TY - JOUR
T1 - Performance of the RB3-LYP, RMP2, and UCCSD(T) procedures in calculating radical stabilization energies for •NHX radicals
AU - Wood, Geoffrey P.F.
AU - Henry, David J.
AU - Radom, Leo
PY - 2003/10/2
Y1 - 2003/10/2
N2 - N-H bond dissociation energies (BDEs) and radical stabilization energies (RSEs) associated with the •NHCF3, •NHCHO, •NHCOCH3, and •NHCONH2 radical have been calculated at a number of theoretical levels. These include UHF, RHF, UB3-LYP, RB3-LYP, UMP2, RMP2, UCCSD(T), and URCCSD(T) with a variety of basis sets, as well as the high-level composite methods W1, CBS-QB3, andG3X(MP2)-RAD. For these systems, particular care must be taken to ensure convergence to the lowest-energy solution of the self-consistent-field (SCF) equations. We have assessed the performance of the various levels of theory in calculating the BDEs and RSEs of the •NHX radicals and find that, although there are somewhat larger errors for the simpler methods, the performance generally parallels that observed previously for •CH2X radicals. In particular (and in contrast to a recent report), RB3-LYP and UCCSD(T) consistently produce very good RSEs for •NHX radicals, provided that the lowest-energy solutions are correctly identified. The RMP2 RSEs, while not as good as those for •CH2X radicals, do not show the previously claimed large errors.
AB - N-H bond dissociation energies (BDEs) and radical stabilization energies (RSEs) associated with the •NHCF3, •NHCHO, •NHCOCH3, and •NHCONH2 radical have been calculated at a number of theoretical levels. These include UHF, RHF, UB3-LYP, RB3-LYP, UMP2, RMP2, UCCSD(T), and URCCSD(T) with a variety of basis sets, as well as the high-level composite methods W1, CBS-QB3, andG3X(MP2)-RAD. For these systems, particular care must be taken to ensure convergence to the lowest-energy solution of the self-consistent-field (SCF) equations. We have assessed the performance of the various levels of theory in calculating the BDEs and RSEs of the •NHX radicals and find that, although there are somewhat larger errors for the simpler methods, the performance generally parallels that observed previously for •CH2X radicals. In particular (and in contrast to a recent report), RB3-LYP and UCCSD(T) consistently produce very good RSEs for •NHX radicals, provided that the lowest-energy solutions are correctly identified. The RMP2 RSEs, while not as good as those for •CH2X radicals, do not show the previously claimed large errors.
UR - http://www.scopus.com/inward/record.url?scp=0142123474&partnerID=8YFLogxK
U2 - 10.1021/jp035945s
DO - 10.1021/jp035945s
M3 - Article
SN - 1089-5639
VL - 107
SP - 7985
EP - 7990
JO - Journal of Physical Chemistry A
JF - Journal of Physical Chemistry A
IS - 39
ER -