For decades, contamination was traditionally characterized through sample mass loss and deposit masses. Significantly improving our physical understanding and quantitative predictions of contamination now requires more detailed characterization for outgassing and contaminant deposition. Important progress was performed recently along this line with the separation and characterization of the chemical species outgassed by space materials, relying on the assessment of thermogravimetric analysis peaks (TGA) by mass spectrometry (MS). The determination of m/z fragments specific of a species allowed characterizing outgassing of each species, by following the corresponding MS signal during the outgassing phase. It allowed showing that in the vast majority of the studied cases, outgassing was ruled by diffusion laws, such as Fick's law.

The study reported here aimed at still going one step beyond, by characterizing several samples of the same material, with (very) variable diffusion lengths (i.e. effective material thicknesses), typically from 100 µm to cm scale. We report such experimental results for EC9323-2 epoxy glue, Kapton polyimide and QS1123 silicone glue. The modeling of the MS outgassing signals that are big enough compared to the MS baseline proved consistent with diffusion physics, with an effect of the diffusion length consistent with material geometry: for each material and species, one single set of diffusion parameters proved sufficient to correctly fit the outgassing of all samples by only changing its geometry (its diffusion length). This paves the way for a really physical modelling of outgassing, and an improved validity of long term predictions for arbitrary temperature history in flight.