Introduction To achieve prolonged lifetime of polymer electrolyte fuel cell (PEFC), hydrogen-peroxide (H2O2)-induced chemical degradation of polymer electrolyte membrane (PEM) is a crucial problem. In general, oxygen (O2) crossover from cathode to anode through the PEM yields H2O2 generation on Pt/C anode catalyst surface through 2-electron pathway of O2 reduction reaction. Thus, mitigation of H2O2 generation on Pt/C surface is a key to suppress the PEM degradation. Trogadas and Ramani reported tungsten trioxide (WO3) addition to Pt/C suppressed H2O2 generation.1 The results suggest that the cooperation of Pt nanoparticles and WO3 located near-by should be effective for suppression of H2O2 generation. However, influence of the oxidation states of tungsten oxide (WO x ) on H2O2 generation at the Pt surface is not fully resolved. Furthermore, it was reported that, although fully oxidized WO3 is stable in acid solution (pH < 2),2 sub-oxides of tungsten (WO x ) could dissolve in Pt-WO x electrochemical system.3 Thus, electrochemical dissolution behaviors of WO x located on Pt substrate surface should be investigated in detail. In this study, we prepare Pt(111) surface that modified by WO x having different oxidation states, and investigated the electrochemical H2O2 generations. Experimental Pt(111) single crystal substrate surface was cleaned by repeated cycles of Ar+ sputtering and annealing in an ultrahigh vacuum (UHV; ~10-8Pa). Subsequently, WO x was deposited on the cleaned Pt(111) via an arc-plasma deposition (APD) method using a W target under partial pressures of O2 (p(O2) = 1×10-1 or 1×10-3 Pa) at ca. 298 K. Deposition amounts of the WO x was estimated to be ca. 1.5 µg/(cm2 of substrate) by using a quartz crystal microbalance installed in the UHV chamber. Then, WO x -modified Pt(111) were annealed at 703 K for 10 min. in UHV. Hereafter, the fabricated samples are referred to as p(O2)-WO x /Pt(111). X-ray photoelectron spectroscopy (XPS) was conducted for the as-fabricated samples in situ in UHV.H2O2 generation amounts were evaluated by the tip generation/substrate collection mode of scanning electrochemical microscope (SECM)4 in O2-saturated 0.1 M HClO4. Generated H2O2 was detected by a Pt tip micro-electrode (diameter: ca. 20 µm, tip potential: 1.26 V vs. RHE) positioned ca. 50 µm at surface normal of the sample electrodes, with sweeping sample electrode potential (E S) at sweep rate of 2 mV/s in negative going direction. Thereafter, potential cycles (PCs: 0.05–1.0 V vs. RHE, sweep rate: 100 mV/s, 100 cycles) were applied to the sample surfaces; then, H2O2 generation was re-evaluated. Finally, the PCs-applied sample surfaces were re-introduced to the UHV chamber to perform XPS measurements. Results and Discussion XP spectra of W4f bands for the as-fabricated and PCs-applied sample surfaces are summarized in Figure. 1 (A). Considering peak areas of deconvoluted components (solid lines), the oxidation states of WO x as-fabricated can be judged mainly to be W6+ and Wx+ (x < 4) for 10-1 Pa-WO x /Pt(111) (a) and 10-3 Pa-WO x /Pt(111) (c), respectively.Figure. 1 (B) shows SECM-estimated H2O2 generations of the WO x /Pt(111) surfaces before (as-fabricated) and after the 100 PCs. As for the as-fabricated surfaces (solid lines), H2O2 detection current (i T) of the Pt tip micro-electrode normalized by respective sample electrode current (i S) is lower for 10-1 Pa-WO x /Pt(111) than 10-3 Pa-WO x /Pt(111), yet normalized i T/|i S|-values for both surfaces are smaller than that of clean Pt(111). The results show that surface modification of Pt(111) by the small amount of WO x , particularly with higher oxidation states, effectively suppressed H2O2 generations. However, the suppression effect degraded by the 100 PCs application for both the electrode surfaces (dotted lines). Compared of W4f bands for the PCs-applied surfaces (Figure. 1 (A) (b, d)) with corresponding as-fabricated ones (a, c), the bands shifted to negative- and positive-binding-energies for 10-1 Pa-WO x /Pt(111) and 10-3 Pa-WO x /Pt(111), respectively, indicating respective reduction and oxidation of the surface WO x by the PCs. In addition, W4f bands for both surfaces decreased in intensity by the PCs application, suggesting decrease in surface amount of the WO x , probably through electrochemical dissolution during the PCs. The PCs-induced changes in oxidation states and amounts of surface WO x should correlate to the degraded H2O2 generation suppressions. Acknowledgement This study was supported by the new energy and industrial technology development organization (NEDO) of Japan, JSPS KAKENHI Grant Number JP21H01645, and JST SPRING Grant Number JPMJSP2114. References P. Trogadas and V. Ramani, J. Electrochem. Soc., 155, B696–B703 (2008).M. Anik and K. Osseo-Asare, J. Electrochem. Soc., 149, B224 (2002).A. J. Martín, A. M. Chaparro, and L. Daza, J. Power Sources, 196, 4187–4192 (2011).C. M. Sánchez-Sánchez and A. J. Bard, Anal. Chem., 81, 8094–8100 (2009). Figure 1
Read full abstract