Synergetic proton conducting effect in acid–base composite of phosphonic acid functionalized polystyrene and triazolyl functionalized polystyrene

Abstract

The synergetic proton conducting effect with three orders of magnitude improvement in proton conductivity was observed in an acid–base composite composed of phosphonic acid functionalized polystyrene (PS-PA) and triazolyl functionalized polystyrene (PS-Tri). In addition, a new method for the development of proton conducting materials by the combination of different acidic and basic polymers is proposed. The PS-PA was synthesized by the bromination of polystyrene on the para-position of the phenyl ring followed by phosphonation and hydrolysis. The PS-Tri was synthesized by the chloromethylation of polystyrene on the para-position of the phenyl ring followed by azidation and 1,3-dipolar cycloaddition or ‘click’ reaction. A maximum proton conductivity of 11.2 mS cm⁻¹, which is three orders of magnitude higher than that of pristine PS-PA or PS-Tri, a tensile strength of 16.3 MPa, and a minimum water uptake of 15.1% (90 °C, 90% RH) were observed in the PS-PA/PS-Tri composite composed of 66.7% PS-PA. Finally, a mosaic-like morphology model and space-charge effects were proposed to explain the synergetic proton conducting effect.

---

1. Introduction

With an increase in serious environmental pollution and energy consumption, there is rising interest to exploit novel and renewable energy resources for substituting fossil fuels, including natural gas, petroleum, and coal, or to develop new energy technologies for utilizing fossil fuels more cleanly and efficiently. Among the many technologies, fuel cells that convert chemical energy directly into electrical energy without combustion are considered to be competitive energy conversion technologies because of their environmental friendliness and high efficiency.¹ Especially, proton exchange membrane fuel cells (PEMFCs) are believed to be a real green energy technology because PEMFCs consume only hydrogen and air and release only water.^2,3 In fact, PEMFCs have been found to have potential applications in transportation and as portable and stationary power sources because of their low operation temperature, high power density, high conversion efficiency, long service life, and high reliability.^4–8

However, the large-scale commercialization of PEMFCs depends significantly on the cost affordability and the supply, distribution, and storage technologies of hydrogen. As one of the two essential materials of PEMFCs, proton exchange membrane (the other material is electrocatalyst) is not only the cause of the high cost of PEMFCs, but also a technological bottleneck of high-temperature PEMFCs (HT-PEMFCs).^9–12 For example, the HT-PEMFCs possess significant advantages when compared with low to medium temperature PEMFCs because the electrocatalyst activity, especially its tolerance to impurities in fed-gas, is considerably enhanced at high temperature, and thus the supply obstacle of hydrogen can be overcome.^13–15 However, the development of HT-PEMFCs relies on the development of high-temperature proton exchange membranes possessing high proton conductivity to support high current, adequate mechanical strength and stability, chemical and electrochemical stability under operation conditions and reasonable prices for large-scale commercialization.^11,16,17 Although the most widely accepted proton exchange membranes based on perfluorosulfonic acid, for example NAFION, could meet almost all these characteristics except for a reasonable price for low to medium temperature applications, the degradation of proton conductivity at high temperatures and dehydrated states has hindered their applications in HT-PEMFCs.^18,19 Therefore, it is important to search proton exchange membranes with a high performance at elevated temperatures and under dehydrated or even anhydrous states.

In the search for high temperature proton exchange membranes, new types of proton exchange membranes based on acid–base composites have attracted significant attention for their improved proton conductivity at elevated temperatures and under dehydrated or even anhydrous states.²⁰ In an acid–base composite, the base component acts as the hydrogen-bonding (H-bonding) acceptor and acid component acts as the H-bonding donor. In addition, the H-bonding bridges provide proton transport pathways under dehydrated states, and the ionic cross-linking between acidic groups and basic groups depresses the water uptake and swelling and improves the mechanical properties.

For example, Liang et al. developed an acid–base composite membrane composed of sulfonated polymers as acidic components and polyetherimide (PEI) as basic components, demonstrating better resistance to swelling and improved thermal stability, but slightly reduced proton conductivity compared to the corresponding pristine sulfonated polymers.²¹ Fu et al. prepared an acid–base composite membrane by blending polysulfone bearing benzimidazole side groups with sulfonated SPEEK, exhibiting better performance in PEMFC at 90 and 100 °C compared to the pristine SPEEK and NAFION 111 membranes.²² Kufacı synthesized a novel anhydrous proton conducting polymer electrolyte based on poly(ethyleneglycol methacrylate phosphate) (PEGMAP) and heterocycles and obtained a proton conductivity of 0.2 mS cm⁻¹ at 160 °C and anhydrous states.²³ Wu et al. blended sulfonated poly(2,6-dimethyl-1,4-phenylene oxide) with (3-aminopropyl)triethoxylsilane through a sol–gel process resulting in high proton conductivity and low methanol permeability as compared to NAFION 117.²⁴

For the preparation of acid–base composite proton exchange membranes, two types of polymers, i.e. the phosphonic acid functionalized and nitrogenous heterocyclyl functionalized polymers, have attracted significant attention as potential functional polymers because of their chemical and thermal stability, versatile hydrogen-bonding formation ability, and potentially high proton conductivity. Phosphonic acid not only possesses improved thermal and chemical stability, but also shows improved proton conductivity at high temperature under anhydrous states compared to sulfonic acid.^25–27 The nitrogenous heterocycles or immobilized nitrogenous heterocycles were first proposed to substitute water as H-bonding formation solvents at high temperature to avoid the evaporation of solvent due to their high boiling point temperature.^9,28 Based on these arguments, Yan et al. investigated the hydrogen bonding and proton transport characteristics of complexes composed of phosphonic acid and nitrogenous heterocycles by density functional theory calculations, molecular dynamics simulations, and ¹H NMR spectroscopy, and proposed and verified the possible synergetic proton transport in acid–base complexes or composites.^29–31 Bozkurt et al. prepared a series of acid–base composite membranes by blending poly(vinylphosphonic acid) (PVPA) with poly(1-vinylimidazole) (PVIm),³² poly(1-vinyl-1,2,4-triazole),³³ poly(2,5-benzimidazole) (ABPBI),³⁴ etc., and satisfactory results were obtained.

