Ability to control number and orientation of unpaired spins on individual atoms?

From NWChem

Viewed 130 times, With a total of 4 Posts
Jump to: navigation, search

Just Got Here
Threads 1
Posts 3
Hello,

I am wondering if there is a way to set the spin states (# of unpaired spins and direction up/down) of individual atoms that would provide more control over specifying an overall system multiplicity (ie. using mult or nopen depending on the theory used). I am interested in studying electron transfer (ET) between two species. For my system the pre- and post-ET states have the same overall multiplicity. I would like to run energy calculations (and perhaps geometry optimizations) with the transferred electron in the pre- and post- ET configurations.

My system, PuO2 (2+) and HS (1-) has the following conditions...

Pre-ET: 2 unpaired spin (up) on the Pu(6+) and no unpaired spin on the S. This would be mult=3 or nopen=2 depending on the theory used (dft vs hf).

Post-ET: 3 unpaired spin (up) on Pu (now 5+) and 1 unpaired spin (down) on the S (part of a now neutral HS). nopen for this system would be 4 and the multiplicity would actually be 3 I think... (3 up - 1 down +1 = 3). I have tested mult=4 just to see what happens too.

For my pre-ET configuration, odft calculations show Mulliken spin values of 1 on the S and a lower than expected values for Pu. uhf calculations show S spin of -1 with Pu much higher than expected. I want to see what the energy looks like if there is ~0 spin on S...

If I run the post-ET configuration as an uhf calculation with nopen 4, I get a positive/up spin of 1 on the S instead of -1 and a lower than expected value on Pu. If I do a odft calculation with mult=4 I get the same thing. If I do mult=3 I just get the same result as my pre-ET state.

Is there a way, using either with dft or hf, to specifically designate the # of unpaired spins and their directions (up/down) for individual atoms? At the moment the treatments I have tried with the overall mult/nopen are not providing enough control.

Thank you,

Will
Edited On 1:13:13 PM PDT - Mon, Jun 12th 2017 by Benderwm

Forum Vet
Threads 7
Posts 1355
Have a look at this example for O3 that makes use of the fragment guess option
http://www.nwchem-sw.org/images/O3_af.nw

For more details

http://www.nwchem-sw.org/index.php/Release66:Hartree-Fock_Theory_for_Molecules#Superpositi...

http://www.nwchem-sw.org/index.php/Release66:Density_Functional_Theory_for_Molecules#ODFT_...

Just Got Here
Threads 1
Posts 3
Hi Edoapra,

Thanks for your quick reply.

I have tried this fragment guess strategy before and have just tried again to be sure. However, this process as implemented via the vectors input fragment frag1.mo frag2.mo... command only provides the algorithm with a guess. It does not continue to hold constant the spin conditions (magnitude or orientation) for a given fragment, it will use these as a starting point for a new calculation that uses a whole system multiplicity/nopen value.

As a result, I get the exact same energy, charge, and spin values for calculations that are run with and without the use of fragments.

Are there any other ways to fix the spin and its orientation for specific atoms?

Thank you,

Will

Forum Vet
Threads 7
Posts 1355
You might try Constrained DFT
http://www.nwchem-sw.org/index.php/Release66:Density_Functional_Theory_for_Molecules#CDFT_...

Just Got Here
Threads 1
Posts 3
OK, this looks like it could be promising.

I will give it a go.

Thank you!


Forum >> NWChem's corner >> Running NWChem



Who's here now Members 0 Guests 0 Bots/Crawler 1


AWC's: 2.5.10 MediaWiki - Stand Alone Forum Extension
Forum theme style by: AWC