Employing Raman gain of optical materials is appealing for a variety of laser applications, to include fiber laser combiners. In order for a Raman combiner to be efficient, the Raman material must have high gain, low loss figure at 1st Stokes wavelength, and high loss figures at higher order Stokes wavelengths. This paper demonstrates an efficient double-clad fiber Raman combiner utilizing fused silica core as gain material with microstructured cladding designed with filtering properties implemented for suppression of higher order Stokes propagation in the core. Comprehensive study results of this Raman combiner will be presented.
Bulk laser damage variability in deuterated potassium di-hydrogen phosphate (DKDP) crystals is well known and makes
online conditioning of multiple-beam laser systems difficult to optimize. By using an empirical model, called Absorption
Distribution Model (ADM), we were able to map the damage variability of the crystals (boule to boule as well as within
the same boule) in terms of defect populations using a damage probability test. Furthermore, a relationship on defect
density and the relative damage behavior of a boule based on its late growth behavior have been found and has been used
successfully to predict laser damage/conditioning using a damage probability test only.
The presence of defects in optical materials can lead to bulk damage or downstream modulation and subsequent surface
damage in high fluence laser systems. An inclusion detection system has been developed by the National Ignition
Facility Optics Metrology Group. The system detects small inclusions in optical materials with increased sensitivity and
speed over previous methods. The system has detected all known inclusions and defects and has detected previously
undetected defects smaller than 5 microns.
Previous work concluded that plasma scalds on laser-conditioned multilayer dielectric mirror coatings are a stable,
benign damage morphology. Recent large-aperture measurements indicate that plasma scalds may lead to fratricide of
down-stream optics by increasing beam contrast. This paper describes the results of measurements performed to
examine the effect of quasi-periodic plasma scalds covering the entire clear aperture on downstream beam modulation.
A collimated, linearly-polarized 1053-nm beamline was constructed that irradiated approximately 5 cm2 of the plasma
scalded region. This beam was propagated ~8 meters and sampled with a 10-bit, megapixel CCD camera and analyzed
for contrast (peak/average intensity). A lineout across the sample was built up by translating the optic across the beam.
The contrast results were compared to a baseline wedged flat with surface figure of λ/100 and a contrast adder for the plasma scalds calculated. This was defined by. In all, optics with average plasma scald fractions of 0.9, 2.3, 4 and 14% were measured. Preliminary results indicate that plasma scald fractions of 4% and below contribute a contrast adder of less than 2.5%.
A multi-wavelength laser based system has been constructed to measure defect induced beam modulation (diffraction) from ICF class laser optics. The Nd:YLF-based modulation measurement system (MMS) uses simple beam collimation and imaging to capture diffraction patterns from optical defects onto an 8-bit digital camera at 1053, 527 and 351 nm. The imaging system has a field of view of 4.5 x 2.8 mm2 and is capable of imaging any plane from 0 to 30 cm downstream from the defect. The system is calibrated using a 477 micron chromium dot on glass for which the downstream diffraction patterns were calculated numerically. Under nominal conditions the system can measure maximum peak modulations of approximately 7:1. An image division algorithm is used to calculate the peak modulation from the diffracted and empty field images after the baseline residual light background is subtracted from both. The peak modulation can then be plotted versus downstream position. The system includes a stage capable of holding optics up to 50 pounds with x and y translation of 40 cm and has been used to measure beam modulation due to solgel coating defects, surface digs on KDP crystals, lenslets in bulk fused silica and laser damage sites mitigated with CO2 lasers.
The large-aperture (up to 40 cm × 80 cm) mirrors required for the National Ignition Facility have very stringent specifications. The specifications include requirements for transmitted and reflected wavefront over a wide spectral frequency, surface quality, laser resistance, spectral characteristics, etc. In order to validate optic performance, metrology tools were fielded at optic fabrication vendors to assure production control. These tools include interferometers, large-area conditioning stations, and photometers. Of the 1800 large-aperture mirrors required for the NIF, approximately 35% have been completed. This presentation will review the types of large-aperture mirrors used on NIF along with the performance of NIF optics as measured and received from our vendors.
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.
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.
