Two interesting questions are addressed in a focal-point computational study of t-butyl radical and the t-butyl anion from the Schaefer group.1 First, is the radical planar? EPR and PES studies from the 1970s indicate a pyramidal structure, with an inversion barrier of only 0.64 kcal mol-1. The CCSD(T)/cc-pCVTZ optimized structure of t-butyl radical shows it to be pyramidal with the out-of-plane angle formed by one methyl group and the other three carbons of 22.9°, much less than the 54.7° of a perfect tetrahedron. Focal point analysis give the inversion barrier 0.74 kcal mol-1, in outstanding agreement with experiment.
Second, what is the electron affinity (EA) of the t-butyl radical? Schleyer raised the concern that the alkyl anions may be unbound, and suggested that the electron affinity of t-butyl radical was -9.6 kcal mol-1; in other words, the anion is thermodynamically unstable. This focal-point study shows just how sensitive the EA is to computational method. The HF/CBS value of the EA is -39.59 kcal mol-1 (unbound anion), but the MP2/CBS value is +41.57 kcal mol-1 (bound anion!). The CCSD/aug-cc-pVQZ value is -8.92 while the CCSD(T)/aug-cc-pVQZ value is +4.79 kcal mol-1. The estimated EA at CCSDT(Q)/CBS is -1.88 kcal mol-1, and inclusion of correction terms (including ZPE and relativistic effect) gives a final estimate of the EA as -0.48 kcal mol-1, or a very weakly unbound t-butyl anion. It is somewhat disconcerting that such high-level computations are truly needed for some relatively simple questions about small molecules.
References
(1) Sokolov, A. Y.; Mittal, S.; Simmonett, A. C.; Schaefer, H. F. "Characterization of the t-Butyl Radical and Its Elusive Anion," J. Chem. Theory Comput. 2012, 8, 4323-4329, DOI: 10.1021/ct300753d.
Henry Rzepa responded on 18 Dec 2012 at 11:41 am #
Re: disconcerting. Yes, often its the smaller molecules that do cause the problems. Think of the fun and games to be had with F2 (it is clearly bound, but by what kind of bond? Its not the obvious covalent bond!). Or C2, and the constant too and fro about its bonding (the discussion about whether it has a 4th bond holding the two atoms together is about to burst upon the stage again shortly). I am also reminded of the famous debates about the sign of the dipole moment on CO, and of course the singlet-triplet splitting in CH2. Even cyclobutadiene continues to generate heat (see also in new year) and I am not even going to mention the 25 year saga of the structure of the norbornyl cation (OK, that is not quite a small molecule).
There does seem an assumption that state-of-the-art research can only be with BIG molecules, but the small ones must not be neglected.