Modélisation de la décomposition thermique des solides

Abstract : The work presented in this thesis focuses on understanding, describing and finally modelling the degradation, under heat exposure, of commonly used furnishing materials (mainly natural or synthetic polymers). This is not a new research field, as many models exist nowadays (such as Arrhenius models associated with varied kinetics functions). However, unlike for development related to gas phase chemical kinetics, most current solid phase chemical kinetics related models suffer from insufficient theoretical backgrounds. Moreover, laboratory experiments, more precisely small scale ones, reveal some behaviours, that are not only unpredicted by current model, but also invalidate some of the underlying assumptions these models are based on. This work context is more precisely related to solid anaerobic decomposition, at small scale, for mass in the range of one milligram, in order to focus on both mass loss kinetics and, to a lesser extent, on thermal transfer in solid phase. This work was structured into three main sections, each of them being a chapter of this thesis. First part concerns numerous experiments based on experimental plans and performed with a thermogravimetric analyzer (TGA). This aims at, on the one hand, determining most influent variables on thermal decomposition, and, on the other hand, identifying behaviour discrepancies between experimental data and predictions of current models available in the literature. Test campaign main result is that there is no mass convergence with TGA, even for very low masses. In addition, the effect of both heating rate and solid phase dilution were explored. This led to some behaviours significantly different from theoretical predictions. Second part concerns the modelling of material thermal degradation in a TGA analysis. It also highlights problems related to heat transfer inside solid material. It is demonstrated in this work that Dirichlet boundary conditions, which are usually used to describe TGA device, are not adequate to reproduce such experimental specificities. Third and last part is dedicated to the elaboration of a kinetic model relying on assumptions specific to the solid phase. First and foremost this model is on based on Eyring’s work on transition state theory. Then exploring thermodynamic formalism related to transition state allowed the formulation of a new model of thermal degradation called “outside Ellingham assumption”. This model is compared to two commonly used ones, so-called (A, E, 1) and (A, E, n) and led to best prediction for single reaction mechanisms. Finally, the prospects inspired by this work concern, first, a better characterisation of thermal boundary condition in TGA, in order to be able to reproduce experiments by modelling. Secondly, it may be relevant to extend the develop model to multiple reaction mechanisms, in order to increase its predictive capabilities for the targeted applications
Complete list of metadatas

Cited literature [156 references]  Display  Hide  Download

https://hal.univ-lorraine.fr/tel-02096047
Contributor : Thèses Ul <>
Submitted on : Thursday, April 11, 2019 - 9:29:34 AM
Last modification on : Friday, April 12, 2019 - 1:24:07 AM

File

DDOC_T_2018_0294_THIRY.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : tel-02096047, version 1

Collections

Citation

Aurélien Thiry-Muller. Modélisation de la décomposition thermique des solides. Thermique [physics.class-ph]. Université de Lorraine, 2018. Français. ⟨NNT : 2018LORR0294⟩. ⟨tel-02096047⟩

Share

Metrics

Record views

34

Files downloads

197