Recent work has shown that the damage resistance of both ICF-class (1600 cm2) DKDP tripler crystals and SiO2 components (lenses, gratings and debris shields) benefits from laser raster scanning using pulsed lasers in the 350 nm range. For laser raster scanning to be a viable optical improvement tool for these large optics, damage improvement must be optimized while maintaining scan times of less than 8 hours/optic. In this paper we examine raster scanning with small beams from tabletop laser systems. We show that 120 Watts of average power is required for a tabletop scanning system at one optic/day. Next, we develop equations for total scan time for square and round top heat beams and round and rectangular Gaussian beams. We also consider the effect of packing geometry (square vs. hexagonal), examine the deviations from uniform coverage with each scan geometry and show that hexagonal packing yields lower scan times but is less efficient in coverage than square geometry. We also show that multiple passes at low packing densities are temporally equivalent to a single pass with higher packing density, and discuss the advantages of each method. In addition, we show that the differences between hexagonal and square scan geometries are negated when pointing errors and fluence fluctuations from the laser are considered.
In this paper we present the results of bulk damage experiments done on Type-II DKDP triple harmonic generator crystals that were raster conditioned with 351 - 355 nm wavelengths and pulse durations of 4 and 23.2 ns. In the first phase of experiments 20 different scan protocols were rastered into a sample of rapid growth DKDP. The sample was then rastered at damage-causing fluences to determine the three most effective protocols. These three protocols were scanned into a 15-cm sample of conventional-growth DKDP and then exposed to single shots of a 1-cm beam from LLNL's Optical Sciences Laser at fluences ranging from 0.5 - 1.5X of the 10% damage probability fluence and nominal pulse durations of 0.1, 0.3, 0.8, 3.2, 7.0 and 20 ns. The experiment showed that pulse durations in the 1 - 3 ns range were much more effective at conditioning than pulses in the 16.3 ns range and that the multiple pass "peak fluence" scan was more effective than the single pass "leading edge" scan for 23.2 ns XeF scans.
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).
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.
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.
Laser conditioning by raster scanning DKP and DKDP crystals using Nd:YAG and XeCl excimer laser systems was demonstrated. The laser systems were evaluated to determine their respective feasibility of improving the damage thresholds of the harmonic materials for use on the National Ignition Facility (NIF). Crystals were first evaluated using an Nd:YAG laser (355 nm, 7.6 ns) by scanning 2 x 2 cm2 areas with sub-damage threshold fluences and then performing unconditioned (S/1) damage tests at 355-nm in the respectively scanned regions. Subsequently, five KDP and DKDP samples of various damage quality were raster scanned in a similar fashion at MicroLas GmbH (Goettingen, Germany) using a commercial Lambda Physik Excimer system (XeCl, (lambda) equals 308 nm, 20 ns). The samples treated in Germany were then tested at Livermore National Laboratory (LLNL) at 355 nm to demonstrate the excimer's potential as an alternative conditioning source.
Over the course of testing a substantial number of KDP and DKDP crystals from rapid and conventional growth processes, we have discovered that there is a consistent difference in the value of the damage resistance between z-cut and tripler, x-cut and y-cut crystals for a given test fluence. This increase in damage probability for tripler, x and y-cut crystals is consistent for both conventional and rapid growth KDP and well as DKDP. It also holds for unconditioned (S/1) and conditioned (R/1) tests and has values of 2.1+/- 0.6 and 1.5+/- 0.3 respectively. Testing has also revealed that there is no sensitivity to incident laser polarization. This is in direct contradiction to models based on simple, non-spherical absorbers. This result plus new information on the size and evolution of bulk damage density (see Runkel et al., this proceedings) has led to a reinterpretation of the growth parameter data for rapid growth NIF boules. It now appears that variations in impurity concentration throughout the boule do not affect the damage probability curve as dramatically as previously thought, although this is still a topic of intensive investigation.
