I have been working on isoconversional methods for quite a time now. Have used all kinds of methods for kinetic analysis of my pyrolysis data: KAS, FWO, Vyazovkin, Kissinger etc. Going through literature I found, Friedman method has certain advantages over these integral methods (exclude Vyazovkin: it works quite well), but the d(a)/dT (a is normalized conversion) term has to be calculated judiciously. I have tried with forward/backward difference simply in excel ((a2-a1)/T2-T1)) and an almost similar thing in MATLAB too. But the results are not plausible enough.

I haven't tried calculating the slope at the different required points of (alpha) vs T curve, as it makes the whole analysis complex.

So what do I really do with the d(alpha)/dT term to compute kinetic parameters for my data?

An elaborate answer would be highly appreciated.

Thanks!!