Reaction barrier heights for cycloreversion of heterocyclic rings: An Achilles’ heel for DFT and standard ab initio procedures

Abstract

a b s t r a c tWe introduce a database of 20 accurate cycloreversion barrier heights of 5-membered heterocyclic rings (to be known as the CRBH20 database). In these reactions, dioxazole and oxathiazole rings are fragmented to form isocyanates, isothiocyanates, and carbonyls. The reference reaction barrier heights are obtained by means of the high-level, ab initio W1-F12 and W1w thermochemical protocols. We evaluate the performance of 65 contemporary density functional theory (DFT) and double-hybrid DFT (DHDFT) procedures. The CRBH20 database represents an extremely challenging test for these methods. Most of the conventional DFT functionals (74%) result in root-mean-square deviations (RMSDs) between 10 and 81 kJ mol À1 . The rest of the DFT functionals attain RMSDs = 5 À 10 kJ mol À1 . Of the 12 tested DHDFT functionals, only five result in RMSDs < 10 kJ mol À1 . The CRBH20 dataset also proves to be a surprisingly challenging target for composite and standard ab initio procedures.

The INV24 test set: how well do quantum-chemical methods describe inversion and racemization barriers?

“…62,65 Therefore, this level of theory is ideal to obtain accurate benchmarks for the assessment of lower-level methods, as previously shown for barrier heights in different chemical processes. 56,57,59 In fact, it has been demonstrated that W2-F12 outperforms other popular composite wave-function approaches for barrier heights of pericyclic reactions. 56 For the water molecule, we can directly compare the W2-F12 result with an estimate for the experimental barrier to linearity, which has been given as 10 967 cm -1 (31.4 kcal/mol).…”

“…To facilitate a better understanding of this method zoo, comprehensive benchmark databases have been developed that cover several chemical problems simultaneously, such as the Gn test sets, [44][45][46][47] the GMTKN24 and GMTKN30 databases for general main group thermochemistry, kinetics and noncovalent interactions, 48,49 and a variety of other comprehensive sets, such as the ones reviewed in Peverati and Truhlar. 50 In the context of reaction barrier heights, it is worthwhile to particularly highlight stand-alone test sets for hydrogen transfer, heavy-atom transfer, nucleophilic substitution, unimolecular, and association reactions (the HTBH38, 51 NHTBH38, 52 and DBH24/08 [53][54][55] sets), for pericyclic reactions (BHPERI 48,56 and CRBH20 57 ), for watercatalyzed proton-transfer reactions, 58 for proton exchange in water, ammonia, and hydrogen fluoride clusters, 59 and rotational barriers in bifuranes and their analogues. 60 Except for a limited study of fullerene fragments in 1999, 61 no comprehensive benchmark study of inversion barriers has been published.…”

Time-Dependent Double-Hybrid Density Functionals with Spin-Component and Spin-Opposite Scaling

“…[12][13][14] DHDFAs are currently regarded as the most sophisticated and most accurate DFAs for molecular thermochemistry, surpassing popular wave-function approaches, such as Møller-Plesset perturbation theory or even the composite complete-basis-set (CBS-QB3) approach. 8,[15][16][17][18][19] Due to its success for electronic ground states, the DHDFA approach has also been extended to the computation of electronic excitation energies. A first solution suggested by Grimme and Neese was to transfer the configuration interaction singles with perturbative doubles 20 (CIS(D)) approach to a time-dependent DFT (TD-DFT) formalism in 2007 (TD-DHDFAs).…”

A computational investigation of the sulphuric acid‐catalysed 1,4‐hydrogen transfer in higher Criegee intermediates

“…G4(MP2) theory has been found to produce thermochemical properties (such as reaction energies, bond dissociation energies, and enthalpies of formation) with a mean absolute deviation of 4.4 kJ mol −1 from the 454 experimental energies of the G3/05 test set . It has also been found that G4(MP2) shows a similarly good performance for reaction barrier heights . Finally, we note that the main text primarily looks at Δ H 298 values, however, similar trends in the reactivity and catalytic activity are observed on the Gibbs free energy (Δ G 298 ) reaction profiles (for further details see Figures S2‐S4 of the Supporting Information).…”