This paper discusses the results of thermal annealing and in-situ second harmonic generation (SHG) damage tests performed on six rapid growth KDP type 1 doubler crystals at 1064 nm (1(omega) ) on the Zeus automated damage test facility. Unconditioned (S/1) and conditioned (R/1) damage probability tests were performed before and after thermal annealing, then with and without SHG on six doubler crystals from the NIF-size, rapid growth KDP boule F6. The tests revealed that unannealed, last-grown material from the boule in either prismatic or pyramidal sectors exhibited the highest damage curves. After thermal annealing at 160 degree(s)C for seven days, the prismatic sector samples increased in performance ranging from 1.6 to 2.4X, while material from the pyramidal sector increased only modestly, ranging from 1.0 to 1.3X. Second harmonic generation decreased the damage fluence by an average of 20 percent for the S/1 tests and 40 percent for R/1 tests. Conversion efficiencies under test conditions were measured to be 20 to 30 percent and compared quite well to predicted behavior, as modeled by LLNL frequency conversion computer codes.
A set of twenty-three 20-L crystallizer runs exploring the importance of several engineering variables found that growth temperature is the most important variable controlling damage resistance of DKDP over the conditions investigated. Boules grown between 45 degree(s)C and room temperature have a 50% probability of 3(omega) bulk damage that is 1.5 to 2 times higher than boules grown between 65 and 45 degree(s)C. This raises their damage resistance above the NIF tripler specification for 8 J/cm2 operation by a comfortable margin. Solution impurity levels do not correlate with damage resistance for iron less than 200 ppb and aluminum less than 2000 ppb. The possibility that low growth temperatures could increase damage resistance in NIF- scale boules was tested by growing a large boule in a 1000-L crystallizer with a supplemental growth solution tank. Four samples representing early and late pyramid and prism growth are very close to the specification as best it is understood at the present. Implications of low temperature growth for meeting absorbance, homogeneity, and other material specifications are discussed.
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.
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 LLNL's 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.
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.
Laser induced materials modifications in the bulk and on the surface of KDP and DKDP are studied using fluorescence image and spectroscopy. Photoluminescence is observed at damaged regions following above threshold exposure with an emission peak centered at 550-nm. In addition, surfaces exposed to > 100 high power, 355-nm laser pulse reveal a reduced surface finishing quality as evidenced by an associated emission under UV photoexcitation. The emission spectra from the laser-induced damage sites and the laser degraded surfaces are similar suggesting the generation of similar defect species.
KEYWORDS: Interference (communication), Crystals, Signal processing, Failure analysis, Diagnostics, Control systems, National Ignition Facility, Statistical analysis, Algorithm development, Data analysis
Automated damage testing of KDP using LLNLs Zeus automated damage test system has allowed the statistics of KDP bulk damage to be investigated. Samples are now characterized by the cumulative damage probability curve, or S-curve, that is generated from hundreds of individual test sites per samples. A HeNe laser/PMT scatter diagnostic is used to determine the onset of damage at each test site. The nature of KDP bulk damage is such that each scatter signal may possess many different indicator of a damage event. Because of this, the determination of the initial onset for each scatter trace is not a straightforward affair and has required considerable manual analysis. The amount of testing required by crystal development for the National Ignition Facility (NIF) has made it impractical to continue analysis by hand. Because of this, we have developed and implemented algorithms for analyzing the scatter traces by computer. We discuss the signal cleaning algorithms and damage determination criteria that have lead to the successful implementation of a LabView based analysis code. For the typical R/1 damage data set, the program can find the correct damage onset in more than 80 percent of the cases, with the remaining 20 percent being left to operator determination. The potential time savings for data analysis is on the order of approximately 100 X over manual analysis and is expected to result in the savings of at least 400 man-hours over the next 3 years of NIF quality assurance testing.
Over the past tow years extensive experiments has been carried out to determine the nature of bulk damage in KDP. Automated damage testing with small beams has made it possible to rapidly investigate damage statistics and its connection to growth parameter variation. Over this time we have built up an encyclopedia of many damage curves but only relatively few samples have been tested with large beams. The scarcity of data makes it difficult to estimate how future crystal will perform on the NIF, and the campaign nature of large beam testing is not suitable for efficient testing of many samples with rapid turn-around. It is therefore desirable to have analytical tools in place that could make reliable predictions of large-beam performance based on small-beam damage probability measurements. To that end, we discuss the application of exponential and power law damage evolution within the framework of Poisson statistics in this memo. We describe the result of fitting these models to various damage probability curves on KDP including the heavily investigated KDP214 samples.
