8
$\begingroup$

When I do a CASSCF calculation I can include the card expec2,lxx,lyy,lzz in order to calculate the two-electron operators (in the x, y, and z-directions) for orbital angular momentum. This is described here.

However, when I try to include the same instruction in an MRCI calculation the run is terminated, and the output says "Unknown command or directive EXPEC2". So clearly this is not the right way to evaluate angular momentum in a CI calculation. But what is? There seems to be no mention of this in the manual.

$\endgroup$
3
$\begingroup$

In the first sentence of this Introduction to MOLPRO, MOLPRO is described as a "system of programs", not just one program.

The first link you provided is for the program called MULTI for MCSCF calculations (which includes multiple types of MCSCF such as CASSCF and RASSCF). The second link you provided is for the program called MRCI. Each program in MOLPRO can have different authors, and you can see that the 9 papers listed in the MULTI program bibliography don't overlap (not even in the years published!) with the 8 papers listed in the MRCI program bibliography, except for the 1987 book chapter which is more of a review of papers previous to that. How things work in one MOLPRO program won't always be the same as in other MOLPRO programs. Some names of things are overloaded, so that the same string is used in multiple programs, but not necessarily with exactly the same meaning.

When I search expec2 in the documentation link you provided for the MRCI program, the only appearance of that string on that page occurs in a MULTI command:

{multi;
occ,3,1,1;frozen,1;wf,6,1;state,3;lquant,0,2,0;wf,6,4;lquant,2;
tran,lz;
expec2,lzlz;}

Interestingly, they used lzlz instead of lzz for the $\hat{L}_z^2$ operator, as described here.

I don't see any example of how to use expec2 with the MRCI program, and the introduction to the MRCI documentation to which you provided the link, says

"The EXPEC directive allows to compute expectation values over one-electron operators, and the TRAN directive can be used to compute transition matrix elements for one-electron properties"

but nothing about 2-electron properties such as the ones available in MULTI by using the EXPEC2 card. If you search the word "properties" in the MRCI documentation, you'll find that there's a sub-section about one-electron properties, but nothing suggesting that two-electron properties are possible.

$\endgroup$
2
  • $\begingroup$ Do you know where or how I can submit feature requests for the MRCI program? $\endgroup$ Jul 8 at 8:44
  • $\begingroup$ @stack-delay I'm extremely shocked that you seem to have downvoted both this answer and Tatiana's answer on your other question, and have upvoted zero posts on this entire site. If people knew this, I personally think it's likely that they would be strongly discouraged from answering any of your questions. $\endgroup$ Jul 9 at 9:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.