No description available.

Cette thèse est consacrée à l’étude du potentiel des cristaux biaxes dans le but de réaliser un laser de puissance. Les cristaux biaxes possèdent deux axes optiques ce qui donnent lieu à un effet appelé "réfraction (ou diffraction) conique". Cet effet, connu depuis 1832, est intensé- ment étudié depuis une décennie. Les progrès en croissance et découpe cristalline permettent d’obtenir des échantillons de longueurs et de qualités optiques suffisantes pour observer ce phénomène. L’utilisation de cet effet en cavité laser ayant déjà été réalisée [79, 78], l’objectif de cette thèse fut dans un premier temps de confirmer ces résultats. Ainsi, un système utilisant des cristaux de KGd(WO4)2 (KGW) dopés au Néodyme a été préalablement testé. Ces premiers résultats furent décevants. La stabilité, la qualité de faisceau ainsi que l’efficacité étaient plus que mé- diocres. L’efficacité maximale fut d’environ 40 % contre 74 % dans la publication [77]. Par ailleurs, le faisceau de sortie présentait un profil elliptique indiquant la présence d’une lentille thermique avec un fort astigmatisme. Ainsi, cette orientation, qui devrait être proche d’une direction athermale (ne présentant pas d’astigmatisme d’origine thermique) selon Biswal [10], ne l’est pas dans une cavité laser pour ce cristal. De plus ce laser est très peu stable et difficile à aligner. Les fluctuations importantes de l’efficacité de celui-ci ont abouti à une interrogation quant à la dépendance des propriétés optiques aux alentours de l’axe optique. Ainsi, dans les premiers chapitres de cette thèse, figurent une étude détaillée des principales propriétés optiques d’absorption et d’émission. Les ré- sultats de cette étude montrent que ces propriétés varient non seulement suivant la polarisation mais également suivant l’orientation aux alentours de l’axe optique. Etant donné que l’état de polarisation autour de l’axe optique varie avec l’orientation, les dépendances en orientation et polarisation sont liées. [...] Enfin, un montage innovant a été testé afin d’utiliser la réfraction conique. La puissance de sortie ainsi que l’efficacité de ce montage étaient très faibles, environ 800 mW en sortie pour 4 W de puissance de pompe. / This thesis is devoted to the study on the potential of biaxial crystals in order to increase laser output power. Biaxial crystals have two optic axes and an effect called conical refraction (or diffraction) can occur. This effect is known since 1832, and intensively studied since the last decade. Thanks to the progress of crystal growth and crystal cutting, it is possible to have long samples of good optical quality in order to ob serve the conical refraction. This effect has already been used in a laser cavity [79, 78]. The aim of this thesis was to confirm the results previously obtained. Thus, Neodymium-doped KGd(WO4)2 (KGW) crystals, cut along the optic axis, have been tested. These first results were disappointing. Stability, beam quality and efficiency were very low. The maximum efficiency achieved was arount 40% compared to the 74% claimed in publication [77]. Thus, this orientation which should be an athermal direction using the calculation of Biswal [10] shows astigmatism when there is a temperature gradient. Furthermore, this laser being difficult to align with strong intensity fluctuations, rise interrogations about the optical property variations around the optic axis. Thus, in the first Chapters of this thesis a study of those properties around the optic axis is given. It shows a strong variation of the optical properties depending on the polarization and orientation around the optic axis. The study of the refraction along the optic axis helps to understand this complex dependency with the orientation. In a second time, holmium-doped KY(WO4)2 (KYW) crystals have been tested for laser emission at 2 µm. The choice of this ion has been done on several criteria. The first one is that the ISL laboratory is used to work with this wavelength and has a lot of equipment. The second one is that Holmium ions have a small "quantum defect" (pump wavelength 1960 nm and lasing wavelength 2074 nm). This low "quantum defect" limits the dispersion between the pump and lasing wavelengths. This decreases the separation between the optic axes of both wavelengths. Thus, when the pump is aligned for conical refraction the lasing wavelength is also aligned along the optic axis. Thanks to a better setup (with new mechanical parts) the polarization and orientation dependencies of the laser efficiency have been investigated. Up to 3 W of output power with a slope efficiency of 70% has been reached near the optic axis. The doping concentration of the Holmium-doped KY(WO4)2 (KYW) being more appropriated for high power laser, no thermal lens effect has been observed. Furthermore, the first conical refraction laser with a quasi-three-level system has been realized. An efficiency of 50% has been achieved with such a conical refraction laser. Finally, a new innovative setup is proposed using the External Conical Refraction. An output power of 800 mW for 4 W of incident pump power has been reached.

The research that is presented in this thesis can be divided into two major parts. The first part concerns longitudinally pumped, bulk Er-Yb lasers. In these lasers, the main limitation is the thermal shortcomings of the phosphate glass host material. From the laser experiments and the spectroscopic measurements on crystalline host materials, as well as an investigation to bring further light to the physical background of the involved dynamics, the thesis presents some novel results that contribute to the search for a crystalline replacement. The second part concerns novel laser concepts applied to Yb-doped double tungstate lasers. Different crystal orientations are investigated, such as an athermal orientation for reduced thermal lensing and a conical refraction orientation for complete polarization tuning. Furthermore, the introduction of volume Bragg gratings in the cavity enables wide spectral tuning ranges and extremely low quantum defects. Regarding the first part, the main results are the achievement of 15 % slope efficiency in a monolithic, continuous-wave Yb:GdCOB laser and the achievement of Q-switching of the same laser. The Q-switched pulse durations were around 5-6 ns and the Q-switched slope efficiency was 11.6 %. For both lasers, a maximum output power of 90 mW was obtained, which is close to ordinary glass lasers under similar conditions. A spectroscopic investigation into the Er,Yb-codoped double tungstates was also performed and the results have enabled mathematical modeling of the fluorescence dynamics in these materials. Finally, the temperature dependence of the dynamics in Er,Yb:YAG was studied and the results have given some insight into the physical background of the mechanisms involved. Regarding the second part, different end-pumped Yb:KReW laser cavities were constructed to demonstrate the different concepts. With a laser crystal cut for propagation along the athermal direction at 17º angle clockwise from the dielectric direction Nm, the thermal lens could be reduced by 50 %. In these experiments the maximum output power was 4 W at 60 % slope efficiency. In another cavity incorporating a volume Bragg grating in a retroreflector set-up, the wavelength could be continuously tuned between 997 - 1050 nm. The spectral bandwidth was 10 GHz and the peak output power was 3 W. The same output power could also be obtained at 1063 nm with the grating positioned as an output coupler instead. If, on the other hand, the grating was positioned as an input coupler, 3.6 W output power at 998 nm was obtained at a quantum defect of only 1.6 %. Furthermore, using a crystal oriented for propagation along an optic axis, internal conical refraction could be used to establish arbitrary control of the polarization direction as well as the extinction ratio. Even unpolarized light could be enforced despite the highly anisotropic medium. With this configuration, the maximum output power was 8.6 W at 60 % slope efficiency which equals the performance of a reference crystal with standard orientation. The completely novel concepts of laser tuning with Bragg grating retroreflectors, of low quantum defect through Bragg grating input couplers and of polarization tuning by internal conical refraction can all easily be applied to several other laser materials as well. / QC 20100713