Electrochemical Oxidation of Hydrolyzed Poly-Oxymethylene-Dimethylether by Pt and PtRu Catalysts on Ta-Doped SnO₂ Supports for Direct Oxidation Fuel Cells

Electrochemically active surface areas of each catalyst were estimated conventionally by the CO stripping method. Each catalyst was maintained at 0.05 V for 30 min. in 0.1 mol dm⁻³ HClO₄ solution saturated with CO by bubbling mixed gas (1% CO + 99% N₂) to ensure that the Pt sites were sufficiently covered by CO. Then, the CO stripping charge was measured in N₂-saturated 0.1 mol dm⁻³ HClO₄ solution (CO-free condition) at 10 mV s⁻¹ to oxidize the adsorbed CO (CO_ads) on the working electrode. The h-POMM_n (n = 3∼8) fuels are oxidized much more quickly than the original, unhydrolyzed ones.^8,9 Based on these previous works, a simulated fuel mixture of methanol and formaldehyde was used as a convenient model for the 100% hydrolyzed fuel of 0.25 mol dm⁻³ POMM₃ (mixture of methanol and formaldehyde, 0.5 and 0.75 mol dm⁻³, respectively, dissolved in 0.1 mol dm⁻³ HClO₄ solution), which is equivalent to either 1 mol dm⁻³ methanol alone or 1.5 mol dm⁻³ formaldehyde alone in the amount of electric charge (6 equivalents) for complete oxidation. The electrochemical activities of Pt/Ta-SnO₂, PtRu/Ta-SnO₂ and c-Pt₂Ru₃/CB for the oxidation of h-POMM₃ fuel, as well as the pure formaldehyde and methanol solutions, were evaluated from hydrodynamic voltammograms at various temperatures in the 0.1 mol dm⁻³ HClO₄ electrolyte solutions by use of the half cell. We calculated the Pt based mass activity of these catalysts for those fuels based on the activities. Various catalysts mentioned above were uniformly dispersed on each working disk electrode (5 mm in diameter) at a constant loading of platinum, 11 μg cm⁻². Nafion films were cast on the catalyst layer (CL) with an average film thickness of 0.1 μm. A platinum mesh was used as the counter electrode. The electrolyte solution of 0.1 mol dm⁻³ HClO₄ was prepared from reagent grade chemicals (Kanto Chemical Co.) and Milli-Q water (Milli-Pore Japan Co., Ltd.) and purified in advance by conventional pre-electrolysis methods. The electrolyte solution was saturated with N₂ or O₂ gas bubbling for at least 1 h prior to the electrochemical measurements. All electrode potentials were controlled by a potentiostat (HZ-5000, Hokuto Denko Co.) with respect to a reversible hydrogen electrode (RHE).

The Ta-SnO₂ nano-sized particle supports with fused-aggregate network structure were synthesized by the flame oxide synthesis method; details of the method are described in earlier papers.^30–33 Briefly, reagent grade 2-ethylhexanoate salts of Sn and Ta (Nihon Kagaku Sangyo, Co. Ltd.) were mixed at desired ratios into turpentine oil. The mixtures were supplied into the burner (>1000°C) of a flame combustion system (FCM-Mini, Hosokawa Micron Co.). The particles obtained were sintered at 800°C for 2 h in air in a rotary kiln furnace (Alfa Engineering, Inc.). The surface area of the Ta-SnO₂ support was 35 m² g⁻¹, estimated by the Brunauer-Emmett-Teller adsorption method (BET, BELSORP-max, MicrotracBEL Co.). The Pt and Ru colloid particles were loaded on the support by a colloidal method.²⁹ Pt/Ta-SnO₂ was heat-treated at 800°C in nitrogen for 2 h and then quenched at room temperature. The PtRu/Ta-SnO₂ was heat-treated at 400°C in air for 0.5 h, 150°C in 1% H₂ (N₂ balance) for 1 h, 800°C in nitrogen for 2 h, and then quenched at room temperature. A commercial Pt₂Ru₃ catalyst supported on carbon black (denoted as c-Pt₂Ru₃/CB, TEC61E54, mean particle size d = 3.5 nm, Pt 29.7 wt%, Ru 23.0 wt%, Pt:Ru = 2.00:2.99 (atomic ratio), Tanaka Kikinzoku Kogyo K. K.) was used as a reference. The crystalline phases of these catalysts were characterized by X-ray diffraction measurements (XRD, Ultima 4, Rigaku Co.) with Cu Kα radiation (0.15406 nm, 40 kV, 40 mA). Contents of the Pt and PtRu catalysts on the Ta-SnO₂ support were quantitatively analyzed by use of an inductively coupled plasma-mass spectrometric analyzer (ICP-MS, 7500CX, Agilent Technologies Inc.). After a pretreatment of these catalysts by the alkaline fusion method (sodium carbonate and sodium peroxide), the analyses were performed with the aqueous hydrochloric acid solutions. The chemical compositions of Pt and Ru were 16.2 wt% (Pt in Pt/Ta-SnO₂), 13.0 wt% (Pt in PtRu/Ta-SnO₂) and 6.3 wt% (Ru in PtRu/Ta-SnO₂, Pt:Ru = 1.00:0.94 (atomic ratio in PtRu/Ta-SnO₂)).

