Abstract

Disodium-3,3′-disulfonate-4,4′-dichlorodiphenylsulfone (SDCDPS) and 5,5′-bis[2-(4-hydroxyphenyl)benzimidazole] (HPBI) monomers were synthesized. Binding these monomers via nucleophilic aromatic polycondensation reaction, a sulfonated poly(arylene ether benzimidazole) copolymer was synthesized. Structures of monomers and copolymer were confirmed by proton nuclear magnetic resonance spectroscopy (1H NMR) and Fourier transform infrared (FTIR) spectroscopy analyses. Proton exchange membrane was prepared by dissolving copolymer in dimethylacetamide (DMAc) and casting onto a glass plate. Copolymer membrane was doped with sulfuric acid to ensure proton exchange character. Single cell performance of the copolymer membrane was tested in a polymer electrolyte membrane fuel cell test station. The highest power density of the membrane was measured as 23.7 mW cm−2 at 80°C. Thermogravimetric analysis (TGA) showed that as the degree of disulfonation is increased thermal stability of the copolymer is increased.

1. Introduction

Environmental challenges of fossil fuel use, effects of harmful emissions on human health, and dependence of industrial nations on oil that leads to oil crises have induced the development of fuel cell technologies in past 25 years [1, 2]. Fuel cells are considered to be the solution of environmentally friendly and highly efficient electrical energy production of the future. Among fuel cell types, polymer electrolyte membrane fuel cells (PEMFCs) attract most attention due to their high power density, high energy transformation efficiency, and wide range of applications in stationary and portable devices [3, 4]. Perfluorosulfonic acid based polymer membranes such as Nafion® are considered as state-of-the-art membranes and used frequently in PEMFCs owing to their outstanding chemical and physical stabilities and high proton conductivity at moderate operating temperatures. However, studies on development of alternative polymer electrolyte membranes continue because of the drawbacks of Nafion membranes like decrease in proton conductivity at temperatures above 80°C due to dehydration, humidification requirement, and high prices [57].

Studies about polymer based sulfonated proton exchange membrane materials such as poly(arylene ether sulfone)s [811], poly(ether ether ketone)s [1214], poly(arylene thioether) [1517], poly(phenylene)s [1820], polyimides [2124], and other types of polymers have largely taken place in the scope of polymer electrolyte membrane researches so far.

Polybenzimidazole (PBI) membranes doped with strong acids predominate over other proton exchange membranes due to their stable proton conductivity at temperatures higher than 100°C. However, synthesis of high molecular weight PBI polymer is difficult and cost is quite high. Besides, dissolution of PBI in common organic solvents is difficult due to its rigid molecular structure. On the other hand, synthesis of poly(arylene ether sulfone)s is easier and cost is lower. However, they have low mechanical stability due to water swelling and dissolution of the poly(arylene ether sulfone) membranes in water that limits their practical applications [6, 7]. In this study, a sulfonated poly(arylene ether benzimidazole) copolymer was synthesized [2530] and a copolymer membrane was prepared by solution casting method. The membrane was treated with sulfuric acid to strengthen the proton conductivity character. Finally single cell performance of the copolymer membrane was tested in a PEMFC test station.

2. Experimental

2.1. Materials

4,4′-Dichlorodiphenyl sulfone (DCDPS) and 4-hydroxybenzoic acid phenyl ester were purchased from ABCR and vacuum dried at 50°C for 24 h before use. Fuming sulfuric acid (65% SO3, extra pure), phenylsulfone, and 1-methyl-2-pyrrolidone (+99.5%) were used as received from Merck. Sodium chloride, 3,3′-diaminobenzidine (DAB), and DMAc were purchased from Sigma Aldrich. DAB was vacuum dried at 50°C for 24 h before use. Sodium hydroxide, toluene, and isopropanol were used as received from Riedel de Haen. Potassium carbonate (Alfa AESAR) was vacuum dried at 50°C for 24 h prior to use. Sulfuric acid (95–97%) was used as received from Fluka.