In this paper, acid–base composite proton exchange membranes prepared based on phosphonic acid functionalized and triazolyl functionalized polystyrene are reported. ¹H NMR spectroscopy was applied to characterize molecular structures, and Thermogravimetric analysis to characterize thermal stability. The synergetic proton conducting effect in the acid–base composite is observed, and a new method for the development of proton conducting materials is proposed.

2. Experimental section

2.1 Materials

Polystyrene (PS) was purchased from Aladdin Reagent Inc. with a molecular weight of 100 000 and was used as received. Zinc chloride (ZnCl₂, Sinopharm Chemical Reagent Co., Ltd (SCRC), >98%) was dehydrated in thionyl chloride (SOCl₂, SCRC, >99%) at 85 °C under reflux for 1 hour. Chloroform (CHCl₃, SCRC, >99%) was pretreated prior to use according to the following process: repeatedly washed using sulfuric acid (H₂SO₄, SCRC, 95–98%) and deionized water for three times, dried using anhydrous calcium chloride (CaCl₂, Aladdin Reagent Inc.), and finally distilled. Sodium ascorbate (NaAsc, 99%), liquid bromine (Br₂), anhydrous ferric chloride (FeCl₃), sodium azide (NaN₃), and 2-methyl-3-butyn-2-ol were purchased from Aladdin Reagent Inc. and were used as received. Diethyl phosphate (DEP, 99%), diphenyl ether (DPE), N,N-dimethylformamide (DMF), triethylamine (Et₃N, >99%), copper sulfate pentahydrate (CuSO₄·5H₂O, >99%), and hydrochloric acid (HCl, 36–38%) were provided by SCRC and were used as received. Tris(dibenzylideneacetone) dipalladium adduct (Pd₂(dbac)₃) was provided by Yurui Reagent Inc. and was used as received. Chloromethyl methyl ether (ClCH₂OCH₃) was purchased from Shanghai Jiyan Chemical Reagent Co. and was used as received.

2.2 Bromination of PS

The bromination of PS was carried out based on the procedure reported by Subianto et al.³⁵ PS (2.8848 g) and anhydrous FeCl₃ (0.3165 g) were first added in a flask equipped with a reflux column and a magnetic stirrer, and 30 mL CHCl₃ was subsequently added to dissolve these materials. Third, 1.5 mL bromine (dissolved in 20 mL CHCl₃) was added very slowly (in about 15 minutes) because of the extremely exothermic nature of the reaction. The reaction mixture was kept in the dark, protected under a continuous N₂ flow, and stirred for 2 h at 60 °C. At the end of the reaction, the reaction mixture was poured into methanol, and the precipitated polymer was recovered as the product (PS–Br). The PS–Br was further dissolved in 1,2-dichloroethane and reprecipitated in methanol to wash out any residual bromine. Finally, the PS–Br was filtered and dried at 60 °C in a vacuum oven overnight, and 3.0596 g of a white powder was obtained (yield 60.1%) (Scheme 1).

¹H NMR (500 MHz, CDCl₃, δ): 7.26–6.86 (meta- and para-aromatic H, PS + substituted PS), 6.86–6.18 (ortho-aromatic H, PS + substituted PS), 1.75 (–CH(Ar)–), 1.41 (–CH₂–).

Scheme 1 Synthesis of phosphonic acid functionalized PS.

2.3 Phosphonation of PS–Br

First, 1.0252 g PS–Br was dissolved in diphenyl ether (12 mL) in a flask fitted with a gas inlet and outlet, and Et₃N (0.5 mL), DEP (6 mL), and Pd₂(dbac)₃ (0.0426 g) were injected successively into the flask under flowing N₂. Third, the reaction solution was stirred at 150 °C for 3 days under continuous N₂ flow. Finally, the phosphonated PS (PS-DEP) was precipitated in a 95/5 v/v methanol/water solution, isolated as a pale yellow solid, and dried under vacuum at 60 °C for 1 day with a yield of 78% (1.0379 g).

¹H NMR (500 MHz, CDCl₃, δ): 7.26–6.86 (meta- and para-aromatic H, PS + substituted PS), 6.84–6.00 (ortho-aromatic H, PS + substituted PS), 4.16 (–O–CH₂–), 2.25–1.60 (–CH(Ar)–), 1.60–1.25 (–CH₂–, –CH₃).

2.4 Hydrolysis of PS-DEP

First, 0.4457 g PS-DEP was added in 10 mol dm⁻³ HCl (50 mL) and stirred at 100 °C for 24 h, and the product was then washed with deionized water for three times to eliminate any residual free acid. Finally, the product was dried at 60 °C in a vacuum oven for 12 h, and 0.3429 g (yield 77%) phosphonic acid functionalized polystyrene (PS-PA) was obtained.

¹H NMR (500 MHz, DMSO-d₆, δ): 7.33–6.87 (meta- and para-aromatic H, PS + substituted PS), 6.87–6.36 (ortho-aromatic H, PS + substituted PS), 3.26–3.17 (–OH–), 1.86–1.00 (–CH(Ar)–, –CH₂–).

2.5 Chloromethylation of PS

First, 2.1327 g of PS and 5.5780 g of anhydrous ZnCl₂ were added in 100 mL pretreated CHCl₃ under N₂ protection in a flask equipped with a gas inlet and an outlet, a reflux column, and a magnetic stirrer. When the PS was completely dissolved, 7 mL of ClCH₂OCH₃ (n_ZnCl2 : n_ClCH2OCH3 : n_PS=2 : 5 : 1; n_PS represents the moles of repeat unit of PS) was added into the solution. Then, the reaction proceeded at 60 °C for 7 h and a pink solution was obtained. This pink solution was precipitated in a 9 : 1 v/v methanol–water mixture, and a white flocculent precipitate was recovered. Finally, the precipitate was redissolved in CHCl₃, and reprecipitated in a methanol–water mixture for 3 times. The resulting polymer (PS–CH₂–Cl) was dried in a vacuum oven at 60 °C for 24 h with a yield of 74% (2.3161 g) (Scheme 2).

