Dear colleague.

I don't think I can help you, but I will write anyway.

In magnetic resonance, usually, when calculating the relaxation time, as in calculating the width of the resonance line, for the correlation function inside the integral, either the Lorentzian form (if the spin-spin interaction is exchange) and Gaussian (if the spin-spin interaction is dipole-dipole) are taken. I think you should use one equation and use either the Lorentz approximation or the Gaussian (normal) one.

Sincerely,

E. Khalvashi

Can you please show the graphs?

There are no graphs in my articles. Due to the pandemic, I cannot see the books. Perhaps you will find graphs in the books:

Abragam and M. Goldman, Nuclear Magnetism: Order and Disorder (Clarendon Press, Oxford, 1982),

Goldman, Spin Temperature and Nuclear Magnetic Resonance in Solids (Clarendon Press, Oxford, 1970) &

M. Mehring, High Resolution NMR Spectroscopy in Solids (Springer-Verlag, Berlin-Heidelberg-New York, 1976)

You said

"I tried to fit the frequency ", please show the fits that you have made :)

You can find graphs, for example, in the article J. Deisenhofer, M.V. Eremin, D.V. Zakharov, V.A. Ivanshin, R.M. Eremina, H.-A. Krung von Nidda, A.A. Mukhin, A.M. Balbashov, and A. Loidl, Phys. Rev. B 65, 104440 (2002); cond-mat/0108515/. In “1996_Engl_e_083_02_0379” paper 1/T2 is equation (10), the second term in equations (15) - (17), (20) and third term in (25). You may be interested in equations (1) and (2) in “IJESIT1909201603_2784 Acceptance Letter”. First equation in p. 131 in publication “p129-133”. Maybe you will find something useful in chapters 4 - 6 of the article “JETP398-En_p. 12”.

Yes, Sir. I am attaching the fitted graphs.@Valentin Bogatu

Thanking you