Proton-Conducting Glass Electrolyte - Analytical Chemistry (ACS

Dec 15, 2007 - This is the first time such a high proton conductivity value has been ... The glass was applied as the electrolyte for an H2/O2 fuel ce...
1 downloads 0 Views 93KB Size
Anal. Chem. 2008, 80, 506-508

Correspondence

Proton-Conducting Glass Electrolyte Thanganathan Uma and Masayuki Nogami*

Department of Materials Science and Engineering, Nagoya Institute of Technology, Showa, Nagoya, 466-8555, Japan

A new porous glass electrolyte consisting of heteropolyacids, i.e., phosphotungstic acid (PWA) and phosphomolybdic acid, was investigated and was found to yield a remarkably high proton conductivity of 1.014 S cm-1 at 30 °C and 85% relative humidity. This is the first time such a high proton conductivity value has been reported for a heteropolyacid glass membrane. The glass was applied as the electrolyte for an H2/O2 fuel cell, and a maximum power density of 41.5 mW/cm2 at 32 °C was attained using this new PWA-containing electrode. Numerous reports exist on highly proton-conductive solid materials,1-4 and recently, many research studies have been devoted to proton-conducting materials containing heteropolyacids, such as phosphotungstic acid (PWA) and phosphomolybdic acid (PMA).5,6 Heteropolyacids and related compounds are known to be interesting catalysts, in both homogeneous and heterogeneous systems.7 The temperature dependence of the electrical conductivities of PWA and PMA are 0.18 and 0.17 S cm-1, respectively, at 25 °C. These values are remarkably larger than those of other proton conductors.8 PWA, with a proton conductivity of ∼2 × 10-1 S cm-1 at 25 °C,9 has been incorporated as a proton source in a silicophosphate amorphous gel structure. Our group was the first to report on the synthesis and characterization of porous P2O5-SiO2 glass, which displayed a maximum conductivity of 2.2 × 10-2 S cm-1 at 50 °C.10 This glass was applied to an H2/O2 fuel cell operating at 50 °C and 90% relative humidity (RH), and the output power was found to be ∼6 mW/cm2. In addition, this mesoporous glass material was found to exhibit high thermal and mechanical stabilities. Recently, we reported on the synthesis and characterization of phosohosilciate glass membrane doped with heteropolyacids * Corresponding author. Tel.: +81 52 735 5285. Fax: +81 52 735 5285. E-mail: [email protected]. (1) Colomban, Ph. Ann. Chim. Sci. Mater. 1999, 24, 1-18. (2) Nogami, M.; Miyamura, K.; Abe, Y. J. Electrochem. Soc. 1997, 144, 21752178. (3) Nogami, M.; Nagao, R.; Cong, W.; Abe, Y. J. Sol-Gel Sci. Technol. 1998, 13, 933-936. (4) Nishino, A. J. Power Sources 1996, 60, 137-147. (5) Staiti, P.; Freni, S.; Hocevars, S. J. Power Sources 1999, 79, 250-255. (6) Corma, A.; Chem. Rev. 1995, 95, 559-614. (7) Otake, M.; Onoda, T. Shokubai 1976, 18, 169-179. (8) Nakamura, O.; Kodama, T.; Ogino, I.; Miyake, Y. Chem. Lett. 1979, 1, 1718 (9) Maruyama, T.; Saito, Y.; Matsumoto, Y.; Yano, Y. Solid State Ionics 1985, 17, 281-286. (10) Nogami, M.; Matsushita, H.; Goto, Y. Adv. Mater. 2000, 12, 1370-1372.

506 Analytical Chemistry, Vol. 80, No. 2, January 15, 2008

(i.e., PMA and PWA) for applications as H2/O2 fuel cell electrolytes. A maximum proton conductivity of 9.1 × 10-2 S cm-1 at 90 °C and 30% RH, as well as a good cell performance (i.e., a power density of 22.9 mW/cm2) was obtained at 30 °C. These materials were also found to exhibit a good thermal stability and a high porosity. Nitrogen sorption isotherms were used to determine the pore accessibility of glasses prepared with PMA11 and PWA,12 and the results displayed specific surface areas ranging from 289334 to 240-640 m2/g as well as average pore sizes of ∼4.8 and ∼3.2 nm as determined by the Barrett-Joyner-Halenda method. The pore size of the silica glass appeared to be the main factor influencing the proton conductivity. The high proton conductivity was attributed to a glass network containing PWA or PMA acid ions, acting as proton donors, and to strongly hydrogen-bonded hydroxyl groups bound to them creating P-O-Si bonds as proton conduction paths. HPAs have a fairly high thermal stability. By mixing polyatoms of the heteropolyanions Mo and W, the reduced polyanions can be reoxidized by oxygen molecules,13 and it has been suggested that the reoxidation proceeds by multielectron transfer in the complex of the reduced species with O2. The stability against hydrolysis in aqueous solution follows the order W > Mo.13 Mixing PWA with PMA increases the stability of the latter, and in the present case, a mixture of the heteropoly acids, PWA and PMA, with phosohosilciate was employed to improve the stability of glass membranes for fuel cell tests. As it seems that the interaction between the Keggin anion and the support occurs through shared protons, a determination of the HPA protonation sites has become mandatory.14 The present report demonstrates that a high proton conductivity can be obtained for such a glass membrane, rendering it a suitable candidate for low-temperature H2/O2 fuel cells. The clearest direct application of solid-state proton conduction occurs in situations where there is a requirement to transmit hydrogen across some intervening barrier. In the H2/O2 fuel cell technology, materials with good proton-conducting abilities, but a blocking effect on electrons and an insolubility in water, would be of great value.15 (11) (12) (13) (14)

