Hello!
I am simulating some EPR spectra of a transition metal cluster that exhibits a spin transition from S = 1/2 to S = 3/2 as the temperature increases. The signal for the 1/2 and 3/2 states can be both seen in the EPR spectrum, and I am trying to correlate the "weights" of each spin state to SQUID magnetometry data. Originally I was using the double integral of each signal to calculate the ratio of high spin to low spin. I then used this value to calculate the expected XmT value and compared it to the DC SQUID data. They nicely matched, however, I then realized that it might not be physically reasonable to compare the 3/2 and 1/2 signal for this purpose.
Next, I tried to simulate the 2 spin systems using the weight variable. I can get decent-looking simulations (see attached image), however now my weighting scheme (that corresponds to an accurate simulation when compared to the experimental spectrum) does not nicely match the SQUID data like the double integral values did. Is there a way I can make my weighting scheme more accurate? Also is it possible to compare the area under each signal for the previously mentioned purpose?
Simulations and code below:
For 4 K data
% spin system 1
sys1.S = 1/2;
sys1.g = [2.029 1.95 1.870];
sys1.weight = 0.95;
sys1.HStrain = [190 290 240];
vary1.g = [0.001 0.01 0.001];
vary1.HStrain = [10 10 10];
vary1.weight = 0.05;
% spin system 2
sys2.S = 3/2;
sys2.g = [4.958 3.643 3.550];
sys2.weight = 0.05;
sys2.HStrain = [941 1680];
vary2.g = [0.01 0.01 0.01];
vary2.HStrain = [150 150];
vary2.weight = 0.005;
sys = {sys1 sys2};
vary = {vary1 vary2};
spc2fit = spc1;
exp2fit = exp;