Highly proton-conducting non-perfluorinated hybrid electrolyte/non-platinum catalyst for H₂/O₂ fuel cells

Abstract

A class of hybrid proton-conducting membranes consisting of a polymer blend with heteropolyacids, poly(vinyl alcohol) (PVA), poly(vinylpyrrolidone) (PVP), silicotungstic acid (SiWA) and orthoethoxysilicate, was prepared. A maximum conductivity value of 0.4 S cm⁻¹ was obtained at 100 °C with excellent thermal and mechanical stabilities. This is the cutting edge that such a high proton conductivity value was achieved for a PVA/PVP/SiWA hybrid composite membrane. A maximum current density of 1200 mA cm⁻² for the membrane electrode assembly, consisting of a carbon black based catalyst and the hybrid membrane, was obtained at 80 °C and 100% relative humidity. To the best of our knowledge, no reports have been published using the same non-Pt catalyst and hybrid membrane electrolyte providing high current density at low temperatures. In this communication, we highlight two systems that present large proton conductivity and current density at relatively low temperatures by the combination of a new class of polymer blend hybrid membranes and a non-Pt catalyst.

---

Proton exchange membrane fuel cells (PEMFCs) using proton conductive materials operating at low temperatures (50–80 °C) are highly attractive for applications such as electric vehicles and residential and portable devices.¹ Fuel cells, which are devices for efficiently transforming chemical energy directly into electricity, are regarded as promising future clean power sources. At present, no known material or storage devices exist in the fuel cell industry that could satisfy all the requirements to enable high-volume automotive application; however, materials do exist that would satisfy requirements for short-term non-vehicular PEM fuel cell applications.

We must pay special attention to membrane technology to develop a class of advanced low cost and highly proton-conducting hybrid materials. The development of material technology has led to novel organic–inorganic hybrid materials for obtaining electrolytes to be used in fuel cell applications. Several researchers have reported on various kinds of proton-conducting composites as alternatives to the Nafion® membrane. Such perfluorosulfonic acid polymers have many attractive properties, including good mechanical strength, good chemical stability, and high proton conductivity, and are widely used in commercial applications.^2,3 However, perfluorosulfonic acid membranes display several drawbacks such as complex interaction with water and difficulty in controlling it, CO poisoning of the Pt catalyst at the anode, and high cost.⁴ To replace perfluorosulfonic acid membranes, like Nafion®, one should introduce novel non-fluorinated materials with similar properties. The main idea is to develop hybrid materials based on the design of hybrid polymers with special emphasis on structural hybrid materials.

On the other hand, presently, the state-of-the-art PEMFCs are operating at low temperatures (less than 100 °C). CO poisoning of anode precious metal catalysts is a major barrier for the commercialization of PEMFCs utilizing hydrocarbon feedstock.⁵ Considering this to solve the problem, there is interest in introducing non-precious metal catalysts for fuel cells operating at low temperatures. One of the main challenges facing the development of PEMFCs is the preparation of catalyst layers for the cathodic reaction in PEMFCs. This challenge can be approached in several ways. One way is to develop alternative catalysts that provide a larger electroactivity at a lower cost than the present catalysts. Our present aim is to completely replace the Pt-based catalyst by a non-Pt catalyst and to increase the performance and properties smoothly for H₂/O₂ PEMFCs operating at low temperatures.

Poly(vinyl alcohol) (PVA) is one of the foremost polymers used in membrane technology because of its good mechanical properties, chemical stability, low cost, film forming ability, and highly hydrophilic behavior.^6,7 PVA is a cheap polymer widely used for the preparation of hybrid membranes for fuel cells. The presence of functional –OH groups in the PVA chains allows the reaction with aldehydes to form chemical cross-links between macromolecular chains, resulting in an extended thermal and mechanical stability of the polymer matrix.⁸ Cross-linking of the hydroxyl functional groups available in PVA can effectively control the water uptake and hence the degree of swelling of the membranes. Optimizing the cross-linking density of the membranes can provide membranes with good mechanical properties.

