Preparation and evaluation of crosslinked sulfonated polyphosphazene with poly(aryloxy cyclotriphosphazene) for proton exchange membrane

Journal of Energy Chemistry 25 (2016) 472–480

Contents lists available at ScienceDirect

Journal of Energy Chemistry journal homepage: www.elsevier.com/locate/jechem

http://www.journals.elsevier.com/ journal-of-energy-chemistry/

Preparation and evaluation of crosslinked sulfonated polyphosphazene with poly(aryloxy cyclotriphosphazene) for proton exchange membrane Yan Dong a, Hulin Xu b, Fengyan Fu a, Changjin Zhu a,∗ a b

Department of Applied Chemistry, Beijing Institute of Technology, Beijing 100081, China Beijing Qintian Science & Technology Development Co., Beijing 100070, China

a r t i c l e

i n f o

Article history: Received 9 September 2015 Revised 16 November 2015 Accepted 16 November 2015 Available online 16 March 2016 Keywords: Polyphosphazene Proton exchange membrane Phase separation Direct methanol fuel cell

a b s t r a c t Several crosslinked proton exchange membranes with high proton conductivities and low methanol permeability coefficients were prepared, based on the sulfonated poly[(4-fluorophenoxy)(phenoxy)] phosphazene (SPFPP) and newly synthesized water soluble sulfonated poly(cyclophosphazene) (SPCP) containing clustered flexible pendant sulfonic acids. The structure of SPCP was characterized by fourier transform infrared spectroscopy (FTIR) and 1 H NMR spectra. The membranes showed moderate proton conductivities and much lower methanol permeability coefficients when compared to Nafion 117. Transmission electron microscopy (TEM) results indicated the well-defined phase separation between the locally and densely sulfonated units and hydrophobic units, which induced efficient proton conduction. In comparison with SPFPP membrane, the proton conductivities, oxidative stabilities and mechanical properties of crosslinked membranes remarkably were improved. The selectivity values of all the crosslinked membranes were also much higher than that of Nafion 117 (0.74×105 S· s/cm3 ). These results suggested that the cSPFPP/SPCP membranes were promising candidate materials for proton exchange membrane in direct methanol fuel cells. © 2016 Science Press and Dalian Institute of Chemical Physics. All rights reserved.

1. Introduction In these years, direct methanol fuel cells (DMFCs) have gained great attention as candidate for future portable power devices for their advantages, such as low emission and simple system [1,2]. The proton exchange membrane (PEM), which transfers protons from the anode to the cathode to produce electricity and prevents the passage of electrons and a fuel gas crossover between the electrodes, is a key component of DMFCs [3]. Perfluorosulfonate membranes, such as Nafion membranes, have been widely applied in DMFCs, because they have high proton conductivities and excellent thermal as well as chemical stabilities [4]. However, there are still several technical problems limiting their applications, such as expensive cost, high methanol crossover and difficulty in preparing. One major obstacle is the high methanol crossover, which causes not only catalyst poisoning but also fuel consumption and energy efficiency loss [4,5]. To overcome these problems, many efforts have been devoted to improving Nafion-based membrane or developing new proton ex-

∗

Corresponding author. Tel: +86 10 68918506. E-mail address: [email protected] (C. Zhu).

change membranes [6,7]. In recent years, many polymeric membranes have been examined for use in fuel cells, including sulfonated poly(phenylene)s [8], sulfonated polysulfones (SPSFs) [9], sulfonated poly(ether ether ketone)s (SPEEKs) [10,11], sulfonated polyimides (SPIs) [12,13] and sulfonated polyphosphazenes [14]. Among these materials, sulfonated polyphosphazenes have attracted considerable attention because of their low cost, acceptably high proton conductivities and low methanol permeability coefficients [15,16]. In the case of sulfonated polyphosphazene, adequate acidic groups are required to obtain sufficient proton conductivity. However, high degree of sulfonation (DS) always results in the increase of water uptake and methanol crossover, and solubility in methanol/water solution [17,18]. Crosslinking of polymer membrane has been thought to be an effective approach to overcome these problems, which can not only suppress excess water uptake, methanol permeability and swelling ratio but also improve the chemical stability [19,20]. Many research groups have made great progress on the crosslinking of sulfonated polymers [21–23]. Self-crosslinked sulfonated polyphosphazenebased PEMs show relatively low proton conductivity and poor thermal stability [24]. In the present study, to achieve the PEM with sufficient proton conductivity and thermal property while maintaining the low

http://dx.doi.org/10.1016/j.jechem.2016.03.013 2095-4956/© 2016 Science Press and Dalian Institute of Chemical Physics. All rights reserved.

