Difficulty using esfit for 4-coordinate Cu2+ system

[cjames17504]

I am able to access the interface for my script; however, MatLab always freezes when I try to use the Monte Carlo method.

Here is my script:

%Data input script for
clear all
[B,spc]= eprload('USU4_20db45K.spc'); %creats array of data for Bruker SPC/PAR
plot(B,spc); %plots data

%Defining the spin system

Sys.S = 1/2; %spin system for Cu2+
Sys.g = [2.058 2.058 2.260]; %g-values for spectrum
Sys.lw = 3.5;% lw in millitesla
Sys.Nucs = '63Cu,14N,14N,14N,14N';
Sys.A = [52, 52, 520; 34, 34, 24; 34, 34, 24; 34, 34, 24; 34, 34, 24]; %Hyperfine values for one Copper nuclei,and three nitrogen nuclei
params.Range = [0, 500];

%variation of paramaters

Vary.g = [0.02 0.02 0.04]; %vary the g-value
Vary.lw = 0.5; %vary the linewidth parameter by 0.5
params.Range = [0, 500];

%Experimental parameters
Exp.mwFreq = 9.4; % X-band
Temperature = 4.5; %Temperature in Kelvin

%
esfit('pepper',spc,Sys,Vary,Exp)


I don't know if it's a code problem, or if the system is too complex (no offense meant) for Easyspin, and it just hangs up Matlab? Just need a little guidance on this.

[Matt Krzyaniak]

So this is a mixture of code and system complexity.
In terms of the code you're including equivalent nuclei which leads to a lot of degeneracies and a difficulty in convergence.
You can get around this problem by making the nuclei slightly different, such as:
Sys.A = [52, 52, 520;34.03, 34.03, 24.03;34.02, 34.02, 24.02;34.01, 34.01, 24.01;34, 34, 24];
the above is sufficient for the algorithm to converge.

Now in terms of the complexity of your system, on my computer one simulation of the system takes ~140s. Couple that time with the number of iterations that need to be performed in some of the optimization routines and you get what appears to be a frozen esfit/matlab.

Unless you have resolved nitrogen hyperfine(as you have the system setup right now you're not going to see anything since lw>a_N) you are probably better off using HStrain or gStrain(http://easyspin.org/easyspin/documentat ... nings.html) rather than including the nitrogen nuclei.

[cjames17504]

Thanks for the input. I will try your suggestions.