¹H NMR (500 MHz, CDCl₃, δ): 7.29–6.86 (meta- and para-aromatic H, PS + substituted PS), 6.86–6.36 (ortho-aromatic H, PS + substituted PS), 4.55–4.42 (Ar–CH₂–Cl), 1.80–1.28 (–CH(Ar)–, –CH₂–).

Scheme 2 Triazolyl functionalization of PS.

2.6 Azidation of PS–CH₂–Cl

PS–CH₂Cl (0.9683 g) and excess NaN₃ (0.5183 g) were mixed in 25 mL DMF and stirred at 90 °C for 12 h. The product was precipitated in water yielding a pale yellow solid (PS–CH₂–N₃, 0.8951 g, yield 89%), and then washed three times in water to remove excess NaN₃.

¹H NMR (500 MHz, CDCl₃, δ): 7.26–6.81 (meta- and para-aromatic H, PS + substituted PS), 6.82–6.44 (ortho-aromatic H, PS + substituted PS), 4.21–4.20 (Ar–CH₂–N₃), 1.78–1.74 (–CH(Ar)–), 1.54–1.40 (–CH₂–).

2.7 The ‘click’ reaction

PS–CH₂–N₃ (0.8951 g, 1.0 equiv.), 2-methyl-3-butyn-2-ol (0.5309 g, 1.1 equiv.), CuSO₄·5H₂O (0.2067 g, 0.1 equiv.), and NaAsc (0.3615 g, 0.3 equiv.) were dissolved in 20 mL DMF and stirred at room temperature for 24 h under N₂ protection. The yellow solution was then precipitated in water and dried yielding 1.1032 g triazolyl functionalized PS (PS-Tri) at a yield of 92%.

¹H NMR (500 MHz, DMSO-d₆, δ): 7.95 (N–CH=C), 7.12–6.72 (meta- and para-aromatic H, PS + substituted PS), 6.72–6.34 (ortho-aromatic H, PS + substituted PS), 5.44–5.40 (Ar–CH₂–N), 5.14–5.12 (–OH), 1.71–1.27 (–CH(Ar)–, –CH₂–, –C(CH₃)₂).

2.8 Preparation of acid–base composite membranes

The PS-PA and PS-Tri were first separately ground into fine powders, and then mixed ( = 0.0%, 50.0%, 64.3%, 66.7%, 68.7%, 71.4%, 75.0%, 100.0%, where n_PS-PA and n_PS-Tri represent the moles of phosphonic acid group and triazolyl group, respectively). Finally, the mixed powders were hot-pressed using a FM1202 vacuum pressing machine at 160 °C for 30 minutes under evacuation.

2.9 Characterization methods

2.9.1 ¹H NMR spectroscopy. The molecular structures of the polymers were characterized by ¹H NMR spectroscopy on a Bruker Avance 500 MHz NMR spectrometer using tetramethylsilane (TMS) as the internal reference and CDCl₃ or DMSO-d₆ as the solvent.

2.9.2 Proton conductivity. Proton conductivity was measured by two-electrode impedance spectroscopy (rounded Pt electrodes with diameter of 6 mm) at 90% relative humidity in a KR-K humidity-temperature controller using a Solartron 1255B response analyzer. To establish the hydration equilibrium, the membranes were kept in the humidity-temperature controller preset at a controlled temperature and humidity for at least four hours before each measurement. All the experimental impedance spectra were fitted to the equivalent circuit as shown in Fig. 1, where R₀ represents the membrane resistance, R₁ and R₂ are interface resistances, CPE₁ and CPE₂ are constant phase elements, and L₀ is inductance (it is significant only at high frequency). The conductivity of the membrane is calculated as, σ = d/R₀A, where d and A are the thickness and area of the membrane, respectively, between two electrodes. Fig. 1 also shows two typical experimental impedance spectra and their corresponding fitted impedance spectra with satisfactory fitting results. Table 1 summarized the fitted parameters of the impedance spectra in Fig. 1 as examples.

Fig. 1 Equivalent circuit and typical impedance spectra; the curves represent the fitted impedance spectra and data points represent the experimental results.

Table 1 Fitted parameters to the equivalent circuit; Q and n are parameters for constant phase elements with

T (°C)	60	80
L0 (μH)	32.46	4.94
R0 (Ω)	1821	63.6
R1 (kΩ)	22.270	7.521
	9.654	4.579
n1	0.415	0.631
R2 (kΩ)	197.2	8.9 × 10^11
	61.01	80.12
n2	0.857	0.577

---

2.9.3 Mechanical property. The mechanical properties of the membranes were characterized using a MTF-100 microcomputer controlled electronic micro-tensile machine stretched at a constant speed of 0.1 mm min⁻¹. Tensile strength was evaluated as ultimate tensile strength at break at 25 °C in air.

2.9.4 Thermogravimetric analysis (TGA). TGA was carried out in a nitrogen atmosphere from 33 to 500 °C at a heating rate of 10 °C min⁻¹. All the samples were pre-heated overnight at 60 °C before testing.

2.9.5 Water uptake. During the measurement of water uptake, the membranes were first dried in an oven at 110 °C for 12 hours, and then weighed (w_d). The weighed membranes were left overnight in a humidity-temperature controller at 90% relative humidity and a preset temperature. After removal from the humidity-temperature controller, the membrane was immediately weighed (w_w)and water uptake (w_u) was calculated as .

