Problems computing LUMO energy for a non-periodic metallic cluster using PM6

39 views
Skip to first unread message

Richard Marchese Robinson

unread,
Jul 29, 2019, 12:25:42 PM7/29/19
to cp2k

Dear all,

 

Firstly, thanks to the developers for making this tool available to the community.

 

I have only recently started using CP2K and I am hoping to be able to compute HOMO – LUMO gaps and individual energies for non-periodic, metallic clusters using different levels of theory. (It is my understanding from the literature that this gap should tend to zero as the clusters become larger.)

 

I am trying to perfom this calculation at the PM6 level of theory.

 

I provide my input and output files.

 

It is my understanding that the 22 eigenvalues in the output files refer to the energy levels of the occupied valence orbitals, with the Fermi energy corresponding to the energy of the HOMO.

 

However, as far as I can tell, I cannot obtain the LUMO energy. Am I overlooking something?

 

I have tried to address this using both the MO and MO_CUBES settings in the input files. I get a warning about the printing of MO cube files not being implemented for semi-empirical methods, e.g. PM6. Since this section is the only section which explicitly refers to the LUMO in the input file, does this mean PM6 LUMO values cannot be computed using CP2K?

 

I am running cp2k.7.0.190717.for.winxp.fftw3 on Windows 7.

 

Thanks in advance for any assistance you have time to provide.

 

Best regards,

 

Richard

 

-----

Dr Richard Marchese Robinson

Post-Doctoral Research Associate

BioRima Project [https://www.biorima.eu/]

University of Leeds

United Kingdom

ICSD_CollCode53759.cif_Avogadro.xyz_PM6_MOs.inp
ICSD_CollCode53759.cif_Avogadro.xyz_PM6_MOs_cp2k.out

Patrick Gono

unread,
Jul 29, 2019, 12:47:31 PM7/29/19
to cp...@googlegroups.com
Dear Richard,

Try to switch off the printing of the cube files, while maintaining NHOMO and NLUMO equal to 1 in the PRINT section of DFT:
&DFT
...
    &PRINT
        &MO_CUBES
            NHOMO  1
            NLUMO  1
            WRITE_CUBE .FALSE.
        &END MO_CUBES
    &END PRINT
&END DFT

Does it help?

Yours sincerely,
Patrick Gono

--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/2c10e923-9c2c-42dc-bc1c-82662b696665%40googlegroups.com.

Richard Marchese Robinson

unread,
Jul 29, 2019, 1:03:31 PM7/29/19
to cp...@googlegroups.com

Hi Patrick,

 

Thanks for the suggestion, but this does not solve the problem.

 

Best regards,

 

Richard

Matt W

unread,
Jul 29, 2019, 2:53:26 PM7/29/19
to cp2k
Try using the MO section rather than MO_CUBES...it may not work.

The semi-empirical methods do not have the full functionality of the DFT methods. I'd suggest trying DFT (for very small systems, but then you know the input works), then seeing if you can do it with PM6 or similar.

Matt

To unsubscribe from this group and stop receiving emails from it, send an email to cp...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "cp2k" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cp...@googlegroups.com.

hut...@chem.uzh.ch

unread,
Jul 29, 2019, 3:26:03 PM7/29/19
to cp...@googlegroups.com
Hi

you need "ADDED_MOS 100" in the &SCF section. Together with an
optimiser with diagonalization (default) and the section

&PRINT
&MO
EIGENVALUES
&END
&END

you will get the HOMO-LUMO value.

regards

Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: "cp2k" <cp...@googlegroups.com>
From: "Matt W"
Sent by: cp...@googlegroups.com
Date: 07/29/2019 08:53PM
Subject: Re: [CP2K:12007] Problems computing LUMO energy for a non-periodic metallic cluster using PM6
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/d154286b-3c7e-4a47-bb20-386487cc2b5a%40googlegroups.com.

Richard Marchese Robinson

unread,
Jul 30, 2019, 9:35:25 AM7/30/19
to cp...@googlegroups.com, Richard Marchese Robinson
Dear all,

Thank you for your assistance.

The answer from Juerg Hutter solved the problem.

Best regards,

Richard
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/OF9E1A43E6.4D9990CB-ONC1258446.006AC063-C1258446.006AC066%40lotus.uzh.ch.
Reply all
Reply to author
Forward
0 new messages