CCL:G: ONIOM calculations in Gaussian



Dear Petra,

we encountered a similar problem a few weeks ago. I appears that Gaussian09 does sometimes automatically create link atoms or checks for atoms that should be replaced by link atoms (although the online manual explicitly states that it does not do that: "Note: All link atoms must be specified by the user. Gaussian 09 does not define them automatically or provide any defaults."). A possible workaround might be using GaussView to set up the system.


In the case of diamond, however, since you probably require multivalent link atoms (i.e., one link atom bound to two or more border carbons), you might need to resort to more advanced schemes than simple hydrogen saturation. Just visualise your problematic model, the coordinates are given in the log file. You should be able to see where the problem comes from.


Cheers,

Jan


>
> Sent to CCL by: "Petra  Matunova" [petra.matunova]|[seznam.cz]
> The surface is hydrogen-terminated and there where should be a
> continuation to the bulk are hydrogens too (it is a corner of a
> nanoparticle).
>
>
>> "zborowsk zborowsk[a]chemia.uj.edu.pl"  wrote:
>>
>> Sent to CCL by: zborowsk [zborowsk%chemia.uj.edu.pl]
>>
>> I must say I am a bit suprised by presence of hydrogen in your
>> diamond....
>>
>>
>> W dniu 2015-11-04 12:06, Petra Matunova petra.matunova]![seznam.cz
>> napisa(a):
>> > Sent to CCL by: "Petra  Matunova" [petra.matunova%%seznam.cz]
>> > I am performing an ONIOM optimization calculation of a diamond
>> > nanoparticle, using b3lyp/6-31g(d):b3lyp/3-21g method. The system
>> > contains 206 C atoms and 130 H atoms in the high level and 488 C
>> > atoms
>> > 188 H atoms in the low level. When I set the charge and multiplicity
>> > as following: 0 1 0 1 0 1, the system do the low level SCF
>> > calculation
>> > without problems, but when starts with the  high level calculation on
>> > the model system, it ends with error:
>> >
>> > The combination of multiplicity 1 and  1455 electrons is impossible.
>> >
>> > It only works when I set  0 1 0 2 0 2. In the high level system there
>> > are 1366 electrons plus 89 link electrons from H atoms used as link
>> > atoms.
>> >
>> > I had a problem with the specification of the link atoms on the
>> > boundary between the low level and high level system. The low level C
>> > atom close to the boundary  has two bonds to another 2 high level C
>> > atoms. I haven't known how to deal with that, as far as I know
>> > Gaussian allows to specify only one bond to for every low level atom
>> > to a high level atom. So I specified for the low level C atom linked
>> > to 2 high level C atoms only one link H atom and set the
>> > corresponding
>> > scale factor to 1, e.g.:
>> >
>> >  C      -1     3.953927    3.329465    3.558583 L H 18  1.
>> >
>> > Like this, I got only a following warning in output, e.g.:
>> >
>> > WARNING: Link atom host   21 is close to both atoms 18 and  123.
>> > Should it be oligovalent?
>> >
>> > Is there a way how to specify a link of the low level C to both of
>> > the high level C to which it is bonded? Or how to solve that, would
>> > it
>> > be right to leave the multiplicity to one?
>> >
>> > Thank you for your help!
>> > Best regards,
>> > Petra
>> >
>> >
>> >
>> > -= This is automatically added to each message by the mailing script
>> > =-
>> > To recover the email address of the author of the message, please
>> > change> Conferences:
>> > http://server.ccl.net/chemistry/announcements/conferences/--
>> https://www.youtube.com/watch?v=IGyfx3Pmeyk
>> Krzysztof K. Zborowski
>> Faculty of Chemistry
>> Jagiellonian University
>> 3 Ingardena Street
>> 30-060 Krakow
>> Poland
>> phone: +48(12)663-2067
>> email: zborowsk]~[chemia.uj.edu.pl
>> gg 3817259
>> skype kzys70
>> www.chemia.uj.edu.pl/~zborowsk
>>
>>
>
>
>
> -= This is automatically added to each message by the mailing script =-
> To recover the email address of the author of the message, please change
>
>
>
>
>       http://www.ccl.net/cgi-bin/ccl/send_ccl_message
>
>
>       http://www.ccl.net/cgi-bin/ccl/send_ccl_message
>
> Subscribe/Unsubscribe:
>       http://www.ccl.net/chemistry/sub_unsub.shtml
>
>
>
> Job: http://www.ccl.net/jobs
> Conferences: http://server.ccl.net/chemistry/announcements/conferences/
>
>
>
>
>      
>
>
>
>
>


Jan Philipp Götze
School of Chemistry
University of St Andrews
North Haugh
St Andrews, Fife
KY16 9ST, Scotland, United Kingdom

Tel.: +44 (0)1334-464748 (ext. 14141)