3. Results and discussion

3.1 ¹H NMR spectroscopy and degree of substitution

The ¹H NMR spectra of PS, PS–Br, and PS-DEP are shown in Fig. 2. For the ¹H NMR spectrum of PS, a broad peak at 6.95–6.31 ppm is from the ortho-aromatic proton H_o and a peak at about 7.27–6.95 ppm is from the meta- and para-aromatic protons H_m and H_p of the phenyl ring. After bromination and phosphonation, these peaks slightly shift to 6.86–6.18 ppm and 7.26–6.86 ppm, or to 6.84–6.00 ppm and 7.26–6.84 ppm, respectively. For the PS-DEP, a new peak appears at 4.16 ppm attributed to the –OCH₂– of the ethoxyl group, and the peak of the –CH₃ of the ethoxyl group appears in the same range as the –CH₂– of the polymer backbone at 1.60–1.25 ppm. In addition, the peaks of aromatic H_m and H_p for PS-DEP, which overlap for PS and PS–Br, split into two groups because of the presence of ³¹P nearby. From the ¹H NMR spectra of PS-DEP and PS-PA in DMSO-d₆, as shown in Fig. 3 (PS-PA does not dissolve in CDCl₃, and the ¹H NMR of PS-DEP in DMSO-d₆ is slightly different from that in CDCl₃), the peak at 3.95 ppm attributed to the –OCH₂– of the ethoxyl group disappears and the peak at 3.26–3.17 ppm attributed to the hydroxyl group, which overlaps with the water peak, confirms the complete hydrolysis of PS-DEP.³⁶

Fig. 2 ¹H NMR spectra of PS, PS–Br, and PS-DEP dissolved in CDCl₃.

Fig. 3 ¹H NMR spectra of PS-DEP and PS-PA dissolved in DMSO-d₆.

The degree of bromination DB (or phosphonation, DP), the number of Br atoms (or phosphonic acid groups) grafted per repeat unit of PS, was evaluated by peak integration of the corresponding ¹H NMR spectrum. First, the peak integration from the ortho-aromatic protons H_o was normalized to 2, and then the peak for the meta- and para-aromatic protons H_m and H_p was integrated as I_m,p. The degree of bromination DB was calculated as DB = 3 − I_m,p. The degree of phosphonation DP was evaluated by the peak integration from the –OCH₂– units at 4.16 ppm with DP = I_ethoxyl/4.

Fig. 4 shows the ¹H NMR spectra of PS, PS–CH₂–Cl, PS–CH₂–N₃, and PS-Tri. The ¹H NMR for PS–CH₂–Cl shows three broad peaks at 6.86–6.36 ppm attributed to the ortho-aromatic proton H_o, 7.29–6.86 ppm attributed to the meta- and para-aromatic protons H_m and H_p, and 4.55–4.42 ppm attributed to the Ar–CH₂–Cl protons H_b. These peaks slightly shift to 6.81–6.44 ppm, 7.26–6.81 ppm, and 4.21–4.20 ppm after azidation, and shifted further to 6.72–6.34 ppm, 7.12–6.72 ppm, and 5.44–5.40 ppm after the “click” reaction.

Fig. 4 ¹H NMR spectra of PS, PS–CH₂Cl, and PS–CH₂N₃ in CDCl₃, and PS-Tri in DMSO-d₆.

The degree of chloromethylation, DC, the number of chloromethyl groups grafted per repeat unit of PS, was evaluated from the peak integration of the peak at 4.55 ppm (I_b) for PS–CH₂–Cl, DC = I_b/2. The degree of triazolylation, DT, the number of triazolyl groups grafted per repeat unit of PS, was evaluated from the peak integration of methyl protons around 1.4 ppm (I_methyl) for PS-Tri, DT = (I_methyl − 3)/6.

In a catalyzed reaction, the bromination degree depends on the amount of catalyst used. When the mass ratio of catalyst to PS increases from 0.068 to 0.095, the degree of bromination increases from 0.36 to 0.51 (Table 2). In the same manner, the degree of phosphonation depends on the amount of catalyst used during the phosphonation reaction. When the molar ratio of catalyst to repeat unit of PS increases from 8.0 × 10⁻³ to 9.2 × 10⁻³, the degree of phosphonation increases from 0.36 to 0.43 (Table 2), which is significantly higher than the literature values of 0.33.³⁵

Table 2 Degrees of bromination and phosphonation

Catalyst^a	t (h)	DB	Catalyst^b	t (h)	DP
a Mass ratio of catalyst to PS.b Molar ratio of catalyst to the repeat unit of PS.
0.068	2	0.36	8.0 × 10^−3	72	0.36
0.071	2	0.47	8.9 × 10^−3	72	0.38
0.095	2	0.51	9.2 × 10^−3	72	0.43

---

The chloromethylation of PS is extremely sensitive to moisture, and the entire reaction must be operated under an anhydrous state. The highest degree of chloromethylation, DC, is 60% in this work, and the degree of triazolylation, DT, is almost equivalent to DC because of the completeness of the ‘click’ reaction.

3.2 Proton conductivity

Fig. 5 shows the impedance spectroscopy of a membrane (composed of 66.7% PS-PA) measured at 10 mV, 20 mV, and 30 mV under 100 °C and 90% relative humidity. It is clear that impedance spectroscopy is almost invariant, especially at a high frequency, revealing a linear relationship between current and voltage (ohmic behavior) of the membrane.

Fig. 5 Impedance spectroscopy of 66.7% PS-PA membrane measured at different voltage (100 °C, RH 90%, 10⁶–0.1 Hz).

The proton conductivities of pristine PS-PA and PS-Tri are very low at about 10⁻³ mS cm⁻¹. For the acid–base composites of PS-PA and PS-Tri, proton conductivities are significantly improved (Fig. 6). The equimolar acid–base composite, composed of 50.0% PS-PA, shows very low conductivity at about 2 × 10⁻³ mS cm⁻¹. It is observed that if a more acidic component is added to the composite, a synergetic proton conducting effect, the proton conductivity of the acid–base composite, improves remarkably when compared with both the acidic component and the basic component. In addition, the maximum proton conductivity, three orders of magnitude improvement, is observed in the composite composed of 66.7% PS-PA (PS-PA : PS-Tri = 2 : 1) at 11.2 mS cm⁻¹ (100 °C and RH 90%). However, proton conductivity decreases when the PS-PA component exceeds 66.7%.

Fig. 6 Proton conductivities of the composites of PS-PA and PS-Tri.

