Next: 20.9.3 Non-adiabatic coupling matrix Up: 20.9 Coupled-perturbed MCSCF Previous: 20.9.1 Gradients for SA-MCSCF

20.9.2 Difference gradients for SA-MCSCF

For computing difference gradients, use

CPMCSCF,DGRAD,state1,state2,[ACCU=thresh],[RECORD=record]

where state1 and state2 specify the two states considered. (e.g., 2.1,3.1 for the second and third states in symmetry 1) The gradient of the energy difference will be computed. Both states must have the same symmetry. record specifies a record on which the gradient information is stored (the default is 5101.1). thresh is a threshold for the accuracy of the CP-MCSCF solution. The default is 1.d-7.

The gradients are computed by a subsequent call to FORCES or OPTG.



P.J. Knowles and H.-J. Werner
molpro@tc.bham.ac.uk
Jan 15, 2002