In this talk I will review recent developments in understanding and controlling light-matter interaction and material response associated with laser powder bed fusion additive manufacturing. As part of the critical assessment of the physics of the process, validated hydrodynamic finite element model simulations have proven to be extremely valuable and can be used to inform rapid solidification microstructural models. I will also discuss new approaches to process optimization that have emerged from our modeling efforts which can improve material properties and part performance. Prepared by LLNL under Contract DE-AC52-07NA27344.
We present a simple and scalable method for the production of optics with incorporated metasurfaces, resulting in durable all-dielectric based meta-optics. The scalability and robustness of this method overcome limitations imposed by current technology when fabricating metasurfaces for high power laser applications, while the simplicity of the fabrication process makes it an exciting technique for metasurface generation. This talk will describe the method, show resultant fabricated metasurfaces and the sensitivity introduced by processing parameters – i.e. control over generated surfaces, and discuss the laser damage performance of these engineered large-scale metasurfaces.
3D printing of multi-material objects enables the design of complex 3D architectures such as printed electronics and devices. The ability to detect the composition of multi-material printed inks in real time is an enabling feature in a wide range of manufacturing sectors. In this study, dielectric properties of microscale embedded metal particles in a dielectric matrix have been characterized using impedance measurements as a function of particle size, shape, volume percentage and frequency. Measurements were found to agree well with calculations based on an anisotropic Maxwell-Garnett dielectric function model. Despite the metal loading exceeding the theoretical percolation threshold, a percolation transition was not observed in the experimental results. With this data, a calibration curve can be established to correlate metal loading with impedance or capacitance, which can be used with an in situ sensor for ink composition measurements during extrusion-based 3D printing. We demonstrate how an in situ sensor can locally measure the composition of the ink, allowing greater control over the resulting properties and functionality of printed materials.
Laser Powder Beam Fusion (LPBF)processes use laser beams to selectively melt powder layers and build three dimensional parts layer by layer. Usually, the beam has a Gaussian profile and the melt temperature peaks near the beam center. For typical conditions this temperature is well over the boiling point and drives intensive evaporation. Evaporation-driven recoil momentum can produce detrimental material spattering and keyhole porosity. Evaporation itself consumes a significant amount of energy thus degrading the process efficiency. It can therefore be beneficial to alter the beam shape so as to have the temperature distribution in the melt pool close to that of a flat top. We determined with a simple thermal model the beam shape providing a relatively flat temperature distribution . The optimal is found to be doughnut mode-like, skewed in the scan direction.
We did high fidelity simulations of the melt pool produced by the optimized beam and evaluated the possible benefits, including the efficiency increase.
We started the experiments with doughnut shape beam far from the optimal but also far from the Gaussian one. The experimental data will be compared with simulation results.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.
Irregularities in process conditions during powder bed fusion (PBF) additive manufacturing (AM) can lead to localized defects and poor part quality. At the same time, because PBF-AM is a layer-by-layer process, material properties such as defect concentration can be characterized in situ, thus providing an opportunity to ‘qualify as you build.’ Here we review methods based on high speed, multi-wavelength optical measurements of melt pool evolution, material ejection and layer-to-layer height variation during PBF processing, and compare the results to ex situ x-ray tomographic measurements. We also discuss the complex mechanisms related to the interaction between the melt pool, laser beam, and powder bed which ultimately drive defect formation. Along with providing process monitoring data to facilitate part certification, data provided by in situ optical diagnostics can help validate process models. The practical implementation of these high speed diagnostics into commercial platforms is also discussed. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.
Localized heating of roughened steel surfaces using highly divergent laser light emitted from high-power laser diode arrays was experimentally demonstrated and compared with theoretical predictions. Polarization dependence was analyzed using Fresnel coefficients to understand the laser-induced temperature rise of HY-80 steel plates under 383- to 612-W laser irradiation. Laser-induced, transient temperature distributions were directly measured using bulk thermocouple probes and thermal imaging. Finite-element analysis yielded quantitative assessment of energy deposition and heat transport in HY-80 steel using absorptivity as a tuning parameter. The extracted absorptivity values ranged from 0.62 to 0.75 for S-polarized and 0.63 to 0.85 for P-polarized light, in agreement with partially oxidized iron surfaces. Microstructural analysis using electron backscatter diffraction revealed a heat affected zone for the highest temperature conditions (612 W, P-polarized) as evidence of rapid quenching and an austenite to martensite transformation. The efficient use of diode arrays for laser-assisted advanced manufacturing technologies, such as hybrid friction stir welding, is discussed.
