DFT Screening on the Geometries and Electronic Properties of Organic Mixed-Valence Systems: Impact of Exchange-Correlation Functionals
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This work evaluates the performance of 26 exchange-correlation functionals across three Robin-Day class II/III borderline organic mixed-valence (MV) systems (TBQ, HBQ, and DphD) focusing on geometries, redox potentials, IVCT excitation energies, and electronic couplings (H12). Calculated geometries demonstrated a tendency for most functionals to favor delocalized structures even for systems experimentally classified as class II or borderline class II/III. BMK-D3 exhibited most reliable performance of planar and linear systems but fails to reproduce π-stacked system. Global hybrid functionals with moderate exact exchange (27%–54%) provided superior redox potential predictions, with M06-D3 being the most reliable. Functionals with high exact exchange (54%) often predicted potential inversion, limiting their utility for MV systems. For IVCT excitation energies, low-to-moderate exact exchange (15–35%) functionals, such as M06-D3 and mPW1PW91, were most accurate. These findings highlight the importance of functional selection in modeling MV systems and position M06-D3 as a versatile choice for balancing accuracy across properties.
Abstract
This work evaluates the performance of 26 exchange-correlation functionals across three Robin-Day class II/III borderline organic mixed-valence (MV) systems (TBQ, HBQ, and DphD) focusing on geometries, redox potentials, IVCT excitation energies, and electronic couplings (H12). Calculated geometries demonstrated a tendency for most functionals to favor delocalized structures even for systems experimentally classified as class II or borderline class II/III. BMK-D3 exhibited most reliable performance of planar and linear systems but fails to reproduce π-stacked system. Global hybrid functionals with moderate exact exchange (27%–54%) provided superior redox potential predictions, with M06-D3 being the most reliable. Functionals with high exact exchange (54%) often predicted potential inversion, limiting their utility for MV systems. For IVCT excitation energies, low-to-moderate exact exchange (15–35%) functionals, such as M06-D3 and mPW1PW91, were most accurate. These findings highlight the importance of functional selection in modeling MV systems and position M06-D3 as a versatile choice for balancing accuracy across properties.