Difficulties with Simulating Hyperfines and esfit

[bookit]

I've been ripping my hair out trying to simulate a spectra I have of a Copper(II) bipyridine nitrate sample. I can simulate it fairly well if I ignore the hyperfine interactions:

but the moment I try to fit to the hyperfines that are occurring at around 330mT, it gets really ugly. The hyperfines are most likely from the copper and have a distance of about 4 Gauss (~115MHz) separating each. When I try to fit using esfit, it ends up ignoring the hyperfines and moving towards the plot shown above. However, when I try to fit it manually, I can't get it close to where it should be.

Here's my code for esfit:

Code: Select all

clear
%Calling data
[bipyFieldG,bipySpec]=eprload('cu_bipy_25db_cw_rt_t2.DTA');
bipyFieldmT=bipyFieldG./10;

% Spin System Values
Sys0.S=3/2;
Sys0.g=[2.2279 2.1425 2.0210];
Sys0.gStrain=[.0868 .0264 .0383];
Sys0.lw=[.5 .5];
Sys0.Nucs='Cu';
%Hyperfines
Sys0.A=[115 115 115];
Sys0.AStrain=[10 10 10];

%Experimental Values
Exp.mwFreq=[9.637389];
Exp.CenterSweep=[326.715,100];
Exp.nPoints=10000;

%Fitting
Vary.g=[.01 .01 .01];
Vary.gStrain=[.025 .025 .025];
Vary.lw=[.25 .25];
Vary.A=[25 25];
Vary.AStrain=[10 10];


esfit('pepper',bipySpec,Sys0,Vary,Exp)

It results in this fitting:


Does anyone have suggestions on how to go about this? I'm fairly new to EasySpin so I could easily be missing something.

[Stefan Stoll]

It looks like you are trying to fit 12 parameters at the same time. That's too many. Fit without hyperfine, and then fix the g values and linewidth, and let only the hyperfine values vary. Then alternate.

[bookit]

I've done that, it still just tries to fit ignoring the hyperfines. It ends up changing the A and AStrain values so that the hyperfines aren't visible at all.

[Stefan Stoll]

Probably your search ranges are too wide. esfit won't automatically find the global minimum, it's a tool to help you find it, so it needs to be steered.

[katarkon]

Bookit, I see an error in your script. The fields A and AStrain in Vary structure have less elements than in Sys one. So, the last elements of the fields will not vary during fitting routine.
Also, the Sys.S=3/2 probably is incorrect. For copper(II) d9 ion the S=1/2 value is expected, for my opinion.
At last, I propose an explicit hyperfine splitting only on the first (weak-field) component of the spectrum.
Good luck.