Skip to Main content Skip to Navigation

Innovative design of polarizing matrices for the early detection of cancer by MRI

Abstract : The goal of this PhD project was to develop a new methodology allowing to immobilize paramagnetic species on mesoporous silica in order to optimize the dynamic nuclear polarization process. For this purpose, different polarizing matrices, denoted as HYPSO for “HYbrid Polarizing SOlids”, were prepared through a sol-gel process. These solids are highly porous and paramagnetic species are covalently attached onto the surface. One of the noticeable benefit of this method lies in the possibility to remove the “glass forming agent” because of a homogeneous and random repartition of the radicals onto the surface of these materials. Two types of polarizing matrices, HYPSO 2 and HYPSO 3 gave distinct polarizations: P(1H)= 50% and P(1H)= 63%. This difference was explained by the influence of the silica matrix which provides better performances when its porosity is interconnected in the 3 dimensions of the space. Following these results, new materials (HYPSO 5) were prepared. The influence of the pore sizes as well as those of the silica grains on the polarization were studied. Remarkable performances: P(1H)= 99% (± 5%) et P(13C)= 51 % were recorded, thereby demonstrating the efficiency of this new method. Moreover, the first MRI pictures using HYPSO 5 as polarizing matrices were obtained after hyperpolarization of a sodium acetate solution. Finally, we were interested to a new formulation based on silica xerogels. Preliminaries results showed good polarizations and the ability to use only one monolith to polarize different liquids
Document type :
Complete list of metadatas

Cited literature [222 references]  Display  Hide  Download
Contributor : Abes Star :  Contact
Submitted on : Monday, February 18, 2019 - 11:44:06 AM
Last modification on : Wednesday, November 20, 2019 - 3:20:03 AM


Version validated by the jury (STAR)


  • HAL Id : tel-02022867, version 1



Matthieu Cavailles. Innovative design of polarizing matrices for the early detection of cancer by MRI. Other. Université de Lyon, 2018. English. ⟨NNT : 2018LYSE1204⟩. ⟨tel-02022867⟩



Record views


Files downloads