2.2. Synthesis of Monomers

Synthesis of HPBI has been previously reported by several researchers [29, 30]. DAB (0.467 mole), 4-hydroxybenzoic acid phenyl ester (0.934 mole), phenylsulfone (500 g), and toluene (150 mL) were added to a jacketed four-necked cylindrical reactor equipped with a Dean-Stark trap, a condenser, a mechanical stirrer, and a nitrogen inlet-outlet. Reaction temperature was controlled by oil bath circulator. Reactor was isolated to prevent heat loss. Dean and Stark trap was filled with toluene to remove water azeotropically. Temperature was raised to 150°C and the solution was allowed to reflux for 6 hours. Toluene was removed and the solution was heated to 280°C for 2 h. Generated water and phenol were removed from Dean and Stark trap and the monomer precipitated. For further removal of phenol, outlet gas stream was connected to two serial gas washing bottles and vacuum was applied for 2 hours. Solution was left to cool and excess ethanol was added at 150°C. Product was left at room temperature overnight and filtered to separate the phenylsulfone that crystallized out of solution. Synthesized tan monomer was washed with acetone and dried at 120°C for 24 hours.

3,3′-Disulfonated-4,4′-dichlorodiphenyl sulfone (SDCDPS) monomer [3033] synthesized in a four-necked round bottom flask equipped with a thermometer, a condenser, a mechanical stirrer, and a nitrogen inlet-outlet. DCDPS (99 mmol) was dissolved in fuming sulfuric acid (65% SO3). Temperature was raised to 110°C and the mixture was stirred for 6 hours to produce a homogeneous solution. Then solution was cooled to room temperature and sodium chloride was added. The dark brown color of the solution turned into white and the product was neutralized to pH 6-7 with sodium hydroxide and excess sodium chloride was added to precipitate the monomer as its sodium form. Synthesized SDCDPS was filtered, recrystallized from isopropanol and water mixture for further purification. Resulted white monomer dried at 120°C for 24 hours.

2.3. Disulfonated Poly(Arylene Ether Benzimidazole) (SPAEB) Copolymers Synthesis

Einsla has previously reported the synthesis of disulfonated poly(arylene ether benzimidazole) copolymer [30]. SDCPS (7 mmol) and equimolar DCDPS and HPBI (14 mmol) were introduced into a 250 mL four-necked flask equipped with a thermometer, mechanical stirrer, a Dean-Stark trap, a condenser, and a nitrogen inlet-outlet. Potassium carbonate (16 mmol) and 75 mL of NMP were introduced to the reaction flask. As an azeotroping agent 40 mL of toluene was introduced to the flask and the Dean-Stark trap was filled with toluene. Temperature was raised to 140°C for dehydration of the system for 4 hours and Dean and Stark trap was emptied. Temperature was raised to 170°C and reaction was carried out for 20 hours. Generated toluene during reaction was removed via Dean and Stark trap. The dark brown viscous solution poured into deionized water and the precipitated copolymer was filtered. The yellowish-brown fibrous copolymer was dried at 120°C for 24 h and grounded.

2.4. Postsulfonation of SPAEB Copolymers

A postsulfonation method was used for further sulfonation of synthesized SPAEB copolymers to analyze the effect of degree of sulfonation to thermal stability and proton conductivity. Postsulfonation was carried out by dissolving 5 g of the copolymer in 100 mL fuming sulfuric acid (65%) and stirred for 10 hours at 30°C. Resulting product was denoted as SPAEB_oleum (SO3%65). After the reaction, the reaction mixture was poured into a large volume water to precipitate the product. The sulfonated polymer material was recovered and washed with water until the wash was pH neutral and it was then dried for 48 h at 100°C.

2.5. Membrane Preparation and Acid Doping

Synthesized copolymer (0.75 g) was dissolved in 31.88 mL of DMAc (2.5%, w w−1) in a 250 mL, two-necked flask equipped with a mechanical stirrer, and a condenser. The flask was inserted in an ultrasonic water bath. Temperature was heated to 80°C for 4 hours. After filtration, solution was cast onto petri dishes and dried. Resulting transparent membranes were immersed in boiling 0.5 M H2SO4 for 2 hours and dried to remove absorbed water. The same membrane preparation procedure was performed for SPAEB_oleum (SO3%65) too.

2.6. Characterization

The 1H NMR spectra of the monomers and copolymer were recorded on a 500 MHz Bruker AVANCE II NMR spectrometer. Grounded monomers and copolymer were dissolved in deuterated dimethylsulfoxide (DMSO-d6) and chemical shifts were measured against tetramethylsilane (TMS) as an internal standard. We used the integral area belonging to the protons adjacent to the sulfonate groups and the integral area of the protons next to the carbon of the imidazole moiety degrees of disulfonation obtained from 1H NMR:

Fourier transform infrared (FTIR) spectra were recorded using a Perkin Elmer Frontier spectrometer. All the samples were dried at 105°C for 24 h prior to analyses.