Proton conductivity is also dependent on temperature, proton conductivity at 40 °C is approximately one order of magnitude lower than that at 100 °C revealing the activation of proton conducting in the acid–base composites.

3.3 Thermal stability of PS-Tri and PS-PA

TGA (Fig. 7) shows that PS-PA undergoes 3.5 wt% initial mass loss below 191 °C because of the evaporation of water, and 8.3 wt% additional mass loss because of the formation of anhydride by the creation of P–O–P linkages, and the evaporation of low molecular weight materials below 312 °C.³⁷ Between 312 °C and 414 °C, the PS-PA is stable and there is insignificant mass loss. Finally, in the narrow temperature range from 414 °C to 443 °C, large mass drop, with only about 35.5 wt% residual mass left, is observed because of the decomposition of PS-PA.

Fig. 7 Thermogravimetric analysis of the PS-Tri and PS-PA; note that the PS-PA curve is offset down by 20 wt% for clarity.

The PS-Tri is thermally stable up to 267 °C under N₂ protection due to its aromatic structure. The mass loss below 267 °C is attributed to the loss of physisorbed and chemisorbed water and other low molecular weight masses. Decomposition starting at 267 °C until 385 °C is attributed to the loss of triazolyl group. Final decomposition starts at 385 °C and finishes at 444 °C with only about 37.3 wt% total mass left.

3.4 Water uptake

Fig. 8 summarizes the water uptake of acid–base composite membranes, as well as the acid and base components at temperatures between 40 to 100 °C and at a relative humidity of 90%. The phosphonic acid group and triazolyl group are highly hydrophilic, and exhibit greater water uptake than the composite membranes. Because the base component PS-Tri is composited into the PS-PA, its hydrophilicity decreases significantly and a lowered water uptake of 15.1% at 100 °C at 90% RH is observed in the PS-PA/PS-Tri membrane containing 66.7% PS-PA.

Fig. 8 Water uptake of the acid–base composite membranes (40–100 °C, 90% RH).

If the composition of a composite is kept constant, the water uptake increases with temperature and the highest water uptake of 34.6% is observed at 100 °C for PS-Tri.

3.5 Mechanical property

Tensile characteristics of pristine PS-PA, PS-Tri, and acid–base composites (dry membranes) are shown in Fig. 9. Tensile strengths at break for pristine PS-PA and PS-Tri are 15.1 MPa and 12.6 MPa, respectively, with the acidic PS-PA significantly better than the basic PS-Tri because of the formation of an anhydride by the creation of P–O–P linkages, resulting in a compact structure.³⁸

Fig. 9 Tensile characteristics of pristine PS-PA and PS-Tri, and the acid–base composites (dry membranes).

The acid–base composite composed of 50% PS-PA shows a tensile strength of 12.9 MPa, which is between the pristine acidic PS-PA and basic PS-Tri components. For the composite composed of 66.7% PS-PA, a tensile strength of 16.3 MPa is observed showing a significant synergetic effect due to the ionic-bridges between the phosphonic acid groups and the triazolyl groups.

The tensile moduli at zero stress are 39.8 and 22.6 MPa for PS-PA and PS-Tri, respectively. For the equimolar PS-PA and PS-Tri composite, the tensile modulus increases to 49.0 MPa, showing a significant ionic interaction between these two components. As more of the acidic component PS-PA is blended into the composite, the tensile modulus decreases to 34.3 MPa in 66.7% PS-PA.

The improvement in the mechanical properties of the PS-PA/PS-Tri composite may be attributed to the ionic cross-links between the acid groups and base groups; however, the number of ionic cross-links, depending on the completeness of the mixing of interaction groups, is sensitive to the process of preparation of the composite. We have to admit that the mechanical properties of the composites are sensitive to the preparation process and often vary from batch to batch in our experiments; therefore, these results are not conclusive.

3.6 Mosaic-like morphology model

Although both phosphonic acid groups and triazolyl groups are amphoteric and can self-dissociate resulting in relatively high proton conductivity especially at high temperature, their proton conductivities are significantly lower than that of hydrated sulfonic acid. When the acidic PS-PA particles and the basic PS-Tri particles were blended and hot-pressed into the acid–base composite membrane, a heterogeneous system is formed. Although we could not exclude the possible formation of a homogeneous system, hot-pressing the blend of PS-PA particles and PS-Tri particles below their melting points was more likely to result in a heterogeneous system, and a heterogeneous system was consistent with (or not contradict) SEM observation (Fig. 10a).

Fig. 10 (a) SEM of a cross-section of the membrane (the membrane was bent and broken immediately after removal from liquid nitrogen). (b) Mosaic-like morphology model of the acid–base composite consisting of PS-PA domains (yellow), deprotonated PS-PAdH⁻ domains (black), and protonated PS-TriH⁺ domains (red), and space-charge layers (white dots represent the negative charges, blue dots the positive charges) are formed at the domain interfaces. (c) Pictorialization of the acid–base neutralization reaction between the PS-PA and PS-Tri and the formation of PS-PAdH⁻and PS-TriH⁺ domains.

Furthermore, the heterogeneous system could be depicted by a percolating mosaic-like morphology model consisting of acidic PS-PA domains and basic PS-Tri domains (Fig. 10b).^39,40 At the domain interfaces, proton transport from the phosphonic acid groups to the triazolyl groups because of the acid–base neutralization reaction, and an acid–base equilibrium as well as space-charge layers are established between the PS-PA domains and PS-Tri domains. At equimolar mixing regions (domains), all the phosphonic acid groups and triazolyl groups are depleted in the acid–base neutralization reaction, and a polymeric salt is formed. At the center of the PS-PA domains, where the PS-PA and PS-Tri are blended at a molar ratio of more than 1 : 1, only the triazolyl groups are depleted and phosphonic acid groups remain intact. In addition to equimolar and PS-PA domains, transition domains, PS-PAdH⁻ domains from the deprotonation of PS-PA and PS-TriH⁺ domains from the protonation of PS-TriH also exist because of the neutralization reaction between PS-PA and PS-Tri, as shown in Fig. 10c.

