Extreme temperature sensing using Brillouin scattering in optical fibers
Stimulated Brillouin scattering in silica-based optical fibers may be considered from two different and complementary standpoints. For a physicist, this interaction of light and pressure wave in a material, or equivalently in quantum theory terms between photons and phonons, gives some glimpses of the atomic structure of the solid and of its vibration modes. For an applied engineer, the same phenomenon may be put to good use as a sensing mechanism for distributed measurements, thanks to the dependence of the scattered light on external parameters such as the temperature, the pressure or the strain applied to the fiber. As far as temperature measurements are concerned, Brillouin-based distributed sensors have progressively gained wide recognition as efficient systems, even if their rather high cost still restricts the number of their applications. Yet they are generally used in a relatively narrow temperature range around the usual ambient temperature; in this domain, the frequency of the scattered light increases linearly with the increase in temperature. The extension of this range toward higher (up to 800°C) and lower (down to 1 K) temperature is the main aim of this thesis. In both cases, our measurements on various fiber samples show that the aforementioned linearity does not hold. Most notably, the characteristics of Brillouin scattering at low and very low temperature are strikingly different from those observed under normal conditions; they are directly related to the disordered nature of the silica that constitutes the fiber, what motivates the large place we here devote to the physics of amorphous solids. Despite the observed nonlinearities, our results demonstrate the feasibility of thermometry based on Brillouin scattering over the full investigated temperature range. We can thus swap in the last part of this work the physicist's point of view for that of an engineer and present the setup and the performances of a sensor, or rather of a whole family of sensors we have tested and improved in the course of the last few years; moreover, some real measurement examples are given. Finally, we propose some possibilities of further enhancement, that remain right now at the preliminary stage of their development.
Faculté des sciences et techniques de l'ingénieur
Institut de transmissions, ondes et photonique
Jury: Marc Ilegems, Eric Picholle, Walter Scandale, Jean-Jacques Simond, Luc Thévenaz
Public defense: 2003-2-21
Record created on 2005-03-16, modified on 2016-08-08