Charged interstitial atom

Clicked A Few Times

12:53:35 AM PDT - Fri, Jun 26th 2015
Hello! I want to calculate charged interstitial atom(Oi) in my cluster. INPUT I found keywords "set atomscf:z -1" and "set atomscf:list_z Oi", but can't understand how to use them properly. Can you please help me to make proper input.

Forum Regular

1:05:58 PM PDT - Mon, Jun 29th 2015
In a standard calculation it is not possible to directly specify the total charge of a single atom in a molecule (note this is not a limitation of the code, just a consequence of the quantum nature of electrons). If you are interested in modifying the nuclear charge of an atom, this is most easily accomplished in the geometry block. For example, Oi -0.54181487 0.29839834 -0.75904939 charge 7 would set the nuclear charge of your interstitial oxygen atom to 7. You would then use the charge directive to set the total charge of the whole system to fix the number of electrons in the system. If you want to calculate the system with a specific total charge for the interstitial atom, then you would need to use the constrained DFT formalism (http://nwchemgit.github.io/index.php/Release65:Density_Functional_Theory_for_Molecules#CDFT_...). Best, Sean

Clicked A Few Times

5:02:58 AM PDT - Tue, Jun 30th 2015
Thank you, Sean! Now I`ll try to use CDFT, it looks more clearer to me. Changing nuclear charge looks a bit confusing. Doesn't this change oxygen to nitrogen? If it is so, can't I use fluorine atom with charge 8 and oxygen basis set, to get negatively charged oxygen?

Forum Regular

9:03:16 AM PDT - Tue, Jun 30th 2015
Yes, in effect changing the nuclear charge like that would be equivalent to changing the element. I was more pointing out a functionality rather than suggesting a course of action. The constrained DFT approach sounds like it will be better for achieving what you are after. Of course if in the physical system you are trying to simulate the interstitial oxygen atom has a negative charge, then presumably if the level of theory is sufficient, you should be able to run a standard calculation and that interstitial oxygen atom would end up being approximately negatively charged. Best, Sean

Clicked A Few Times

So I added these strings to input:
charge -1
/

/
DFT
convergence nolevelshifting

odft

mult 2 // adding one alpha-electron to system

cdft 1 1 charge -1.0 // placing negative charge on interstitial atom

Calculation started without errors. Will see what I'll get

Forum Vet

1:28:50 PM PDT - Wed, Jul 29th 2015
I have just discovered that the CDFT implementation had a pretty problematic bug since version 6.3. I have found the fix for it, and the patch for 6.5 is available at http://nwchemgit.github.io/images/Cdft.patch.gz

Clicked A Few Times

12:51:46 AM PDT - Wed, Sep 16th 2015
Could you please describe this bug? When I got the results of calculations, I found that formation energy of negatively charged defect is negative, that has no sense. For a short period of time I'm not able to perform new calculation (ran out of CPU time), so can't test same calculations with patch. If patch won't affect total energies, how should I solve problem with defect negative formation energy?