Poly(vinylpyrrolidone) (PVP) is a highly soluble polymer in polar solvents, such as alcohols. The sol–gel reaction often uses an alcohol as a solvent and produces an alcohol as a reaction product; therefore it is preferable to avoid phase separation in the reaction. It is necessary to have a homogeneous solution to produce a uniform film with a low scattering loss.^9,10 As a water-soluble polymer, PVP has a beneficial effect on protection, viscosity, absorbency, and solubilization, with its most significant features being excellent solubility and biocompatibility. Considering the film forming properties of PVP, it was blended with pectin to improve its mechanical properties. Cassu and Felisberti¹¹ attributed the compatibility of PVA and PVP to hydrogen bonding that may take place between the proton-accepting carbonyl moiety in pyrrolidone rings and the hydroxyl side groups of PVA. Hydrogen bonding is also responsible for the solubility of both PVA and PVP in water. By blending PVP with PVA, the properties of the membrane would be improved. Recently, we reported that PVA-based composite membranes demonstrated a maximum proton conductivity of 1.7 × 10⁻² S cm⁻¹ at 140 °C and a relative humidity (RH) of 50%.¹²

Heteropoly acids (HPAs) are used in hybrid composite membranes for fuel cells because of their good proton conductivity at room temperature (0.02–0.1 S cm⁻¹). HPAs as aqueous solutions have already been used in low-temperature fuel cells with excellent performance, but the risk of their continuous leakage during cell operation is not negligible. To overcome this problem and to increase the lifetime of the cell, cross-linking will also solve the problem of acid leaching from the membranes under hydrated conditions by effectively immobilizing the acid within the polymer matrix through formation of the cross-linked networks. Therefore, significant economic advantage provides a motivation to develop and commercialize PEMFCs operating in these conditions. Silicotungstic acid (SiWA) has shown interesting properties as a promoter in the electrochemical oxidation of methanol.¹³ The solid proton conductors (superacids) under evaluation include phosphotungstic acid, SiWA, zirconium hydrogen phosphate, zeolite, and silica. These materials are brittle inorganic substances and therefore are generally incorporated into a composite structure containing flexible polymeric ionomer phases. However, although to a lesser extent than Nafion®, they also lose water at high temperatures with reduced proton conductivity.

The alternative materials for membranes can be organic–inorganic hybrids; they are a remarkable family of isotropic, flexible, and amorphous materials.¹⁴ In addition to the blending properties of the hybrids, they can also be easily synthesized through processing routes with low cost and reduced environmental impact. One approach under active consideration is the use of alternative hybrid membrane composites containing organic non-fluorinated polymers and inorganic oxide gels. Alternative membrane chemistries are being investigated to enable sufficient proton conductivity at lower relative humidity. To our knowledge, no previous study has investigated the conductivity properties of PVA/PVP/SiWA based hybrid membranes and performances of single cells made by using such membranes.

A class of PVA/PVP/SiWA hybrid membranes were prepared by the following steps: PVA and PVP (80/20 wt%) were separately dissolved in distilled water with stirring for 6 h, at 80 °C for PVA (Mw: 100 000 g mol⁻¹, Nacalai Tesque, Japan) and at 50 °C for PVP (Mw: 100 000 g mol⁻¹, Nacalai Tesque, Japan) to obtain homogenous solutions. After cooling to room temperature, both solutions were mixed under stirring. TEOS (Si(OC₂H₅)₄, 99.9%, Colcote) was hydrolyzed in an acid solution (0.15 N–HCl aq) and ethanol (1 : 2 ratio) and stirred for 30 min. A 2 g quantity of SiWA (silicotungstic acid, Chameleon reagent) was dissolved in the desired amount of distilled water and added to the above solution under similar conditions under stirring for 24 h. Subsequently, the obtained homogeneous solution was poured into Petri dishes and allowed to dry at room temperature for few weeks. The resultant membranes were heat-treated at 150 °C for 6 h. The thickness of the obtained hybrid membrane was 0.85 mm (Fig. 1).

Fig. 1 A photograph of the PVA/PVP/SiWA hybrid membrane.

Both cathode and anode catalyst layers consisted of carbon black (Vulcan XC-72, CABOT corporation, MA, USA), Nafion® (perfluorinated ion-exchange resin, 5 wt% solution in lower aliphatic alcohols/H₂O mix, Aldrich), polytetrafluoroethylene (PTFE, Aldrich) and 2-propanol (Chameleon reagent). Catalyst inks were prepared using the ultrasonic method.^15,16 The prepared ink was sprayed on the carbon sheets, and then they were dried by heating stepwise to 100 °C for 5 h. Cathode and anode catalyst layers were prepared from these inks and then membrane electrode assemblies (MEAs) were prepared by hot-pressing (pressure of 1 MPa for 2 min at 120 °C). The catalyst (anode and cathode) coated membranes had an electrode geometrical surface area of 1 cm². The thickness of the MEAs was 0.8 mm.

