A Study of H₂O₂ with Threshold Photoelectron Spectroscopy (TPES) and Electronic Structure Calculations: Redetermination of the First Adiabatic Ionization Energy (AIE)

Abstract

In this work, hydrogen peroxide has been studied with threshold photoelectron (TPE) spectroscopy and photoelectron (PE) spectroscopy. The TPE spectrum has been recorded in the 10.0-21.0 eV ionization energy region, and the PE spectrum has been recorded at 21.22 eV photon energy. Five bands have been observed which have been assigned on the basis of UCCSD(T)-F12/VQZ-F12 and IP-EOM CCSD calculations. Vibrational structure has only been resolved in the TPE spectrum of the first band, associated with the X̃(2)Bg H2O2(+) ← X̃(1)A H2O2 ionization, on its low energy side. This structure is assigned with the help of harmonic Franck-Condon factor calculations that use the UCCSD(T)-F12a/VQZ-F12 computed adiabatic ionization energy (AIE), and UCCSD(T)-F12a/VQZ-F12 computed equilibrium geometric parameters and harmonic vibrational frequencies for the H2O2 X̃(1)A state and the H2O2(+) X̃(2)Bg state. These calculations show that the main vibrational structure on the leading edge of the first TPE band is in the O-O stretching mode (ω3) and the HOOH deformation mode (ω4), and comparison of the simulated spectrum to the experimental spectrum gives the first AIE of H2O2 as (10.685 ± 0.005) eV and ω4 = (850 ± 30) and ω3 = (1340 ± 30) cm(-1) in the X̃(2)Bg state of H2O2(+). Contributions from ionization of vibrationally excited levels in the torsion mode have been identified in the TPE spectrum of the first band and the need for a vibrationally resolved TPE spectrum from vibrationally cooled molecules, as well as higher level Franck-Condon factors than performed in this work, is emphasized.

Active Thermochemical Tables: The Adiabatic Ionization Energy of Hydrogen Peroxide

“…Note that the present determination is in fact very close to an ab initio calculation documented in ref . On the basis of the UCCSD­(T*)-F12 approach, these workers obtained a best estimate adiabatic ionization energy of 10.66 eV, which is in good agreement with the present calculation.…”

“…Simulated photoelectron spectra calculated at 200 and 300 K are shown in Figure a,b, respectively, along with the data from ref . In both cases, the origin band from the simulation was shifted to coincide with the band designated as “ a ” at 10.649 eV from the original study, which represents the alternative assignment of Schio et al discussed in the Introduction. Clearly, agreement with the experimental spectrum is qualitatively excellent at both temperatures, although the 200 K simulation appears to provide better quantitative agreement.…”

“…The ATcT value for the ionization energy of HOOH from this (provisional) version of the TN is 10.642 ± 0.006 eV. While considering both values derived from the photoelectron study of Schio et al, 29 the ATcT statistical analysis finds that the ionization energy derived from the preferred value (10.678 ± 0.008 eV) is still much too high (in spite of the fact that it was additionally shifted to lower energy by 7 meV) and could be made consistent with the other determinations in the TN only if its uncertainty is augmented to ±0.037 eV. Thus, the ATcT analysis independently confirms the conjecture that the alternate value of Schio et al is a significantly better choice.…”

“…The latter overlaps the ATcT error range, although just barely so, but is outside the ranges given by refs and . However, the authors of ref state that their determination is far from certain and hope that their study will “stimulate ... interest in this problem to test the assignments presented and the derived AIE”. This work takes up that challenge.…”

“…The long progression in the bending mode ν 3 begins with feature a′ . This interpretation of the spectrum having its origin at 10.649 eV was originally rejected in ref for the following reasons. Assuming this position of the origin, the authors of ref assigned features b and a′ to 4 0 1 (−) and 3 0 1 (+) (as is consistent with the present assignment) but discarded this possibility, since there are “no clearly resolved features” to the red of b and a′ that can be attributed to the corresponding torsional hot bands.…”