[molpro-user] Triplet Phi States

[Allan East]

Hi Edmund;

I can answer this one, since we worked on the doublet phi state for NO 
this past year.

The wf card is for the state wavefunction, not individual orbitals.  
Also, the program will work in C2v rather than C-inf-v, so you'll have 4 
symmetry classes (A1, B1, B2, A2) to consider.

For a triplet phi state you want either:
wf,e#,2,2 (the x component) or wf,e#,3,2 (the y component)
together with an LQUANT=3 flag (for the phi rather than the pi).  
Careful choice of the OCC card (number of occupied orbitals of each 
symmetry in C2v) might be required.  During the run, MOLPRO will then 
decide what the optimal singly occupied orbitals should be as it 
converges the wavefunction --- an ORBPRINT or maybe GPRINT card will 
spit out the LCAO-MO's to tell you what those are.  Sounds like you 
shouldn't need f functions to find the phi state, although they might 
help if you're looking for a super-accurate energy.

Sincerely,
Allan East
University of Regina

MEYER EDMUND RICHARD wrote:

>Dear users,
>
>I am trying to construct triplet phi (as well as quintet phi) states in
>MOLPRO for PtH+ potential energy curves.  Making a triplet phi combines
>one d electron and one p electron. So, how do I define that wavefunction?
>Is it just wf,e#,2,2; for the p and another wf,e#,4,2; for the d?
>Is it possible even though there is no f-basis set for Pt?  Thank you for
>your time and patience and have a nice day.
>
>Edmund Meyer
>JILA/Univ. CO, Boulder
>Office S370
>303-492-2548
>  
>