The electrodes (anode and cathode) were attached onto both sides of the membrane, with a carbon cloth on each side, by hot-pressing to form MEAs with three phase boundaries. The MEAs with active geometric areas of 1 cm² were sandwiched between two gas diffusion layers and assembled into a single cell possessing straight flow-through channels. The uniformity of cell compression was verified using a torque wrench (dial indicating Torque Wrench, Japan). The single cell was evaluated in a fuel cell test station (RFC-1500, Japan).

The proton conductivities of the polymer hybrid membranes were measured using the ac impedance spectroscopy technique with a Solartron 1260 frequency response analyzer in the range 1 Hz to 1 MHz at a voltage of 100 mV, employing a transverse two-electrode configuration. The proton conductivities (σ) of the samples were calculated from the impedance data, using the relationship σ = d/Rtl, where d is the distance between the electrodes, t and l are the thickness and width of the films, respectively, and R is derived from the low intersect of the high frequency semi-circle on a complex impedance plane with the Re(Z) axis, where Re refers to ‘Real’ in the complex impedance plane.

The polarization measurements were carried out using a gas supply system RFC-1500 model fuel cell test station. The gases were allowed to pass through stainless steel humidifiers before entering the fuel cell inlets, and the flow rates (10–50 mL min⁻¹) were controlled using mass flow controllers. Pure hydrogen and oxygen gases were supplied, respectively, to the anode and cathode sides of the single cell. The performance of the single cell was evaluated over the temperature range of 50–80 °C at 1 atm pressure (anode/cathode).

Fig. 2 displays the proton conductivity of the PVA/PVP/SiWA hybrid membrane, measured at various temperatures (30–100 °C) and a relative humidity of 50%. The proton conduction changed as a function of temperature. The maximum conductivity of 0.4 S cm⁻¹ was obtained at 100 °C for the PVA/PVP/SiWA hybrid membrane with a composition of 80/20 wt%/2 g. As a result, the proton conductivity increased by the same order of magnitude (0.1–0.4 S cm⁻¹) in the temperature range from 30 to 100 °C with 50% RH, clearly indicating that the Grotthuss mechanism^17–19 of proton hopping via water molecules dominated the proton conduction. The molecular water in the composites was also presumed to provide large proton mobility, contributing to an increase and stabilization of the proton conductivity. Reports indicated that an increase in water content can lead to greater proton conductivity at a low PVP content.²⁰ A similar finding by Lu et al.²¹ shows that PVA–PVP interpenetrating polymer network membranes achieve an exceptional selectivity in sorption of water over methanol in terms of swelling ratio. The present conductivity result was comparable to the reported results for PVA/PVP and PVA/SiWA composite membranes,^22–24. Qiao et al. studied the highly proton-conducting membrane poly(vinylpyrrolidone)(PVP)/(PVA–PAMPS), and found it reached a maximum of 0.088 S cm⁻¹ by varying the mass ratio of PVA/PVP.²² Moreover, the highest room temperature conductivity of 5.02 × 10⁻³ S cm⁻¹ among the PVACO-silicotungstic acid based composites was observed for the membrane.²³ Similarly, the PVA/SiWA/MPTS/H₃PO₄ membrane demonstrated a maximum conductivity of 8.5 × 10⁻³ S cm⁻¹ under identical conditions.²⁴ Also compared with the conductivity of the PVA/PVP/SiWA hybrid membrane, it proved to be greater than that of commercial Nafion®,²⁵ and the composite membrane containing SiWA exhibited a conductivity value of 10⁻² S cm⁻¹ in the temperature range of 333–353 K,²⁶ and a maximum conductivity of 4.71 × 10⁻³ S cm⁻¹ was obtained for the membrane with 30 wt% of SiWA in the PVA matrix.²⁷ This conductivity value was compared with the conductivities reached of 0.257 and 0.32 S cm⁻¹ at 298 K for the 60 wt% phosphotungstic acid and silicotungstic acid solutions, respectively.²⁸

Fig. 2 The proton conductivity of the PVA/PVP/SiWA hybrid membrane as a function of temperature.

