You can estimate the activation energy Ea from the temperature dependence of resistance R(T) as a slope of the plot of log R vs. reciprocal temperature using the Arrhenius law: R(T)=R0*exp(Ea/kT). More interestingly, you can also estimate the hopping energy Eh (related to Mott like behavior) by using Eq.(13) for conductivity from the attached paper.

You should try to use the "master curve" method. Plot the electrical quantites vs. log frequency then thry to shift the curves obtained at various temperatures along the log frequency curve so as to obtain a single "master" curve. Finally plot the shif factor values vs. 1/T - an you get the activation energy form the slope. If you need more instruction, please contact me. It will work if you have a single activation energy for the process.

Thanks for your valuable answer Mr. Banhegyi Sir and can u plzz. explain it or instruct me for master curve method but we dont have the value of a single activation energy.

I would also need some help for calculating the Arrhenius relationship for glass fiber epoxy composites.

Please help!

How can i help you Mr. Garg feel free to ask coz now i calculated the activation energy using arrhenius plot.

Thanks Manindra for offering me help. I have data for the wt. gain of the specimens for different time periods say 15, 30 days. I have plotted the graph of %M vs sqrt (time). It shows linear, Fickian, behavior for certain time and then the slope is not linear anymore (non-Fickian) behavior I suppose.

Firstly, how do I calculate D(diffusion coefficient) from this data and then use Arrhenius eqn to find out the remaining life of the composite?

Please suggest

There are many methods by which you can calculate the activation energy, i recommend you, Flynn Wall Ozawa Method. 

from the Arrhenius relationship