how to read wavefunction(.wfn file) from old calculation into new?
I don’t think that was the intention of the .wfn file. That file is created for an old interface to AIMPAC. Unclear to me if anyone still uses that interface.
1 Like
Suppose I calculated LDA energy of one system and and want to use the wave function from LDA calculation to PBE . What shall I do in this case ?
Use SCF_GUESS = READ in a 2nd calculation.
$molecule
0 1
O
H 1 0.95
H 1 0.95 2 104.5
$end$rem
method lda
basis sto-3g
$end@@@
$molecule
read
$end$rem
method pbe
basis sto-3g
scf_guess read
$end
Although as written, this is sort of a silly calculation insofar as the 2nd job just reads in the LDA orbitals and uses those as a guess for the PBE calculation. The energy on the first cycle of the 2nd job should be the PBE energy computed using LDA orbitals, if that is what you are looking for. Otherwise I would need to know more about what you are trying to do.
Note also that you don’t need to run this as two jobs in same input file, you could save the scratch files and come back later. That procedure is described in the manual but basically it’s
qchem -nt <no_threads> -save input output scratch
and the scratch files with then be saved for later use, e.g.,
qchem -nt <no_threads> -save input2 output2 scratch
hi can you help me saying the rem tag for non-self consistent calculation ?
Okay, so you want to evaluate the PBE energy using self-consistent LDA orbitals, or generally just the energy of functional X using self-consistent orbitals from functional Y, correct? What you can do is set SCF_GUESS=READ in the 2nd job (with METHOD = X, following a self-consistent calculation with METHOD = Y), and then also set MAX_SCF_CYCLES = 1 in the 2nd job. That will generate a crash (because the SCF won’t converge in a single cycle), but I think that the energy on the first cycle should be meaningful, i.e., it should be the energy of Y using the SCF orbitals from X.
1 Like
Thank you so much it’s great help to me .