The particle size distributions, dispersion conditions and element distributions of the catalysts thus obtained were characterized by images with high-angle annular dark field-scanning transmission electron microscopy (HAADF-STEM) from a scanning transmission electron microscope (STEM, HD-2700, Hitachi High-Technologies Co.) with an energy dispersive X-ray spectroscopic attachment (EDX, Quantax XFlash 5010, Bruker AXS K. K.). The electronic states of Pt and Sn in the Ta-SnO₂ supported catalysts were characterized with in-situ X-ray photoelectron spectroscopy (XPS, JPS-9010, JEOL Ltd.) without exposure to the atmosphere after the heat-treatment mentioned above.

Three types of MEAs were fabricated by using Pt/Ta-SnO₂, PtRu/Ta-SnO₂, and c-Pt₂Ru₃/CB as the anode catalysts and a commercial Pt/CB (TEC10E50E, Tanaka Kikinzoku Kogyo K. K.) as the cathodes, respectively. The CLs were formed on the polymer electrolyte membrane (Nafion NRE 117, Du Pont Co.) by spraying ink containing each catalyst, which was described in previous papers.²⁹ The Pt loading amounts on anode and cathode were 0.3 mg cm⁻² and 0.5 mg cm⁻², respectively. The volume ratio of perfluorosulfonic acid binder (PFSA; equivalent weight = 909 g eq⁻¹, Asahi Glass Co., Ltd.) and support in the ink was 0.7 (volume ratio, dry basis). The catalyst-coated membranes obtained (CCMs: electrode area, 29.2 cm²) were heat-treated at 120°C for 30 min. Then, the CCMs were sandwiched with two gas diffusion layers (PX30, ZOLTEK Co.) to form the MEAs. The MEAs were mounted in single cells with carbon plates that included a multi-channel serpentine flow field. The cell voltage (V) as a function of current density (I) was measured by supplying an aqueous solution in which was dissolved h-POMM₃ (mixture of methanol and formaldehyde, 2 and 3 mol dm⁻³, respectively), 2 mol dm⁻³ methanol or 3 mol dm⁻³ formaldehyde fuel to the anode at 80°C. The flow rates of each fuel into the anode were controlled at 3 × 10⁻³ dm³ min⁻¹ by a plunger fixed-quantity pump (IP-7100; Lab-Quatec Co.). The oxygen was bubbled through a hot water reservoir so that it was humidified at 100% relative humidity (RH) and was supplied to the cathode (1 atm, 0.30 dm³ min⁻¹). The I-V curves were measured galvanostatically in steady-state operation, with a measurement time of 5 min for each point by use of an electronic load (PLZ-664WA, Kikusui Electronics Co.) controlled by a measurement system (PEMTest 8900, Toyo Co.).

TEM images of Pt/Ta-SnO₂ and PtRu/Ta-SnO₂ are shown in Figs. 1a and 1b. Each of the Ta-SnO₂ nanoparticles (diameter > 10 nm) were partially fused with nearest neighbors and formed a fused-aggregate network structure. This type of network structure is desirable to form electronically conductive pathways.^28,30–33 Such a structure also constructs open pores from 10 nm to 50 nm in diameter, surrounded by fused particles, which should act as mass transport pathways for reactants such as oxygen and the present fuels.^28,30–33 The Pt and PtRu particles (diameter < 10 nm) were highly dispersed on the supports, as shown in the TEM images (Figs. 1a and 1b). As shown in Figs. 1c and 1d, Pt/Ta-SnO₂ exhibited a broader size-distribution and a larger mean-particle size (d) than PtRu/Ta-SnO₂, i.e., d = 6.3 ± 1.0 nm for the former and 4.1 ± 0.6 nm for the latter, respectively.

