A new chapter in the never ending story of cycloadditions: The puzzling case of SO₂ and acetylene

Abstract

A comprehensive study of the different classes of cycloaddition reactions ([3+2], [2+2], and [2+1]) of SO2 to acetylene and ethylene has been performed using density functional theory (DFT) and composite wavefunction methods. The [3+2] cycloaddition reaction, that was previously explored in the context of the cycloaddition of thioformaldehyde S‐methylide (TSM) to ethylene and acetylene, proceeds in a concerted way to the formation of stable heterocycles. In this paper, we extend our study to the [2+2] and [2+1] cycloadditions of SO2 to acetylene, which would produce 1,1‐oxathiete‐2‐oxide and thiirene‐1,1‐dioxide, respectively. One of the main conclusions is that cyclic 1,1‐oxathiete‐2‐oxide can open through a relatively easy breaking of the SO single bond and rearrange toward sulfinyl acetaldehyde (SA). The SA molecule can easily undergo several internal rearrangements, which eventually lead to sulfenic acid and sulfoxide derivatives of ethenone, 1,2,3‐dioxathiole, and CO plus sulfinylmethane. The most probable path, however, produces 2‐thioxoacetic acid, whose derivatives (or those of the corresponding acetate) are usually obtained by Willgerodt–Kindler‐type sulfuration of acetates. This product can in turn decompose, leading to the final products CO2 and H2CS. Comparison of this decomposition path with that of 2‐amino‐2‐thioxoacetic acid shows that the process occurs through different H‐transfer processes.

Energetics of the OH radical H-abstraction reactions from simple aldehydes and their geminal diol forms

“…The other two DFT methods, M06-2X-D3 and PW6B95-D3, are over this threshold on more occasions. This shows that, while the DFT methods here used to obtain the equilibrium geometry of the stable bound species on top of which the other calculations are performed are equivalent (provided there are no multi-conformation or degeneracy problems, as we showed in previous publications [44,63]) not all of them are equally useful to obtain relative energies, unless further corrections are carried on. The size of the average error of the three DFT procedures, B2PLYP (3.0 kJ mol -1 ), M06-2X-D3 (3.5 kJ mol -1 ), and PW6B95-D3 (7.5 kJ mol -1 ), shows that these methods are reasonably appropriate for this type of study on larger molecules for which the coupled-cluster calculations cannot be performed, although B2PLYP should be the method of choice, resources allowing.…”