measuring the volume change of the membrane before and after PA doping.

The calculation was calculated as follows in eqn (3),

r_s = [(V₁ − V₀)/V₀] × 100% (3)

where V₀ is the pristine volume of dry membrane, and V₁ is the volume of the membrane immersed in PA solutions and dried in an oven.

Proton conductivity

The proton conductivity of membranes was calculated through plane by a two-electrode AC impedance method over a frequency range of 1–10⁵ Hz from 80 to 160 °C on an electrochemical workstation (Zahner Im6ex, Germany). The test was carried out under anhydrous conditions. The proton conductivities (σ, S cm⁻¹) of the membranes were obtained using eqn (4),

σ = d/(RS_m) × 100% (4)

where d is the distance between the two electrodes (the thickness of the membrane), and S_m represents the area of the membrane that contacts the two electrodes. The R is ohmic resistance, which was measured by electrochemical workstation.

Fenton test

The chemical oxidation stability of membranes was measured by oxidative radicals. By immersing in 3% H₂O₂ and 4 ppm Fe²⁺ (added as FeSO₄) to test membrane degradation, the Fenton test were kept at 60 °C up to 192 h, and the Fenton's reagent was refreshed every 24 h. In addition, the membrane samples were collected, cleaned with deionized water several times after 24 h, and dried at 110 °C for 20 h in a vacuum oven. The degradation of the membranes was evaluated by the weight loss.

Results and discussion

Structure characterization

The synthetic routes of the BPY–PBI-x serial membranes are shown in Scheme 1, where ‘x’ signifies the molar feed percent of 4,4′-[(4,4′-bipyridine)-2,6-diyl]dibenzoic acid (BPY for short) in gross diacid monomers. The calculation was shown in eqn (5).

x = [m/(m + n)] × 100% (5)

Scheme 1 The synthetic procedure of polybenzimidazole copolymers containing bipyridine unit (BPY–PBI-x).

In the process of polycondensation, the BPY monomer was added into reaction vessel by scale. By adjusting the molar ratio of BPY, a series of BPY–PBI-x with different molar contents of BPY monomers can be gained, whose proportion have ranged from 0% to 100%. The structures of resultant polybenzimidazoles were confirmed by FT-IR (Fig. 1).

Fig. 1 FT-IR spectra of BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%).

Fig. 1 gives the FT-IR spectra of BPY–PBI-x serial copolymers. The characteristic absorption bands at around 3187 cm⁻¹ are attributed to the stretching vibrations of N–H. The peaks at 1680 cm⁻¹ have not residual carbonyl peak, which indicates that the cyclization to form benzimidazole rings is complete. The characteristic peaks of benzimidazole backbone are clearly visible at 1295 cm⁻¹ and 1450 cm⁻¹, suggesting the existence of the absorption of breathing vibration of imidazole ring and in-plane ring vibration of 2,6-bisubstituted benzimidazole. In addition, another specific peak at 803 cm⁻¹ can be identified as the existence of C–H out-of-plane bending vibration in heterocyclic ring. Undoubtedly, these characteristic peaks provide strong evidence of the polymerization.

To identify the successful introduction of BPY monomer as well as its influence on the polymer structure, the gel permeation chromatography (GPC) was carried out. The molecular weight data of GPC about BPY–PBI-x serial copolymers were shown in Table 2. According to Table 2, the molecular weight of the prepared BPY–PBI-x have little significant change, implying that the addition of BPY monomer has no impact on the polymerization degree of PBI, which assures the high molecular weight of the BPY–PBI-x polymers. For example, the weight-average molecular weight (M_w) variation range of BPY–PBI-x from 1.2385 × 10⁶ to 1.6563 × 10⁶ g mol⁻¹. So, all prepared BPY–PBI-x polymers meet the requirement of membrane formation.