The proton transfer phenomenon follows two principal mechanisms where the proton remains shielded by electron density along its entire diffusion path. Under the present conditions of low humidity (50%), water molecules were quickly chemisorbed in the defect sites located on the surface, presenting a high local charge density and a strong electrostatic field upon exposure of the membrane to the atmosphere. Moreover, the amount of these molecules, once absorbed, did not change further by exposure to the humidity. The large proton conductivity can be achieved by raising the temperature. This was mainly attributed to the membrane's ability to retain water, thus complying with the Grotthuss mechanism.^17–19 These active sites promoted water dissociation to provide protons as charge carriers, 2H₂O ↔ H₃O⁺ + OH⁻, for the hopping transporting mechanism known as the Grotthuss chain reaction.^29,30 Charge transport occurs when the hydronium (H₃O⁺) releases a proton to a neighboring water molecule, which accepts it while releasing another proton. The protons are transferred from one vehicle to the other one by hydrogen bonds (proton-hopping mechanism).^31,32 Simultaneous reorganization of the proton environment,³³ consisting of a reorientation of individual species or even more extended ensembles, leads to the formation of an uninterrupted path for proton migration. The most trivial case of proton migration, known as the vehicle mechanism,^34,35 requires the translational dynamics of a larger species. In this mechanism, the proton diffuses through the medium together with a “vehicle carrier” (for example, with H₂O as H₃O⁺).

A highly proton-conducting membrane electrolyte and non-Pt electrode was introduced in the MEA, and the cell performance at low temperatures was studied. The relationship between voltage and current was displayed in the polarization curves and the recorded current density values at various cell temperatures (50, 60, 70, and 80 °C) are shown in Fig. 3. The trend is an increase in current density when increasing the cell temperature with a relative humidity of 100% at constant pressure. The overall measurements thus demonstrated a maximum current density of 1200 mA cm⁻² at 100% RH with a cell temperature of 80 °C. In this case, the main part of the MEAs consisting of the electrolyte was synthesized as a highly proton-conducting membrane. According to our calculations, the cell performances were superior at high humidity conditions. A similar observation was also made for some of the newly synthesized hybrid membranes.^36,37 Moreover, the results were compared with our previous reports,^24,38 confirming that this hybrid membrane and non-Pt catalyst combination is better than a PVA-based hybrid membrane electrolyte.

Fig. 3 The current–voltage relationship of the MEA consisting of a PVA/PVP/SiWA hybrid membrane and non-Pt catalyst operating at various temperatures and constant relative humidity of 100%.

The aim of this study was to investigate a class of proton-conducting non-fluorinated hybrid membranes with a non-Pt catalyst for MEAs, suitable for H₂/O₂ fuel cells operating at low temperatures. Experiments were conducted on the development of low temperature fuel cells using the hybrid membrane with a high proton conductivity of 0.4 S cm⁻¹ at 100 °C and current density of 1200 mA cm⁻² at 80 °C and 100% relative humidity. However, another approach would be improving the fabrication process of the hybrid membrane and catalyst layers for MEAs performing in low temperature H₂/O₂ fuel cells. Studies in our laboratory have shown promising results on conductivity at low temperatures and low relative humidity. These goals will only be achieved through the development of new polymeric or organic–inorganic hybrid materials for H₂/O₂ fuel cells operating at low temperatures.

I am grateful for the financial support by the Ministry of Education, Sport, Culture, Science and Technology (MEXT) and the Special Coordination Funds for Promoting Sciences and Technology of Japan.

