From chemistry-request "-at-" server.ccl.net Mon Oct 2 05:30:57 2000 Received: from mel.ruc.dk (mel.ruc.dk [130.225.220.27]) by server.ccl.net (8.8.7/8.8.7) with ESMTP id FAA31574 for ; Mon, 2 Oct 2000 05:30:56 -0400 Received: from smtprelay.ruc.dk (smtprelay.ruc.dk [130.225.220.22]) by mel.ruc.dk (8.9.3/8.9.1) with ESMTP id LAA22444; Mon, 2 Oct 2000 11:30:36 +0200 (MET DST) Received: from virgil.ruc.dk (virgil.ruc.dk [130.225.220.110]) by smtprelay.ruc.dk (8.9.3+Sun/8.9.3) with ESMTP id LAA26559; Mon, 2 Oct 2000 11:20:05 +0200 (MEST) Received: from VIRGIL/SpoolDir by virgil.ruc.dk (Mercury 1.44); 2 Oct 00 11:30:36 +0100 Received: from SpoolDir by VIRGIL (Mercury 1.44); 2 Oct 00 11:30:16 +0100 From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: "Kiniu WONG Kin-Yiu" Date: Mon, 2 Oct 2000 11:29:59 +0100 Subject: CCL:How to calculate excitation energy in high accuracy by CC: chemistry- at -ccl.net Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-ID: <101DC80A2ACA -x- at -x- virgil.ruc.dk> Kiniu WONG Kin-Yiu: > I couldn't get an accurate excitation energy by Gaussian, even a simple > atom such as Hydrogen and Beryllium. The excitation energy of H is 10.2eV, but > by UCIS(50-50)/6-31+G(d), it gives the excitation energy = 26.1eV: > > CIS wavefunction symmetry could not be determined. > Excited State 1: ?Spin -?Sym 26.1001 eV 47.50 nm f=0.0000 Dear Kiniu, calculation of excitation energies of the H atom with the 6-31+G(d) basis set is meaningless. The electronic transitions of atomic hydrogen are of Rydberg-type, corresponding to a change in the main quantum number n: 1s->2s, 1s->2p, and so forth. But the basis set 6-31+G(d) does not include any "diffuse functions" on hydrogen, and is unsuitable for description of excited states. Inclusion of a 2s-type diffuse function leads to a drastic decrease in the computed transition energy: 6-31++G(d): 1s->2s = 10.4 eV However, this basis set does not predict the three optically allowed 1s->2p transitions, that are energetically degenerate with the optically forbidden 1s->2s transition. Computation of 1s->2p transitions requires basis sets with an accurate representation of 2p functions. But degeneracy with 1s->2s is not easily predicted with standard sets, for example: AUG-cc-pVTZ: 1s->2s = 10.2 eV, 1s->2p = 11.2 eV AUG-cc-pVQZ: 1s->2s = 10.2 eV, 1s->2p = 10.9 eV AUG-cc-pV5Z: 1s->2s = 10.2 eV, 1s->2p = 10.7 eV Evidently, different angular momenta are treated at different levels of accuracy. Yours, Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Phone: +45 4674 2000 (RUC) Department of Chemistry +45 4674 2710 (direct) Roskilde University (RUC) Fax: +45 4674 3011 P.O.Box 260 E-Mail: JSL # - at - # virgil.ruc.dk DK-4000 Roskilde, Denmark http://www.rub.ruc.dk/dis/chem/psos/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=