 |
SEARCH
TOOLBOX
LANGUAGES
Forum Menu
How to get the open shell singlet state (singlet biradical) of ozone (O3)
From NWChem
Viewed 8720 times, With a total of 14 Posts
|
|
|
|
| 11:09:47 PM PST - Mon, Dec 26th 2011 |
|
I want to get the open shell singlet state of O3 using DFT method but failed. I have searched the internet and got some answer like:
For example, you could generate vectors
with a UHF triplet calculation and read these in to the DFT singlet
calculation with 'odft' enabled and it should strongly bias towards a
symmetry-broken UKS singlet wavefunction
I followed the example. First, I got the triplet vectors. Second I read the triplet vectors and use "odft" and "mult 1". However in the end the wavefunction still come back to the close shell singlet and <S2> = 0.0000.
By using Gaussian package, I can get the open shell state of O3 by using "guess=mix". The open shell state is 0.003a.u. lower in energy than the close shell singlet and <S2>=0.414265.
Is there some method similar to do this?
|
|
|
-
Bert Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
|
|
|
Forum Vet
Threads 4
Posts 597
|
|
| 2:00:51 PM PST - Wed, Dec 28th 2011 |
|
Yes, you could do something similar to this. I don't know what the level of mixing is, but what you can do is the following:
Before your second dft task (where you want to compute the open-shell singlet), add to the input deck:
set mixang 30
You can play with the value to get the state you want.
Also, you may want to look at setting max_ovl in the dft input block to maximize the overlap.
Bert
Quote: Dec 27th 6:09 amI want to get the open shell singlet state of O3 using DFT method but failed. I have searched the internet and got some answer like:
For example, you could generate vectors
with a UHF triplet calculation and read these in to the DFT singlet
calculation with 'odft' enabled and it should strongly bias towards a
symmetry-broken UKS singlet wavefunction
I followed the example. First, I got the triplet vectors. Second I read the triplet vectors and use "odft" and "mult 1". However in the end the wavefunction still come back to the close shell singlet and <S2> = 0.0000.
By using Gaussian package, I can get the open shell state of O3 by using "guess=mix". The open shell state is 0.003a.u. lower in energy than the close shell singlet and <S2>=0.414265.
Is there some method similar to do this? |
|
|
|
|
|
|
| 3:47:59 AM PST - Fri, Dec 30th 2011 |
|
title "oss_read_hf_mixang_2"
SCRATCH_DIR /state/partition1/
memory 3000 mb
geometry noautoz nocenter
symmetry c1
O 0.00000000 0.00000000 -0.50788777
O 0.00000000 -1.07796685 0.25394388
O -0.00000000 1.07796685 0.25394388
end
basis
* library 6-31G**
end
dft
xc b3lyp
ADAPT off
convergence nolevelshifting
odft
mult 1
grid nodisk
max_ovl
end
set mixang 30
task dft
|
|
|
|
|
|
|
| 3:52:14 AM PST - Fri, Dec 30th 2011 |
|
Dear Bert,
I have tried "set mixang 30" to "set mixang 100", but failed to get the open shell singlet state of O3.
I also tried to combine "set mixang 30" with reading vectors from triplet, but still failed. The final result always leads to the close shell singlet state.
I have put in the input information of mine above.
Have you tried ozone before and successful? If you were successful, can you paste your input file for me?
Thanks a lot!
|
|
|
-
Bert Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
|
|
|
Forum Vet
Threads 4
Posts 597
|
|
| 3:36:54 PM PST - Wed, Jan 4th 2012 |
|
Hmmm, weird state. Can you post or send me or post the specific alpha and beta orbitals so I can understand the character you're trying to get to and understand what needs to be done. With a 0.4 S2, wonder if this a spin-contamination and the state is artificially low.
Also, you are not using symmetry in your Gaussian run either?
Bert
Quote: Dec 30th 10:52 amDear Bert,
I have tried "set mixang 30" to "set mixang 100", but failed to get the open shell singlet state of O3.
I also tried to combine "set mixang 30" with reading vectors from triplet, but still failed. The final result always leads to the close shell singlet state.
I have put in the input information of mine above.
Have you tried ozone before and successful? If you were successful, can you paste your input file for me?
