A powerful tool for comparing different test procedures to measure the probability and density of laser induced...

The determination of the laser damage resistance of optics in the nanosecond regime is based on statistical tests and approaches because the response of the components is mainly related to the presence of defects randomly distributed in the optics and is therefore probabilistic in nature. For practical reasons, the tests are mostly carried out with beams of small dimensions (several tens of micrometer), that make it possible to determine a damage probability from which a laser damage threshold is extracted. This threshold is, however, highly correlated with the size of the test beam and the sampling of the test procedure. Some measurements are also made with beams of large dimensions (several millimeters) from which a damage density is determined. However, the relationship between the damage probability and the damage density is not trivial. It is based on assumptions that are difficult to verify because the experimental validations are carried out on different laser installations. In order to study accurately the coherence between these tests with small and large beams, as well as the link between damage probability and damage density, it is necessary to perform measurements on the same laser installation. We propose here, to compare for the first time, the results obtained with the same laser source with a large beam and also with small beams. The small beams are shaped from phase objects specifically implemented to obtain several small beams from a single larger beam. The consistency of the laser damage that results from both sets of measurements is demonstrated here. It validates the assumptions made and the specific mathematical treatment implemented to establish the link between the two approaches. In fine, it also validates and strengthens the approach previously developed from the rasterscan procedure [Lamaignère et al., Rev. Sci. Instrum. 78, 103105 (2007)] used to measure damage densities from the scanning of optics with beams of small dimensions. The reported original work based on phase objects thus makes it possible to replicate small beam tests with a large beam facility. The comparison between the results from the small beams and the results from the large beam experiments definitively makes the link between damage probabilities and damage densities. This also shows that small beam tests are reasonable representative of tests carried out with large beams.