Zoom In Zoom Out Reset image size

The XRD peaks for the Pt in Pt/Ta-SnO₂, which were located around 2θ = 39.8°, 46.4° and 67.2° were slightly shifted to higher 2θ compared to those of pure Pt (Fig. 2), because the catalyst was sintered at 800°C in N₂ atmosphere, which allowed elemental tin originating from the Ta-SnO₂ support to diffuse into the Pt catalyst particles, converting them to a PtSn alloy, as seen below. No XRD peaks for the PtRu alloy on Ta-SnO₂ were detected in Fig. 2. As discussed later, Sn diffused from the Ta-SnO₂ to the catalyst during the heat-treatment procedure. Sn-rich Pt-Sn and Ru-Sn phases are known to exhibit low melting points (under 800°C).³⁴ It is likely that the lack of clear crystallinity is due to the PtRu alloy particles, which were sintered at 800°C and quenched to room temperature, being originally in a disordered state, and the disordering was further promoted by the additional diffusion of elemental Sn into the particles from the oxide substrate.

Zoom In Zoom Out Reset image size

The XPS spectra of Pt and Sn are shown in Fig. 3 for Pt/Ta-SnO₂ (a, c) and PtRu/Ta-SnO₂ (b, d), respectively. Almost no shifts (≤0.1 eV) of Pt 4f binding energies for PtRu/Ta-SnO₂ as well as Pt/Ta-SnO₂ were observed, in comparison with those of pure Pt metal. No noticeable shifts from a standard reference were observed for the Sn 3d binding energy peaks of the SnO₂ substrates on either of the supported catalysts, probably due to the large excess of the Ta-SnO₂ compared with the catalyst particles. However, clear peaks for metallic Sn appeared on both catalysts, even though no corresponding XRD peaks were observed. On the other hand, the line analysis of STEM-EDX with the HAADF-STEM images of Pt/Ta-SnO₂ and PtRu/Ta-SnO₂ (Figs. 4a, 4b) indicated that a relatively high content of Sn existed in each catalyst particle, and it gradually decreased with increasing distance from the contact with the support toward the outer particle surface; this was commonly observed in several tens of particles examined. It was also confirmed from the analysis that Pt and Ru were dispersed uniformly in the particles, forming the alloy but without exhibiting XRD peaks, as mentioned above, indicating disordering of the lattice structures due to the mixing with Sn. Also, the relative intensity (Sn versus Pt (Sn/Pt) of PtRu/Ta-SnO₂ at the interface and center of the catalyst was higher in Pt. The amount of diffused Sn was larger in PtRu/Ta-SnO₂ (Fig. 4b) than that in Pt/Ta-SnO₂ (Fig. 4a). From these observations, shown in Figs. 2 to 4, we conclude that the metallic Sn diffuses from the Ta-SnO₂ support into the noble metal catalyst during the sintering procedure and that the surface or a near-surface layer of the particles could involve Sn, Pt and Ru in the metallic state, as illustrated schematically in the inset of Fig. 4.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The CO stripping voltammograms for each catalyst in 0.1 mol dm⁻³ HClO₄ solution saturated with N₂ at 25°C are shown in Fig. 5. The onset potentials and peak potentials for CO oxidation reveal significant differences between these catalysts. The onset potentials decreased in the order c-Pt₂Ru₃/CB > Pt/Ta-SnO₂ > PtRu/Ta-SnO₂, and the peak potentials decreased in the order Pt/Ta-SnO₂ > PtRu/Ta-SnO₂ ≈ c-Pt₂Ru₃/CB. The lower onset potentials were associated with the presence of Sn, most likely due to the much greater ease with which Sn is oxidized to SnO₂ (−0.106 V vs. RHE)³⁵ compared with that with which Ru is oxidized to Ru₂O₃ (0.738 V vs. RHE).³⁶ In contrast, the lower peak potentials were associated with the presence of Ru. Thus, the best overall behavior is obtained when both are present. In the bifunctional mechanism, both Sn and Ru can act to donate OH to assist in the oxidation of CO, with Sn doing so at a lower potential, but with Ru having higher intrinsic activity.³⁷

