Extended Data Fig. 3: Temperature-controllable photocatalytic OWS system and reaction condition optimization.

a, Temperature-controllable photocatalytic OWS system. b, Schematic illustration of the temperature-controllable photocatalytic OWS system. A double-layer chamber was used to perform the temperature-controllable photocatalytic OWS. The circulating water provided by a PolyScience 7L heated circulator was used to control the temperature of the reaction chamber. c, STH of InGaN/GaN NWs with different Rh/Cr₂O₃/Co₃O₄ precursor volumes at 70 °C. x (x = 2, 3, 4, 5, 6) μl of 0.2 mol l⁻¹ Na₃RhCl₆, x μl of 0.2 mol l⁻¹ K₂CrO₄, x μl of 0.2 mol l⁻¹ Co(NO₃)₂·6H₂O were used in the photodeposition of cocatalyst. The photodeposition method was shown in the experimental section. The optimized contents (µg cm⁻²) of cocatalyst obtained by ICP was shown in the Extended Data Table 1. d, STH of Rh/Cr₂O₃/Co₃O₄-InGaN/GaN NWs under different light intensity at 70 °C. 1 sun: 100 mW cm⁻². The STH of Rh/Cr₂O₃/Co₃O₄-InGaN/GaN NWs first increased with cocatalyst content and then reached a maximum value. A further increase on the cocatalyst content could not improve the STH. The optimized cocatalyst content was utilized in all subsequent experiments. The light intensity on the photocatalyst wafer was adjusted from 1,000 mW cm⁻² to 3,800 mW cm⁻² at 70 °C. The results showed that the STH was not observably changed with the light intensity larger than 13 suns at the same temperature (70 °C). Error bars indicate standard deviation for three measurements.