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

F

Cl *

N

HO

P

n

Cl

NaH,

,

HO

Quattro-II triple quadrupole mass spectrometer (WATERS, USA). Molecular weight measurements were performed via Viscotek IMBHMW-3078 gel permeation chromatography (GPC) equipped with Viscotek VE 1122 solvent delivery system (Tetrahydrofuran was used as a solvent at 25 °C. Mn and Mw were calibrated by standard polystyrene samples). Thermogravimetic analysis (TGA) was performed with TGA-Q500 (TA, USA) at a heating rate of 10 ºC/min under nitrogen in the range of 25–700 ºC. Transmission electron microscopic (TEM) observations were performed with a JEM-2010 (JEOL, Japan) transmission electron microscope.

O

F N

1,4-dioxane

P

n

O

473

1 F

O N

P O

Conc. H2SO4

n

N

F

F

O

O

P N P O n-x O

x

SO3H

2 Scheme 1. Synthetic route of polymers 1 and 2 (SPFPP).

water swelling and methanol crossover, we designed a new sulfonated poly(aryloxy cyclotriphosphazene) copolymer and combined it with the sulfonated polyphosphazenes by crosslinking in the membrane preparation. As a result, the obtained membranes showed higher proton conductivities than sulfonated polyphosphazene (SPFPP) membrane and oxidative stabilities and mechanical properties were improved. Additionally, the properties of the crosslinked membranes such as water uptake, swelling ratio, methanol permeability, oxidative stability and thermal stability are also discussed in detail. 2. Experimental 2.1. Materials The synthesis and characterization of sulfonated poly[(4fluorophenoxy)(phenoxy)] phosphazene (SPFPP) were reported by our previous work with minor modification [25]. The structure of SPFPP 2 was shown in Scheme 1. 2,6-bis(hydroxymethyl)-4methylphenol (BHMP) was prepared according to the previous report [26]. Hexachlorocyclotriphosphazene (NPCl2 )3 was purchased from LanYin Chemical, China. Tetrahydrofuran (THF), dioxane, dimethylsulfoxide (DMSO), concentrated sulfuric acid (98%), were purchased from Aldrich Chemical Co. (NPCl2 )3 was purified by two recrystallizations from hexane and vacuum sublimation prior to use. 4,4 -(hexafluoroisopropylidene) diphenol (6F-BPA), boron tribromide (BBr3 ), 1,4-butanesultone were purchased from Sigma-Aldrich. Dioxane and THF were freshly distilled from sodium benzophenone ketyl. Other chemical reagents and the organic solvents were purchased from Beijing Chemical Reagent and were purified by conventional methods.

2.2.1. Synthesis of tetra-substituted cyclotriphosphazene monomer (N3 P3 (OC6 H4 OCH3 )4 Cl2 ) (3) (NPCl2 )3 (3.5 g, 0.0101 mol) was dissolved in 15 mL of THF. 4Methoxyphenol (5.01 g, 0.0404 mol) was dissolved in 50 mL of THF and then the solution was added dropwise to the suspension of NaH (1.04 g, 0.0424 mol) in 10 mL of THF. This reaction mixture was stirred at room temperature for 24 h under the nitrogen atmosphere. The resultant sodium phenoxide solution was added dropwise to the stirred solution of (NPCl2 )3 in THF at 0 ºC. This mixture was warmed to room temperature to perform the reaction for 12 h. After the reaction was completed, the solvent was evaporated at reduced pressure, and then the oily resultant was dissolved in 50 mL of CH2 Cl2 . This solution was washed three times with 50 mL of 3% aqueous NaHCO3 , dried over Na2 SO4 , and concentrated by rotary evaporation. Column chromatography (silica gel, CH2 Cl2 :nhexane = 7:5) was carried out for further purification to give offyellow oil as 3 (4.434 g, 63.0 wt% yield). IR (KBr, v, cm−1 ); 1234.6. 1 H NMR (300 MHz, DMSO-d ): δ = 3.756 (m, 3H, OCH ), 6.795– 6 3 7.178 (m, 4H, ArH). MS: m/z 697.81, [M + H]+ . 2.2.2. Synthesis of cyclolinear poly(aryloxycyclotriphosphazene) containing methoxy groups (4) A mixture of 6F-BPA (1.736 g, 0.0076 mol) and NaH (0.382 g, 0.0156 mol) in THF (20 mL) was stirred and refluxed at room temperature for 12 h, and afterward a solution of 3 (5.305 g, 0.0076 mol) in 10 mL of freshly distilled THF was added dropwise to this reaction mixture to perform the reaction with vigorous stirring at room temperature for 24 h under a nitrogen atmosphere. Then the yellow precipitate was filtered off, and the solution was poured in to 50 mL of deionized water with vigorous stirring. The resulting copolymer was washed with deionized water and hot methanol three times and dried at 70 ºC under vacuum for 24 h. 1 H NMR (300 MHz, DMSO-d6 ): δ = 3.734 (m, 1H, OCH3 ), 6.776–7.071 (m, 2H, ArH). 31 P-NMR (162 MHz, DMSO-d6 ): δ = 9.475 (s, 2P, POC6 H4 OCH3 ), 9.601 (s, 1P, P-OC6 H4 C(CF3 )2 OC6 H4 ). 2.2.3. Synthesis of cyclolinear poly(aryloxycyclotriphosphazene) containing hydroxyl groups (5) A sample of 2.0 g of 4 was dissolved into 50 mL of CH2 Cl2 in a 100 mL three-neck flask equipped with mechanical stirrer and a nitrogen inlet. BBr3 (2 mL) was mixed with CH2 Cl2 (20 mL), and the resulting solution was added dropwise to the 4 solution at 0 ºC (ice bath). After 12 h, the mixture was poured into water with stirring, the polymer was washed with boiling water, recovered, and then dried under vacuum at 70 ºC for 24 h. IR (KBr, v, cm−1 ); 3166.5. 1 H NMR (300 MHz, DMSO-d ): δ = 6.596–7.264 (m, 6H, ArH), 9.448 6 (m, 1H, OH). (Mn = 68069 Da, Mw = 103772 Da, Mw / Mn = 1.525).

