Analyses multivariées et multiblocs d'images hyperspectrales pour la prédiction de la corrosion sur le long terme : application à la corrosion atmosphérique d'alliages ferreux historiques

Abstract : This work aims to develop a methodology in order to understand atmospheric long term corrosion mechanism with chemometric tools on iron materials. Structural and elementary data are acquired with Raman microspectrometry and Scanning Electron Microscopy coupled to Energy Dispersive X-Ray spectroscopy (SEM-EDX) on two samples, one extracted from Amiens’ cathedral and one from Metz’s cathedral. These two datasets are coupled together then studied with several multibloc approaches. The first studies spectral data in order to avoid information loss and the second use the same reduced data in order to simplify the study. These two approaches give same results which leads to: minor elements are colocalized with some crystalline phases. For example, ferrihydrite is colocalized with minor elements (silicon, phosphorus and potassium). This can influence crystalline phase’s physicochimico reactivity properties in corrosion product layer and the corrosion rate.
Document type :
Theses
Complete list of metadatas

Cited literature [87 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-01795165
Contributor : Abes Star <>
Submitted on : Friday, May 18, 2018 - 11:00:21 AM
Last modification on : Thursday, February 7, 2019 - 1:32:26 AM
Long-term archiving on : Monday, September 24, 2018 - 11:31:09 AM

File

2017PA066472.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-01795165, version 1

Citation

Robin Le Penglau. Analyses multivariées et multiblocs d'images hyperspectrales pour la prédiction de la corrosion sur le long terme : application à la corrosion atmosphérique d'alliages ferreux historiques. Matériaux. Université Pierre et Marie Curie - Paris VI, 2017. Français. ⟨NNT : 2017PA066472⟩. ⟨tel-01795165⟩

Share

Metrics

Record views

205

Files downloads

204