Using the absorption frequencies measured by Robinson and DiGiorgio, a rotational analysis of the 0+←0 and 1+←0 bands of the 3A2←1A1 transition in H2CO has been performed. Each band contains a number of overlapping subbands which are themselves made up of several branches consisting of individual rotational lines. Despite the overlapping, it was found possible to give assignments to a large number of lines. All branches are q-type branches; that is, only ΔK = 0 transitions are observed. The positions of the triplet-state rotational levels relative to the lowest rotational level of the ground vibronic state were obtained with the aid of the very accurate ground-state rotational levels determined interferometrically by Parkin and Poole. These triplet-state levels were found to fit accurately the Hamiltonian proposed by Van Vleck for an asymmetric-rotor molecule in a triplet electronic state provided that further terms were added to take account of rotational distortion. Values of the rotational constants, rotational distortion constants, and spin-interaction constants of the 3A2(0+) and 3A2(1+) states have been determined. The rotational constants for the 3A2(1+) state are as follows: A = 8.6275 (±0.0008) cm−1; B = 1.1491 (±0.0001) cm−1; and C = 1.0343 (±0.0001) cm−1. The results for the 3A2(0+) state are very similar. From these constants it is possible to show that the equilibrium nuclear geometry of formaldehyde in this state is highly nonplanar, the geometrical parameters having the following values: r(C–H) = 1.10 Å; r(C–O) = 1.28 Å; ∠HCH = 118°; out-of-plane angle = 36°. The spin-interaction constants, which determine the extent of the triplet splitting of the rotational levels, take account not only of spin—molecular rotation, spin—orbit, spin—other-orbit, and spin—spin interactions, but also of the effect of rotational asymmetry upon these interactions. The observed relative intensities of the rotational lines agree very well with those predicted by the recent theory of Hougen. In particular, the almost complete absence in the spectrum of transitions to the F2 component of the triplet state is understandable in the light of this theory. As a consequence of the fact that only ΔK = 0 transitions are observed (the theoretically possible ΔK = ±2 transitions were not detected), it is possible to deduce with the aid of Hougen's theory that it is a 1A1 state that, by spin—orbit coupling with the 3A2 state, permits the formally forbidden triplet←singlet bands to occur. This result is in complete agreement with the theoretical deductions of Sidman made some years ago.
Read full abstract