Considerable attention has been paid over the years to the problem of growing high purity KDP and KD*P to meet damage threshold requirements of inertial confinement fusion lasers at LLNL. The maximum fluence requirement for KD*P triplers on the NIF is 14.3 J/cm2 at 351 nm in a 3 ns pulse. Currently KD*P cannot meet this requirement without laser (pre)conditioning. In this overview, recent experiments to understand laser conditioning and damage phenomena in KDP and KD*P are discussed. These experiments have led to a fundamental revision of damage test methods and test result interpretation. In particular, the concept of a damage threshold has given way to measuring performance by damage distributions using beams of millimeter size. Automated R/1 damage test have shown that the best rapidly grown KDP crystals exhibit the same damage distributions as the best conventionally grown KD*P. Continuous filtration of the growth solution and post-growth thermal annealing are shown to increase the damage performance as well.
Lynn Sheehan, Sheldon Schwartz, Colin Battersby, Richard Dickson, Richard Jennings, James Kimmons, Mark Kozlowski, Stephen Maricle, Ron Mouser, Michael Runkel, Carolyn Weinzapfel
The Laser Program at LLNL has developed automated facilities for damage testing optics up to 1 meter in diameter. The system were developed to characterize the statistical distribution of localized damage performance across large- aperture National Ignition Facility optics. Full aperture testing is a key component of the quality assurance program for several of the optical components. The primary damage testing methods used are R:1 mapping and raster scanning. Automation of these test methods was required to meet the optics manufacturing schedule. The automated activities include control and diagnosis of the damage-test laser beam as well as detection and characterization of damage events.
KEYWORDS: Monte Carlo methods, Silica, Failure analysis, Rhodium, Data modeling, National Ignition Facility, Optical testing, Error analysis, Crystals, Computer simulations
In this paper, a Monte Carlo computer analysis of the R/1 automated damage test procedure currently in use at LLNL is presented. This study was undertaken to quantify the intrinsic sampling errors of the R/1 ADT method for various types of optical materials, particularly KDP and fused silica, and to provide a recommended minimum number of test sites. A gaussian/normal distribution of 10 J/cm2 average fluence was used as a damage distribution model. The standard deviation of the distribution was varied to control its shape. Distributions were simulated which correspond to the damage distribution of KDP and fused silica. A measure of the variability in test results was obtained by random sampling of these distributions and construction of the cumulative failure probability 'S' curves. The random samplings were performed in runs of 100 'tests' with the number of samples per test ranging from 2 to 500. For distributions with (mu) /(sigma) equals 5-10, the study found an intrinsic error of 3 to 5 percent in the maximum deviation from the distribution average when using 100 sites test. The computations also showed substantial variation in the form of the CFD for any given test.
Recently reported experiments have investigated the statistics of laser damage in KDP and KD*P. Automated damage tests have allowed cumulative failure and damage probability distributions to be constructed. Large area tests have investigated the feasibility of on-line laser conditioning and damage evolution for tripler harmonic generation crystal on the NIF. These test have shown that there is a nonzero probability of damage at NIF redline fluence and that the damage pinpoint density evolves exponentially with fluence.
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.