2.2. Characterizations and measurements Fourier transform infrared (FTIR) spectra of the membranes were measured on a Nicolet Nexux 470 (Nicolet, USA). 1 H and 31 P nuclear magnetic resonance (NMR) spectra were recorded on an AVANCE-300 spectrometer (Bruker Biospin International AG, Switzerland). Mass spectrum (MS) was collected using a Micromass

2.2.4. Synthesis of sulfonated poly(cyclophosphazene) (SPCP) (6) 2.0 g of sample 5 and NaH (0.382 g, 0.0156 mol) were dissolved into 30 mL of DMSO at room temperature and stirred for 24 h under nitrogen atmosphere. 1 mL of 1,4-butanesultone was added, the reaction temperature rise to 100 ºC and kept at the temperature for another 24 h. The viscous solution was precipitated into 50 mL of

474

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

Table 1. Compositions of membranes and their oxidative stabilities, IEC, water uptake and swelling ratio of membranes. Membrane

SPFPP (wt%)

SPFPP cSPFPP/SPCP 5 cSPFPP/SPCP 10 cSPFPP/SPCP 15 Nafion 117 a b

SPCP (wt%)

100 95 90 85 0

Oxidative stability

0 5 10 15 0

IEC (mmol/g)

RWa (%)

tb (h)

89 98 95 93 98

2 5 4 3.5 7

0.90 1.07 1.31 1.57 0.90

Water uptake (wt%)

Swelling ratio (%)

25 °C

80 °C

25 °C

80 °C

19.3 27.7 35.1 39.8 34.5

33.5 45.6 49.3 54.5 40.3

4.4 3.8 7.2 8.6 23.5

9.2 6.7 9.4 10.9 32.4

Remaining weights of membranes after treating in Fenton’s reagent (3% H2 O2 containing 2 ppm FeSO4 ) at 80 ºC for 1 h. Dissolution time of polymer membranes.

H 3CO

OCH 3 CF3

Cl N P

Cl

Cl OCH 3

N

N

Cl

O HO

P

N

P

Cl

NaH, THF

Cl

Cl

P O

HO

O P

N

OH

C CF3

N

NaH, THF

Cl

P O

H3 CO

OCH3

3

HO

H3 CO

OH

O O P N N P

* O

N

CF3

P

O

O O P N N

1. CH2 Cl2 / BBr3 O

C

O

*

n

P

*

2. H2 O

OCH 3

O

N

CF3 O

P

C

O CF3

CF3 HO

H3 CO

OH

5 O O O S

4

OCH 3

NaH, DMSO

R

R OH

*

H 3C

O O P N N

CF3

P

C

O

N

R

P O O

CF3

OH O

*

n

Crosslinker

R

6 R= O

OH

SO 3Na Scheme 2. Synthetic route of polymers 3–6.

O

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

isopropanol and washed with n-hexane three times before being dried under vacuum at 100 ºC for 24 h. IR (KBr, v, cm−1 ); 3448.6, 1055.8, 1011.2. 1 H NMR (300 MHz, DMSO-d6 ): δ = 1.750 (m, 16H, CH2 ), 2.48 (m, 8H, CH2 ), 3.906 (m, 8H, CH2 ), 6.770–7.172 (m, 24H, ArH).

