The calculation of properties of open-shell 5f molecules is a challenge for the methods of quantum chemistry : these complexes have many low lying configurations, spin-orbit effects are important and correlation effects must be taken into account. The SO-CASPT2 method gives results that compare well to experimental data : it is a two-step wave function based method. The multiconfigurational nature of the wave functions is described by starting with a CASSCF calculation, correlation effects are calculated by 2nd order theory and spin-orbit effects are introduced in the very last step by a state interaction procedure.
The first principle calculation of the magnetic properties of actinide complexes is one hand a helpful tool for the interpretation of the experimental data and on the other hand, permits the calculation of the parameters of the model Hamiltonian.
We will show by several examples how calculations have become a complementary tool to the experimental data in order to get information about the nature and the magnetization of the ground and excited states.
- calculation of EPR parameters : the ground state and excited states of actinyl complexes depend strongly on the nature of the equatorial ligands. All calculations are rationalized using a model based on crystal field theory.
- calculation of magnetization of single molecule magnets : complexes of U(III) show a blocking of the magnetization.
- determination of the parameters for paramagnetic NMR : the presence of a paramagnetic atom modify the NMR spectrum bringing informations about the structure and the bonding in the complex. Two mechanisms contribute to the shift : the contact shift results from the spin polarization while the pseudocontact shift is due to the direct interaction with the anisotropic magnetic properties of the paramagnetic ion. The latter becomes important in the case of heavy elements. The analysis of the experimental measurements and our calculations permit to distinguish between the two mechanisms.