The formation of a mosaic-like morphology causes the redistribution of charge carriers, increases the concentration of charge carriers at the domain interfaces and improves the overall conductivities of the membranes. This improvement in overall conductivities is similar to the space-charge effect in the heterogeneously doped systems or in the heterojunctions of the two-phase systems.^41–43 In the heterojunctions composed of CaF₂ and BaF₂, ionic conductivity is improved up to two orders of magnitude due to the redistribution of fluoride ions at the interfaces.⁴¹ In fact, the ionic conductivity of a two-phase mixture may exceed those of pure constituent phases even if the second phase is a “chemically inert” phase such as alumina and silica.⁴³ The conductivity of heterogeneously doped systems or heterojunctions increases with decreasing sizes of heterostructures or heterojunctions.^41–43 In our PS-PA and PS-Tri blend, domain sizes are in the range of 0.5 μm (Fig. 10a) within the upper limit of the heterojunctions of CaF₂ and BaF₂.⁴¹ The PS-PA and PS-Tri are soft matters, gradual transition from PS-PA to PS-Tri are expected.⁴¹ Therefore, the domain sizes at 0.5 μm in the PS-PA and PS-Tri blend might correspond to significantly smaller equivalent heterostructures compared to the heterojunctions of CaF₂ and BaF₂.

The mosaic-like morphology model is consistent with the temperature dependence of proton conductivity. By fitting proton conductivity σ to Arrhenius equation (Fig. 11a),

σ = σ₀ exp(−E_a/RT)

where R and T are gas constant and temperature, respectively, the activation energy, E_a, and preexponential factor, σ₀, are obtained, as summarized in Table 3. Watanabe, et al. reported that the overall conductivity change with temperature follows the Vogel–Fulcher–Tammann equation in ionic liquids.^44,45 In our case, no significant improvement was observed if the Vogel–Fulcher–Tammann equation is used because of nonlinear curving fitting, and more parameters to be determined when compared with the Arrhenius equation.

Fig. 11 (a) Arrhenius plot for the proton conductivities of the PS-PA/PS-Tri composites. (b) Composites of PS-SA and PS-Tri also show synergetic proton conducting effect.

Table 3 Activation energy and preexponential factor for proton conductivity

Membranes	Ea (kJ mol^−1)	σ0 (mS cm^−1)
50.0% PS-PA	9.34	7.11 × 10^−2
64.3% PS-PA	28.2	3.73 × 10^3
66.7% PS-PA	32.8	5.18 × 10^5
68.8% PS-PA	57.0	6.38 × 10^8
71.4% PS-PA	75.3	2.44 × 10^10
75.0% PS-PA	70.8	3.17 × 10^9

---

Based on the mosaic-like morphology model and space-charge effect, overall conductivity is determined by the concentration of transportable protons (charge carriers) and transport barrier in the domain interfaces. The preexponential factor increases with the concentration of transportable protons, and the activation energy increases with the transport barrier. In homogeneous systems, such as pure PS-PA and PS-Tri systems, there are no domain interfaces and no space-charge effect, and the preexponential factor is low. In a 50.0% PS-PA membrane, the volumes of acidic and basic domains are almost equivalent, and phosphonic acid groups are readily neutralized by the triazolyl groups in the domain interfaces; the overall concentration of transportable protons is limited. As more and more PS-PA component is blended into the composite, more and more PS-PA domains remain unneutralized and the overall concentration of transportable protons increases. The overall concentration of transportable protons increases until reaching a maximum value in the 71.4% PS-PA system, and then decreases as excessive PS-PA is blended into the system. Correspondingly, the preexponential factor increases from 7.11 × 10⁻² mS cm⁻¹ in the 50.0% PS-PA system to a maximum value of 2.44 × 10¹⁰ mS cm⁻¹ in the 71.4% PS-PA system, and then decreases to 3.17 × 10⁹ mS cm⁻¹ in the 75.0% PS-PA system (Table 3).

From Table 3, it can be seen that the transport barrier also increases with increasing PS-PA content. In a homogeneous system, all the transportable protons occupy potential wells with similar depths and need to overcome the same transport barrier or activation energy. In a heterogeneous system, the transportable protons occupy various potential wells (from shallow wells to deep wells), and the transport barrier is determined by the well depth it occupies. The transportable protons occupy shallow wells and the untransportable protons occupy the deepest wells and are actually trapped. In the domain interfaces, there are a spectrum of potential wells with different depths, which is consistent with the observations in the CaF₂–BaF₂ heterostructures where activation energy continuously changes.⁴¹ In a system with low overall concentration of transportable protons, only a small portion of protons occupy shallow wells and are transportable, most of the other protons occupy deep wells and are untransportable. As the concentration of transportable protons increases, the transportable protons not only occupy shallow wells but also occupy deeper wells, resulting in higher overall transport barriers (Table 3). In a system with the highest concentration of transportable protons, the overall transport barrier also reaches its highest value. Finally, the overall transport barrier decreases as the concentration of transportable protons decreases. Under the joint effect of concentration change of transportable protons and transport barrier, overall proton conductivity increases significantly as more and more PS-PA is blended into the heterogeneous system and reaches the maximum value in the 66.7% PS-PA system.

To verify if the synergetic proton conducting effect could be observed in other acid–base composites, sulfonated polystyrene (PS-SA, the degree of sulfonation (DS) is 0.30) was synthesized and acid–base composites were prepared from PS-SA and PS-Tri. From the proton conductivity shown in Fig. 11b, it could be concluded that synergetic proton conducting effect is also observed in the acid–base composites of PS-SA and PS-Tri; however, it is not as great as that in the composites of PS-PA and PS-Tri.

Nowadays, the most important method for the development of better proton conducting materials is to functionalize existing materials with proton conducting groups, to modify the structure of proton conducting materials, and to dope additives to the existing material. However, synergetic proton conducting effect allows us to develop better proton conducting materials by combining different acidic and basic polymers, and opens a new route to the development of new proton conducting materials.

