Equilibrium Structures of Heterocyclic Molecules with Large Principal Axis Rotations upon Isotopic Substitution

Abstract

Equilibrium structures, r(e), of the heterocyclic molecules oxirane, furazan, furan, ethylene ozonide, and 1,3,4-oxadiazole have been determined using three different, somewhat complementary techniques: a completely experimental technique (r(m)), a semiexperimental technique (r(e)(SE), whereby equilibrium rotational constants are derived from experimental effective ground-state rotational constants and corrections based principally on an ab initio cubic force field), and an ab initio technique (r(e)(BO), whereby geometry optimizations are usually performed at the coupled cluster level of theory including single and double excitations augmented by a perturbational estimate of the effects of connected triple excitations [CCSD(T)] using quadruple-ζ Gaussian basis sets). All these molecules are asymmetric tops with the moment of inertia I(c) much larger than the other two moments of inertia, I(a) and I(b). Molecules of this shape experience a large rotation of the principal axis system upon certain isotopic substitutions. For such isotopologues it is difficult to obtain a good structural fit to the semiexperimental moments of inertia I(a) and I(b), which may significantly reduce the accuracy of the r(e)(SE) structural parameters. The origin of this difficulty is explained. For the heavy-atom skeleton of these molecules it was possible to determine a rather accurate empirical mass-dependent structure without a priori knowledge of the equilibrium structure.

Why it is sometimes difficult to determine the accurate position of a hydrogen atom by the semiexperimental method: Structure of molecules containing the OH or the CH3 group

“…Thus, it should be difficult to determine an accurate SE structure. However, the C and CD 2 isotopic substitutions do not produce a rotation of the PAS and the first‐moment equations make up for the lack of isotopic substitution for fluorine. Furthermore, it is a planar molecule with C 2 v symmetry and, therefore, its structure is completely determined by only five parameters while accurate ground‐state rotational constants are available for five isotopologues .…”

“…The potential hindering the internal rotation of the methyl group is not high, V 3 = 372.7 cm −1 ; thus it produces large splittings of the rotational transitions. The lowest‐order experimental vibration‐rotation interaction constants ( α ‐constants) have been determined for the torsional mode of HCOOCH 3 and HCOOCH 3 . They were also obtained from an ab initio cubic force field calculated at the MP2/VTZ level and a good agreement was found.…”

“…The potential energy surface and the torsional spectrum were calculated at the MP4/VQZ level of theory and the molecule was found moderately rigid with two low‐frequency vibrations, the angle CCO bend at 213 cm −1 and the OH torsion at 294 cm −1 . Recently, due to its astrophysical interest, the rotational spectrum was extensively studied, including that of the C species, the O species, and the deuterated species . We now have a complete set of accurate ground‐state rotational constants for nine isotopologues.…”

“…The error of the rovibrational correction can then be as large as, or even larger than, the isotopic shift of the equilibrium moment of inertia and the corresponding coordinate is then ill determined as, for instance, for the hydrogen atoms of dimethyl sulfide, (CH 3 ) 2 S or dimethyl sulfoxide, and (CH 3 ) 2 SO A third reason is a large rotation of the PAS on isotopic substitution which amplifies the errors of the rovibrational corrections . This is particularly important for oblate tops.…”

Computational Evidence Suggests That 1-Chloroethanol May Be an Intermediate in the Thermal Decomposition of 2-Chloroethanol into Acetaldehyde and HCl

“…Meanwhile, the best experimental and theoretical determination of the structure of oxirane can be found in a paper by Demaison et al [38]. Our B97X-D/cc-pVTZ calculations afford results for the CC and CH bonds very near the experimental ones, but a too short CO bond which, on the contrary, is well predicted at the highest level of theory used by Demaison et al [38].…”