Problems with Brueckner orbitals

[Dave Moore]

Hi,

There appear to be some problems with the Brueckner orbitals in Molpro
2002.3.  Specifically, when using the SAVE directive in a BCCD calculation,
it seems that the core orbital coefficients are not transferred to the
specified record, which makes further manipulation of the orbitals using
MATROP difficult.  I have not checked if this is also true of the density
matrix (i.e. that it is a valence density only).  In addition, my old MOLPRO
input files using the BRUECKNER directive (that used to work fine) now
suffer from the same problem.  Another, more minor problem is that the
occupation numbers are not stored on the SAVE record, so the orbitals cannot
be output in MOLDEN format without manipulation using MATROP.
	To help illustrate these issues, I have included two input files below.
The first one uses the new format (BCCD with SAVE), and the second uses the
old format (BRUECKNER directive and subsequent MATROP manipulation).  The
old format used to give correct MOLDEN output files with MOLPRO version
2002.1.  I would appreciate any help you can give.  Thanks,

Dave Moore
FOM Rijnhuizen
Edisonbaan 14
3439 MN, Nieuwegein
The Netherlands
+31+30 609-6826
-------------- next part --------------
A non-text attachment was scrubbed...
Name: example.new.inp
Type: application/octet-stream
Size: 940 bytes
Desc: not available
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20021007/62be5d55/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: example.old.inp
Type: application/octet-stream
Size: 1127 bytes
Desc: not available
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20021007/62be5d55/attachment-0001.obj>