Thermogravimetric curves are largely affected by experimental conditions, especially by the heating rate. So the degradation experiments of samples were carried out at the same scanning rate, 10°C/min, because it is a medium scanning rate among those usually employed for thermal degradations [34, 35]. Thermogravimetric analysis (TGA) was carried out in dynamic heating conditions under flowing nitrogen (100 mL−1) in the temperature range 35–600°C, using a Netzsch STA 449 F3 thermogravimetric analyzer. Samples of about 9.30 × 10−3 g for SPAEB and 9.59 × 10−3 g for SPAEB_oleum (SO3%65) were used for degradation experiments. Their masses as a function of temperature were monitored and the experimental data were used to plot the percentage of undegraded copolymer as a function of temperature.

A Cannon Ubbelohde dilution viscometer was used to measure inherent viscosity of the copolymer in DMAc at a concentration of 0.5 g dL−1 at 30°C.

2.7. Proton Conductivity Measurements

The membrane samples were prepared as rectangular sheets and placed in a Teflon conductivity cell (Bekktech 112) which is connected to four Pt electrodes and the distance between the two inner electrodes was used to measure the potential difference trough the sample. Proton conductivities of the membranes were measured for planar direction sending hydrogen (relative humidity (RH) = 0%) into a Scribner 850 fuel cell test system connected to the proton conductivity cell at 80°C. A Solartron 1287 electrochemical interface was used to measure voltage and current values. Proton conductivity of a sample of commercial Nafion 212 membrane was measured at 80°C with 100% RH to see the reliability of the results.

2.8. Single Cell Performance Test

The standard fuel cell test station was described elsewhere [36]. Copolymer membrane was sandwiched between the catalyst coated electrodes to produce the membrane electrode assembly. The Pt loadings of cathode and anode were approximately 0.5 mg cm−2. SGL-10BC was used as gas diffusion layer and the active area of the single cell was 25 cm2. The current and power densities of the fuel cell were recorded. The working temperature of the fuel cell was kept at 80°C. Hydrogen and oxygen were both fed to the fuel cell at a rate of 80 sccm without external humidification at ambient pressure.

3. Results and Discussion

3.1. Structural Analysis of SDCDPS

The FTIR spectrum of SDCDPS is shown in Figure 1. C-H stretching of aromatic ring is assigned at 3095 cm−1. The absorption peak at 1638 cm−1 represents the C=C stretching vibrations bands. Peaks at 813–824 cm−1 are attributed to the C-Cl stretching bands. In-plane bending of aliphatic C-H is assigned at 1288 cm−1. Peak observed at 594 cm−1 is due to C-S aromatic ring stretching. The absorption peaks at 1024 and 1084 cm−1 indicates the S=O stretching vibrations of sulfonic acid groups. This result confirms the sulfonation of DCDPS.

1HNMR spectrum of SDCDPS is shown in Figure 2. The existence of the resonance at 8.34 ppm represents the protons adjacent to the sulfonate groups. This result confirms that the sulfonation of DCDPS was achieved.

3.2. Structural Analysis of HPBI

In the FTIR spectrum of HPBI represented in Figure 3, the large peaks observed at 2500–3200 cm−1 are assigned to R2NH stretching vibration and Ar-OH vibration bands. Peaks at 1593 cm−1 and 1620 cm−1 show C=C stretching and C=N stretching-vibration bands (characteristic band of imidazole groups).

Figure 4 shows the 1H NMR spectrum of HPBI. The peak observed at 13.3 ppm is assigned to N-H groups. Peak at 10 ppm confirms the existence of the protons of Ar-OH groups. This result shows that HPBI monomer synthesis was performed successfully.

3.3. Characterization of Copolymer

Inherent viscosity of the copolymer was measured as 0.92 dL g−1 in DMAc at a concentration of 0.5 g dL−1 at 30°C. In FTIR spectrum of sulfonated poly(arylene ether benzimidazole) copolymer shown in Figure 5, peaks at 3000–3500 cm−1 are attributed to the N-H vibration bands of amide groups. Peaks observed at 1500–1600 cm−1 are due to the aromatic C=C and C=N stretching bands and they arise from in-plane bending of N-H and conjugation vibration between benzene and imidazole rings. The peak at 1149 cm−1 indicates aromatic sulfone groups. Peaks at 1027 cm−1 and 1243 cm−1 indicate stretching vibration of O=S=O sulfonic acid groups.

