SiH 2 , a critical study APOSTOLOS KALEMOS 1,2,3 , THOM H. DUNNING JR 1,2,3* and ARISTIDES MAVRIDIS 4 1 Joint Institute for Computational Sciences, University of Tennessee-Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA 2 Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996, USA 3 Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA 4 Laboratory of Physical Chemistry, Department of Chemistry, National and Kapodistrian University of Athens, PO Box 64 004, 157 10 Zografou, Athens, Greece (Received 25 May 2004; accepted 4 August 2004) The first four spectroscopic states of the silylene molecule SiH 2 , namely, ~ X 1 A 1 , ~ a 3 B 1 , ~ A 1 B 1 and ~ B 1 A 1 were examined theoretically using multireference methods coupled with very large correlation consistent basis sets. Our aim is understanding why SiH 2 has a singlet ground state ( ~ X 1 A 1 ) as opposed to the ~ X 3 B 1 state of the isovalent carbene CH 2 , as well as the rationalization of its geometric and bonding characteristics. The interpretational philosophy followed is based on strictly calculable quantities in an effort to reduce to a minimum the always present but not well-defined ‘chemical intuitionism’. All of our calculated quantities are in excellent agreement with existing experimental results. 1. Introduction ‘Why is methylene a ground state triplet while silylene is a ground state singlet?’ is the title of a recently published article by Apeloig et al. [1] that tries to shed some light in the quite surprising reversal of the ground state multiplicities in going from CH 2 to SiH 2 based on a rather complicated energy decomposition scheme. Along the same vein is the study by Gaspar et al. [2] entitled ‘The quest for triplet ground state silylenes’. The authors believe that the larger size of the valence silicon orbitals in comparison with those of carbon is the main reason behind the elusiveness of the triplet ground state silylenes. They claim that the larger size of these orbitals leads to a decrease in the repulsion of the non-bonding electrons in the singlet state, hence their energy lowering separation in the triplet state is less capable of compensating an attendant promotion energy. Influenced by Gordon [3], who as early as 1985 suggested that bulky substituents may open the angle in the singlet sufficiently to invert the ordering of the two states, Gaspar and collaborators [2] have been trying to synthesize such a substituted silylene hoping that the increased bond angle would reduce the energy difference between the in-plane and out-of-plane non-bonding orbitals, thus decreasing the promotion energy required to reach the triplet state. In 2001 Jiang and Gaspar [4] reported the end of a long quest for a triplet silylene based on the preparation of a product that could not arise from a singlet silylene at room temperature, and thus the reaction could be regarded as chemical evidence for its triplet ground state. Yoshida and Tamaoki [5], based on density functional theory (DFT) calculations over an extended range of substituted silylenes, sug- gested that the quest for a triplet ground state silylene must go on until a direct electron spin resonance (ESR) observation is recorded, a declaration also made by Gaspar et al. [2]. The quest is not finally over because of the lack of either an ESR signal or a chemically induced dynamic nuclear polarization nuclear magnetic resonance (NMR) experiment, or by detection of the silylene by kinetic ultraviolet (UV) spectroscopy. It seems that the singlet ground state SiH 2 as opposed to the triplet ground state of the isovalent carbene, CH 2 , is an afflictive question analogous to the distressing and rather troublesome issues that shaded for a long time the (correct) bent structure of the ~ X 3 B 1 state and ~ a 1 A 1 – ~ X 3 B 1 splitting of the CH 2 molecule [6]. *Author for correspondence. e-mail: dunning@jics.utk.edu MOLECULAR PHYSICS, 10–20 DECEMBER 2004, VOL. 102, NO. 23–24, 2597–2606 Molecular Physics ISSN 0026–8976 print/ISSN 1362–3028 online # 2004 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/00268970412331293802