2.2.8. Water uptake measurements Water uptake (WU) was measured by immersing the membrane into deionized water for 24 h. Then the membrane was taken out, wiped with a tissue paper, and quickly weighed on a microbalance. Water uptake was calculated according to the following equation:

Water uptake(% ) = 2.2.5. Crosslinked membrane formation and proton exchange The crosslinked membranes were prepared according to previous report [23]. SPFPP 2, the sulfonated copolymer 6 and the crosslinker BHMP (5 wt% to the copolymers) were dissolved in DMSO (5 wt%) overnight. After filtration, a drop of methanesulfonic acid was added to the solution and then the solution was cast directly onto clean Petri dish. After being carefully dried at 120 ºC for 1 h and vacuum-dried at 60 ºC for 24 h, the membranes were obtained. The membrane was immersed into water to remove the residual solvent. Proton exchange treatment was performed by immersing the crosslinked membrane in 1 M H2 SO4 for 24 h. Finally, the membranes were taken out, and rinsed with deionized water till the rinsed water became neutral to obtain the acid form membranes. The membranes thus prepared were designated as cSPFPP/SPCP X, where X is the SPCP content (wt%) in the membrane. Compositions of the membranes are listed in Table 1. All the membranes were approximately 100–150 μm thick. 2.2.6. Proton conductivity The proton conductivity of the membrane was determined using electrochemical impedance analyzer (CHI660D) over the frequency from 1 Hz to 100 MHz. A four-point-probe cell with two pairs of platinum plate electrodes pressed with a sample membrane was mounted in a sealed Teflon cell. The distance between two electrodes was 1 cm. The cell was placed in deionized water for measurement. The membranes were hydrated in deionized water at room temperature for 1 day prior to the measurement. The conductivity (σ ) of the samples was calculated from

σ = L/RS where L is the distance between the electrodes to measure the potential. R is the membrane resistance, and S is the cross sectional area of the membrane sample.

2.2.7. Methanol permeability Methanol permeability (P) measured through a reported method was carried out using a liquid diffusion cell composed of two compartments containing solutions A and B. solution A (VA = 50 mL) was 1 mol/L methanol solution, and solution B (VB = 50 mL) was deionized water. The membrane under test was immersed in deionized water for hydration before measurements and then vertically placed between the two compartments by a screw clamp [14]. Both solutions were stirred during testing to keep them homogenous. A gas chromatograph (Shimadzu, GC-14B) was used to monitor the concentration of methanol diffusion from compartment solution A to B across the membrane over time. Peak areas were converted into methanol concentration with a calibration curve. The methanol permeability coefficient was calculated by the following equation:

P = (k × VB × h )/(A × CA ) where P is the methanol permeability (cm2 /s), k is the slope of the straight-line plot of methanol concentration in solution B versus testing time, VB is the volume of solution B (mL), CA is the concentration of methanol in A (mol/L), A is the membrane areas (cm2 ) and h is the thickness (cm) of wet membrane, respectively.

475







Wwet − Wdry /Wdry × 100%

where Wwet and Wdry are the weights of the wet and dry membranes, respectively. The weights of dried membranes were measured after drying in vacuum at 70 ºC for 8 h. Swelling ratio (SW) was determined by immersing membranes samples into water for 24 h and measuring the change in length before and after the swelling according to the following equation:

Swelling ratio(% ) = [(Lwet − Ldry )/Ldry ] × 100% where Lwet and Ldry are the lengths of wet and dry membranes, respectively. 2.2.9. Oxidative stability Oxidative stability of the membranes was tested by immersing the dry membrane into hot Fenton’s reagent (3% H2 O2 containing 2 ppm FeSO4 ) at 80 ºC for 1 h [12]. The stability was evaluated by changes in weight. 2.2.10. Ion exchange capacity (IEC) IEC of the sulfonated polymers was measured using a typical titration method. The membranes in acid form were equilibrated with 50 mL NaCl solution of 2 mol/L for 24 h at room temperature. The amount of the H+ released from the membranes was determined by titration of 0.01 mol/L NaOH aqueous solution using pH meter to monitor the end points. The moles of the proton were equal to those of sulfonic groups and the IEC was calculated from the titration data using the following equation:

IEC = (CNaOH × VNaOH )/Ws where CNaOH is the concentration of NaOH solution, VNaOH is the consumed volume of NaOH solution, and Ws is the weight of the dry membrane sample. 2.2.11. Mechanical property The mechanical properties of the dry membranes were determined from stress-strain curve measured by SHIMADZUAG-I 1KN at a strain rate of 2 mm/min. The size of membrane specimens was 30 mm×10 mm. Each sample was measured 3 times and their average value was calculated. 2.2.12. Transmission electron microscopic (TEM) observations The membranes were stained with lead by ion exchange of the sulfonic acid groups by immersing it in a large excess of Pb(NO3 )2 aqueous solution for 24 h and then rinsed with water, and dried at room temperature. The stained membranes were embedded in epoxy resin and sectioned to give 70 nm thick membranes. TEM observations were performed with a JEOL JEM-2010 transmission electron microscope. 3. Results and discussions 3.1. Synthesis and characterization of copolymer 6 For potential application in the DMFCs, the polymer used for PEM should not only possess high proton conductivity and low methanol permeability but also proper thermal and chemical stability. The new sulfonated block copolymer 6 was designed (Scheme 2) as the cyclophosphazene ring is a good monomer for the preparation of polymer due to its good properties such as

476

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

Fig. 1. FTIR spectra of synthetic products: (1) 3, (2) 4 (–OCH3 ), (3) 5 (–OH) and (4) 6 (SPCP).