Considerable attention has been paid over the years to the problem of growing high purity KDP and KD*P to meet damage threshold requirements on succeeding generations of inertial confinement fusion lasers at LLNL. While damage thresholds for these materials have increased over time, the current National Ignition Facility (NIP) maximum fluence requirement (redline) for KD*P frequency triplers of 14.3 J/cm2 at 351 nm, 3 ns has not been reached without laser (pre)conditioning. It is reasonable to assume that. despite the rapid increase in damage thresholds for rapidly grown crystals, ·a program of large scale conditioning of the 192 NIF triplers will be required. Small area ramp (R/1) tests on single sites indicate that KDP damage thresholds can be raised on average up to 1.5X the unconditioned values. Unpublished LLNL 3m raster conditioning studies on KDP, however, have not conclusively shown that off-line conditioning is feasible for KD*P. Consequently, investigating the feasibility of on-line conditioning of NIF triplers at 3m has become a high priority for the KDP damage group at LLNL. To investigate the feasibility of on-line conditioning we performed a series of experiments using the Optical Sciences Laser (OSL) on numerous samples of conventional and rapid growth KDP and KD*P. The experiment entailed exposing sites on each sample to a range of ramped shot (Nil) sequences starting at average fluences of -2 J/cm2 (in a 7 mm "top hat" beam @ 351 nm, 3 ns) up to peak fluences of approximately 13 J/cm2• Test results indicated that the most effective conditioning procedure entailed a 7-8 shot ramp starting at 2 J/cm2 and ending at 12-13 J/cm2• The pinpoint onset fluence for the 8/1 tests was 1.4 times that of the unconditioned site. Damage evolution appears to be exponential as a function of increasing fluence. When damage occurs after conditioning however, pinpoint density evolution exhibits a greater slope than less conditioned sites. The overall reduction in the total pinpoint number can be as high as 300X. Despite laser conditioning , the pinpoint onset for the samples considered is below the NIF redline fluence of 14.3 J/cm2• In addition, the exponential pinpoint evolution curves indicate that damage levels at NIF redline fluences will be on the order of 1 Q4 pinpoints/mm3• 1bis suggests that there will be significant damage in NIF triplers, however, substantial damage has not been observed in the large Beamlet tripler (conventionally grown KD*P) under similar exposure conditions. By applying the OSL damage evolution curves to model NIF THG output spatial profiles it is possible to show damage in NIF triplers will be slight, consisting of isolated clusters with a few pinpoints at high fluence portions of the beam. This prediction has been verified by scatter mapping the 37 cm Beamlet tripler crystal. These results will be discussed in a future memo. These results indicate the feasibility of on-line conditioning for the NIF laser. Keywords: KDP, DKDP, KD*P, bulk laser damage, laser conditioning
Interest in producing high damage threshold KH2PO4 (KDP) and (DxH1-x)2PO4 (DKDP) for frequency conversion and optical switching applications is driven by the requirements of the National Ignition Facility (NIF) at the Lawrence Livermore National Laboratory. At present only the best crystals meet the NIF system requirements at the third harmonic and only after a laser conditioning process. Neither the mechanism for damage in bulk KDP nor the mechanism for conditioning ins understood. As part of ta development effort to increase the damage thresholds of KDP and DKDP, we have been developing techniques to pinpoint the locations where damage will initiate in the bulk material. After we find these locations we will use other measurement techniques to determine how these locations differ from the other surrounding materials and why they cause damage. This will allow crystal growers to focus their efforts to improve damage thresholds. Historically, damage thresholds have increased it is believed as a consequence of increased purity of the growth solution and through the use of constant filtration during the growth process. As a result we believe that damage is caused by defects in the crystals and have conducted a series of experiments using light scatter to locate these defects and to determine when and where damage occurs. In this paper we present results which show a low correlation between light scatter from bulk defects in KDP and the initiation sites for damage. We have also studied the effects of thermal conditioning on light scatter, strain induced birefringence and damage threshold. We have seen evidence that regions of high strain also exhibit lower damage threshold than the surrounding lower strain material. When thermally conditioned, these crystals show a decrease in some of the strong linear scattering features and a decrease in the strain birefringence while the damage threshold in these regions increased to that of the surrounding bulk material.
We report the experimental results of impurity contamination and laser-induced damage investigations on rapidly grown potassium dihydrogen phosphate (KDP) crystals. Using absorption spectroscopy and chemical analysis, we determined the impurity distribution in the different growing sectors of KDP single crystals. The level of impurity was dependent on the starting materials and growth rate. We also studied the influence of impurities on the laser-induced damage in fast grown KDP. The laser damage threshold in the impurity- rich prismatic sector is same as in the high purity pyramidal sector within the experimental error. Meanwhile, the laser damage threshold (LDT) at the boundary of the prismatic and pyramidal sectors is less than half of that in the bulk. Furthermore, we found that the thermal annealing of the crystal eliminated the weakness of this sector boundary and increased its LDT to the same level as in the bulk of the crystal. Our result suggests that laser damage occurred in the vicinity of a high, localized strain field.
