Morea, I have not read this paper thoroughly, since linear polymer binders in paints studied were chemically quite different from the cross-linkable polymer binders we expect in steel primer paints (epoxy, linseed-alkyds, polyesters etc.).
Anyway, DSC (thermal behavior) of all polymer binders would be roughly similar, so this paper is still valuable for us. And not so many papers on this topic (DSC of polymers at high/degradation temperatures) are available.
Notably, authors observed released heat/enthalpy in the range ca 3 to 8 kJ/mol, which is much lower (5x to 10x times lower) than the theoretical heat released by complete thermal oxidative degradation of these polymers, see Table 2 "Estimated enthalpies of thermal decomposition" (theoretical values are between ca 20 to 42 kJ/mol). Btw, measured released heats (overal exothermic effects) are in a good agreement with the heats reported in Bentham paper
Italian authors have not discussed in detail the reasons for these differences between theory and experiments (some of them can be connected with DSC method itself), but it is not so important. For us, they simply provided one of the indirect experimental clues that paints can show similar DSC traces under air as reported in Bentham paper for alleged "nanothermite"
Novadays, DSC and TGA curves can be recorded simultaneously using the same apparatus (e.g. entry Thermogravimetric analysis on Wiki), which is of course an ideal instrumentation for such measurements on thermal degradation of polymers.
(Btw, my link from yesterday to the paper does not work today, so
this should be a better link (?)