thermal stability, rigidity and chemical stability [27]. Moreover, in this study, the side groups of the cyclophosphazene ring can be changed specifically into alkylsulfonic acids by chemical grafting, which could provide sufficient proton conductivity for the polymer. Scheme 2 shows the synthesis of copolymer 6. A tetrasubstituted cyclotriphosphazene monomer 3 is first synthesized, and its chemical structure is characterized by FTIR, 1 H NMR and MS. The FTIR spectrum of 3 (Fig. 1) shows two absorption peaks at 880.6 and 1172.9 cm−1 , indicating the formation of a P–O–C bond via substitution reaction. The 1 H NMR spectrum of 3 exhibits the signal at 3.756 ppm correspond to –OCH3 . The other signals are observed at 6.795–7.178 ppm which can be assigned to aromatic protons. In addition, the MS analysis result is in good agreement with the data in terms of the 3. The novel cyclotriphosphazene copolymer 4 is synthesized from 3 and 6F-BPA. Compared with FTIR spectrum of 3 (Fig. 1), the disappearance of characteristic P–Cl absorption bands at 531.3 and 615.2 cm−1 in the FTIR spectrum of 4 proves the whole substitution of chloride atoms [28]. The 1 H NMR spectrum of 4 shows the expected resonance signals for the aromatic protons of both the substituted methoxyphenol groups on cyclotriphosphazenes and the phenyl groups on the backbone (Fig. 2b). The 31 P-NMR spectrum of 4 shown by Fig. 3 displays the intensive double resonance signals at 9.475 and 9.601 ppm, corresponding to P-(OC6 H4 OCH3 ) and P-(OC6 H4 C(CF3 )2 OC6 H4 ) on the phosphazene ring. The chemical shifts for these two types of phosphorus atom demonstrate a slight gap of 0.126 ppm due to the similar substituent environments. In FTIR spectrum of 5 (Fig. 1), as an important feature of the spectrum, the absorption peak at 3166.5 cm−1 is attributed to the –OH stretching. This provides the evidence for the production of the hydrogen-bond O–H. Comparative 1 H NMR spectra of 4 and 5 copolymers confirm that complete demethylation occurs: –OCH3 proton signals disappear, and –OH proton signal appears at 9.413 ppm, as shown in Fig. 2. Scheme 2 shows the grafting reaction of sulfobutyl groups onto 5 by a nucleophilic reaction with 1,4-butanesultone using NaH. The chemical structure of 6 is confirmed by FTIR and 1 H NMR spectra. As shown in Fig. 1, the absorption peak at 3448.6 cm−1 is assigned to O–H vibration of sulfonic acids groups, and the absorption at 1055.8 cm−1 is due to symmetric O=S=O stretching vibrations of sulfonic acid groups [29,30]. In the 1 H NMR spectrum of 6 (Fig. 2d), the –OH proton signal at 9.413 ppm disappears, while the signals for the four sulfobutyl methylene groups (Ha , Hb , Hc and Hd ) appears, which further supports the formation of 6 bearing sulfobutyl groups.

Fig. 2. 1 H NMR spectra of synthetic polymers: (a) 3, (b) 4 (–OCH3 ), (c) 5 (–OH) and (d) 6 (SPCP) in DMSO-d6 .

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

477

3.2. Preparation of cSPFPP/SPCP membranes

Fig. 3.

31

The preparation of the cSPFPP/SPCP membranes is based on the reaction of SPFPP, phenyl groups in the SPCP and BHMP as the crosslinker in the presence of methanesulfonic acid according to previous report (Scheme 3) [23,31,32]. The cSPFPP/SPCP membrane was prepared as follows: a DMSO solution of SPFPP 2, the sulfonated copolymer 6 and the crosslinker BHMP (5 wt% to the copolymers), a drop of methanesulfonic acid was cast directly onto clean Petri dish. After being carefully dried at 120 ºC for 1 h and vacuum-dried at 60 ºC for 24 h, the membranes were obtained. The obtained membranes were immersed in water to remove any residual solvent. After the crosslinking, the membrane became insoluble in water and DMSO, indicating the success of the crosslinking reaction in the membranes.

P-NMR spectra of 4 (–OCH3 ) in DMSO-d6 .

F

O

O N

P O

N

O

P

x

N

n-x-y

P

O

y

O

H2 C

F

SO3H

F

R1

R

R

H2C O O P N N

CF3

P

C

O

N

P O O

O

n

CF3

R

R CH2

F

SO3H

R1

F

CH2 O N

P

O

O N

y

O

P

N

n-x-y

O

P O

x

F

R:

O

SO3 Na

R1:

H3C

Scheme 3. Chemical structures of the crosslinked membrane.

OH

478

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

Fig. 4. FT-IR spectra of (1) SPFPP, (2) cSPFPP/SPCP 5.

