The IPEA shift in CASPT2

Posted by Björn Roos on February 09, 2007 at 14:47:23:

Dear MOLCAS users,
Some users have noted that CASPT2 in MOLCAS-6.4 does not give the same energy as earlier versions. This is correct and is due to a modification of the zeroth order Hamiltonian, which has been introduced in order to reduce the systematic error of the method. This error has been giving somewhat to low relative energies comparing closed shell and open shell states (for example in dissociation or excitation). The correction is called the "IPEA shift" and contains a parameter that has been set to 0.25 by default. Users should avoid playing with this parameter unless they want to use CASPT2 as a true empirical program.
For more details see: CPL 396,142 (2004).

Follow Ups:

Post a Followup

Name:
E-Mail:

Subject:

if B is 1s22s22p1, what is Li?

Passfield:

Comments:
: Dear MOLCAS users, : Some users have noted that CASPT2 in MOLCAS-6.4 does not give the same energy as earlier versions. This is correct and is due to a modification of the zeroth order Hamiltonian, which has been introduced in order to reduce the systematic error of the method. This error has been giving somewhat to low relative energies comparing closed shell and open shell states (for example in dissociation or excitation). The correction is called the "IPEA shift" and contains a parameter that has been set to 0.25 by default. Users should avoid playing with this parameter unless they want to use CASPT2 as a true empirical program. : For more details see: CPL 396,142 (2004).

[ Follow Ups ] [ Post Followup ] [ Molcas user's WWWBoard ]