Table 2 Molecular weight of the prepared BPY–PBI-x

	PBI–BPY-0%	PBI–BPY-10%	PBI–BPY-30%	PBI–BPY-50%	PBI–BPY-70%	PBI–BPY-100%
Mn (g mol^−1)	6.5807 × 10^5	1.1093 × 10^6	8.5578 × 10^5	6.2254 × 10^5	9.8094 × 10^5	1.0716 × 10^6
Mw (g mol^−1)	1.3872 × 10^6	1.5253 × 10^6	1.2612 × 10^6	1.2385 × 10^6	1.6563 × 10^6	1.3666 × 10^6

---

Morphology can be more intuitive response microstructure and usually defined as the form and texture of a material. The microstructures of BPY–PBI-0% membrane and BPY–PBI-100% membrane were investigated in detail through scanning electron microscopy (SEM) and the results are displayed in Fig. 2. The surface of BPY–PBI-0% membrane and BPY–PBI-100% membrane are smooth, compact and uniform at 100 00× magnification (see Fig. 2(a) and (b)).

Fig. 2 The SEM image of (a) BPY–PBI-0% membrane surface at 100 00× magnification, (b) BPY–PBI-100% membrane surface at 100 00× magnification.

Solubility of BPY–PBI-x

The solubility of BPY–PBI-x membranes was tested by solutions of each polymer (1.5 wt%) in DMAc, which was prepared at room temperature and kept for 1 h (shown in Fig. 3). Undoubtedly, the solubility of the polymer is a crucial parameter, yet the pristine PBI embody poor solubility and processing ability. Hence, many attempts, containing the modification of main chain or the substitution of different groups in imidazole ring,^12,39 have been made to improve the solubility of pristine PBI. The direct incorporation of hetero atoms (oxygen, nitrogen) in the polymer backbone is known to enhance the solubility.^16,40 What's more, the bulky unit can make the polymer skeleton form non-planar structure, which enhance the solubility due to the disruption of the rigid structure and decrease of intermolecular hydrogen bonds.^23,41 Moreover, the mPBI and ABPBIs were hard to dissolve in DMAc at room temperature. Thus, comparing with the mPBI and ABPBIs, better solubility of BPY–PBI-x serial membranes is due to the existence of bulky BPY monomer structure, which the bulky structure decreased intermolecular hydrogen bonds. Meanwhile, the existence of the N atom of bipyridine monomer enhances the solubility.

Fig. 3 The solubility of BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%) in DMAc.

Tension test

The mechanical properties of the copolymer membranes have been researched by adjusting the proportion of BPY monomer. As shown in Table 3, all of the membranes measured at room temperature and ambient atmospheric humidity show comparable mechanical stabilities, with a tensile strength that varies from 76 to 100 MPa and the elongation at break of 5% to 7%, which are much higher than other PEMs reported in literatures.^42,43 With the addition of the BPY monomer proportion, we can discover that the tensile strength of BPY–PBI-x membranes have a declining trend on the whole. It maybe relate to the reduction of intermolecular hydrogen bonds among main chains due to the addition of the bulky monomer content. Nevertheless, the doped membranes become more plastic at high acid doping levels. i.e. the polymer chains are more flexible.^44,45 Although the tensile strength of BPY–PBI-x serial doped membranes decreases from 25.7 to 7.2 MPa while the elongation at break increases (47.8–285.6%) with the increasing of doping level, the tensile strength are still higher than other membranes reported in some literature.^46,47 On the whole, the tensile strength of BPY–PBI-x serials membranes is in line with our expectations.

