International audienceIn order to resolve problems concerning the understanding and the control of laser-induced damage of silica optical elements, a collaboration between the CEA and different university laboratories has been undertaken. Ultra-pure silica model samples, seeded with gold nanoparticles whose diameter did not exceed 5 nm, were prepared. The aim in using these samples was to observe the mechanism of damage initiation that could be attributed to inclusions of nanometric size. This paper presents the different steps encountered during this study: preparation of the samples, the laser-induced damage tests, the Nomarski and atomic-force microscope observations of this damage and a series of experiments using a time-of-flight mass spectrometer at Argonne National Laboratory. The experimental data are then interpreted, and, in particular, compared to numerical simulations. A very encouraging result is the existence of a pre-damage phase at very low fluences that is not detectable by classical optical devices. The experimental means developed for such model samples should be transposable to the analysis of industrial glasses

The laser damage community has long been searching for the reproducibility of damage measurements in the nanosecond (ns) regime. Here we show that laser-induced bulk damage density of frequency conversion crystals can be measured with high accuracy and repeatability in the range from 2 to 20 ns. The rasterscan test procedure (Lamaignère et al 2007 Rev. Sci. Instrum. 78 103105), previously developed in order to determine laser damage density of large aperture UV fused silica optics, has been adapted to bulk damage measurement. The large volume scanned during tests permits us to measure very low damage densities. For smaller optical components, small volumes are tested using the normalized 1/1 test procedure. Whatever test procedures, accuracy and repeatability are obtained by means of a suitable data reduction. For comparison between different procedures, the classical damage probability plot has to be converted in terms of damage density. The consideration of error bars on damage site distributions is compulsory to compare experimental data. A special emphasis is put on damage detection tools. When tests are carried out on diverse facilities, pulse duration, spatial distribution and beam overlap are the key parameters which are to be taken into account to compare experimental data. We describe the equipment, test procedures and data analysis to perform these damage tests with small beams (Gaussian beams, about 1 mm @1/e, and top-hat beams). Other tests are realized with larger beams (cm sized) which are compared with small beam results. The consistency of all the results gives confidence in the measurements. Reproducibility of measures is discussed in connection with current theoretical understanding of laser damage.

No abstract