The water uptakes and swelling ratios of the memebranes are also important properties for the fuel cell operation due to their important effects on the proton conductivity and methanol permeability [34]. As shown in Table 1, the water uptakes of cSPFPP/SPCP membranes are found to increase with the increase of SPCP content because the incorporation of the sulfonic groups can increase the hydrophilic property and water absorbability of the polymer matrix [35]. As expected, due to the hydrophilic property of sulfonated groups, the water uptake increases with elevated temperature. In this experiment, when the content of SPCP is increased to 20%, the crosslinked membrane could be broken in hot water due to high SPCP loading. Hence, in this study, we choose 15% as the maximum of SPCP content. The swelling ratios of crosslinked membranes increase slowly with the increase of temperature. The swelling ratio of cSPFPP/SPCP 15, with the IEC of 1.57 mmol/g, is only 10.9% at 80 ºC, still less than that of Nafion 117. Crosslinked membranes show much better swelling ratios in comparison with other sulfonated polyphosphazene membrane [21]. 3.5. Oxidative stability

Fig. 5. TGA curves of SPFPP and crosslinked membranes.

3.3. FTIR absorption spectra and thermal analysis of the crosslinked membranes Fig. 4 shows the FTIR spectra of SPFPP and cSPFPP/SPCP 5 membranes, respectively. The FTIR spectrum of cSPFPP/SPCP 5 shows some differences compared with that of SPFPP. The characteristic peak of P=N groups at 1179.4 cm−1 is overlapped and enhanced by the bending vibration of P=N originated from SPCP [28]. The new absorption peak at 1044.9 cm−1 could be observed owing to symmetric O=S=O stretching vibration of sodium sulfonate group in SPCP [29,30]. The thermal stabilities of SPFPP and crosslinked membranes are determined by thermogravimetric analysis (TGA). As shown in Fig. 5, the crosslinked membranes exhibit three major weight loss steps from room temperature to 700 ºC. The first weight loss below 150 ºC can be assigned to the evaporation of free water in the polymer. The second weight loss from 150 to 350 ºC is attributed to the decomposition of sulfonic acid groups. The last weight loss step represents the thermal decomposition of the main chains of crosslinked membranes. 3.4. Ion-exchange capacity (IEC), water uptake and swelling ratio of membranes IEC represents the amount of exchangeable protons in membranes. It is a key parameter for evaluating membrane performance [33]. In the present study, IEC values of the membranes with different content of SPCP are listed in Table 1. As expected, the IEC values of the membranes increase from 0.90 to 1.57 mmol/g with increasing the SPCP content from 0 to 15%.

It is essential for PEMs to possess good oxidative stability. Fenton’s test is a common method which has been widely used for evaluating the oxidative stabilities of PEMs. In this study, the oxidative stabilities of the membranes are evaluated in hot Fenton’s reagent (80 ºC). Oxidative stabilities of SPFPP and crosslinked membranes are shown in Table 1. Weight retention for all the samples was above 89% after treatment in Fenton’s reagent at 80 ºC for 1 h, and most samples remained undissolved in Fenton’s reagent in 2 h of treatment at 80 ºC. It can be seen that all the prepared crosslinked membranes display better oxidative stabilities against Fenton’s reagent compared to the SPFPP membrane and other sulfonated polymers with similar IEC values, such as sulfonated polyphosphazene and sulfonated poly(ether ether ketone) [24,33]. The improved oxidative stabilities of the crosslinked membranes may be attributed to: 1) crosslinking structure makes the membranes more compact, which possess higher resistance to degradation [33]; 2) The introduction of cyclophosphazene can effectively improve the oxidative stability [27]. For all the crosslinked membranes, less than 10% of weight losses are observed after the test, indicating they possess excellent oxidative stabilities. 3.6. Proton conductivity and methanol permeability The proton conductivities of the membranes measured at different temperatures are listed in Table 2. During the measurement, the membranes are fully hydrated and all of the membranes are soaked in water for hydration. Fig. 6 shows the proton conductivities of membranes as a function of temperature ranging from 25 to 80 °C. As shown in Fig. 6, it’s found that that the proton conductivities of all the membranes increase with increasing temperature or the content of SPCP. These results are consistent with the water uptake and swelling ratio (Table 1). As shown in Table 2, the introduction of SPCP in polymer results in improvement of proton conductivity. Meanwhile, the introduction of more sulfonic acids groups originated from SPCP, which leads to more water absorption and enables protons to pass through the ionic clusters more easily, the proton conductivity therefore is enhanced [36]. The methanol permeability coefficients of crosslinked membranes gradually increase from 1.35×10−7 to 5.01×10−7 cm2 /s, which are much lower than Nafion 117 (12.1×10−7 cm2 /s). This tendency is the same as the results of water uptake. The increase of SPCP content leads to incorporation of more sulfonic acids groups, resulting in formation of more water absorption channels. Hence water and methanol could relatively easily go through the

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

479

Table 2. Proton conductivities, methanol permeability, selectivity and mechanical properties of membranes. Membrane