Zoom In Zoom Out Reset image size

The electrochemically active surface areas of each sample were estimated conventionally from the CO stripping peak areas (ECA_CO), assuming the oxidation of a monolayer of adsorbed CO on polycrystalline Pt requires 420 μC cm⁻². The ECA_CO values for Pt/Ta-SnO₂, PtRu/Ta-SnO₂ and c-Pt₂Ru₃/CB were 28.1, 35.6 and 87.6 m² g_Pt⁻¹, respectively, referred to Pt weight. On the other hand, based on the mean particle sizes determined by TEM and their compositions, the specific surface areas can be calculated to be 44.4, 68.3 and 80.0 m² g_cat⁻¹, which can be reduced to Pt weight-base values (ECA_TEM) of 44.4, 46.0, 45.0 m² g_Pt⁻¹, respectively. These values are not consistent with the ECA_CO values but nevertheless might be reasonable, because there were large difference among the catalysts in the components, compositions, surface states of the catalyst particles and their support materials, resulting in the large differences in the amounts of adsorbed CO on their surfaces. In other words, it is difficult to obtain reliable data of the specific surface area on such catalysts by methods applying gas-adsorption behavior, including ECA_CO. It is well known that the CO coverage and/or adsorption strength on Pt and Pt alloys depend upon various structural and electronic factors. This has been documented experimentally and theoretically in previous papers, for pure Pt, and for Pt-M alloys,^42–45 and thus, it has been found that the apparent CO coverage on Pt and Pt-M surfaces (θ, %) does not always reach 100%.^37–41 Therefore, in the present work, we have simply compared the catalytic behavior among them based on the Pt weight, or the geometric electrode surface area of the RDE, and the MEAs loaded with the same amounts of Pt.

The hydrodynamic voltammograms for the respective oxidation reactions of the h-POMM₃ (mixture of 0.75 mol dm⁻³ formaldehyde and 0.5 mol dm⁻³ methanol, h-POR), 1.0 mol dm⁻³ of methanol (MOR), and 1.5 mol dm⁻³ of formaldehyde (FOR) in 0.1 mol dm⁻³ of HClO₄ solutions saturated with oxygen at 80°C are shown in Figs. 6a–6c. The current density was based on the geometric electrode area, which corresponds to the same amounts of Pt loading, 11 μg cm⁻². Notice, however, that the current densities for the MOR were multiplied by 10 in order to facilitate the comparison of the onset potentials, which were defined as the potentials observed at a current density of 10 μA cm⁻². The onset potentials for the h-POR, MOR and FOR on Pt/Ta-SnO₂ and PtRu/Ta-SnO₂ shifted noticeably to lower potentials in comparison with those on c-Pt₂Ru₃/CB, except for the MOR activity being inferior on Pt/Ta-SnO₂ and superior on c-Pt₂Ru₃/CB. The onset potentials for each fuel on these catalysts are summarized as a function of measurement temperature in Fig. 7. Those for the h-POR and FOR on Pt/Ta-SnO₂ and PtRu/Ta-SnO₂ were at least 0.2 V lower than those for c-Pt₂Ru₃/CB at measurement temperatures from ambient to 80°C, although the onset potentials for the MOR were more or less at the same level on each catalyst over the whole temperature range.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Thus, the onset potentials for the h-POR and FOR appear to be consistent with those for CO oxidation already discussed. The major effect is due to the presence of Sn, while there is an additional effect of Ru. In contrast, the onset potentials for the MOR do not appear to be related to those for CO oxidation, and thus the rate-determining step is likely to be an earlier step in the overall process, for example, CO formation.³⁶

As seen in Fig. 6, the current densities for the h-POR and FOR on Pt/Ta-SnO₂ and PtRu/Ta-SnO₂ were much higher than that on c-Pt₂Ru₃/CB over the whole potential region measured. However, the current density for the MOR, being one order of magnitude lower than those for the h-POR and FOR, was exceptionally lower on Pt/Ta-SnO₂ than c-Pt₂Ru₃/CB.

