During preparation of sample, it may have some thermal history. By having first heating, we are removing all thermal history.  In next heating your sample shows real thermal response and it is free from any thermal history.

If you want to compare number of samples for DSC, we need to analyse them under same conditions.  That case its better to delete all thermal history and compare them under similar condition.

There is another reason to evaluate the second heating run.

If you put solid material in the DSC crucible, the thermal connection between the solid material and the bottom of the DSC crucicle changes during the first melting. Thus, the data of the first cycle differs from succeeding cycles. Typically, the melting peak of the first run is shifted to higher temperatures.

My comment to Johan: as far as i know, second heating doesn't always remove thermal history of the sample,so it is not always reproducible. For example in case of shellac, heating above its melting point creates cross link (aging) and the cross link can not be removed by heating even it improve the cross link.

Thank you all , really kind

thanks Johan for more details

you said that allows crystalline memory erases the polymer in 5 min , as I know there is no crystalline at this temperature > Tm , could you please explain more

Also , why the DSC curve rose up after melting point ?

The industry standard for DSC on a cured sample is to perform two heat cycles. Typically ramped from 0-200°C at 10°C/min.

The first heat gives you the thermal properties of the sample as it was originally cured. Think of it as the cure history of the sample. There are number of things to learn from a 1st heat. The more a sample is cured the higher the Tg value will be in the first heat. Residual cure can also be observed in the 1st heat by looking at enthalpy values above the glass transition. As the dH value disappears the degree of cure increases.

The second heat provides information on the ultimate thermal properties of the cured sample (provided that you have the optimum stoichiometry for epoxy: curing agent). Since the sample gets ramped all the way up to 200°C during the first heat scan, it is now fully cured as it starts the 2nd heat cycle. You may note that there is no residual enthalpy in 2nd heat scans and any signs of stress relaxation disappear as well.