Michael J. Runkel

NIF Optical Materials and Fabrication Technologies: An Overview

The high-energy/high-power section of the NIF laser system contains 7360 meter-scale optics. Advanced optical materials and fabrication technologies needed to manufacture the NIF optics have been developed and put into production at key vendor sites. Production rates are up to 20 times faster and per-optic costs 5 times lower than could be achieved prior to the NIF. In addition, the optics manufactured for NIF are better than specification giving laser performance better than the design. A suite of custom metrology tools have been designed, built and installed at the vendor sites to verify compliance with NIF optical specifications. A brief description of the NIF optical wavefront specifications for the glass and crystal optics is presented. The wavefront specifications span a continuous range of spatial scale-lengths from 10 μm to 0.5 m (full aperture). We have continued our multi-year research effort to improve the lifetime (i.e. damage resistance) of bulk optical materials, finished optical surfaces and multi-layer dielectric coatings. New methods for post-processing the completed optic to improve the damage resistance have been developed and made operational. This includes laser conditioning of coatings, glass surfaces and bulk KDP and DKDP and well as raster and full aperture defect mapping systems. Research on damage mechanisms continues to drive the development of even better optical materials.

Laser-induced damage in deuterated potassium dihydrogen phosphate

Laser-induced pinpoint bulk damage of deuterated potassium dihydrogen phosphate at 351 nm is shown to depend on the propagation direction relative to the crystallographic axes and on growth temperature in addition to the previously reported dependence on continuous filtration. Pulse-length scaling is also consistent with earlier reports. The leading hypothesis for the cause of pinpoint damage is absorbing nanoparticle impurities, and our results are consistent with but not conclusive for that model. Advances in technology have led to greatly improved damage resistance.

NIF Pockels cell and frequency conversion crystals

The National Ignition Facility (NIF) at the Lawrence Livermore National Laboratory is a stadium-sized facility containing a 192-beam, 1.8-Megajoule, 500-Terawatt, ultraviolet laser system together with a 10-meter diameter target chamber with room for nearly 100 experimental diagnostics. Each beam line requires three different large-aperture optics made from single crystal potassium dihydrogen phosphate (KDP). KDP is used in the plasma electrode pockels cell (PEPC) and frequency doubling crystals, while deuterated KDP (DKDP) crystals are used for frequency tripling. Methods for reproducible growth of single crystals of KDP that meet all material requirements have been developed that enable us to meet the optics demands of the NIF. Once material properties are met, fabrication of high aspect ratio single crystal optics (42 × 42 × 1 cm) to meet laser performance specifications is the next challenge. More than 20% of the required final crystal optics have been fabricated and meet the stringent requirements of the NIF system. This manuscript summarizes the challenges and successes in the production of these large single-crystal optics.

Improving 351-nm damage performance of large-aperture fused silica and DKDP optics

A program to identify and eliminate the causes of UV laser- induced damage and growth in fused silica and DKDP has developed methods to extend optics lifetimes for large- aperture, high-peak-power, UV lasers such as the National Ignition Facility (NIF). Issues included polish-related surface damage initiation and growth on fused silica and DKDP, bulk inclusions in fused silica, pinpoint bulk damage in DKDP, and UV-induced surface degradation in fused silica and DKDP in a vacuum. Approaches included an understanding of the mechanism of the damage, incremental improvements to existing fabrication technology, and feasibility studies of non-traditional fabrication technologies. Status and success of these various approaches are reviewed. Improvements were made in reducing surface damage initiation and eliminating growth for fused silica by improved polishing and post- processing steps, and improved analytical techniques are providing insights into mechanisms of DKDP damage. The NIF final optics hardware has been designed to enable easy retrieval, surface-damage mitigation, and recycling of optics.

Effect of Vacuum on the Occurrence of UV-Induced Surface Photoluminescence, Transmission Loss, and Catastrophic Surface Damage

Vacuum degrades the transmittance and catastrophic damage performance of fused-silica surfaces, both bare and silica-sol anti-reflective coated. These effects may be important in certain space application of photonics devices. When exposed to hundreds of 355-nm, 10-ns laser pulses with fluences in the 2 - 15 J/cm2 range, transmittance loss is due to both increased reflectance and absorption at the surface. Spectroscopic measurements show that the absorbed light induces broadband fluorescence from the visible to infrared and that the peak photoluminescence wavelength depends cumulative fluence. The effect appears to be consistent with the formation of surface SiOx4/ (x < 2 with progressively lower x as cumulative fluence increases. Conversely, low fluence CW UV irradiation of fluorescent sites in air reduces the fluorescence signal, which suggests a photochemical oxidation reaction back to SiO2. The occurrence of catastrophic damage (craters that grow on each subsequent pulse) also increases in a vacuum relative to air for both coated and uncoated samples.

Engineered defects for investigation of laser-induced damage of fused silica at 355 nm

Embedded gold and mechanical deformation in silica were used to investigate initiation of laser-induced damage at 355 nm (7.6 ns). The nanoparticle-covered surfaces were coated with between 0 and 500 nm of SiO2 by e-beam deposition. The threshold for observable damage and initiation site morphology for these engineered surfaces was determined. The gold nanoparticle coated surfaces with 500 nm SiO2 coating exhibited pinpoint damage threshold of <0.7 J/cm2 determined by light scattering and Nomarski microscopy. The gold nanoparticle coated surfaces with the 100 nm SiO2 coatings exhibited what nominally appeared to be film exfoliation damage threshold of 19 J/cm2 via light scattering and Nomarski microscopy. With atomic force microscopy pinholes could be detected at fluences greater than 7 J/cm2 and blisters at fluences greater than 3 J/cm2 on the 100-nm-coated surfaces. A series of mechanical indents and scratches were made in the fused silica substrates using a non-indentor. Plastic deformation without cracking led to damage thresholds of approximately 25 J/cm2, whereas indents and scratches with cracking led to damage thresholds of only approximately 5 J/cm2. Particularly illuminating was the deterministic damage of scratches at the deepest end of the scratch, as if the scratch acted as a waveguide.

