Triplet Electronic States of Acetylene: Introduction

Introduction and Background Information

We have presented a number of studies of triplet states of acetylene to assist in the interpretation of spectroscopic experiments by the Field group(MIT) and others. Acetylene plays an important role in fundamental studies of physical chemistry, since it represents the simplest possible alkyne and is one of the most thoroughly studied tetra-atomic molecules. Moreover, excited states of acetylene have an important place in history after landmark papers by Ingold, King, and Innes [C. K. Ingold and G. W. King, J. Chem. Soc. (London) 1953, 2702, 2704, 2708, 2725, 2745; K. K. Innes, J. Chem. Phys. 22, 863 (1954)] showed that the first excited singlet state (labeled A ¹A_u) is trans-bent; this was the first was experimental demonstration that an excited state could have a qualitatively different geometry than the ground state.

Electronic Structure Predictions from Walsh Diagrams

Based on the Walsh diagram for HAAH molecules, one would expect a pi->pi* transition to lead to four low-lying triplet and four low-lying singlet minima, and moreover that all of these minima ought to be bent. The triplets are lower-lying in energy than the singlets, and simple group theory in conjunction with the Walsh diagram suggests that the lowest excited potential energy surface (T₁) will have cis and trans minima with term symbols ³A₂ and ³A_u, respectively. The next-higher triplet potential energy surface (T₂) is predicted to have cis and trans minima ³B₂ and ³B_u, respectively, according to the Walsh diagram. These predictions based on very rudimentary electronic structure considerations were ultimately overturned, as discussed below.

Early Experimental Work: Burton and Hunziker

One of the first experimental studies of triplet acetylene was by Burton and Hunziker [J. Chem. Phys. 57, 339 (1972)], who deduced that a metastable triplet state of acetylene was formed via mercury photosensitization. Arguing by analogy to other systems, these workers expected that they had formed the most stable of the possible triplet species in their experiment. Since they failed to observe any absorption by this triplet state between 220 and 540 nm, they argued that it must be linear or trans-bent, which would make upward transitions to any other triplet states forbidden by symmetry. However, this contradicted other theoretical studies performed around that time [W. E. Kammer, Chem. Phys. Lett. 6, 529 (1970); R. Ditchfield, J. Del Bene, and J. A. Pople, J. Am. Chem. Soc. 94, 4806 (1972); D. Demoulin, Chem. Phys. 11, 329 (1975)], which seemed to indicate that (1) in contrast to the Walsh predictions, the ³B₂ and ³B_u states were actually lower than ³A₂ and ³A_u, and that (2) the cis ³B₂ state was a few tenths of an eV below the trans minimum ³B_u. Given the very modest levels of theory possible at that time, however, it was commonly anticipated that better theoretical treatments would come into harmony with the exeperimental interpretations.

More Accurate Theory Clarifies the Situation

In 1978, Wetmore and Schaefer [J. Chem. Phys. 69, 1648 (1978)] presented the first theoretical study sufficiently reliable to settle the question of the character of the lowest-lying triplet state of acetylene. Their double-zeta plus polarization configuration interaction singles and doubles (DZP CISD) study of the four lowest-lying triplet minima of acetylene predicted the ³B₂ cis minimum to be the lowest energy, at 0.35 eV below the ³B_u trans form and 3.49 eV above the ground state. Moreover, because the cis form was found to be the lowest in energy, an upward transition to the ³A₂ minimum would be symmetry allowed and was predicted to appear at 8630 cm^-1. Rising to the challenge, Wendt, Hippler, and Hunziker [J. Chem. Phys. 70, 4044 (1979)] made triplet acetylene and found a single, strong band in its absorption spectrum at 7388 cm^-1. The lack of a vibrational progression indicated very similar geometries for the ³A₂ and ³B₂ minima, consistent with the theoretical predictions. Moreover, the rotational constants extracted from the spectrum were in good agreement with the computed values.

Next: Potential difficulties for ³B₂ predictions Up: Contents Previous: Contents

C. David Sherrill
28 Jan 2000