Table 3 Mechanical properties and acid leaching of BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%)

Samples	Thickness (μm)	Tensile strength (MPa)		Elongation at break (%)		Acid leaching (wt%)
		Undoped	Acid doped	Undoped	Acid doped
BPY–PBI-0%	81	86.5	25.7	6.8	47.8	45.8
BPY–PBI-10%	84	99.7	21.4	6.2	62.7	44.3
BPY–PBI-30%	80	76.1	16.5	5.6	95.4	40.6
BPY–PBI-50%	82	82.3	13.8	5.9	146.2	37.9
BPY–PBI-70%	79	72.2	12.6	7.2	210.5	36.1
BPY–PBI-100%	80	68.7	7.2	5.4	285.6	33.7

---

Acid leaching

The acid leaching test of the BPY–PBI-x serial membranes was carried out according to the previous method.^48–50 The acid leaching values of the BPY–PBI-x serials membranes are summarized in Table 3. From Table 3, we can find the acid leaching decreases from 45.8% to 33.7% with the addition of the BPY monomer, which attributes to the increase of N atom of proton acceptor from bipyridine unit. The bipyridine unit offers numerous basic sites to interact with more PA molecules. Moreover, the free acid between the polymers chains is a major factor for facilitating proton transfer via rearrangement of hydrogen bonds, and can form H₂PO₄⁻/HPO₄²⁻ anionic chains.^51,52 Thus, the acid retention ability of the BPY–PBI-x serials membranes also may attribute to the complicated hydrogen bonds network and anionic chains under the higher doping level.

Thermal stability

Fig. 4 shows the TGA curves of BPY–PBI-x serial membranes recorded under nitrogen atmosphere at a heating rate of 15 °C min⁻¹ from 40 to 800 °C. All of the BPY–PBI-x membranes show two evident weight losses. The initial weight loss, which obvious appears before 200 °C, is due to evaporation of the absorbed water within the copolymer and the residual solvents on the membranes. The membranes samples suffer only 6% weight loss in this way. The second weight loss is observed at approximately 580–650 °C corresponded to the degradation of the polymer backbone. As depicted in Fig. 4, the good thermal properties of BPY–PBI-x membranes indicate ones are excellent candidates as proton conducting materials.

Fig. 4 Thermal gravity analysis curves of BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%) in N₂ atmosphere at a heating rate of 15 °C min⁻¹ in the temperature range from 40 to 800 °C.

PA-doping level and swelling ratio of BPY–PBI-x

The BPY–PBI-x serial membranes were easily fabricated by casting polymer solution using DMAc as solvent. All of the membranes were soaked into a H₃PO₄ solution (80 wt%) at room temperature for 3–4 days to obtain acid-doping PBI membranes. Hence the membranes of doping levels ranging from 151.4% to 342.7% (shown in Fig. 5) were acquired. It is well known that the proton conductivity of membrane rest with the phosphoric acid loading of the membrane. According to the Fig. 5, the acid-doping levels will increases with the increasing content of BPY monomer, which implies the introduction of the monomer could enhance the acid absorbing ability of membranes. As a result of this, the extra N atom consisting in the BPY monomer enhances the basic sites and hence facilitated the membranes to absorb higher content of phosphoric acid. Anything else, intermolecular hydrogen bonds among main chains were reduced and the rigid among main chains were reduced and the rigid structure of BPY–PBI-x serial membranes also would be destroyed due to the addition of bulky monomer content, which result in the existence of twisted non-coplanar and the enhancement the free volume of inter-chain.^23,53,54 As a consequence, the larger free volume will make the membranes absorb higher content of phosphoric acid. However, the higher PA-doping level will lead to another problem about swelling ratio of membranes. Thus, how to keep the membranes good dimensional stability still needs to continuously explore.

Fig. 5 Comparison of PA-doping level/volume swelling ratio BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%).

Proton conductivity

The proton conductivity of the BPY–PBI-x serial acid-doping membranes was measured at a temperature range of 80–160 °C. The conductivity of the acid-doping PBI would be reduced above 160 °C due to the dehydration of phosphoric acid, which turned into H₄P₂O₇.⁵⁵ As depicted in Fig. 6, the proton conductivity of BPY–PBI-x serial membranes at low temperature is not very ideal. Even so, this situation can be improved with the increasing of temperature and PA-doping level. The reason is that the proton conductivity of the PA-doping membranes is a primary function of the doping acid content and temperature.⁵⁶ In view of the unique polymer structure and the larger free volume, some references show the similar higher PA-doping level compared with us,^23,54 and embody higher proton conductivity. The proton conductivity of the BPY–PBI-x membranes increased with the increase of the BPY content, which indicated the advantage of the N atom of proton acceptor and the enhancement of free volume. The conductivity of doping levels of BPY–PBI-100% membrane with the highest doping level (342.7%) reaches 78.6 mS cm⁻¹ at 160 °C.