SPFPP cSPFPP/SPCP 5 cSPFPP/SPCP 10 cSPFPP/SPCP 15 Nafion 117

Proton conductivity (S/cm)

Methanol permeability coefficient

Selectivity

Tensile strength

Maximum elongation

25 °C

80 °C

(10− 7 cm2 /s)

(105 S· s/cm3 )

(MPa)

(%)

0.031 0.044 0.061 0.072 0.090

0.069 0.091 0.112 0.134 0.170

1.35 1.99 2.48 5.01 12.1

2.30 2.21 2.46 1.44 0.74

0.97 3.69 3.41 3.29 14.17

13.43 5.73 5.57 3.26 190.22

5 to cSPFPP/SPCP 15, the tensile strengths are in range of 3.69– 3.29 MPa. The tensile strengths of membranes decrease with the rise of SPCP content. The cSPFPP/SPCP 5 membrane has slightly better mechanical property than cSPFPP/SPCP 15 membrane due to the higher sulfonic acid content in the latter one [20]. In addition, the mechanical properties of the cSPFPP/SPCP x membranes are greatly superior to the pristine SPFPP membrane. This result means the crosslinking can help to enhance the mechanical stability effectively, which agrees well with results in literature [39]. 3.8. TEM observations

Fig. 6. Proton conductivity of SPFPP and crosslinked membranes at elevated temperature (25–80 °C).

channels, water uptakes and methanol permeability coefficients are enhanced [37]. Selectivity, which is defined as the ratio of proton conductivity to methanol permeability, is a significant value for PEM. The higher the selectivity of the membrane is, the better the membrane can perform as both a good conductor and a good separator in DMFC. The selectivity values of these crosslinked membranes are higher than that of Nafion 117 (0.74×105 S· s/cm3 ). Compared with reported SPEEK membrane with similar IEC value [38], the proton conductivity of the cSPFPP/SPCP 15 is higher than that of SPEEK in the range of 25 to 80 ºC. The methanol permeability coefficient of cSPFPP/SPCP 15 (5.01×10−7 cm2 /s) is lower than that of SPEEK (11.6×10−7 cm2 /s) at 25 ºC. In addition, the selectivity value of cSPFPP/SPCP 15 is three times higher than that of SPEEK. These results suggest that the crosslinked membrane possesses excellent comprehensive properties and have the potential to be used in DMFCs. 3.7. Mechanical property The mechanical properties of the membranes are evaluated and listed in Table 2. For the membranes prepared from cSPFPP/SPCP

The morphology of the crosslinked membranes is investigated by transmission electron microscopy (TEM). Proton conductivity of the membranes is closely related to their morphology. TEM images of membranes SPFPP and cSPFPP/SPCP 15 are shown in Fig. 7. The dark and bright areas can be assigned to the hydrophilic and hydrophobic regions, respectively. Both SPFPP and cSPFPP/SPCP 15 show clear phase separation structures. cSPFPP/SPCP 15 with higher IEC value is observed to have more continuous and larger hydrophilic domains than that of SPFPP. The larger and connected hydrophilic domain is favorable for the effective proton transportation via the membrane. 4. Conclusions New copolymer sulfonated poly(cyclophosphazene) containing clustered flexible pendant sulfonic acids is synthesized. The crosslinked cSPFPP/SPCP blend membranes show high proton conductivities and low methanol permeability coefficients. The effects of crosslinking and SPCP content on the performance of membranes have been investigated. The water uptakes, proton conductivities and methanol permeability coefficients of the crosslinked membranes increase with increasing SPCP content. The large difference in polarity between the locally and densely sulfonated units and hydrophobic units of the polymers results in the formation of well-defined phase-separated structures, which enables efficient proton conduction. The cSPFPP/SPCP membranes show not only excellent thermal properties but also improved oxidative stabilities and mechanical properties. Moreover, the cSPFPP/SPCP membranes

Fig. 7. TEM images of (a) SPFPP and (b) cSPFPP/SPCP 15.

