C2 Triplet: A non-trivial DFT Computation

67 views
Skip to first unread message

Tanvir ur Rahman Chowdhury

unread,
Aug 23, 2021, 9:48:27 PM8/23/21
to NWChem Forum
Hello,
I have a DFT problem that needs to be done it two steps. First, I have to occupy a C2 dimer with the following occupation and perform a self-consistent DFT calculation:
     1.0          1.0
     1.0          1.0
     1.0          1.0
     1.0          1.0
     1.0          0.5
     1.0          0.5
     1.0          0.0
Second, I have to perform a non-self-consistent calculation of a regular C2 dimer with integer occupation using the Kohn-Sham potential/density from step one. In other words, I have to do just one iteration starting from the self-consistent Kohn-Sham potential or density from (b) self-consistent ensemble.

I wrote the following two codes for step one and two. I would really appreciate it if you let me know if this is the right way to do it.

Step one:
geometry  
  C   0.0 0.0 0.0
  C   0.0 0.0 1.243   
end  

basis  
 * library 6-311++G(3df,3pd) 
end

occup
 7 6 
 1.0 1.0 
 1.0 1.0 
 1.0 1.0 
 1.0 1.0 
 1.0 0.5
 1.0 0.5
 1.0 0.0

end

dft
 odft
 grid xfine  
 convergence energy 1d-8
 iterations 100  
 xc scan 
 vectors output c2.movecs
end

task dft

Step Two:
geometry  
  C   0.0 0.0 0.0
  C   0.0 0.0 1.243   
end  

basis  
 * library 6-311++G(3df,3pd) 
end


dft
 odft
 mult 3
 grid xfine
 xc scan 
 vectors input c2.movecs
 noscf
end

task dft

Niri Govind

unread,
Aug 24, 2021, 12:26:40 AM8/24/21
to nwchem...@googlegroups.com
Hi Tanvir,

Your two inputs look good. 

Best,
-Niri

Niri Govind
PNNL

--
You received this message because you are subscribed to the Google Groups "NWChem Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nwchem-forum...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nwchem-forum/3400cd92-e05d-493d-8fa1-637adcc34147n%40googlegroups.com.

Tanvir ur Rahman Chowdhury

unread,
Aug 24, 2021, 4:49:14 PM8/24/21
to NWChem Forum
Hi Niri,
Thank you very much for getting back. In the first step, I am not sure about the first two numbers of the occup directive. They are the numbers of alpha and beta electrons, as you mentioned earlier.

occup
 7 6 
 1.0 1.0 
 1.0 1.0 
 1.0 1.0 
 1.0 1.0 
 1.0 0.5
 1.0 0.5
 1.0 0.0

end

However, if I set the numbers 7 and 6, respectively. Then the spin multiplicity becomes 2. Alternatively, if I set the numbers to 7 and 5, that is:
occup
 7 5
 1.0 1.0 
 1.0 1.0 
 1.0 1.0 
 1.0 1.0 
 1.0 0.5
 1.0 0.5
 1.0 0.0

end
Then the output orbital analysis shows that there are only 5 beta orbitals and NWChem completely ignores the 6th orbital with 0.5 occupancies.

How can I occupy all alpha and beta orbitals with correct occupancy while keeping the spin multiplicity to 3?
Thanks in advance.

Edoardo Aprà

unread,
Aug 24, 2021, 6:30:52 PM8/24/21
to NWChem Forum
As mentioned earlier, the first line of the occup input field refers to the number of orbitals.
Let's keep as 7 7 (7 alpha orbitals and 7 beta orbitals) for simplicity.
I believe the following input works  for your case

#Step One
start c2_focc

geometry
  C   0.0 0.0 0.0
  C   0.0 0.0 1.243
end

basis  spherical

 * library 6-311++G(3df,3pd)
end

occup
 7 7

 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 0.5
 1.0 0.5
 1.0 0.0
end

dft
  mulliken

 odft
 mult 3
 grid xfine
 convergence energy 1d-8
 iterations 100
 xc scan
 vectors output c2.movecs
end

task dft

#Step Two:


dft
 mult 3

 vectors input c2.movecs
 noscf
end

occup
 7 7

 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 0.0
 1.0 0.0
end

task dft

Tanvir ur Rahman Chowdhury

unread,
Aug 24, 2021, 8:36:46 PM8/24/21
to NWChem Forum
Hi Edorado,
Thank you very much for your assistance. The codes ran successfully. However, the total DFT energy in step one was expected to be higher than in step two. I find the opposite.
In any case - thanks again.

Edoardo Aprà

unread,
Aug 24, 2021, 9:03:08 PM8/24/21
to NWChem Forum
The second step shows a rather poor integration (Numeric. integr. density =       11.878083456587)
If you disable symmetry in the second step, the integration improves (Numeric. integr. density =       11.999999995444) and the energy value changes (drops by ~ 0.08 hartrees), too.
Here is the modified input file.

 sym off
 adapt off

 vectors input c2.movecs
 noscf
end

occup
 7 7

 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 1.0
 1.0 0.0
 1.0 0.0
end

task dft
Reply all
Reply to author
Forward
0 new messages