4. Conclusion

In conclusion, acid–base composite proton exchange membranes were prepared from PS-PA and PS-Tri. The degree of phosphonation at 43% is significantly higher than that reported in literature of about 33%;³⁵ a degree of triazolylation of 60% is reported for the first time.

Synergetic proton conducting effect is observed in the acid–base composite proton exchange membrane composed of PS-PA and PS-Tri with a maximum proton conductivity of 11.2 mS cm⁻¹ at 100 °C and RH 90%, three orders magnitude improvement compared to that of the acidic or basic component. In addition, the synergetic proton conducting effect is explained in terms of space-charge effect based on the mosaic-like morphology model consisting of free PS-PA domains, deprotonated PS-PAdH⁻ domains, and protonated PS-TriH⁺ domains. Moreover, the acid–base composite membrane shows reduced hydrophilicity with a lowered water uptake of 15.1% (90 °C and 90% RH) and an improved tensile strength at 16.3 MPa due to ionic bridges between the acidic and basic components.

Finally, the synergetic proton conducting effect opens a new route to the development of new proton conducting materials by combining different acidic and basic polymers.

Acknowledgements

The authors thank financial support from the Chinese National Science Foundation (nos 21073118, 21376147), the Innovation Program of Shanghai Municipal Education Commission (13ZZ078), and the 085 Knowledge Innovation Program, and they acknowledge the High Performance Computing Center and Laboratory for Microstructures, Shanghai University, for computing and structural characterization support.

