[CP2K:2943] Re: bug in cp_fm_invert ?

[Hanning Chen]

Hi, Flo

  Besides the issue on the sign of determinant, I have another concern on
the et_coupling subroutine. As shown in the attached file, the one-electron
operator coupling between two Slater determinants is the summation of the
corresponding couplings between the orbitals weighted by the inverse of
overlap matrix. The derivation in details is given in "Methods of Molecular
Quantum Mechanics, Second Edition. (R. McWeeny), Page 61-66".

  Apparently, there is an index swap between the orbital coupling and
overlap matrix. Therefore, in "et_coupling.F",


 b(i)=b(i)+rest_MO(2)%matrix%local_data(k,j)*inverse_mat%local_data(k,j)*tttt

should be changed to


  b(i)=b(i)+rest_MO(2)%matrix%local_data(k,j)*inverse_mat%local_data(j,k).

Similar modification ought to be applied on a(i) too.

Could you please check this issue too ?

Thanks.

Hanning


On Wed, Nov 24, 2010 at 11:08 AM, flo <fsch... at pci.uzh.ch> wrote:

> Hi Hanning,
>
> I agree, that it should matter there. But indeed I wrote something
> wrong in my first reply. It should give the correct sign of the
> determinant. As your test indicates it does not, that is definitely a
> bug. I will have a look into this as soon as I got a little time.
>
> Thank you
> Flo
>
> --
> You received this message because you are subscribed to the Google Groups
> "cp2k" group.
> To post to this group, send email to cp... at googlegroups.com.
> To unsubscribe from this group, send email to
> cp2k+uns... at googlegroups.com <cp2k%2Bun... at googlegroups.com>.
> For more options, visit this group at
> http://groups.google.com/group/cp2k?hl=en.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20101124/9bd05c79/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cp2k_et_coupling.pdf
Type: application/pdf
Size: 156296 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20101124/9bd05c79/attachment.pdf>