Results of pulse-scaling experiments on rapid-growth DKDP triplers using the Optical Sciences Laser at 351 nm

Results are reported from recently performed bulk-damage, pulse-scaling experiments on DKDP tripler samples taken from NIF-size, rapid-growth boule BD7. The tests were performed on LLNLs Optical Sciences Laser. A matrix of samples was exposed to single shots at 351 nm (3(omega) ) with average fluences from 4 to 8 J/cm2 for pulse durations of 1, 3 and 10 ns. The damage sites were scatter-mapped after testing to determine the damage evolution as a function of local beam fluence. The average bulk damage microcavity (pinpoint) density varied nearly linearly with fluence with peak values of approximately 16,000 pp/mm3 at 1 ns, 10,000 pp/mm3 at 3 ns and 400 pp/mm3 at 10 ns for fluences in the 8-10 J/cm2 range. The average size of a pinpoint was 10(+14,-9) micrometers at 1 ns, 37+/- 20 micrometers at 3 ns and approximately 110 micrometers at 10 ns, although all pulse durations produced pinpoints with a wide distribution of sizes. Analysis of the pinpoint density data yielded pulse-scaling behavior of t0.35. Significant planar cracking around the pinpoint as was observed for the 10 ns case but not for the 1 and 3 ns pulses. Crack formation around pinpoints has also been observed frequently for Zeus ADT tests at approximately 8 ns. The high pinpoint densities also lead to significant eruption of near-surface bulk damage. Measurements of the damage site area for surface and bulk gave ratios (Asurf/Abulk) of 2:1 at 1 ns, 7:1 at 3 ns and 110:1 at 10 ns.

Results of raster-scan laser conditioning studies on DKDP triplers using Nd:YAG and excimer lasers

In this paper we present the results of damage tests performed at 1064 and 355-nm at 8-10 ns on conventional and rapid growth DKDP tripler crystals. The crystals were laser conditioned prior to damage testing by raster scanning using either Nd:YAG (1064 and 355 nm, 8-10 ns) or excimer lasers at 248, 308 or 351 nm with pulse durations of approximately 30-47 ns. The results show that it is possible to attain increases in 355-nm damage probability fluences of 2X for excimer conditioning at 248 and 308 nm. However, these wavelengths can induce absorption sufficient to induce bulk fracture by thermal shock when impurities such as arsenic, rubidium and sulfur are present in the crystals in sufficient quantity. Tests to evaluate the efficiency of 351-nm conditioning (XeF excimer) show improvements of 2X and that thermal fracture by induced absorption is not a problem. We also discuss our recent discovery that low fluence raster scanning at UV wavelengths leads to 1064-nm damage thresholds of over 100 J/cm2 (10-ns pulses).

Effect of impurities and stress on the damage distributions of rapidly grown KDP crystals

Development of high damage threshold, 50 cm, rapidly grown KD*P frequency triplers for operation on the National Ignition Facility (NIF) in the 14 J/cm2, 351 nm, 3 ns regime requires a thorough understanding of how the crystal growth parameters and technologies affect laser induced damage. Of particular importance is determining the effect of ionic impurities which may be introduced in widely varying concentrations via the starting salts. In addition, organic particulates can contaminate the solution as leachants from growth platforms or via mechanical ablation. Mechanical stresses in the crystals may also play a strong role in the laser-induced damage distribution (LIDD), particularly in the case of large boules where hydrodynamic forces in the growth tank may be quite high. In order to investigate the effects of various impurities and stresses on laser damage we have developed a dedicated, automated damage test system with diagnostic capabilities specifically designed or measuring time resolved bulk damage onset and evolution. The data obtained makes it possible to construct characteristic damage threshold distributions for each samples. Test results obtained for a variety of DKP samples grown form high purity starting salts and individually doped with Lucite and Teflon, iron, chromium and aluminum show that the LIDD drops with increasing contamination content. The results also show that solution filtration leads to increased damage performance for undoped crystals but is not solely responsible for producing the high LIDDs required by the NIF. The highest LIDD measured on a rapidly grown sample indicate that it is possible to produce high damage threshold material using ultrahigh purity, recrystallized starting salts, continuous filtration and a platform designed to minimize internal stress during growth.

Analysis of bulk DKDP damage distribution, obscuration, and pulse-length dependence

Recent LLNL experiments reported elsewhere at this conference explored the pulse length dependence of 351 nm bulk damage incidence in DKDP. The results found are consistent, in part, with a model in which a distribution of small bulk initiators is assumed to exist in the crystal, and the damage threshold is determined by reaching a critical temperature. The observed pulse length dependence can be explained as being set by the most probable defect capable of causing damage at a given pulse length. Analysis of obscuration in side illuminated images of the damaged region yields estimates of the damage site distributions that are in reasonable agreement with the distributions experimentally directly estimated.