Uma, T.; Nogami, M. J. Electrochem. Soc. 2007, 154, B32-B38. Uma, T.; Nogami, M. J. Membr. Sci. 2006, 280, 744-751. Kozhevnikov, I. V.; Matveev, K. I. Appl. Catal. 1983, 5, 135-150. Ganapathy, S.; Fourier, M.; Paul, J. F.; Delevoye, L.; Guelton, M.; Amoureux, J. P.; J. Am. Chem. Soc. 2002, 124, 7821-7828. (15) Bruinink, J. J. Appl. Electrochem. 1972, 2, 239-249. 10.1021/ac0706630 CCC: $40.75

© 2008 American Chemical Society Published on Web 12/15/2007

Figure 1. Logarithm of the conductivity as a function of humidity for the PWA/PMA-P2O5-SiO2 glass composite membrane.

Figure 2. Photocopies of the surface area of the electrode before and after fuel cell operation.

EXPERIMENTAL SECTION The proton conductivity (σ) of the PWA/PMA-P2O5-SiO2 glass composite membrane was calculated from the electrolyte resistance (Rb) obtained from the intercept of the Cole-Cole plot with the real axis, the membrane thickness (l), and the electrode area (A) according to the equation σ ) l/ARb. Polarization measurements were carried out on a membrane electrode assembly with a Solartron SI 1287 electrochemical interface and a Solartron SI 1260 impedance/gain-phase analyzer. RESULTS AND DISCUSSION The conductivity of the glass composite electrolyte containing a mixture of PWA/PMA was measured at a constant temperature of 30 °C and varying humidity conditions, and Figure 1 shows the humidity dependence of the logarithm of the proton conductivity. A proton conductivity of 1.014 S cm-1 was achieved at 30 °C with 85% RH. To the best of our knowledge, this is the highest conductivity obtained for a glass. It is well-known that the protonic conductivity of heteropolyacids is strictly related to the number of water molecules coordinated to the Keggin anions. Kreuer has suggested that the HPA acts as a Bronsted acid toward the hydration water, which is generally loosely bound in the structure, resulting in a high proton conductivity.16 Consequently, the conductivity of HPAs is strictly related to the number of water molecules coordinated to the Keggin unit. Furthermore, glasses with a high concentrations of OH groups are potential proton conductors. According to the so-called “vehicle mechanism”, a large uptake of water is essential for a fast proton conduction.17 It is thought that the water that is physically adsorbed promotes proton transport through the membrane. On the other hand, due to the presence of hydrogen bonds between the surface functionalities and the physically adsorbed water, the conduction mechanism occurs by hopping of the hydrogen ions. A membrane electrode assembly for fuel cell conditioning and testing of the new electrode was prepared with PWA, poly(tetraflouroethylene), Pt/C powder, and Nafion. The experimental details were similar to those in a previous report.12 The thickness of the electrode was 0.2 mm. Figure 2 shows the surface appearance of the prepared electrode before and after fuel cell operation. A polarization plot recorded for the H2/O2 fuel cell with the glass composite membrane, and the Pt/C electrode is shown in (16) Kreuer, K. D. Chem Mater 1996, 8, 610-641. (17) Kruer, K. D. Solid State Ionics 1997, 97, 1-15.

Figure 3. Polarization curve for the PWA/PMA-P2O5-SiO2 glass membrane at 32 °C and 30% RH.

