Dear Esayspin group,
I am a Ph.D. student from NUS. I have some problems regarding the fitting of Cu oxide. I tested a sample of CuGaZrOx mixed oxide and got the Cu2+ EPR spectra. I tried to analyze it but the code I wrote had some problems and the fitting did not work properly. Below are the attached code, DTA file, and DSC file. Can you help me correct the code and tell me the reason why I am wrong? I am looking forward to your reply.
Best wishes,
Xiaoyu
Dear Xiaoyu,
first it looks like you have the magnetic field value in Gauss, whereas EasySpin works in mT, so as a second line I would add the conversion:
B = B./10; % conversion G to mT
The microwave frequency was very approximate with 9.8, better use the full 9.825671 GHz.
Under these conditions A(Cu) will strongly contribute to spectral width, and those values seem small to me at the moment, best check the literature for your class of systems, same for g-values (which will have an offset here unless you have ensured calibration of the magnetic field).
Also, your spectrum is rather broad and anisotropic, so the choice of line broadening model is important.
You can look at the models in the ES collection of examples, e.g. https://easyspin.org/easyspin/documenta ... smain.html -> broaden.m
Or also the Cu examples there, which will have mostly narrower spectra than yours, so you will have to use larger values, e.g. higher gStrain.
Then I would suggest to first simulate a reasonable initial guess and plot by
[Bsim, spc_sim] = pepper(Sys0,Exp);
plot(B,spc./max(spc),B_sim,spc_sim./max(spc_sim));
Note also you typo: Sys0.lwpp = 10;
With this your simulation should work, however, you might need the changes described above to get better results.
Once this looks reasonable fitting with esfit will become much easier. Note that the call to esfit depends on the EasySpin version 5 vs. 6, if I remember correctly. I use ES6 and the call is:
esfit(spc,@pepper,{Sys0,Exp},{SysVary});
Dear Daniel,
Thanks for your reply. After correcting the code following your suggestions, it can work. But it needs more reasonable parameters applied and maybe there are two components. I will try again. Thank you very much.
Best wishes,
Xiaoyu