Fig. 6 The proton conductivity of the BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%) after doped with 80% PA.

To further understand the relationship between proton conductivity and temperature, the Arrhenius plots of BPY–PBI-x serial membranes are plotted between the logarithm of conductivity and temperature (Fig. 7). In terms of proton-hopping dominant mechanism, the linear relationship between conductivity and temperature is well fitted by Arrhenius eqn (6):

σ = σ₀ exp(−E_a/RT) (6)

where E_a is the value of activation energy for conduction, σ₀ is a pre-exponential factor and R is the ideal gas constant.¹⁵ The activation energy of the acid doped BPY–PBI-x serial membranes decreased from 15.1 kJ mol⁻¹ to 10.3 kJ mol⁻¹ when the proton conductivity increased from 9.6 mS cm⁻¹ to 78.6 mS cm⁻¹. In addition, the activation energy of the BPY–PBI-100% membranes was much lower than that of other membranes. This indicates that the proton conductivity was corresponding with the value of typical Arrhenius plots.

Fig. 7 Arrhenius plots of ln σ vs. 1000/T of BPY–PBI-x (x = 0%, 10%, 30%, 50%, 70%, 100%).

Fenton test

The chemical oxidation stability of the membranes is a crucial performance to the durability of PEMFCs, and Fenton reagent is generally used to measure the oxidative stability of membranes. In the test, free oxidative radicals formed from the decomposition of H₂O₂ are reported to attack the backbone of polymers.²⁶ Therefore, in order to evaluate the oxidative stability of the membranes, the membrane samples were investigated by immersing the film in Fenton's reagent (3% H₂O₂ and 4 ppm Fe²⁺, 60 °C, refreshed every 24 h). The results are shown in Fig. 8. It could be found that the degradation of BPY–PBI-0% membrane is about 15% in this oxidation process. However, the BPY–PBI-100% membrane is only less than 7%. After 192 h of testing, all membranes suffer from different levels of weight loss. Meanwhile, the BPY–PBI-x serial membranes degradation gradually become slighter with the increase of the BPY proportion, which may correspond to unique oxidation resistance of the presence of bipyridine unit. Due to the introduction of the bipyridine unit, the anti-oxidative stability of BPY–PBI-x serial membranes will increase because the pyridine unit can quench radicals by the formation of pyridine-N-oxide. Thus the oxidation reactions of imidazole units in PBI will be retarded, and the corresponding durability of PBI membrane will increase as well.

Fig. 8 Oxidation stability of BPY–PBI-x polmyers recorded every 24 h in Fenton test (3% H₂O₂ and 4 ppm Fe²⁺, 60 °C, refreshed every 24 h).

Hence, a possible route of how nitrogen-atoms behaves in the anti-oxidation process is proposed. Firstly, in the electron transfer process, OH˙ or HO₂˙ radicals are often produced. These radicals can initiate chain reactions. No matter when the chain reaction occurs during the running of fuel cell, it will cause a series of chain reactions and do harm to the membranes of fuel cell. Afterwards, nitrogen-atoms of bipyridine units, which could act as antioxidant to terminate these chain reactions by removing free radical intermediates, will retard the oxidation reactions of attacking the H–N of imidazole groups of PBI. The corresponding mechanism is shown in Scheme 2.

Scheme 2 Possible mechanism involved in the antioxidant process of BPY–PBI-x polmyers.