References

1. C. Song, Catal. Today, 2002, 77, 17–49.
2. B. Wahdame, D. Candusso, X. Francois, F. Harel, A. DeBernardinis, J. M. Kauffmann and G. Coquery, Fuel Cells, 2005, 7, 47–62.
3. M. Arita, Fuel Cells, 2002, 2, 10–14.
4. M. Aparicio and A. Duran, J. Sol-Gel Sci. Technol., 2004, 31, 103–107.
5. G. Gottesfeld and Z. Zawodzinski, in Electrochemical Science and Engineering, ed. R. Alkire, H. Gerischer, D. Kolb and C. Tobias, Wiley-VCG, New York, 1994, vol. 5, pp. 195–301.
6. L. Lebrun, A. D. Silva and M. Metayer, J. Appl. Polym. Sci., 2002, 84, 1572–1580.
7. G. Hirankumar, S. Selvasekarapandian, M. S. Bhuvaneswari, R. Baskaran and M. Vijayakumar, Ionics, 2004, 10, 135–138.
8. A. A. Mohamad and A. K. Arof, Ionics, 2006, 12, 57–61.
9. J. Zieba, Y. Zhang, P. N. Prasad, M. K. Casstevens and R. Burzynski, in Sol–Gel Optics II, Proc. Soc. Photo-Opt.: Instrum. Eng, ed. J. D. Mackensie, 1758, 403, 119922.
10. J. Scheirs, S. W. Bigger, E. T. H. Then and N. C. Billingham, J. Polym. Sci., Part B: Polym. Phys., 1993, 31, 287–297.
11. S. N. Cassu and M. I. Felisberti, Polymer, 1997, 38, 3907–3911.
12. T. Uma, S. Kumar, A. Kishimoto and K. Kimura, Mater. Lett., 2012, 72, 81–87.
13. A. S. Arico, H. Kim, A. K. Shukla, M. K. Ravikumar, V. Antonucci and N. Giordano, Electrochim. Acta, 1994, 39, 691–700.
14. B. Schaudel, C. Guermeur, C. Sanchez, K. Nakatani and J. A. Delaire, J. Mater. Chem., 1997, 7, 61–65.
15. U. R. Pillai, E. S. Demessie and R. S. Varma, Appl. Catal., A, 2003, 252, 1–8.
16. S.-J. Shina, J.-K. Leeb, H.-Y. Haa, S.-A. Honga, H.-S. Chunb and I.-H. Oha, J. Power Sources, 2002, 106, 146–152.
17. Proton Conductors: Solids, Membranes and Gels-Materials and Devices, ed. P. Colomban, Cambridge University Press, Cambridge, 1992.
18. S. Slade, S. A. Campbell, T. R. Ralph and F. C. Walsh, J. Electrochem. Soc., 2002, 149, A1556–A1564.
19. S. Li and M. Liu, Electrochim. Acta, 2003, 48, 4271–4276.
20. J. Qiao, T. Hamaya and T. Okada, Polymer, 2005, 46, 10809–10816.
21. J. Lu, Q. Nguyen, J. Zhou and Z. H. Ping, J. Appl. Polym. Sci., 2003, 89, 2808–2814.
22. J. Qiao, T. Hamaya and T. Okada, Polymer, 2005, 46, 10809–10816.
23. A. Anis, S. M. Al-Zahrani, A. K. Banthia and S. Bandyopadhyay, Int. J. Electrochem. Sci., 2011, 6, 2652–2671.
24. U. Thanganathan, J. Parrondo and B. Rambabu, J. Appl. Electrochem., 2011, 41, 617–622.
25. P. Dimitrova, K. Friedrich, U. Stimming and B. Vogt, Solid State Ionics, 2002, 150, 115–122.
26. M. Helen, B. Viswanathan and S. S. Murthy, J. Membr. Sci., 2007, 292, 98–105.
27. A. Anis, A. K. Banthia and S. Bandyopadhyay, J. Power Sources, 2008, 17, 772–779.
28. A. Horky, P. Nazir Kherani and G. Xu, J. Electrochem. Soc., 2003, 150, A1219–A1224.
29. R. E. Lechner, Ferroelectrics, 1995, 167, 83–98.
30. M. Etienne and A. Walcarius, Talanta, 2003, 59, 1173–1188.
31. K. Kordesch and G. Simader, Fuel Cells and Their Applications, VCH, Weinheim, Germany, 1996.
32. K. D. Kreuer, A. Rabenau and W. Weppner, Angew. Chem., 1982, 94, 224–225.
33. M. Mulder, Basic Principles of Membrane Technology, Kluwer Academic Publishers, Dordrecht, 1996.
34. K. Kordesch and G. Simader, Fuel Cells and Their Applications, VCH, Weinheim, Germany, 1996.
35. K. D. Kreuer, W. Weppner and A. Rabenau, Angew. Chem., Int. Ed. Engl., 1982, 21, 208–209.
36. K. Tadanaga, H. Yoshida, A. Matsuda, T. Minami and M. Tatsumisago, Electrochem. Commun., 2003, 5, 644–646.
37. K. Tadanaga, H. Yoshida, A. Matsuda, T. Minami and M. Tatsumisago, Electrochim. Acta, 2004, 50, 705–708.
38. T. Uma, J. Parrondo and B. Rambabu, J. Solid State Electrochem., 2012 , (online).

---