Harmonic Scale Factors of Fundamental Transitions for Dispersion‐corrected Quantum Chemical Methods

This work provides a procedure and database for obtaining the vibrational frequency scale factors that align quantum chemically computed harmonic frequencies with experimental vibrational spectroscopic data. The database comprises 441 molecules of various sizes, from diatomics to the buckminsterfull...

Full description

Saved in:
Bibliographic Details
Published inChemphyschem Vol. 25; no. 23; pp. e202400547 - n/a
Main Authors Tikhonov, Denis S., Gordiy, Igor, Iakovlev, Danila A., Gorislav, Alisa A., Kalinin, Mikhail A., Nikolenko, Sergei A., Malaskeevich, Ksenia M., Yureva, Karina, Matsokin, Nikita A., Schnell, Melanie
Format Journal Article
LanguageEnglish
Published Germany Wiley Subscription Services, Inc 02.12.2024
John Wiley and Sons Inc
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:This work provides a procedure and database for obtaining the vibrational frequency scale factors that align quantum chemically computed harmonic frequencies with experimental vibrational spectroscopic data. The database comprises 441 molecules of various sizes, from diatomics to the buckminsterfullerene C60. We provide scale factors for 27 dispersion‐corrected methods, 24 of which are DF‐Dn/B with DF=BLYP, PBE, B3LYP, PBE0, Dn=D3(BJ), D4, and B=6‐31G, def2‐SVP, def2‐TZVP, and three of them are the 3c‐family composite methods (HF‐3c, PBEh‐3c, and r2SCAN‐3c). The two scale factors are derived for each method: the absolute scaling, minimizing the absolute deviation of the scaled harmonic frequency from the experimental value, and the relative scaling, which minimizes an analogous relative deviation. The absolute type of scaling is recommended for frequencies above 2000 cm−1, while the relative scaling is optimal for frequencies below 2000 cm−1. This study presents two new sets of scaling factors for adjusting theoretical harmonic vibrational spectra computed using dispersion‐corrected density functional theory methods to better match experimental data. The first set (absolute scaling) better aligns with higher frequencies, while the second one (relative scaling) evenly reduces errors across the entire spectral range.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1439-4235
1439-7641
1439-7641
DOI:10.1002/cphc.202400547