When hydrogen is depleted in a polymer electrolyte fuel cell, the anode potential increases (cell reversal), which involves the oxidation of supported carbon (COR) and causes the growth of Pt particles, resulting in the deactivation of the anode catalyst. The addition of the oxygen evolution (OER) catalysts has been reported to suppress the degradation1. Though it has also been reported that the combination of Ti4O7 with Ir catalysts increased the reversal tolerance2, the combination of Pt/Ti4O7 and Ti4O7-supported OER catalysts has not been reported. We have reported that Pt catalysts supported on Ti4O7-C synthesized by the carbothermal process showed excellent performance3. In order to investigate how the distribution of the OER catalyst in the catalyst layer affects the reversal-tolerance, we compared the performance of the physical mixture of Pt/Ti4O7-C and Ti4O7-supported OER catalysts with the composite catalysts (Pt and OER catalyst supported on Ti4O7-C) in this study. The hydrogen starvation was carried out during the hydrogen pump tests in order to clarify the degradation on HOR. Experimental Preparation of catalysts: 20 wt% Pt/Ti4O7-C (11 wt% C) was prepared as previously reported3. RuO2 catalysts Ti4O7-C or Pt/Ti4O7-C was impregnated with RuCl3 solution, and precipitated at pH 8 with NaOH solution. The residue was washed, and dried at 150°C. 21 wt% RuO2/Ti4O7-C and 19 wt% Pt-2.0 wt% RuO2/Ti4O7-C were used for the physical mixture (abbreviated to catalyst-M) and for the composite catalyst, respectively. IrO2 catalysts Ti4O7-C or Pt/Ti4O7-C was impregnated in IrCl3 solution, and hydrothermally treated at 180°C. Afterwards, the solution was washed and dried at 80°C. 3.8 wt% IrO2 and 8 wt% Pt-3.7 wt% IrO2/Ti4O7-C were used for the physical mixture and the composite catalyst, respectively. Electrochemical measurement: The MEA was composed of Nafion NRE211 as the electrolyte membrane, a commercial heat-treated c-Pt/C (Pt loading: 0.5 mg-Pt/cm2) as the cathode, and the above-mentioned catalyst or a commercially available c-Pt/C (Pt loading: 0.2 mg-Pt/cm2) as the anode. The RuO2 and IrO2 loadings were 0.021- 0.042 mg/cm2 and are indicated in parentheses. The performance was evaluated with IV curves, the hydrogen pump, CV and LSV by flowing H2 gas to the cathode under 80°C fully humidified condition. The hydrogen starvation was performed by switching the anode supply gas to N2 (RH 100%) while the hydrogen pump was carried out at 0.2 A/cm2 for 10 minutes at a cutoff voltage of 2.7 V. After the test, a hydrogen pump, CV, LSV, and an IV test were performed. The second hydrogen starvation test was repeated in the same manner. Results and Discussion As shown in Table 1, the voltage of c-Pt/C quickly increased above 1.7 V, however, a plateau was observed at 1.8 V after the rapid voltage increase during the second hydrogen starvation, which indicated COR. After the second hydrogen starvation, the current at 0.4 V for the LSV increased from 7.1 mA/cm2 before the second test to 47.9 mA/cm2, indicating the damage of the membrane. The HOR activity also deteriorated after the hydrogen starvation. No improvement was observed for Pt/Ti4O7-C and Pt/Ti4O7-C+RuO2/Ti4O7-M (0.025) under this condition, however, the voltage rise to 1.7 V on the RuO2 composite catalyst was suppressed for the first starvation, due to the OER. This effect disappeared for the second hydrogen starvation. On the other hand, the catalyst physically mixed with IrO2/Ti4O7-C showed a voltage plateau from 1.50 to 1.75 V during the hydrogen starvation, indicating a significant improvement, and the increase in the IrO2 loading improved the reversal-tolerance. Furthermore, for the IrO2 composite catalyst, a plateau due to the OER was observed at 1.5 - 1.7 V for the two hydrogen starvation tests. The voltage increase of the hydrogen pump after the tests was also slight and the change in ECSA was small. Since the difference from the physical mixture was clearly observed, it was deduced that the OER catalyst in the vicinity of the Pt catalyst improved the reversal-tolerance.Acknowledgments: This work was supported by NEDO.(1) T. R. Ralph et al., Platinum Metals Rev. , 46, 117 (2002). (2) T. Ioroi et al., J. Power Sources, 450, 227656 (2020). (3) M. Chisaka et al., Chem. Comm. , 57, 12772 (2021). Figure 1