The oxidation of nitrogen-atoms in bipyridine units of PBI were further confirmed by a subsequent test. The BPY monomer was firstly transformed to methyl ester structure (sBPY for short). Then, sBPY was oxidized under the condition of glacial acetic acid and hydrogen peroxide at 80 °C for 24 h. The oxidative compound without purification was checked by MS and FT-IR spectroscopic analyses. According to the Fig. 9, the molecular ion peak at 441.1453 [M + H]⁺ can be explained that only one nitrogen in sBPY was oxidized, and the peaks at 457.1403 [M + H]⁺ represents the structure of sBPY oxidized completely. What's more, the existence of N → O bonds also was proved by FT-IR spectra. Fig. 10 shows the comparison of the FT-IR spectra results of the tested samples. As signed in the figure, the characteristic peak at 840 cm⁻¹ proves the formation of N → O bonds, which indicates the transformation from nitrogen to N → O bonds. Thus, the nitrogen-atoms could act as antioxidant to terminate these chain reactions by removing free radical intermediates, and thus retarded other oxidation reactions occurred in the H–N of imidazole groups of PBI. To sum up, the tests above mentioned provide strong evidence for the anti-oxidation mechanism.

Fig. 9 The mass spectrometry analysis of N → O bonds of the sBPY.

Fig. 10 FT-IR spectra of the sBPY before and after oxidation.

Conclusions

The serial polymers of PBI containing BPY moieties were synthesized successfully through a microwave-assisted polycondensation in this paper. The BPY–PBI-x membranes are investigated on the part of solubility, oxidative stability, PA-doping level, swelling ratio, proton conductivity, and so on.

In this research, the introduction of bulky BPY diacid monomer containing bipyridyl moieties makes the main chain of serial PBI form a twisted plane. It increases the free volume of inter-chain, and decreases the intermolecular hydrogen bonding interaction. Based on these advantages, the non-planar BPY–PBI-x serial membranes exhibit satisfactory mechanical properties, thermal stability, solubility, especially the PA-doping level and higher proton conductivity. The proton conductivity of BPY–PBI-100% membrane can reaches 78.6 mS cm⁻¹ at 160 °C. In addition, the BPY–PBI-x serial membranes exhibit better anti-oxidative stability because the pyridine unit can quench radicals by the formation of pyridine-N-oxide. The unique anti-oxidation mechanism of the BPY–PBI-x serial membranes, which firstly was pointed out about the application of pyridine heterocyclic ring in HT-PEMs, provide new insights for novel anti-oxidation membrane design. Therefore, the BPY–PBI-x serial membranes satisfy the three goals of soluble PBI, antioxidative PBI and higher proton conducting PBI.

Acknowledgements

The project is supported by the National Natural Science Foundation of China (Grant No. 21176023, No. 21276021).

