Abstract
This Ph.D. work deals with the silica mechanical resonator used in the design of high-performance gyroscopes. The silica is chosen for its property of extremely low internal damping coefficient, which allows the obtaining of very strong mechanical overvoltage. Despite the use of an industrial manufacturing process adapted to the realization of resonators whose functional characteristics conform with the need, it is observed, on certain parts the appearance, with the wire of time, of a particular typology of defect, which generates a degradation of the properties of resonators. These defects do not correspond to the expectations of theoretical/behavioral models verified on most of the population of resonators produced. It is, therefore, necessary to go further in the understanding of the underlying mechanisms and their contributors, such as residual stresses, storage atmosphere, storage time... The objective of this thesis work is therefore to better understand the state of the surface and sub-surface of silica and its evolution in time depending on the shaping of this surface and the different treatments it may undergo during its production and storage. For this, the thesis work is based on two axes of study. The first one focuses on the defects produced during the manufacturing process of the resonator, with a sampling of the resonators directly on the production line at various stages. The second axis of study is more particularly interested in the origin of the defects, unexpected, which are observed. Different hypotheses are explored to explain them.