Supplementary material for González-Luque (2000) Proc. Natl. Acad. Sci. USA 97 (17), 9379-9384.
Table 3.
S0, S1, S2 state averaged CASSCF and CASPT2 total and relative (DE) energies for the PSB11 model 1Structure | State | CASSCF au | CASPT2 au | w *† | D E eV(kcal mol-1) |
Planar FC | S 0 | -441.14129 | -442.43024 | 0.680 | 0.00 (0.00) |
S 1 | -441.01678 | -442.33585 | 0.663 | 2.57 (59.3) | |
S 2 | -440.98174 | -442.29793 | 0.662 | 3.60 (83.0) | |
Planar SP | S 0 | -441.11745 | -442.41907 | 0.674 | 0.30 (7.0) |
S 1 | -441.03140 | -442.34230 | 0.666 | 2.39 (55.2) | |
S 2 | -440.99894 | -442.31735 | 0.655 | 3.07 (70.8) | |
Twist 8° | S 0 | -441.11771 | -442.41923 | 0.675 | 0.30 (6.9) |
S 1 | -441.03144 | -442.34220 | 0.666 | 2.40 (55.4) | |
S 2 | -440.99895 | -442.31737 | 0.644 | 3.07 (70.8) | |
Twist 18° | S 0 | -441.11491 | -442.41689 | 0.674 | 0.36 (8.4) |
S 1 | -441.03193 | -442.34276 | 0.666 | 2.38 (54.9) | |
S 2 | -440.99632 | -442.31532 | 0.155‡ | 3.12 (72.1) | |
Twist 44° | S 0 | -441.10537 | -442.40855 | 0.673 | 0.59 (13.6) |
S 1 | -441.03556 | -442.34661 | 0.666 | 2.28 (52.6) | |
S 2 | -440.98781 | -442.30735 | 0.512‡ | 3.34 (77.0) | |
Twist 68° | S 0 | -441.08366 | -442.38714 | 0.671 | 1.17 (27.0) |
S 1 | -441.04185 | -442.35860 | 0.666 | 1.95 (45.0) | |
S 2 | -440.97429 | -442.28905 | 0.411‡ | 3.84 (88.6) | |
Twist 90° CI | S 0 | -441.07403 | -442.36984 | 0.672 | 1.64 (37.8) |
S 1 | -441.04919 | -442.37170 | 0.665 | 1.59 (36.7) | |
S 2 | -440.97446 | -442.27591 | 0.224‡ | 4.20 (96.9) |
The basis set used is ANO-S (see Computational Methods for details).
*Weight of the CASSCF reference function (i.e., the zero-order function) in the first-order function.
†
CASPT2 refers to standard CASPT2 results with no level shift applied. In four cases, due to the occurrence of intruder states, calculations using a level shift of 0.1 au also were carried out [CASPT2 (LS = 0.1 au)]. It was done to check how "reasonable" the standard CASPT2 results are when the reference weight is low (‡ results). It was concluded, that in spite of the low weights, the corresponding results can be confidently used. Another possibility is to use the LS = 0.1 au results for those (‡) cases. A full calibration vs. the variation of the LS parameter (LS = 0.0, 0.1, 0.2, 0.3, 0.4) has not been performed.