Figure 8 shows the Pt-based mass activities at 0.50 V (MA_0.50V) for the h-POR, MOR and FOR on the different catalysts as a function of measurement temperature. The MA_0.50V values for the h-POR and FOR on Pt/Ta-SnO₂, and PtRu/Ta-SnO₂ were higher than those for c-Pt₂Ru₃/CB; in particular, those for the h-POR on PtRu/Ta-SnO₂ and Pt/Ta-SnO₂ were more than 4 times larger than that for c-Pt₂Ru₃/CB over the whole temperature range. The higher MA_0.50V values for the MOR on c-Pt₂Ru₃/CB in comparison with those on Pt/Ta-SnO₂ and PtRu/Ta-SnO₂, shown in Fig. 8b, might be due to the larger specific surface area of the former, with mean particle size d = 3.5 nm, and also due to the higher CO tolerance with larger Ru content in the alloy.

Zoom In Zoom Out Reset image size

In the interest of completeness, we also checked the specific activities based on the catalyst area at 0.50 V (j_0.50V) for the h-POR, MOR and FOR at 25°C. Each of the j_0.50V values based on the ECA_CO (j_0.50V,CO, A m⁻²_ECA,CO) and ECA_TEM (j_0.50V,TEM, A m⁻²_ECA,TEM) of PtRu/Ta-SnO₂, Pt/Ta-SnO₂, and c-Pt₂Ru₃/CB are listed in Table I. Both sets of values were fairly consistent. We found that both the j_0.50V,CO and j_0.50V,TEM values for the h-POR, MOR and FOR for PtRu/Ta-SnO₂ and Pt/Ta-SnO₂ were much higher than those for c-Pt₂Ru₃/CB, except for the j_0.50V,TEM for the MOR for Pt/Ta-SnO₂ at 25°C. As already mentioned, this discrepancy might be related to differences in the surface composition of each catalyst. However, in general, the MA_0.50V, j_0.50V,CO and j_0.50V,TEM values for PtRu/Ta-SnO₂ were much higher than those for c-Pt₂Ru₃/CB.

The hydrolysis of POMM_n occurs promptly to produce formaldehyde and methanol under acidic conditions (Eq. 2) and supplies protons via oxidation, as shown below (Eqs. 3–12).

Formaldehyde branch:

Overall:

Methanol branch:

Overall:

In this research, M stands for either Ru or Sn, the latter having diffused into the Pt catalyst particle during the sintering procedure. The kinetics of the FOR and MOR at the anode are strongly affected by the poisoning due to the presence of CO as an intermediate on the Pt surface (Eqs. 3, 8). The adsorbed water or hydroxyl on the other metal (M) surface becomes an oxygen source, which accelerates the oxidation of the adsorbed CO to CO₂ via the bifunctional mechanism (Eqs. 5, 10).^{14,15,45–47} The oxidation of CO adsorbed on the Pt surface of Eqs. 5, 10 is likely to be the rate-determining step (rds) for some of these oxidation reactions. The approximate activation energies for the FOR and h-POR of PtRu/Ta-SnO₂ and Pt/Ta-SnO₂ were 15–16 kJ mol⁻¹, which were approximately half that for c-Pt₂Ru₃/CB (30 kJ mol⁻¹), and which can be assigned to Eqs. 5, 10 as the rds. The activation energies for the MOR for PtRu/Ta-SnO₂ and Pt/Ta-SnO₂ were both 21 kJ mol⁻¹, which was approximately two-thirds that for c-Pt₂Ru₃/CB (30 kJ mol⁻¹). Thus, the rds for the MOR might be different from that for the h-POR and FOR on the PtRu/Ta-SnO₂ and Pt/Ta-SnO₂, as already mentioned, and might involve at least partial control by Eq. 8, which involves the breakage of four bonds, compared with only two for Eq. 3 for formaldehyde. The bond breakage steps would clearly have to be sequential but are shown for brevity here as single overall steps. A detailed mechanistic analysis is beyond the scope of the present work but will be pursued in continuing work. We have not been able to find relevant theoretical work that could help to elucidate the differences in behavior between the oxidation of formaldehyde and that of methanol. It is important to also note that the presence of Sn as well as Ru on the top surface or subsurface layer of the catalyst particles of Pt/Ta-SnO₂ and PtRu/Ta-SnO₂, as depicted in Fig. 4, may weaken the CO bonding strength at Pt sites, resulting in the enhancement of water adsorption and CO oxidation, as shown above.^48–50