Figure 6 shows the 1H NMR spectrum of sulfonated poly(arylene ether benzimidazole) (SPAEB) copolymer. Proton resonances at 7.00, 7.85, and 8.33 ppm confirm that sulfone groups are involved in the structure.

Figure 7 shows 1H NMR spectrum of SPAEB_oleum (SO3%65). 1H NMR spectra were also used to evaluate the degrees of disulfonation of SPAEB and SPAEB_oleum (SO3%65) using a literature method [30] and calculated aswhere is the peak which is attributed to the protons adjacent to the sulfonate groups and is the peak assigned to protons next to the carbon of the imidazole group located at both the sulfonated and nonsulfonated regions. The degrees of disulfonation for SPAEB and SPAEB_oleum (SO3%65) were calculated as 31% and 52%, respectively.

3.4. FTIR Analysis and Thermal Behavior of SPAEB Membrane

FTIR spectrum of SPAEB copolymer membrane is shown in Figure 8. Spectrum shows benzimidazole adsorption bands at 1629 cm−1, 1584 cm−1, and 1465 cm−1. Peaks at 1025 cm−1 and 1097 cm−1 are assigned to stretching of the sulfonate groups.

TGA curves of the SPAEB and SPAEB_oleum (SO3%65) copolymer membranes are shown in Figure 9. The first decomposition step around 300°C is attributed to the decomposition of the sulfone groups according to the literature [37, 38]. Second thermal degradation steps occurring in temperature ranges 450–500°C should be due to polymer main chain. TGA curves reveal that SPAEB copolymer degrades at about 450°C, whereas the postsulfonated SPAEB_oleum (SO3%65) copolymer is thermally stable up to ~475°C. This thermal degradation behavior can be explained with higher percentage of double-bond character due to the increasing number of sulfone groups incorporated in copolymer chain in agreement with the literature [39, 40]. This result also shows that the presence of sulfonic groups involving strong intra- and intermolecular hydrogen bonds increases thermal stability as in the literature [41].

3.5. Proton Conductivity

The resistance, (Ω), of the membranes was calculated as the slope of voltage versus current graphs. Voltage versus current graphs plotted for 30 times to reach a stable resistance value for each membrane. Figure 10 shows the voltage versus current graphs of SPAEB_oleum (SO3%65), SPAEB, and commercial Nafion 212 membranes. The proton conductivity, (S/cm), values were calculated from the equationwhere (0.425 cm) is the distance between the voltage sense electrodes (two inner electrodes), is the membrane thickness which is 0.0051 cm, 0.0144 cm, and 0.0784 cm for Nafion 212, SPAEB_oleum (SO3%65) and SPAEB membranes, respectively, and is the sample width.

Proton conductivities of the membranes at 80°C are calculated as 0.203 S/cm (RH = 100%), 0.0059 S/cm (RH = 0%), and 0.0012 S/cm (RH = 0%) for Nafion 212, SPAEB, and SPAEB_oleum (SO3%65), respectively. This result suggests that, as more sulfonic acid sites are incorporated into the membrane structure, the proton conductivity is decreased. Presence of sulfonic acid groups in the membrane structure decreases proton conductivity interacting with imidazole sites.

3.6. Fuel Cell Performance Test

Figure 11 shows the polarization and power density curves of SPAEB copolymer membrane. Average membrane thickness was measured as 0.0094 cm. The highest power density of the membrane was measured as 23.7 mW cm−2 at 80°C without humidification.

4. Conclusions

SPAEB copolymer was synthesized successfully via nucleophilic aromatic polycondensation of the two monomers synthesized (SDCDPS and HPBI) and the commercial monomer DCDPS. Structures of monomers and the copolymer were confirmed by 1H NMR and FTIR analysis. TGA analyses showed that as the degree of disulfonation is increased thermal stability of the copolymer is increased due to double-bonds and hydrogen bonds of sulfone groups. According to results of proton conductivity tests, as the degree of disulfonation increases, proton conductivity of the membranes decreases. Sulfonic acid groups prevent proton conduction from interacting with imidazole groups. Single cell performance test result shows that SPAEB membranes can be used as a polymer electrolyte membrane.

Since cost of PBI synthesis is quite high, SPAEB copolymer membrane, which includes a cheaper component poly(arylene ether), reduces the cost of the synthesis. There can be further researches to enhance the fuel cell performance of SPAEB membranes.

Competing Interests

The authors declare that they have no competing interests.

Acknowledgments

This research was supported by Anadolu University Scientific Research Projects Committee (Project no. 1206F102).