Improving the accuracy of solid-state nuclear magnetic resonance chemical shift prediction with a simple molecular correction.

摘要: A fast, straightforward method for computing NMR chemical shieldings of crystalline solids is proposed. The combines the advantages both conventional approaches: periodic calculations using plane-wave basis sets and molecular computational approaches. capture nature solids, but level electronic structure calculation limited to general-gradient-approximation (GGA) density functionals. It demonstrated that a correction GGA result calculated on an isolated molecule at higher theory significantly improves correlations between experimental shifts while adding almost no additional cost. Corrections with hybrid functional improved accuracy 13C, 15N 17O shift predictions allowed identifying errors in previously published data. Applications approach isocytosine, methacrylamide, testosterone are presented.