References

1. A. B. Stambouli, Fuel cells: The expectations for an environmental-friendly and sustainable source of energy, Renewable Sustainable Energy Rev., 2011, 15(9), 4507–4520.
2. P. Costamagna and S. Srinivasan, Quantum jumps in the PEMFC science and technology from the 1960s to the year 2000: Part I. Fundamental scientific aspects, J. Power Sources, 2001, 102(1–2), 242–252.
3. P. Costamagna and S. Srinivasan, Quantum jumps in the PEMFC science and technology from the 1960s to the year 2000 Part II. Engineering, technology development and application aspects, J. Power Sources, 2001, 102(1–2), 253–269.
4. B. Blunier and A. Miraoui, Proton exchange membrane fuel cell air management in automotive applications, J. Fuel Cell Sci. Technol., 2010, 7(4), 1007–1017.
5. L. Carrette, K. A. Friedrich and U. Stimming, Fuel cells – Fundamentals and applications, Fuel Cells, 2001, 1(1), 5–39.
6. P. Corbo, F. Migliardini and O. Veneri, Lithium polymer batteries and proton exchange membrane fuel cells as energy sources in hydrogen electric vehicles, J. Power Sources, 2010, 195(23), 7849–7854.
7. P. Corbo, F. E. Corcione, F. Migliardini and O. Veneri, Experimental study of a fuel cell power train for road transport application, J. Power Sources, 2005, 145(2), 610–619.
8. P. Corbo, F. Migliardini and O. Veneri, An experimental study of a PEM fuel cell power train for urban bus application, J. Power Sources, 2008, 181(2), 363–370.
9. K. D. Kreuer, On the development of proton conducting polymer membranes for hydrogen and methanol fuel cells, J. Membr. Sci., 2001, 185(1), 29–39.
10. S. J. Peighambardoust, S. Rowshanzamir and M. Amjadi, Review of the proton exchange membranes for fuel cell applications, Int. J. Hydrogen Energy, 2010, 35(17), 9349–9384.
11. S. Bose, T. Kuila, T. X. H. Nguyen, N. H. Kim, K. T. Lau and J. H. Lee, Polymer membranes for high temperature proton exchange membrane fuel cell: Recent advances and challenges, Prog. Polym. Sci., 2011, 36(6), 813–843.
12. H. Zhang and P. K. Shen, Recent development of polymer electrolyte membranes for fuel cells, Chem. Rev., 2012, 112(5), 2780–2832.
13. C. Yang, P. Costamagna, S. Srinivasan, J. Benziger and A. B. Bocarsly, Approaches and technical challenges to high temperature operation of proton exchange membrane fuel cells, J. Power Sources, 2001, 103(1), 1–9.
14. R. Savinell, E. Yeager, D. Tryk, U. Landau, J. Wainright, D. Weng, K. Lux, M. Litt and C. Rogers, A polymer electrolyte for operation at temperatures up to 200 °C, J. Electrochem. Soc., 1994, 141(4), L46–L48.
15. P. Jannasch, Recent developments in high-temperature proton conducting polymer electrolyte membranes, Curr. Opin. Colloid Interface Sci., 2003, 8(1), 96–102.
16. H. Zhang and P. K. Shen, Advances in the high performance polymer electrolyte membranes for fuel cells, Chem. Soc. Rev., 2012, 41(6), 2382–2394.
17. S. Di, L. Yan, S. Han, B. Yue, Q. Feng, L. Xie, J. Chen, D. Zhang and C. Sun, Enhancing the high-temperature proton conductivity of phosphoric acid doped poly(2,5-benzimidazole) by preblending boron phosphate nanoparticles to the raw materials, J. Power Sources, 2012, 211, 161–168.
18. A.-C. Dupuis, Proton exchange membranes for fuel cells operated at medium temperatures: Materials and experimental techniques, Prog. Polym. Sci., 2011, 56(3), 289–327.
19. J. Ma, H.-J. Ni, D.-Y. Su, M.-Y. Huang and X.-X. Wang, The research status of Nafion ternary composite membrane, Int. J. Hydrogen Energy, 2012, 37(17), 13185–13190.
20. S. Ü. Çelik, A. Bozkurt and S. S. Hosseini, Alternatives toward proton conductive anhydrous membranes for fuel cells: Heterocyclic protogenic solvents comprising polymer electrolytes, Prog. Polym. Sci., 2012, 37(9), 1265–1291.
21. Y. F. Liang, H. Y. Pan, X. L. Zhu, Y. X. Zhang and X. G. Jian, Studies on synthesis and property of novel acid–base proton exchange membranes, Chin. Chem. Lett., 2007, 18(5), 609–612.
22. Y. Fu, A. Manthiram and M. D. Guiver, Blend membranes based on sulfonated poly(ether ether ketone) and polysulfone bearing benzimidazole side groups for proton exchange membrane fuel cells, Electrochem. Commun., 2006, 8(8), 1386–1390.
23. M. Kufacı, A. Bozkurt and M. Tülü, Poly(ethyleneglycol methacrylate phosphate) and heterocycle based proton conducting composite materials, Solid State Ionics, 2006, 177(11–12), 1003–1007.
24. D. Wu, T. Xu, L. Wu and Y. Wu, Hybrid acid-base polymer membranes prepared for application in fuel cells, J. Power Sources, 2009, 186(2), 286–292.
25. I. Dimitrov, S. Takamuku, K. Jankova, P. Jannasch and S. Hvilsted, Polysulfone functionalized with phosphonated poly(pentafluorostyrene) grafts for potential fuel cell applications, Macromol. Rapid Commun., 2012, 33(16), 1368–1374.
26. M. Schuster, T. Rager, A. Noda, K. D. Kreuer and J. Maier, About the choice of the protogenic group in PEM separator materials for intermediate temperature, low humidity operation: A critical comparison of sulfonic acid, phosphonic acid and imidazole functionalized model compounds, Fuel Cells, 2005, 5(3), 355–365.
27. S. J. Paddison, K. D. Kreuer and J. Maier, About the choice of the protogenic group in polymer electrolyte membranes: Ab initio modelling of sulfonic acid, phosphonic acid, and imidazole functionalized alkanes, Phys. Chem. Chem. Phys., 2006, 8(39), 4530–4542.
28. K. D. Kreuer, A. Fuchs, M. Ise, M. Spaeth and J. Maier, Imidazole and pyrazole-based proton conducting polymers and liquids, Electrochim. Acta, 1998, 43(10–11), 1281–1288.
29. B. Yue, L. Yan, S. Han and L. Xie, Proton transport pathways in acid-base complex consisting of phosphonic acid group and 1,2,3-triazolyl group, J. Phys. Chem. B, 2013, 117(26), 7941–7949.
30. L. Yan and L. Xie, Molecular dynamics simulations of proton transport in proton exchange membranes based on acid-base complexes, in Molecular Interaction, ed. A. Meghea, InTech, Rijeka, Croatia, 2012, pp. 335–360.
31. D. Zhang and L. Yan, Probing the acid-base equilibrium in acid-benzimidazole complexes by 1H NMR spectra and density functional theory calculations, J. Phys. Chem. B, 2010, 114(38), 12234–12241.
32. S. Isikli, S. Tuncagil, A. Bozkurt and L. Toppare, Immobilization of invertase in a novel proton conducting poly(vinylphosphonic acid)-poly(1-vinylimidazole) network, J. Macromol. Sci., Part A: Pure Appl. Chem., 2010, 47(7), 639–646.
33. A. Aslan and A. Bozkurt, Development and characterization of polymer electrolyte membranes based on ionical cross-linked poly(1-vinyl-1,2,4-triazole) and poly(vinylphosphonic acid), J. Power Sources, 2009, 191(2), 442–447.
34. O. Acar, U. Sen, A. Bozkurt and A. Ata, Proton conducting membranes based on poly(2,5-benzimidazole) (ABPBI)-poly(vinylphosphonic acid) blends for fuel cells, Int. J. Hydrogen Energy, 2009, 34(6), 2724–2730.
35. S. Subianto, N. R. Choudhury and N. K. Dutta, Palladium-catalyzed phosphonation of SEBS block copolymer, J. Polym. Sci., Part A: Polym. Chem., 2008, 46(16), 5431–5441.
36. N. Y. Abu-Thabit, S. A. Ali and S. M. Javaid Zaidi, New highly phosphonated polysulfone membranes for PEM fuel cells, J. Membr. Sci., 2010, 360(1–2), 26–33.
37. B. Lafitte and P. Jannasch, Phosphonation of polysulfones via lithiation and reaction with chlorophosphonic acid esters, J. Polym. Sci., Part A: Polym. Chem., 2005, 43(2), 273–286.
38. B. Lafitte, M. Puchner and P. Jannasch, Proton conducting polysulfone ionomers carrying sulfoaryloxybenzoyl side chains, Macromol. Rapid Commun., 2005, 26(18), 1464–1468.
39. R. A. Rozendal, T. H. Sleutels, H. V. Hamelers and C. J. Buisman, Effect of the type of ion exchange membrane on performance, ion transport, and pH in biocatalyzed electrolysis of wastewater, Water Sci. Technol., 2008, 57(11), 1757–1762.
40. T. Xu, Ion exchange membranes: State of their development and perspective, J. Membr. Sci., 2005, 263(1–2), 1–29.
41. N. Sata, K. Eberman, K. Eberl and J. Maier, Mesoscopic fast ion conduction in nanometre-scale planar heterostructures, Nature, 2000, 408(6815), 946–949.
42. J. Maier, Space charge regions in solid two phase systems and their conduction contribution — II Contact equilibrium at the interface of two ionic conductors and the related conductivity effect, Ber. Bunsenges. Phys. Chem., 1985, 89(4), 355–362.
43. J. Maier, Space charge regions in solid two-phase systems and their conduction contribution—I. Conductance enhancement in the system ionic conductor-‘inert’ phase and application on AgCl:Al₂O₃ and AgCl:SiO₂, J. Phys. Chem. Solids, 1985, 46(3), 309–320.
44. A. Noda, K. Hayamizu and M. Watanabe, Pulsed-gradient spin-echo 1H and 19F NMR ionic diffusion coefficient, viscosity, and ionic conductivity of non-chloroaluminate room-temperature ionic liquids, J. Phys. Chem. B, 2001, 105(20), 4603–4610.
45. H. Tokuda, K. Ishii, M. A. B. H. Susan, S. Tsuzuki, K. Hayamizu and M. Watanabe, Physicochemical properties and structures of room-temperature ionic liquids. 3. Variation of cationic structures, J. Phys. Chem. B, 2006, 110(6), 2833–2839.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra04514a

---