Water splitting photocatalysts based on semiconducting single-walled carbon nanotubes (s-SWCNTs) have been attracted much attention from the viewpoint of solar fuel production.[1-7] Since s-SWCNTs have chirality dependent bandgap covering a majority of the solar spectrum (400 – 2000 nm), s-SWCNTs are expected for an efficient solar-light absorber.[8] However, upon photoirradiation, the large exciton binding energy in s-SWCNTs prevents mobile carrier generation. Then, in order to make a particulate photocatalyst consisting of s-SWCNTs, an electron-extracting material is necessary. In this context, we have developed CNT-photocatalysts that contain a s-SWCNT/C60 coaxial heterojunction fabricated by a physical modification of s-SWCNTs using fullerodendron.[4] Here, the C60 acts as the electron-extracting material to generate a charge-separated state, s-SWCNT•+/C60 •–, followed by electron transfer to a co-catalyst that produces hydrogen from water. Z-scheme photocatalyst system employing BiVO4 as an oxygen evolution photocatalyst (OEP) and s-SWCNT/C60 coaxial hybrids as a hydrogen evolution photocatalyst (HEP) produced H2 and O2 from water, of which solar-to-hydrogen conversion efficiency (STH) was 0.089%.[7] We conducted an H2 evolution reaction using s-SWCNT/C60/Ru(III) upon monochromatic irradiation at 570, 650, and 680 nm, which are E22 absorption of (6,5), (7,5), and (8,3)tubes, respectively, and estimated the external quantum yields to be 0.22% for (6,5), 0.16% for (7,5), and 0.70% for (8,3). These EQY values were considerably lower than that using 1000-nm-light (E11 absorption of (6,5) and (8,3)tubes) irradiation, 12.8%.[7] The result might be attributed to branching pathways from the higher exciton (E22 excitation) to other than the lower exciton (E11 excitation). Blackburn and co-workers reported such a branching ratio is crucial for the PL quantum yield of s-SWCNTs.[9] Cui and co-workers describe E22 excitation makes interfacial electrotransfer inhibited in the (6,5)tube/C60 heterojunction, although ultrafast electron transfer process can be seen in E11 excitation by a DFT-based nonadiabatic dynamics simulation.[10] In this context, direct electron extraction from E22 excitation, which is faster than other branching pathways, is expected to improve the photocatalytic activity of CNT-photocatalysts. Recently, Blackburn reported higher-energy second excitonic s-SWCNT transitions produce more photocurrent by the use of submonolayer coverages of s-SWCNTs on single crystal TiO2 electrode, demonstrating carrier injection rates are competitive with fast hot-exciton relaxation processes.[11] These backgrounds prompt us to investigate a CNT-photocatalyst containing the s-SWCNT/TiO2 heterojunction. This paper describes fabrication of s-SWCNT/TiO2/Pt nanohybrids and their photocatalytic activity for H2 evolution from water. Interestingly, their EQYs of H2 evolution reaction using E22 excitation, 12% for (6,5)tube/TiO2/Pt and 43% for (8,3)tube/TiO2/Pt is much higher than those of (6,5)tube/C60/Ru(III) and (8,3)tube/C60/Ru(III), 0.22% and 0.70%.