This work applies different thermal analysis methods to polymer based materials degradation, studying the degradation process itself and evaluating the degree of material damage as a consequence of chemical degradation by thermal or radiation effects. On the one hand, thermal degradation in varied atmospheres is investigated by means of thermogravimetric analysis (TGA) in dynamic experiments. The authors find that the evolution of the sample mass follows a mixture of logistics models, and these can fit an overall TGA curve. The fitting parameters have important physical meaning related to the kinetics of the different processes involved and to the relative amount of each component in the sample. The method itself entails separating overlapping processes. Other improvements made by the authors are related to reducing the noise and smoothing the TGA and differential scanning calorimetry (DSC) data, particularly when estimating TGA derivatives through logistic regression.

Analyzing many materials by means of TGA results in more or less complex traces that do not allow a simple parametric fit like the one described above. Although it reproduces asymptoticity at the beginning and end of the reaction, there are times when many processes overlap, resulting in a complex trace that would need a high number of logistic components to be adequately fitted. However, it is possible to use a local polynomial regression model instead. This is also applicable to DSC traces, whose shapes are totally different from those found in TGA. The authors propose a model based on a nonparametric estimation, where the fit's suitability very much depends on the bandwidth selection, especially where derivatives are concerned. The proposed model gives a satisfactory fitting. It smoothes noise and always provides reliable values, different from those obtained by other methods strongly dependent on user choice.

On the other hand, to evaluate the degree of damage by thermal analysis methods, dynamic mechanical analysis (DMA) is applied to polyamides. The glass transition temperature is measured before and after exposure to varying doses of proton radiation, emulating the space environment. Other examples show how exposure over long periods at moderately elevated temperatures results in reduction of some mechanical properties. Additionally, the effect of different nanofillers on styrene-isoprene-styrene block copolymers is evaluated by DMA. A shift in the glass transition temperature seems to be dependent on nanofiller content. The degradation of some materials suitable for space applications, such as polyethylene and polyamide, are also briefly reviewed.