Thanks a lot! |
|
|
|
|
|
|
| 11:44:34 PM PST - Wed, Jan 4th 2012 |
|
Dear Bert,
There is a link on how to calculate the open shell singlet state of ozone using Gaussian.
http://www.cup.uni-muenchen.de/ch/compchem/energy/stability.html
I just followed the procedure of website.
And I think the symmetry is in the Gaussian calculation.
Thanks and regards
|
|
|
|
|
|
|
| 6:14:08 PM PST - Thu, Jan 5th 2012 |
|
I'm interested in your results, so I know what state you are looking at. Can you send me a copy out your Gaussian output with the HOMO and LUMO for both the alpha and beta orbitals of this biradical?
And, can you double check if you are using point group symmetry, and which point group.
Thanks,
Bert
|
|
|
|
|
|
|
| 1:07:54 AM PST - Fri, Jan 6th 2012 |
|
Dear Bert,
Can I know your email address? I can send you the output file generated from Gaussian.
|
|
|
-
Bert Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
|
|
|
Forum Vet
Threads 4
Posts 597
|
|
| 11:32:10 AM PST - Fri, Jan 6th 2012 |
|
bert.dejong@pnnl.gov
Quote: Jan 6th 8:07 amDear Bert,
Can I know your email address? I can send you the output file generated from Gaussian. |
|
|
-
Edoapra Forum:Admin, Forum:Mod, bureaucrat, sysop
|
|
|
Forum Vet
Threads 7
Posts 1110
|
|
| 1:38:01 PM PST - Fri, Jan 6th 2012 |
|
Input files
|
I have upload two input files that gives you the singlet broken symmetry solution for O3.
1) Using the "mixang" directive
http://www.nwchem-sw.org/images/O3_singl.nw
2) Using the fragment guess option
http://www.nwchem-sw.org/images/O3_af.nw
Cheers, Edo
|
|
|
|
|
|
|
| 12:12:42 AM PST - Sat, Jan 7th 2012 |
|
Dear Edo,
The examples are very cool. I will try them. Thanks a lot for your kind help!
|
|
|
-
Niri Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
|
|
|
Forum Regular
Threads 4
Posts 176
|
|
| 6:08:37 PM PST - Mon, Jan 9th 2012 |
|
Hi,
You will get <S2> = 0.5719 with Edo's input files. These are with B3LYP.
I am curious how you got <S2>=0.414265 with Gaussian. Did you use pure HF ?
Cheers,
-Niri
|
|
|
|
|
|
|
| 4:17:39 AM PST - Tue, Jan 10th 2012 |
|
Edo, we need to document the adapt keyword in the manual 
Bert
|
|
|
|
|
|
|
| 1:03:28 AM PST - Thu, Jan 12th 2012 |
|
Quote:Niri Jan 10th 1:08 amHi,
You will get <S2> = 0.5719 with Edo's input files. These are with B3LYP.
I am curious how you got <S2>=0.414265 with Gaussian. Did you use pure HF ?
Cheers,
-Niri
With gaussian, you can use stable=opt in the input file, then the software automatically find the open shell singlet state.
|
|
|
|
|
|
Clicked A Few Times
Threads 4
Posts 9
|
|
| 3:00:59 AM PDT - Tue, Sep 3rd 2013 |
|
Hi Niri,
I notice that, if you optimize the structure with the open-shell singlet, you get nearly the same
spin-contamination <S2>=0.414265 of Gaussian09. I think the structure of open-shell singlet is similar
to triplet than singlet.
Shahab
|
|
Edited On 3:07:23 AM PDT - Tue, Sep 3rd 2013 by Shahabnaghavi
|
|
--
ººººººººººººººººººººººººººººººººººººººººººººººººººººººººº
S. Shahab Naghavi
(SISSA)
Scuola Internazionale Superiore di Studi Avanzati
Via Bonomea 265, Tireste, Italy
Phone: +39-403787448
Email(1): snaghavi@sissa.it
ººººººººººººººººººººººººººººººººººººººººººººººººººººººººº
|
AWC's:
2.5.10 MediaWiki - Stand Alone Forum Extension
Forum theme style by: AWC
| |
