Impact of a plasma mitigation process on the 266 nm pulsed LiDT of RAR nano-textured fused silica

Abstract

The pulsed laser induced damage threshold (LiDT) of Random Anti-Reflective (RAR) nano-textured fused silica optics has been shown to be many times higher than thin-film AR coated optics at wavelengths ranging from the near UV through the NIR. Because an RAR nano-texture is formed by a plasma etch process that removes part of the optic surface, the observed increase in damage resistance has kept track with the LiDT advances attained by low roughness super-polishing and damage pre-cursor mitigation techniques. In this work, nano-second pulse LiDT testing of RAR nano-textured optics was conducted at the deep UV wavelength of 266nm. The effect on 266nm LiDT of the uniform removal of additional surface material from fused silica optics using a dry plasma etch process was investigated. This plasma-polishing (PP), pre-RAR process was varied using fluorine-based chemistries that removed 100-300nm of material from each test surface, with surface roughness then characterized using white-light interferometry. Photothermal interferometry confirmed that no surface absorption was added by the PP, RAR, and PP-RAR plasma etching. Both standard grade, and ultra-low bulk absorption (low-OH) fused silica were included in the tests. RAR nanotextured surfaces showed an average damage threshold of 8.4 J/cm2, a level 3 times higher than a commercially available thin-film AR coated surface. Unexpected from pulsed LiDT testing at many longer wavelengths, all plasma etched surfaces exhibited less than half the damage threshold of the untreated, as-polished fused silica surfaces, and there was no observed correlation with surface roughness or plasma etch depth. From work by others it was theorized that exposure to the deep UV photons generated by the plasma might induce absorptive electronic defects in the fused silica material that could explain the reduced damage resistance relative to non-exposed surfaces. As an initial test of this concept an RAR nano-textured sample was baked at 400C to remove the suspected electronic defect. The subsequent pulsed LiDT of this one annealed sample was found to be 15.5 J/cm2, nearly double that of all other plasma etched samples. Further work to confirm this result is on-going.