EasySpin forum

Hello, Easyspin community!
I would like to fit cw EPR spectrum which should be attributed to some organic radical R-CH2-CH-OR type. Beta-hydrogen atoms are attached to the same carbon atom, so their HFCs have to be connected (the HCH angle is roughly 2pi/3). It is reasonable to use the so-called McConnell rule, so
A(beta-H_1) = const * (cos(t))²
A(beta-H_2) = const * (cos(t+2*pi/3))², where t is the angle between the p-orbital of the unpaired electron and the C-H bond.
I wrote this code:

clear
[B, spc, par] = eprload('DBDECP_12p9b0_21sc');
bckg = ( sum(spc(1:10)) + sum(spc(end-10:end)) )/20; % calculation of the background
spc = real(spc - bckg); % substaction of the background
spc = spc / max(abs(spc)); % normalization

Exp.nPoints = 2500;
Exp.mwFreq = 9.45124;
Exp.CenterSweep = [337.5 25];
Exp.ModAmp = 0.1; % 0.1 mT (1 G) modulation amplitude, peak-to-peak

%1= Radical 1 CH2-CH-OP
Sys1.Nucs = '1H,1H,1H';
Sys1.r1 = pi/1.9; % angle between "p-orbital" and H
Sys1.bbm = 2.8; % Beta B_max
Sys1.bae1 = 0;% Error in the angle between "p-orbital" and H
Sys1.ba1 = Sys1.bbm(cos(Sys1.r1))²;
Sys1.ba2 = Sys1.bbm(cos(Sys1.r1+2pi/3+Sys1.bae1))²;

Sys1.aah = 1.4; % A(Alpha-H)
Sys1.A = mt2mhz([Sys1.ba1; Sys1.ba2 ;Sys1.aah]); % mt
Sys1.g = 2.0025;
%Sys1.g = 2.003;
Sys1.lwpp = [0.55 0.55];
Sys1.weight = 0.5;

%1= Radical 1 CH2-C*H-OP
Vary1.r1 = pi;

SimOpt.Method = 'perturb2';
FitOpt.Method = 'simplex fcn'; % simplex algorithm, integrals of spectra
esfit('pepper',spc,{Sys1},{Vary1},Exp,SimOpt,FitOpt);

However, during the fitting procedure, the variation of r1 does not lead to the adjustment of the model spectrum (screenshots are attached). Is it possible to implement the parameterically defined HFCs using this or maybe some different approach?
Thanks

It looks like you are using EasySpin 5.2.x. Try using 6.0.0-dev.x, where such models are a lot easier to implement.