Require birad_r_recombination to be between a doublet and triplet

[donerancl]

trafficstars

The Birad_R_Recombination family seems to be intended for recombination of a doublet R radical and a triplet biradical with opposite spin, producing a doublet radical.

Ex.

Currently, the family also allows a similar recombination between two triplets.

Ex.

These triplets must have opposite spin to form a bond, forming an overall singlet. However, RMG calculates the multiplicity based on the number of unpaired electrons (2) + 1. It is not clear to me yet how RMG should represent singlet diradicals. Two options are

• Change the Molecule().update() function to assign multiplicity for these species correctly. Their adjacency lists are identical apart from the multiplicity, so this would require inputting information about the reaction producing the diradical. This issue will only be an issue for some families so this might not be a bad option.
• Represent singlet diradicals with a closed shell structure. This requires some code to interconvert between the singlet diradical and closed shell structures to generate reactions.

<=>

For now, I suggest limiting this reaction family to triplet doublet recombinations.