NBO3 vs NBO6 behavior

NBO3 vs NBO6 behavior

Postby gbreton » Thu Aug 23, 2018 2:05 pm

Hello,

I have a relatively large molecule that I am subjecting to NBO analysis to examine hydrogen bonding interactions. When I use Gaussian 09 and embedded NBO 3, the results are aligned with my expectations (i.e., there is an expected carbonyl oxygen lone pair --> sigma* H-O bond interactions representing the hydrogen bond). However, when I use G09 and NBO6, the output file suggests there are two units (but it is a single molecule!) and provides some weird results, including a LP --> LV (on the OH bond) interaction. I can't seem to make it behave as it should. Any suggestions as to why it would think there is an intermolecular interaction where there is not?
gbreton
 
Posts: 13
Joined: Thu Dec 04, 2014 7:50 pm

Re: NBO3 vs NBO6 behavior

Postby ericg » Sun Aug 26, 2018 8:50 pm

Could you post some sample output, or a copy of the Gaussian input file so that I can try this myself?

Eric
ericg
 
Posts: 298
Joined: Sat Dec 29, 2012 9:31 am

Re: NBO3 vs NBO6 behavior

Postby gbreton » Mon Aug 27, 2018 11:11 am

Yes, sent a private message with info included. Thanks for looking it over!
gbreton
 
Posts: 13
Joined: Thu Dec 04, 2014 7:50 pm

Re: NBO3 vs NBO6 behavior

Postby gbreton » Fri Sep 14, 2018 2:19 am

Update: I wondered whether the problem might be the M062X functional we were using for the NBO calculation (since that was what we used for optimizations). I switched to the more widely used B3LYP functional and had the same separation of the molecule into two units with bizarre interaction energies (>400 kcal/mol). Finally, I tried ditching DFT altogether and used HF and the same basis set and it worked fine. I *think* I have seen reference to DFT methods occasionally causing problems in NBO calculations. Does anyone have reference to such problems?
gbreton
 
Posts: 13
Joined: Thu Dec 04, 2014 7:50 pm


Return to General NBO Discussion

Who is online

Users browsing this forum: No registered users and 2 guests

cron