Nanosecond laser ablation modelling applied to LIBS diagnostics. In PFMC-20 Book of Abstracts - ISBN: 9789612643164

Laser Induced Breakdown Spectroscopy (LIBS) is a powerful diagnostic for the determination of the elemental composition of materials. Its inherent advantages, such as in-situ analysis capability and no need for sample pre-treatment, make LIBS a promising technique for monitoring Plasma-Facing Components after exposure to harsh plasma conditions. Within this context, determining the depth resolution of the technique and the heat-affected zone remains a challenging issue. To address this, a finite element model was developed using COMSOL Multiphysics to simulate laser ablation of materials predicting crater geometry and ablation rate. Nanosecond pulse duration was considered, as nanosecond LIBS currently represents the most widespread technology in the field of nuclear fusion [1]. The model solves the heat conduction equation assuming gaussian laser heat source. Depending on the fluence and the sample material two different mass removal mechanisms, normal evaporation and phase explosion, were considered. Additionally, the laser intensity reduction caused by the plasma shielding effect was included. For model validation, three materials were irradiated with nanosecond laser pulses in a low-pressure chamber: the fusion-relevant tungsten and molybdenum, and silicon, chosen for its lower boiling temperature and minimal surface roughness. Laser fluence was varied approximately from 2 J/cm2 to 90 J/cm2 and different numbers of consecutive pulses were applied. Simulations showed an agreement with experimental ablated volumes within 50% for tungsten in the 2-9 J/cm2 fluence interval, assuming material removal by normal evaporation. For silicon, better agreement (within 13%) was achieved in the 20-90 J/cm2 fluence range, considering phase explosion and plasma shielding effects. Scanning Electron Microscopy observations confirmed the presence of phase explosion effects in silicon, while no evidences of that were observed for tungsten in the above-mentioned fluence range. An underestimation of simulated ablation rate (up to 1.7 times for silicon and 1.9 for tungsten) was obtained with respect to experiments, probably due to the non-ideal gaussian spatial distribution of pulses. Since a picosecond LIBS is set to be installed on the upgraded-GyM linear device (BiGyM) the model is currently being updated to become valid for this temporal regime. This will require implementing a Two-Temperature Model to account for separate electron and ion heating dynamics.   Acknowledgements Work carried out in the frame of project NEFERTARI - CUP B53C22003070006, funded by the European Union under the National Recovery and Resilience Plan (NRRP) - NextGenerationEU. References [1] G. S. Maurya, et al., Journal of Nuclear Materials 541, 152417 (2020)