The primary sources of damage on the National Ignition Facility (NIF) Grating Debris Shield (GDS) are attributed to
two independent types of laser-induced particulates. The first comes from the eruptions of bulk damage in a
disposable debris shield downstream of the GDS. The second particle source comes from stray light focusing on
absorbing glass armor at higher than expected fluences. We show that the composition of the particles is
secondary to the energetics of their delivery, such that particles from either source are essentially benign if they
arrive at the GDS with low temperatures and velocities.
The ablation of magnetron sputtered metal films on fused silica substrates by a 1053 nm, picosecond class laser was studied as part of a demonstration of its use for in-situ characterization of the laser spot under conditions commonly used at the sample plane for laser machining and damage studies. Film thicknesses were 60 and 120 nm. Depth profiles and SEM images of the ablation sites revealed several striking and unexpected features distinct from those typically observed for ablation of bulk metals. Very sharp thresholds were observed for both partial and complete ablation of the films. Partial film ablation was largely independent of laser fluence with a surface smoothness comparable to that of the unablated surface. Clear evidence of material displacement was seen at the boundary for complete film ablation. These features were common to a number of different metal films including Inconel on commercial neutral density filters, stainless steel, and aluminum. We will present data showing the morphology of the ablation sites on these films as well as a model of the possible physical mechanisms producing the unique features observed.
A comprehensive study of laser-induced damage associated with particulate damage on optical surfaces is presented. Contaminant-driven damage on silica windows and multilayer dielectrics is observed to range from shallow pitting to more classical fracture-type damage, depending on particle-substrate material combination, as well as laser pulse characteristics. Ejection dynamics is studied in terms of plasma emission spectroscopy and pump-probe shadowgraphy. Our data is used to assess the momentum coupling between incident energy and the ejected plasma, which dominates the laser-particle-substrate interaction. Beam propagation analysis is also presented to characterize the impact of contaminant-driven surface pitting on optical performance.
We study the formation of laser-induced Hertzian fractures on silica output surfaces at high incident influences initiated by surface bound metal particles. Hertzian fracture initiation probability as a function of incidence influence is obtained for two particle materials. The resulting modified damage density curve shows prototypical features determined by the surface-bound particles population. The data is further used to calculate the coupling coefficient between incident energy and the ejected plasma momentum.
We report an investigation on the response to laser exposure of a protective capping layer of 1ω (1053 nm) high-reflector
mirror coatings, in the presence of differently shaped Ti particles. We consider two candidate capping layer materials,
namely SiO2 and Al2O3. They are coated over multiple silica-hafnia multilayer coatings. Each sample is exposed to a
single oblique (45°) shot of a 1053 nm laser beam (p polarization, fluence ~ 10 J/cm2, pulse length 14 ns), in the
presence of spherically or irregularly shaped Ti particles on the surface. We observe that the two capping layers show
markedly different responses. For spherically shaped particles, the Al2O3 cap layer exhibits severe damage, with the
capping layer becoming completely delaminated at the particle locations. In contrast, the SiO2 capping layer is only
mildly modified by a shallow depression, likely due to plasma erosion. For irregularly shaped Ti filings, the Al2O3
capping layer displays minimal to no damage while the SiO2 capping layer is significantly damaged. In the case of the
spherical particles, we attribute the different response of the capping layer to the large difference in thermal expansion
coefficient of the materials, with that of the Al2O3 about 15 times greater than that of the SiO2 layer. For the irregularly
shaped filings, we attribute the difference in damage response to the large difference in mechanical toughness between
the two materials, with that of the Al2O3 being about 10 times stronger than that of the SiO2.
