Singlet-triplet separation experimental determination

The next issue was the singlet-triplet separation, and this story, like the story of the CH2 angle, involves a disagreement between theory and experiment. But first let us take an overview of the situation. Between 1967 and 1972 there were five experimenfal determinations [25-29] of the singlet-triplet splitting and these are listed in Table 1. Two of these [25,27] measured 1-3 kcal/mol and three [26,28,29] deduced 8-9 kcal/mol. I say deduced because backing these numbers out of the experimental data required considerable effort and skill. 

As for the previous two theories, we are interested in W(t)/t — k as a function of the energy separation Eai — Es. Robinson and Frosch have empirically correlated the square of the overlap integral [as in eq. (7)] with the energy separation for a singlet-triplet transition. From experimentally determined rate constants and an arbitrary but reasonable value for pel (0.01 cm-1) an expression relating < 0>2>tal to Eti ESo was deduced,... 

Carbene structures and properties can now be computed with chemical accuracy , despite the difficulties associated with multi-reference species such as singlet carbenes. This is particularly encouraging since the determination of singlet-triplet energy separations and accurate structures of carbenes, which very often are, at best, fleetingly observable intermediates, is extremely difficult experimentally. Transition structures also can be computed apparently with good accuracy and complement experimental studies of the reactions as well as the complex rearrangements commonly found in carbene chemistry. 

---