Polarizability of an open shell system with different occupations.

[Mhmd Al ahmad]

Dear DIRAC users,
I calculated the polarizability of the system Au^(3+), where the electronic configuration is [Xe] 5p^(6) 5d^(8), using the finite-field perturbation scheme.
In the first case, I put the occupation as :
i) .CLOSED SHELL
 74
.OPEN SHELL
 1
 2/4
The value of the polarizability I got is  -30218.443 a.u.
In the second case, I consider the average of configurations, so the occupation put as follows:
ii) .CLOSED SHELL
 68
.OPEN SHELL
 1
 8/10
The value of the polarizability in this case is 6.149 a.u.
The formula I used is : -[(E(+F) + E(-F) -2E(0))/F^(2)].
Please, could you explain why there is such a big difference between the two values.

Thank you very much for your effort and time,
Mohamed  

[Visscher, L. (Luuk)]

Dear Mohamed,

This likely has to do with the symmetry-breaking that occurs in the calculations with the field in case i). This may lower the energy artificially much more than the effect of the field itself.

Best check the orbitals that you got using a population analysis with labels defined explicitly and make sure that you understand what happens to the wave function.

Few more things to check:

- you likely need better convergence of the SCF than the default as you are looking at small energy differences (in case ii).

- for Au(III) the ground state is ^3F_4, you are calculating the weighted average of all d^8 states in the second case and of smaller set of states that includes a J=4 state, but also other states in the first case. A better and still cheap calculation is to use COSCI and get the polarisability of the specific state you are interested in. Still better and doable for an atom is to do Fock space coupled cluster and compute this as the energy of a doubly ionised d^10 state.

Best regards,

Luuk

--
You received this message because you are subscribed to the Google Groups "dirac-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dirac-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/dirac-users/7f5c6691-6552-46e2-bc56-4ae305eaca2en%40googlegroups.com.
<-0.0005_i.out><+0.0005_ii.out><-0.0005_ii.out><0.000_i.out><+0.0005_i.out><0.000_ii.out>