Hello,
I am using EasySpin to simulate a molecule with both second and fourth order zero-field splitting parameters. In order to match the simulated spectrum to my experimental data I have included strain in the second order terms with Sys.DStrain, but there are features that require strain in the fourth order terms as well. I have been able to implement the necessary broadening by calculating spectra for a range of F values and then weighting and summing them, but in order to properly model the data I need to loop hundreds of times, which is very time consuming and makes it very difficult to use the simulation to find the best ZFS parameters.
Is there any way to include fourth order strain in the same way that the second order strain is calculated? Nonzero Sys.DStrain terms do not appreciably slow down the calculation, so I'm wondering if there's something equivalent to a Sys.FStrain term I could use, or somehow implement without looping.
Thanks!
Charles
Fourth order ZFS strain
-
Stefan Stoll
- EasySpin Creator
- Posts: 1073
- Joined: Mon Jul 21, 2014 10:11 pm
- Location: University of Washington
Re: Fourth order ZFS strain
EasySpin does not have built-in support for fourth-order ZFS strain.
The reason EasySpin's implementation for DStrain is so fast is that it involves the approximation that the resonance field positions vary linearly with D and E. This is a good approximation only if the spread of D values is much smaller than D itself, and should therefore be used with care. For fourth-order terms, this linear approximation is likely only valid for a very limited number of cases.
Looping over a distribution of your fourth-order parameter is the best and most accurate you can do. I would try to minimize the number of simulations N you have to do over your parameter distribution. Depending on other broadening mechanisms, it might be possible to get away with as few as N=10. Decrease N until the broadened lines in the spectrum loose their smoothness. Then do you fitting. At the end, increase N again to verify the spectrum is converged with respect to the strain broadening.
The reason EasySpin's implementation for DStrain is so fast is that it involves the approximation that the resonance field positions vary linearly with D and E. This is a good approximation only if the spread of D values is much smaller than D itself, and should therefore be used with care. For fourth-order terms, this linear approximation is likely only valid for a very limited number of cases.
Looping over a distribution of your fourth-order parameter is the best and most accurate you can do. I would try to minimize the number of simulations N you have to do over your parameter distribution. Depending on other broadening mechanisms, it might be possible to get away with as few as N=10. Decrease N until the broadened lines in the spectrum loose their smoothness. Then do you fitting. At the end, increase N again to verify the spectrum is converged with respect to the strain broadening.
Re: Fourth order ZFS strain
Thanks for the quick reply and explanation, that makes sense. I'll try to play around with the number of loops and other broadening terms to find the optimal time.