Page 1 of 1
E/D strain
Posted: Thu Dec 17, 2015 3:10 am
by algen
Hi there,
I am working on an iron metalloprotein system (S=5/2) and need to model ZFS parameters. I understand that these are implemented in chili using Sys.D = [D E]. My question is largely about how to model strain in these types of systems. In previous software that I've used and in most published papers strain is reported and fitted as E/D strain (as in, sE/D = 0.1 cm-1, for example). I see that DStrain specifies a distribution for D and E independently. This might be a totally silly question, but is sE/D just calculated arithmatically from DStrain = [x, y]? Thanks so much for any insight!
Re: E/D strain
Posted: Thu Dec 17, 2015 4:31 pm
by Stefan Stoll
Typically, in these types of systems the spectrum is pretty invariant to D, since D is very large compared to the typical microwave quantum. Therefore, you can assume D as constant and vary E. Then the variation in E/D is just the variation in E multiplied by D.
You could something like
Code: Select all
D = 10000; % insert your value here
EoD = 0.1; % insert your value here
D_fwhm = 0; % insert your value here
EoD_fwhm = 0.02; % insert your value here
Sys.D = [D, D*EoverD];
Sys.DStrain = [D_fwhm, D*EoD_fwhm];
Re: E/D strain
Posted: Tue Dec 22, 2015 2:55 am
by lorenz
I have also a short question on E- and D-strain.
If my expected E (and D) distribution is in the magnitude of 10-20% of the E (and D) value, is the strain-model still valid or would it be better to run a loop over the E/D-distribution.
kind regards
lorenz
Re: E/D strain
Posted: Tue Dec 22, 2015 12:46 pm
by Stefan Stoll
Typically, you should be safe with 10% or lower. However, this strongly depends on the rest of the spin Hamiltonian parameters.
It's always best to at least once check the perturbative strain treatment done by EasySpin with explicit averaging over an E distribution using a for-loop for your given spin system.
Re: E/D strain
Posted: Wed Dec 23, 2015 7:32 am
by lorenz
Thanks for your answer!
Re: E/D strain
Posted: Fri Dec 25, 2015 11:25 am
by algen
Hi Stefan,
Thanks so much for your reply--that's very helpful. I can then report or play with EoD or EoD_fwhm directly, it looks like. A much more straightforward solution than anything I had come up with!
Happy Holidays!
Re: E/D strain
Posted: Thu Dec 15, 2016 11:02 am
by frapart
Dear spectroscopist
I am a bit surprised, I usually use pepper for frozen samples simulation. I tried with chili and easy spin return an error : chili doesnot support strains.
I am trying to make a simple rhombic iron three simulation.
Best regards
Yves
Re: E/D strain
Posted: Fri Dec 16, 2016 11:04 am
by Stefan Stoll
DStrain is indeed only supported for rigid-limit simulations with pepper. In all other case, use an explicit loop over a distribution of D values, simulate each spectrum in turn, and then add them up using appropriate weights.
Re: E/D strain
Posted: Wed Jan 04, 2017 7:27 am
by frapart
Thank you very much answer. That what I was assuming at the begining.
All my best wishes for 2017