The damage threshold specifications for the National Ignition Facility will include a mixture of standard small- area tests and new large-area tests. During our studies of laser damage and conditioning processes of various materials we have found that some damage morphologies are fairly small and this damage does not grow with further illumination. This type of damage might not be detrimental to the laser performance. We should therefore assume that some damage can be allowed on the optics, but decide on a maximum damage morphology limit. A new specification of damage threshold termed 'functional damage threshold' was derived. Further correlation of damage size and type to system performance must be determined in order to use this measurement, but it is clear that it will be a large factor in the optics performance specifications. Large-area tests have verified that small area testing is not always sufficient when the optic in question has defect-initiated damage. This was evident for example on sputtered polarizer and mirror coatings where the defect density was low enough that the features could be missed by standard small-area testing. For some materials, the scale-length at which damage non- uniformities occur will effect the comparison of small-area and large-area tests. An example of this was the sub- aperture tests on KD*P crystals on the Beamlet test station. The tests verified the large-area damage threshold to be similar to that found when testing a small-area. Implying that for this KD*P material, the dominate damage mechanism is of sufficiently small scale-length that small-area testing is capable of determining the threshold. The Beamlet test station experiments also demonstrated the use of on- line laser conditioning to increase the crystals damage threshold.
Interest in producing high damage threshold KH2PO4 (KDP) and (DxH1-x)2PO4 (KD*P, DKDP) for optical switching and frequency conversion applications is being driven by the system requirements for the National Ignition Facility (NIF) at Lawrence Livermore National Lab (LLNL). Historically, the path to achieving higher damage thresholds has been to improve the purity of crystal growth solutions. Application of advanced filtration technology has increased the damage threshold, but gives little insight into the actual mechanisms of laser damage. We have developed a laser scatter diagnostic to better study bulk defects and laser damage mechanisms in KDP and KD*P crystals. This diagnostic consists of a cavity doubled, kilohertz class, Nd:YLF laser (527 nm) and high dynamic range CCD camera which allows imaging of bulk scatter signals. With it, we have performed damage tests at 355 nm on four different `vintages' of KDP crystals, concentrating on crystals produced via fast growth methods. We compare the diagnostic's resolution to LLNL's standard damage detection method of 100X darkfield microscopy and discuss its impact on damage threshold determination. We have observed the disappearance of scatter sites upon exposure to subthreshold irradiation. In contrast, we have seen scatterers appear where none previously excited. This includes isolated, large (high signal) sites as well as multiple small scatter sites which appear at fluences above 7 J/cm2 (fine tracking). However, we have not observed a strong correlation of preexisting scatter sites and laser damage sites. We speculate on the connection between the laser-induced disappearance of scatter sites and the observed increase in damage threshold with laser conditioning.
Single crystals of KH2PO4 (KDP) and (DxHlx)2PO4 (DKDP) will be used for frequency conversion and as part of a large aperture optical switch in the proposed National Ignition Facility (NW) at the Lawrence Livermore National Laboratory (LLNL). These crystals must have good optical properties and high laser damage thresholds. Currently these crystals have a lower laser damage threshold than other optical materials in the laser chain which has forced designers to limit the output fluence of the NIF in order to avoid damaging the crystals. Furthermore, while more efficient frequency conversion schemes are being explored both theoretically and experimentally, the advantages of these schemes can not be fully realized unless the damage thresholds of the conversion crystals are increased. Over the past decade, LLNL has generated an extensive data base on the laser damage in KDP and DKDP crystals both at the first and third harmonics of Nd-YAG.1 While the damage thresholds of these crystals have increased over this time period due, in part, to better filtration of the growth solution,2 the damage thresholds of the best crystals are still far below what is expected from theoretical limits calculated from the band structure of perfect crystals. Thus damage in KDP and DKDP is caused by defects in the crystals. We also rely on a process called laser conditioning to improve the damage thresholds of the crystals. Unfortunately, little is understood about the mechanism of laser induced damage, the conditioning process in the crystals, or the defects which are responsible for damage. We have recently implemented a scatter diagnostic for locating and studying defects in crystals and as a tool for studying the mechanism of laser damage and laser conditioning.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.