480

Y. Dong et al. / Journal of Energy Chemistry 25 (2016) 472–480

exhibit higher selectivity values compared with Nafion 117. The results indicate that these membranes were promising candidates for fuel cell applications. References [1] S.K. Kamarudin, F. Achmad, W.R.W. Daud, Int. J. Hydrog. Energy 34 (2009) 6902. [2] Y. Paik, S.S. Kim, O.H. Han, Angew. Chem, Int. Ed 47 (2008) 94. [3] J.M. Song, D.W. Shin, J.Y. Sohn, Y.C. Nho, Y.M. Lee, J.H. Shin, J. Membr. Sci 430 (2013) 87. [4] K.A. Mauritz, R.B. Moore, Chem. Rev. 104 (2004) 4535. [5] C. Coutanceau, R.K. Koffi, J.M. Leger, K. Marestin, R. Mercier, C. Nayoze, P. Capron, J. Power Sour. 160 (2006) 334. [6] J.J. Sumner, S.E. Creager, J.J. Ma, J. Electrochem. Soc. 145 (1998) 107. [7] D.R Paul, L.M. Robeson, Polymer 49 (2008) 3187. [8] H. Ghassemi, J.E. McGrath, Polymer 45 (2004) 5847. [9] M. Schuster, K.D. Kreuer, H.T. Anderson, J. Maier, Macromolecules 40 (2007) 598. [10] K. Nakabayashi, T. Higashihara, M. Ueda, Macromolecules 43 (2010) 5756. [11] S.M.J. Zaidi, S.D. Mikhailenko, G.P. Robertson, M.D. Guiver, S. Kaliaguine, J. Membr. Sci 173 (20 0 0) 17. [12] K. Matsumoto, T. Higashihara, M. Ueda, Macromolecules 41 (2008) 7560. [13] C. Genies, R. Mercier, B. Sillion, N. Cornet, G. Gebel, M. Pineri, Polymer 42 (2001) 359. [14] M.L. He, H.L. Xu, Y. Dong, Chin. J. Polym. Sci 32 (2014) 151. [15] R. Wycisk, P.N. Pintauro, J. Membr. Sci 119 (1996) 155. [16] Q. Guo, P.N. Pintauro, H. Tang, S.O Connor, J. Membr. Sci. 154 (1999) 175. [17] F. Meier, S. Denz, A. Weller, Fuel Cells 3 (2003) 161. [18] M.L.D. Vona, E. Sgreccia, S. Licoccia, M. Khadhraoui, R. Denoyel, P. Knauth, Chem. Mater 20 (2008) 4327.

[19] H. Luo, G. Vaivars, M. Mathe, J. Power Sources 195 (2010) 5197. [20] S.L. Zhong, X.J. Cui, H.L. Cai, T.Z. Fu, C.J. Zhao, H. Na, J. Power Sour. 164 (2007) 65. [21] Y. Zhang, Y. Wan, G. Zhang, K. Shao, C. Zhao, H. Li, H. Na, J. Membr. Sc 348 (2010) 353. [22] H. Meng, J. Power Sour. 164 (2007) 688. [23] X. Zhang, L. Sheng, T. Higashihara, M. Ueda, Polymer Chem. 4 (2013) 1235. [24] X.Y. Zhou, J. Weston, E. Chalkova, M.A. Hofmann, C.M. Ambler, H.R. Allcock, S.N. Lvov, Electrochim. Acta 48 (2003) 2173. [25] F. Fu, H. Xu, M. He, Y. Dong, P. Liu, X. Hao, C.J. Zhu, Chin. J. Polym. Sci 32 (2014) 996. [26] K.E. Koenig, G.M. Lein, P. Stuckler, T. Kaneda, D.J. Cram, J. Am. Chem. Soc 101 (1979) 3553. [27] S.Y. cho, H.R. Allcock, Chem. Mater 19 (2007) 6338. [28] Y. Bai, X.D. Wang, D.Z. Wu, Ind. Eng. Chem. Res 51 (2012) 15064. [29] M.B. Cui, Z.H. Zhang, T. Yuan, H. Yang, L. Wu, J. Membr. Sci 465 (2014) 100. [30] H.L. Cheng, J.M. Xu, L. Ma, L.S. Xu, B.J. Liu, Z. Wang, H.X. Zhang, J. Power Sour. 260 (2014) 307. [31] L. Sheng, T. Higashihara, R. Maeda, T. Hayakawa, M. Ueda, J. Polym. Sci. Part A: Polym. Chem 51 (2013) 2216. [32] F.Y. Fu, H.L. Xu, Y. Dong, M.L. He, Z. Zhang, T. Luo, Y. Zhang, X. Hao, C.J. Zhu, J. Membr. Sc 119 (2015) 489. [33] M.Y. Li, G. Zhang, S. Xu, C.J. Zhao, M.M. Han, L.Y. Zhang, H. Jiang, Z.G. Liu, H. Na, J. Power Sour. 255 (2014) 101. [34] M.A. Hickner, H. Ghassemi, Y.S. Kim, B.R. Einsla, J.E. McGrath, Chem. Rev. 104 (2004) 4587. [35] J.H. Pang, H. Zhang, X.F. Li, Z.H. Jiang, Macromolecules 40 (2007) 9435. [36] R.K. Nagarale, G.S. Gohil, V.K. Shahi, J. Membr. Sci 280 (2006) 389. [37] S. Zhong, X. Cui, H. Cai, T. Fu, K. Shao, H. Na, J. Power Sour. 168 (2007) 154. [38] N. Zhang, G. Zhang, D. Xu, C. Zhao, W. Ma, Li H, Y. Zhang, S. Xu, H. Jiang, H. Sun, H. Na, Int. J. Hydrog. Energy 36 (2011) 11025. [39] S. Subianto, M. Pica, M. Casciola, P. Cojocaru, L. Merlo, G. Hards, D.J. Jones, J. Power Sour. 233 (2013) 216.