Figure 3. The cell performance was measured at a room temperature of 32 °C with 30% RH. As can be seen in the figure, the open circuit voltage was ∼1 V. For a constant cell potential of 0.6 V, a maximum power density of 41.5 mW/cm2 was observed after operation for 6 h. Despite this, the maximum power density was higher than what was recently reported for a PMA-doped phosphosilicate glass electrolyte.11 The current density reached a value of 157 mA/cm2, and after 10 h, it was found to decrease to 114 mA/cm2. Furthermore, the power density decreased to 30.5 mW/cm2. This drop in power density of the fuel cell after operation for 6 versus 10 h was probably due to a degradation at the anode as a result of interactions with H2. It was thus not a result of interactions with H2O or any lack of dimensional integrity. The drop in fuel cell voltage with increasing current density (Figure 3) had many causes.18 The primary of these, for a relatively thick electrolyte as was used in this experiment (glass thickness, 0.55 mm), was its resistance. For a constant cell potential of 0.6 V, the performance of the cell was observed to be stable during the 6-h operation at a temperature of 32 °C. A total ohmic cell resistance of 3 Ω cm2 was measured by impedance spectroscopy at 0.6 V for an operation time of 6 h. Dramatic improvements in power densities have been achieved in the fuel cell through reduction of the electrolyte thickness and increase of the surface area of the catalyst. Further reduction of the performance losses can be expected by increasing the operating temperature above 50 °C. In prior fuel cell performances studies, operating the fuel cell at higher temperature (100 and 120 °C) (18) Slade, R. C. T.; Omana, M. J. Solid State Ionics 1992, 58, 195-199.

Analytical Chemistry, Vol. 80, No. 2, January 15, 2008

507

effectively reduced the performance losses that were due to CO poisoning.19 A color change from white to blue on the electrolyte surface surrounding the electrode area could be observed after the electrochemical measurements. The chemical reaction of the heteropolyacid (PWA, PMA)-containing glass membrane with hydrogen is known to cause the reduction of the PWA or PMA species to a heteropolyblue compound constituting a mixedvalence intervalence transition. The occurrence of such a chemical reaction between H2 and phosphotungstic acid was strongly supported by the aforementioned color change of the electrolyte coming out from the anodic side of the cell.20 This electrochemical reduction of the tungsten-containing polyanion would render it suitable for use in displays or as inorganic resists.21 Numerous studies have investigated the kinetics of oxygen reduction on such platinum electrocatalysts in acid electrolytes.22 Cell performance is mainly determined by the high conductivity characteristics of the electrolyte and its promoting effect toward the oxygen reduction reaction.23 Due to the use of heteropolyacid solutions, the investigation of the oxygen reduction process was mainly limited to the range of low temperatures in relations to loss of water by evaporation with consequent crystallization of the heteropolyacids.24 Low-temperature fuel cells are in need of improved electrocatalysts for both the anode and the cathode reactions. Since most of the electrochemical performances losses

can be attributed to the sluggish kinetics of the oxygen reduction reaction. A test of heteropoly acids (PWA and PMA) can enhance activity of Pt/C catalyst for the oxygen reduction reaction will be progressed in future. (For details of all experimental procedures, please see the Supporting Information).

(19) Si, Y.; Jiang, R.; Kunz, H. R.; Fenton, J. M. J. Electrochem. Soc. 2004, 151, A1820-A1824. (20) Giordano, N.; Staiti, P.; Arico, A. S.; Passalacqua, E.; Abate, L.; Hocevar, S. Electrochim. Acta 1997, 42, 1645-1652. (21) Tatsumisago, M.; Minami, T. J. Am. Ceram. Soc. 1989, 72, 484-486. (22) Bregolia, L. Electrochim. Acta 1978, 23, 489-492. (23) Giordano, N.; Staiti, P.; Hocevar, S.; Arico, A. S. Electrochim. Acta 1996, 41, 397-403. (24) Giordano, N.; Arico, A. S.; Hocevar, A.; Staiti, P.; Antonucci, P. L.; Antonucci, V. Electrochim. Acta 1993, 38, 1733-1741.

SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

508

Analytical Chemistry, Vol. 80, No. 2, January 15, 2008

CONCLUSIONS The high proton conductivity of a new glass composite membrane was investigated as a low-temperature H2/O2 fuel cell electrolyte, and these materials were found to yield higher OCVs. A loss in the performance was however noticed after extended fuel cell operating times. The fuel cell produced a maximum power density of 41.5 mW/cm2 at 108 mA/cm2. It is expected that the fuel cell performance could be developed by using this electrolyte/ electrode at somewhat higher temperatures under similar operating conditions. One speculative explanation for the better performance can be related to the proton transport mechanism induced by the presence of the heteropolyacids, thus rendering it possible to fabricate new glass composite materials with heteropolyacids for use as electrolytes in H2/O2 fuel cells. ACKNOWLEDGMENT The authors thank the Japan Society for the Promotion of Science (JSPS) fellowship program, Japan, for their financial support of this work.

Received for review April 4, 2007. Accepted October 19, 2007. AC0706630