Daraksha Bano, Shanshan Qin, Hyesung Kim, Xin Zhou, Xin Zhou, Xuemei Zhou, Xuemei Zhou, Patrik Schmuki
Identifying whether platinum (Pt) nanoparticles (NPs) or Pt single atoms (SAs) actually drive the cocatalytic effect in photocatalytic H 2 evolution on TiO 2 is still under debate, in part because trace SAs are difficult to quantify on NP-decorated surfaces. Using sputtered anatase TiO 2 thin films as a planar model platform, we prepared three motifs─SA-only, NP-only, and NP&SA─via reactive deposition from H 2 PtCl 6 (SAs) and thermal dewetting (NPs). Quantitative HAADF-STEM enabled SA counting and NP metrology, while XPS differentiated oxidized Pt δ+ from metallic Pt 0 . Under 365 nm illumination in 50% MeOH/H 2 O, the H 2 evolution rate scales linearly with the measured SA density across all samples. The activity of nominally NP-only films (but trace SAs present) is quantitatively reproduced by an SA-only film with the same SA density, demonstrating that residual SAs fully account for the observed rate. Within our preparation and detection limits, we find no measurable contribution from Pt NPs or SA–NP synergy. These results suggest that undetected SAs may explain a significant portion of apparent NP activity reported in earlier studies. Because truly SA-free NP surfaces are difficult to realize and metastable SAs may form under operating conditions, these results suggest that some reported NP activity on Pt/TiO 2 likely originates from undetected SAs, underscoring the need to report trace SA densities when benchmarking Pt cocatalysts.