Notes and references

1. A. Chandan, M. Hattenberger, A. El-Kharouf, S. Du, A. Dhir, V. Self, B. G. Pollet, A. Ingram and W. Bujalski, J. Power Sources, 2013, 231, 264–278.
2. Q. F. Li, J. O. Jensen, R. F. Savinell and N. J. Bjerrum, Prog. Polym. Sci., 2009, 34, 449–477.
3. J. S. Wainright, J. T. Wang, D. Weng, R. F. Savinell and M. Litt, J. Electrochem. Soc., 1995, 142, L121–L123.
4. E. Quartarone and P. Mustarelli, Energy Environ. Sci., 2012, 5, 6436–6444.
5. R. H. He, Q. F. Li, A. Bach, J. O. Jensen and N. J. Bjerrum, J. Membr. Sci., 2006, 277, 38–45.
6. S. R. Samms, S. Wasmus and R. F. Savinell, J. Electrochem. Soc., 1996, 143, 1225–1232.
7. J. A. Asensio, E. M. Sánchez and P. Gómez-Romero, Chem. Soc. Rev., 2010, 39, 3210–3239.
8. S. Authayanun, K. Im-Orb and A. Arpornwichanop, Chin. J. Catal., 2015, 36, 473–483.
9. S. Angioni, P. P. Righetti, E. Quartarone, E. Dilena, P. Mustarelli and A. Magistris, Int. J. Hydrogen Energy, 2011, 36, 7174–7182.
10. Y. S. Guan, H. T. Pu and D. C. Wan, Polym. Chem., 2011, 2, 1287–1292.
11. J. S. Yang, Q. F. Li, J. O. Jensen, C. Pan, L. N. Cleemann, N. J. Bjerrum and R. H. He, J. Power Sources, 2012, 205, 114–121.
12. S. Maity, A. Sannigrahi, S. Ghosh and T. Jana, Eur. Polym. J., 2013, 49, 2280–2292.
13. J. C. Chen, P. Y. Chen, Y. C. Liu and K. H. Chen, J. Membr. Sci., 2016, 513, 270–279.
14. E. Quartarone, A. Magistris, P. Mustarelli, S. Grandi, A. Carollo, G. Z. Zukowska, J. E. Garbarczyk, J. L. Nowinski, C. Gerbaldi and S. Bodoardo, Fuel Cells, 2009, 9, 349–355.
15. L. Xiao, H. Zhang, T. Jana, E. Scanlon, R. Chen, E. W. Choe, L. Ramanathan, S. Yu and B. C. Benicewicz, Fuel Cells, 2005, 5, 287–295.
16. A. Sannigrahi, S. Ghosh, S. Maity and T. Jana, Polymer, 2010, 51, 5929–5941.
17. J. Fang, X. Lin, D. Cai, N. He and J. Zhao, J. Membr. Sci., 2016, 502, 29–36.
18. Y. Guan, H. Pu, M. Jin, Z. Chang and A. D. Modestov, Fuel Cells, 2012, 12, 124–131.
19. C. H. Shen, S. L. Hsu, E. Bulycheva and N. Belomoina, J. Membr. Sci., 2012, 399, 11–15.
20. J. S. Yang, D. Aili, Q. F. Li, Y. X. Xu, P. P. Liu, Q. T. Che, J. O. Jensen, N. J. Bjerrum and R. H. He, Polym. Chem., 2013, 4, 4768–4775.
21. Q. T. Che, Z. F. Zhu, N. Chen and X. Zhai, Mater. Des., 2015, 87, 1047–1055.
22. J. C. Chen, J. A. Wu and K. H. Chen, RSC Adv., 2016, 6, 33959–33970.
23. X. P. Li, C. Liu, S. H. Zhang, G. P. Yu and X. G. Jian, J. Membr. Sci., 2012, 423, 128–135.
24. J. S. Yang, Y. X. Xu, L. Zhou, Q. T. Che, R. H. He and Q. F. Li, J. Membr. Sci., 2013, 446, 318–325.
25. J. S. Yang, C. Liu, L. P. Gao, J. Wang, Y. X. Xu and R. H. He, RSC Adv., 2015, 5, 101049–101054.
26. Q. F. Li, C. Pan, J. O. Jensen, P. Noyé and N. J. Bjerrum, Chem. Mater., 2007, 19, 350–352.
27. C. H. Shen and S. L. Hsu, J. Membr. Sci., 2013, 443, 138–143.