Figures 9a–9c shows IR-free I-V curves for single cells using Pt/Ta-SnO₂, PtRu/Ta-SnO₂ and c-Pt₂Ru₃/CB as the anode catalysts (the same amounts of Pt loading, 0.30 mg cm⁻²) at 80 °C by supplying the h-POMM₃ (mixture of methanol and formaldehyde, 2 and 3 mol dm⁻³, respectively, dissolved in electrolyte), 2 mol dm⁻³ methanol and 3 mol dm⁻³ formaldehyde, respectively. All of the cells, fed with the three types of fuels, exhibited open circuit voltages (OCVs) of approximately the same value, 0.8 V. It is reasonable to consider that the steady FOR and MOR in the h-POMM₃ cells may continuously proceed at some intermediate cell voltage, i.e., an overvoltage that makes possible a balanced consumption of formaldehyde and methanol, so that there is no progressive accumulation of methanol at a particular current; this supposition is now being tested by use of a monitoring system during long term operation. An additional advantage of the use of the oxide support is discussed below.

Zoom In Zoom Out Reset image size

Unlike the cells that utilized oxide supports, except the methanol cell, the cells that utilized c-Pt₂Ru₃/CB exhibited immediate voltage drops of 0.2–0.3 V from the OCVs as the current was increased, as seen in Figs. 9a–9c. This may be because the surface of the c-Pt₂Ru₃/CB catalyst is mostly covered with poisoning CO, even though a small number of CO-free sites might remain at the OCV of 0.8 V. As a result, the voltage drops quickly, i.e., a significant overpotential is required for the CO removal, as seen in Fig. 8, to enable the fuel oxidation to proceed at reasonable currents. In the methanol cells, an accumulation of poisoning CO presumably occurred at high coverages on the catalyst surfaces, even in the cells utilizing the Ta-SnO₂ support. After the CO was removed from the surfaces in all of the cells utilizing c-Pt₂Ru₃/CB and at the methanol cell utilizing PtRu/Ta-SnO₂ and Pt/Ta-SnO₂, current could be drawn without further large voltage drops, at least up to 300 mA cm⁻²; for the CB-supported catalyst, this might be due to the well-known properties common to CB, i.e., good electronic conductivity and well-developed agglomerate structure.

On the other hand, we have reported the preparation of nano-sized oxide materials with the unique CB-like fused-aggregate network structure, as well as high electronic conductivity.^28,33 The interconnected micropores of the Ta-SnO₂ support in the CLs were expected to play the role of pathways for the supply of the liquid fuels and the efficient removal of product water and CO₂. To our knowledge, gratifyingly superior performance found in this work, e.g., the OCV of 0.8 V and high current density of 500 mA cm⁻² at cell voltages exceeding 0.45 V for the h-POMM₃ cells, with the relatively low Pt loading of 0.3 mg cm⁻², have never been achieved with conventional direct methanol fuel cells (DMFCs) utilizing PtRu/CB catalysts, even at Pt loadings a factor of 10 higher. Formaldehyde has an irritating odor and significant toxicity, but the POMM_n fuels avoid these problems. From the viewpoint of the fuel regulations for European countries (EN 590), the POMM_n (n = 3∼5) fuels are also suitable.⁸ With increasing n, the amount of formaldehyde increases and that of methanol decreases. There might still be further improvements in anodic catalysts for the MOR, but the present work suggests that a viable alternative could be the use of the hydrated POMM_n (n = 3∼5) fuels, since they contain relatively high volumes of formaldehyde with increasing n value for the same volume of methanol, which can lessen the participation of the slower MOR, thus leading to a higher cell performance. Thus, the great merits of these h-POMM_n fuel cells have been demonstrated here, in comparison with conventional DMFCs, in terms of performance, in addition to the intrinsic fuel properties of high volumetric energy density, high flash temperature and low toxicity. The output cell voltage, as well as the current density seen above for the h-POMM_n/air cell, are still inferior to those for H₂/air cells, but, in the former, the liquid fuel can be utilized directly, without the need for a fuel reforming/purification unit or heavy high-pressure gas cylinder for H₂ storage. The achievement of such a high performance for the h-POR and FOR can be ascribed to the utilization of the Pt/Ta-SnO₂ and PtRu/Ta-SnO₂ catalysts described here.