Structure, Energy, and Vibrational Frequencies of Oxygen Allotropes O_n (n ≤ 6) in the Covalently Bound and van der Waals Forms: Ab Initio Study at the CCSD(T) Level

Abstract

Recent experiments on the UV and electron beam irradiation of solid O 2 reveals a series of IR features near the valence antisymmetric vibration band of O 3 which are frequently interpreted as the formation of unusual O n allotropes in the forms of weak complexes or covalently bound molecules. In order to elucidate the question of the nature of the irradiation products, the structure, relative energies, and vibrational frequencies of various forms of O n (n = 1−6) in the singlet, triplet, and, in some cases, quintet states were studied using the CCSD(T) method up to the CCSD(T,full)/cc-pCVTZ and CCSD(T,FC)/aug-cc-pVTZ levels. The results of calculations demonstrate the existence of stable highly symmetric structures O 4 (D 3h ), O 4 (D 2d ), and O 6 (D 3d ) as well as the intermolecular complexes O 2 ·O 2 , O 2 ·O 3 , and O 3 ·O 3 in different conformations. The calculations show that the local minimum corresponding to the O 3 ···O complex is quite shallow and cannot explain the ν 3 band features close to 1040 cm −1 , as was proposed previously. For the ozone dimer, a new conformer was found which is more stable than the structure known to date. The effect of the ozone dimer on the registered IR spectra is discussed.

An exploration of the ozone dimer potential energy surface

“…Their minimum, with C s symmetry, has a dimerization energy of 3.0 kcal/mol. A more recent paper by Gadzhiev et al 21 reproduces satisfactorily the experimental behavior of O 3 and its homodimer at the CCST(T,full)/cc-pCVTZ level. They identified two minima: one similar to that proposed by Slanina and Adamowicz 20 and a C i geometry, more stable by 0.25 kcal/mol.…”

Multiconfigurational short-range density-functional theory for open-shell systems

“…In general, the dependence on µ is moderate (at most a few kJ/mol) in the interval µ = 0.1-0.5. The difference is larger (≈25-30 kJ/mol) between the results obtained with the srLDA and srPBE functionals; thus there is a significant effect Literature CCSD(T)/cc-pVTZ 52 124.66 CCSDT/cc-pVTZ 52 114.34 CCSDTQ/cc-pVTZ 52 99.48 MRCI 51 111.61 MRCI/cc-pVTZ 52 98.40 QMC 53 96.58…”

“…27 It should be emphasized that although O 2 is a prototypical open-shell system for which a wealth of literature data is available, 50 the spin-state splitting between the 3 Σ − g ground-state and the first singlet state, 1 ∆ g , is surprisingly demanding in terms of the theoretical treatment. [51][52][53] Likewise, the spin-state splitting between the lowest electronic states, 6 A g and 4 T 1g (assuming idealized, octahedral symmetry) in [Fe(H 2 O) 6 ] 3+ , has turned out to be challenging; there is currently no consensus on which method is the most accurate, as both CASPT2 and CCSD(T) have been shown to overestimate the spin-state splitting somewhat. 54 Accordingly, both O 2 and [Fe(H 2 O) 6 ] 3+ have low-lying excited states with different spin multiplicity than the ground-state, and for both a high-level method is evidently needed to obtain correct results.…”

Properties of reactive oxygen species by quantum Monte Carlo

“…The quantitative evaluation of the energy gap between the triplet ground state and the first singlet excited state is a challenging for quantum chemistry approaches, and is estimated correctly only by the most accurate methods, as for instance CCSDTQ in Ref. 85. Nevertheless, a correct evaluation of this energy gap is fundamental for the description of the reactivity of singlet oxygen, which is one of the most dangerous and reactive oxygen species.…”