[Columbus] Molden Format

[Thomas Mueller]

On Wed, 14 Oct 2009, Scott Brozell wrote:

> Hi,
>
> Ok, you are using transmo.
> Thomas, can you confirm that transmo.f should be changed to add
>>>> [GEOMETRIES] XYZ
> in this code snippet :
>       write(20,*) '[Molden Format]'
>       write(20,*) '[Atoms] AU'
>
> thanks,
> Scott
>

   Hi,

  seems that the specification in the molden file constructed by
  transmo.f is incomplete. So one should add the missing
  line.

  There are some additional points to remark:

   * if you are just interested in displaying the cartesian
     geometries use  the geomout file, it is much smaller
     and molden can process it quickly. For frequency calculations
     there should be some output file molden.freq containing
     the relevant information excluding the MOs. Finally
     frequency calculations and geometry optimizations of
     linear molecules require a dummy atom at right angles
     to the molecule.

   * with large basis sets, the MO file in molden format
     is huge and mostly useless for molden because the
     algorithm to display the molecular orbitals is rather
     slow

   * I don't think that the molecular orbitals im molden
     format are always right - d and f type orbitals may
     show an incorrect scaling factor

   * the old transmo.f code is from my side mostly obsolete,
     since I have implemented an new version that includes
     a more efficient symmetry transformation scheme
     (including dalton and molcas support) and a basis set
     projection to project the optimized MOs from a smaller
     basis set onto a larger basis set without any restrictions
     (apart from the trivial requirement that there must be reasonable
      overlap between the two basis sets). Support up to
      i-type functions is included but I have not yet done
      the molden interface.

  Regards,

  Thomas


> On Tue, Oct 13, 2009 at 08:05:27PM -0400, Francois Dion wrote:
>> The first calculation was done using Columbus 5.9.0, with Dalton as
>> integration program. C1 symetry and minimal basis set is used. The
>> calculation is a closed shell scf followed by a mcscf. If you need more
>> information about this calculation, I can give them. I show here the header
>> of the 3 files generated in MOLDEN repertory (molden_mo_mc.sp
>> molden_mo_scf.sp  molden_no_mc.sp), wich is the same for all of them :
>>
>>  [Molden Format]
>>  [Atoms] AU
>>   O     1    8     0.0000000000    0.0000000000   -1.8900000000
>>   O     2    8     0.0000000000    0.0000000000    1.8900000000
>>   C     3    6     0.0000000000    0.0000000000    0.0000000000
>>  [5D]
>>  [7F]
>>  [GTO]
>>
>>
>> On Tue, Oct 13, 2009 at 5:12 PM, Scott Brozell <srb at osc.edu> wrote:
>>
>>> On Mon, Oct 12, 2009 at 02:14:16PM -0400, Francois Dion wrote:
>>>> Hello dear Columbus user and developer,
>>>>
>>>> I wanted to use Molden to make pictures of molecular amplitudes produced
>>> by
>>>> a Columbus calculation. My molecular system is very simple : CO_2, but
>>> when
>>>> I look at the output given by Columbus using  Molden, I see 4 atoms!
>>>>
>>>> So, since the molecule is written in Cartesian mode, I think there is a
>>>> missing line in the file header, which would be:
>>>>
>>>> [GEOMETRIES] XYZ
>>>>
>>>> Adding this line to the Molden file created by Columbus, the molecule
>>>> structure now appears correctly. I want to share this with you all.
>>>
>>> What version are you using ?  What protocol did you follow for the calc ?
>>> runc seems to handle this correctly back to at least 5.7.
>>>
>>> Scott
> _______________________________________________
> Columbus mailing list
> Columbus at osc.edu
> http://email.osc.edu/mailman/listinfo/columbus
>

-----------------------------------------------------------
Dr. Thomas Mueller
Institute for Advanced Simulation (IAS)
Juelich Supercomputing Centre (JSC)

Phone:  +49-2461-61-3175
Fax:    +49-2461-61-6656
E-mail: th.mueller at fz-juelich.de
WWW:    http://www.fz-juelich.de/jsc/

JSC is the coordinator of the
John von Neumann Institute for Computing (NIC)
and member of the
Gauss Centre for Supercomputing (GCS)
-----------------------------------------------------------


------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
Forschungszentrum Juelich GmbH
52425 Juelich
Sitz der Gesellschaft: Juelich
Eingetragen im Handelsregister des Amtsgerichts Dueren Nr. HR B 3498
Vorsitzende des Aufsichtsrats: MinDir'in Baerbel Brumme-Bothe
Geschaeftsfuehrung: Prof. Dr. Achim Bachem (Vorsitzender),
Dr. Ulrich Krafft (stellv. Vorsitzender), Prof. Dr.-Ing. Harald Bolt,
Prof. Dr. Sebastian M. Schmidt
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------