28. S. Wang, C. Zhao, W. J. Ma, G. Zhang, Z. J. Liu, J. Ni, M. Y. Li, N. Zhang and H. Na, J. Membr. Sci., 2012, 411, 54–63.
29. S. Kundu, M. W. Fowler, L. C. Simon, R. Abouatallah and N. Beydokhti, J. Power Sources, 2008, 183, 619–628.
30. F. J. Pinar, P. Cañizares, M. A. Rodrigo, D. Ubeda and J. Lobato, RSC Adv., 2012, 2, 1547–1556.
31. D. M. Xing, H. M. Zhang, L. Wang, Y. F. Zhai and B. L. Yi, J. Membr. Sci., 2007, 296, 9–14.
32. P. Trogadas and V. Ramani, J. Power Sources, 2007, 174, 159–163.
33. M. Aoki, H. Uchida and M. Watanabe, Electrochem. Commun., 2006, 8, 1509–1513.
34. P. Trogadas, J. Parrondo and V. Ramani, Electrochem. Solid-State Lett., 2008, 11, B113–B116.
35. C. He, K. F. Han, J. H. Yu, H. Zhu and Z. M. Wang, Eur. Polym. J., 2016, 74, 168–179.
36. M. K. Sharma, I. Senkovska, S. Kaskel and P. K. Bharadwaj, Inorg. Chem., 2010, 50, 539–544.
37. J. Park, J. R. Li, E. C. Sanudo, D. Q. Yuan and H. C. Zhou, Chem. Commun., 2012, 48, 883–885.
38. Y. Y. Xu, S. P. Yu, K. F. Han, J. H. Yu, H. Zhu and Z. M. Wang, Chem. J. Chin. Univ., 2013, 6, 1547–1554.
39. J. R. Klaehn, T. A. Luther, C. J. Orme, M. G. Jones, A. K. Wertsching and E. S. Peterson, Macromolecules, 2007, 40, 7487–7492.
40. S. Ghosh, A. Sannigrahi, S. Maity and T. Jana, J. Phys. Chem. B, 2010, 114, 3122–3132.
41. T. Ma, Y. F. Li, S. J. Zhang, F. C. Yang, C. L. Gong and J. J. Zhao, Chin. Chem. Lett., 2010, 21, 976–978.
42. J. Wang, J. F. Zheng, Z. Zhao and S. B. Zhang, J. Mater. Chem., 2012, 22, 22706–22712.
43. D. C. Seel and B. C. Benicewicz, J. Membr. Sci., 2012, 405, 57–67.
44. J. Lobato, P. Canizares, M. A. Rodrigo, D. Úbeda and F. J. Pinar, ChemSusChem, 2011, 4, 1489–1497.
45. J. Lobato, P. Canizares, M. A. Rodrigo, D. Úbeda and F. J. Pinar, J. Power Sources, 2011, 196, 8265–8271.
46. D. Aili, M. K. Hansen, C. Pan, Q. F. Li, E. Christensen, J. O. Jensen and N. J. Bjerrum, Int. J. Hydrogen Energy, 2011, 36, 6985–6993.
47. L. X. Xiao, H. F. Zhang, E. Scanlon, L. S. Ramanathan, E.-W. Choe, D. Rogers, T. Apple and B. C. Benicewicz, Chem. Mater., 2005, 17, 5328–5333.
48. S. Maity, S. Singha and T. Jana, Polymer, 2015, 66, 76–85.
49. Y. S. Guan, H. T. Pu, M. Jin, Z. H. Chang and D. C. Wan, Fuel Cells, 2010, 10(6), 973–982.
50. N. Uregen, K. Pehlivanoglu, Y. Ozdemir and Y. Devrim, Int. J. Hydrogen Energy, 2016, 1–12.
51. S. Subianto, Polym. Int., 2014, 63, 1134–1144.
52. R. He, J. Membr. Sci., 2003, 226, 169–184.
53. R. A. Potrekar, M. P. Kulkarni, R. A. Kulkarni and S. P. Vernekar, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 2289–2303.
54. X. B. Li, H. W. Ma, H. L. Wang, S. T. Zhang, Z. H. Jiang, B. J. Liu and M. D. Guiver, RSC Adv., 2015, 5, 53870–53873.
55. S. W. Chuang and S. L. C. Hsu, J. Polym. Sci., Part A: Polym. Chem., 2006, 44, 4508–4513.
56. J. Lobato, P. Canizares, M. A. Rodrigo and J. J. Linares, Electrochim. Acta, 2007, 52, 3910–3920.

---