Direct synthesis of hydrogen peroxide (DSHP) from H
2 and O
2 is a promising alternative to the current industrial method for the manufacture of H
2O
2, the Riedl-Pfleiderer process. H
2O
2 does not accumulate to the required industrial-grade concentration with current catalytic technology because of significant secondary reactions (H
2O
2 decomposition and hydrogenation). Most studies focus on increasing H
2O
2 selectivity by alloying Pd with another metals (e.g., Au, Sn, Zn, Pt)
1. Plauck et al.
2 used DFT and found O-O bond break is rate determining for Pd-catalyzed H
2O
2 decomposition, suggesting a secondary KIE (1~2) should be observed. However, a large primary kinetic isotope effect (KIE) of ~7 was measured for H
2O
2 decomposition. A step involving O-O bond breaking would at a minimum require a pair of sites, leading to an inhibitory effect of H
2O
2 at a high H
2O
2 concentration, but our experimental work demonstrates Pd-catalyzed H
2O
2 decomposition follows saturation kinetics. The inconsistency between experiment and theory may be attributed to a solvent-assisted pathway. Proton electron transfer, a reaction involving the concerted transfer of electrons and protons, typically has a large KIE, and is identified as a kinetically relevant step. Larger particles with smaller work function have increased reactivity towards H
2O
2 decomposition due to increased electron transfer to H
2O
2 relative to smaller Pd particles, supporting a proposed mechanism involving kinetically relevant electron transfer steps. The addition of a Pd catalyst to Fenton system (Fe
3+/Fe
2+ catalyzed H
2O
2 decomposition with HO⢠formation) accelerates the formation rate of hydroxybenzoic acid even though Pd catalyst does not catalyze hydroxylation of benzoic acid with H
2O
2. Through the kinetic coupling of Pd-catalyzed H
2O
2 decomposition and Fenton probe chemistry, we demonstrated the most abundant reactive intermediate are *HOO and *H
2O
2, their relative concentrations vary with particle size during reaction.
(1) ACS Catal. 2018, 8, 1520â1527.(2) Proc. Natl. Acad. Sci. 2016, E1973âE1982.