Correlated ab initio calculations of one‐bond ³¹P⁷⁷Se and ³¹P¹²⁵Te spin–spin coupling constants in a series of PSe and PTe systems accounting for relativistic effects (part 2)

Abstract

Synthetic chalcogen–phosphorus chemistry permanently makes new challenges to computational Nuclear Magnetic Resonance (NMR) spectroscopy, which has proven to be a powerful tool of structural analysis of chalcogen–phosphorus compounds. This paper reports on the calculations of one‐bond 31P77Se and 31P125Te NMR spin–spin coupling constants (SSCCs) in the series of phosphine selenides and tellurides. The applicability of the combined computational approach to the one‐bond 31P77Se and 31P125Te SSCCs, incorporating the composite nonrelativistic scheme, built of high‐accuracy correlated SOPPA (CC2) and Coupled Cluster Single and Double (CCSD) methods and the Density Functional Theory (DFT) relativistic corrections (four‐component level), was examined against the experiment and another scheme based on the four‐component relativistic DFT method. A special J‐oriented basis set (acv3z‐J) for selenium and tellurium atoms, developed previously by the authors, was used throughout the NMR calculations in this work at the first time. The proposed computational methodologies (combined and ‘pure’) provided a reasonable accuracy for 31P77Se and 31P125Te SSCCs against experimental data, characterizing by the mean absolute percentage errors of about 4% and 1%, and 12% and 8% for selenium and tellurium species, respectively. The present study reports typical relativistic corrections to 77Se31P and 125Te31P SSCCs, calculated within the four‐component DFT formalism for a broad series of tertiary phosphine selenides and tellurides with different substituents at phosphorus.

Quantum chemical calculations of77Se and125Te nuclear magnetic resonance spectral parameters and their structural applications

“…In the subsequent work, [386] Rusakova and Rusakov examined the performance of J-oriented acv3z-J basis set for tellurium atom on the example of one-bond 31 P- 125 Te NMR SSCCs in a series of phosphine tellurides with different functional groups. The calculations were performed using the composite nonrelativistic scheme implying that the SOPPA(CC2) values were summed with the CCSD corrections.…”

“…The acv3z-J basis set was set on tellurium atom in all NMR calculations. The combined "relativistically corrected" computational scheme with composite nonrelativistic part, applied to the 31 P- 125 Te SSCCs in Rusakova and Rusakov, [386] provided the MAE = 206.3 Hz and MAPE = 12%. The "pure" relativistic four-component DFT(KT1) scheme was examined against that combined computational scheme, and the statistical characteristics of the former occurred to be better than that of the latter.…”

“…The performance of J-oriented acv3z-J basis set for selenium has been recently scrutinized on the example of one-bond 31 P- 77 Se NMR SSCCs of phosphine selenides with different functional groups by Rusakova and Rusakov. [386] The acv3z-J basis set was tested in that work as an integrative part of the computational methodology, based on the idea of separate evaluation of nonrelativistic values and their relativistic corrections (the combined approach). In that way, the proposed methodology implied the usage of accurate and efficient nonrelativistic composite approximation for the basic nonrelativistic values of phosphorus-selenium SSCCs and the fully relativistic four-component DFT-KT1 approach for their relativistic corrections.…”

“…In that way, the proposed methodology implied the usage of accurate and efficient nonrelativistic composite approximation for the basic nonrelativistic values of phosphorus-selenium SSCCs and the fully relativistic four-component DFT-KT1 approach for their relativistic corrections. The nonrelativistic composite approximation, applied to phosphorus-selenium SSCCs in Rusakova and Rusakov, [386] was based on the idea of Reid and Collins, proposed very recently for the NMR shieldings. [387] The composite methodology of Reid and Collins was conceived so as to approximate the value of SC, that could be obtained with a very computationally expensive high-quality method combined with a large basis set, as a combination of the value obtained with a much less demanding method and the large basis set and the corresponding correlation correction, calculated using a smaller basis set.…”

On the Utmost Importance of the Basis Set Choice for the Calculations of the Relativistic Corrections to NMR Shielding Constants

“…It has become common practice to combine high-quality quantum chemical modeling of spectra with NMR experiments. In this respect, much effort has been put into the development of accurate and efficient quantum chemical computational protocols for NMR parameters [ 1 , 2 , 3 ], and special attention has been paid to the effects of special relativity manifesting in the NMR parameters of compounds containing heavy elements [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ]. In particular, these may have a significant impact not only on the NMR shielding constants of heavy nuclei [ 8 , 13 , 14 , 15 , 16 ], but also on those of the light nuclei [ 17 , 18 , 19 , 20 , 21 , 22 , 23 ].…”