Investigation of the effect of solvation on 1J(Metal–P) spin–spin coupling

Abstract

The solvent effect on the indirect ¹J(M–P) spin–spin coupling constant in phosphine selenoether peri-substituted acenaphthene complexes LMCl₂ is studied at the PP86 level of nonrelativistic and four-component relativistic density functional theory. Depending on the metal, the solvent effect can amount to as much as 50% or more of the total J-value. This explains the previously found disagreement between the ¹J(Hg–P) coupling in LHgCl₂, observed experimentally and calculated without considering solvent effects. To address the solvent effect, we have used polarizable continuum and microsolvated models. The solvent effect can be separated into indirect (structural changes) and direct (changes in the electronic structure). These effects are additive, each brings roughly about 50% of the total effect. For the in-depth analysis, we use a model with a lighter metal, Zn, instead of Hg. A much smaller solvent effect on ¹J(Hg–P) for a dimer form of LHgCl₂ is explained. Pilot calculations of ¹J(M–P) couplings in analogous systems with other metals indicate that for metals preferring square planar structures the solvent effect is insignificant because these structures are fairly rigid. Tetrahedral structures are less constrained and can respond more easily to external effects such as solvation.