Surface particulate contamination on optics can lead to laser-induced damage hence limit the performance of high power laser system. In this work we focus on understanding the fundamental mechanisms that lead to damage initiation by metal contaminants. Using time resolved microscopy and plasma spectroscopy, we studied the dynamic process of ejecting ~30 μm stainless steel particles from the exit surface of fused silica substrate irradiated with 1064 nm, 10 ns and 355 nm, 8 ns laser pulses. Time-resolved plasma emission spectroscopy was used to characterize the energy coupling and temperature rise associated with single, 10-ns pulsed laser ablation of metallic particles bound to transparent substrates. Plasma associated with Fe(I) emission lines originating from steel microspheres was observe to cool from <24,000 K to ~15,000 K over ~220 ns as τ-0.22, consistent with radiative losses and adiabatic gas expansion of a relatively free plasma. Simultaneous emission lines from Si(II) associated with the plasma etching of the SiO2 substrate were observed yielding higher plasma temperatures, ~35,000 K, relative to the Fe(I) plasma. The difference in species temperatures is consistent with plasma confinement at the microsphere-substrate interface as the particle is ejected, and is directly visualized using pump-probe shadowgraphy as a function of pulsed laser energy.
We study the formation of laser-induced shallow pits (LSPs) on silica output surfaces and relate these features to optical performance as a function of incident laser fluence. Typical characteristics of the LSPs morphology are presented. Closed-form expressions for the scattered power and far-field angular distribution are derived and validated using numerical calculations of both Fourier optics and FDTD solutions to Maxwell’s equations. The model predictions agree well with the measurements for precise profile micro-machined shallow pits on glass, and for pitting caused by laser cleaning of bound metal micro-particles at different fluences.
Surface modification of fused silica windows caused by the laser ablation of surface-bound microparticles is
investigated. Using optical and electron microscopies between laser pulses, we detail the ablation, fragmentation
and dispersal of 2-150 μm diameter particles of various materials. Following complete ablation and ejection of all
debris material, surface pitting was found to be highly dependent on material type and particle size. Subsequent
light propagation modeling based on pit morphology indicates up to ~4x intensification. Understanding this class of
non-local, debris-generated damage is argued to be important for effective design of high-power optical windows
and debris-mitigation strategies.
We present a method to repair damaged optics using laser-based chemical vapor deposition (L-CVD). A CO2 laser
is used to heat damaged silica regions and polymerize a gas precursor to form SiO2. Measured deposition rates and
morphologies agree well with finite element modeling of a two-phase reaction. Along with optimizing deposition
rates and morphology, we also show that the deposited silica is structurally identical to high-grade silica substrate
and possesses high UV laser damage thresholds. Successful application of such a method could reduce processing
costs, extend optic lifetime, and lead to more damage resistant laser optics used in high power applications.
Chemical vapor deposition (CVD) has been used for the production of fused silica optics in high power laser
applications. However, relatively little is known about the ultraviolet (UV) laser damage threshold of CVD films
and how they relate to intrinsic defects produced during deposition. We present a study relating structural and
electronic defects in CVD films to the 355 nm pulsed laser damage threshold as a function of post-deposition
annealing temperature (THT). Plasma-enhanced CVD, based on SiH4/N2O under oxygen-rich conditions, was used
to deposit 1.5, 3.1 and 6.4 μm thick films on etched SiO2 substrates. Rapid annealing was performed using a
scanned CO2 laser beam up to THT~2100 K. The films were then characterized using X-ray photoemission
spectroscopy (XPS), Fourier transform infrared spectroscopy (FTIR), and photoluminescence (PL). A gradual
transition in the damage threshold of annealed films was observed at THT up to 1600 K, correlating with a decrease
in NB silanol and broadband PL emission. An additional sharp transition in damage threshold also occurs at ~1850
K indicating substrate annealing. Based on our results, a mechanism for damage-related defect annealing is
proposed, and the potential of using high-THT CVD SiO2 to mitigate optical damage is also discussed.
Laser machining of optics to mitigate surface defects has greatly enhanced the ability to process large
optics such as those found in fusion-class lasers. Recently, the use of assist reactive gases has shown promise in
enhancing manifold etching rates relative to ambient conditions for CW-laser exposures. However, these methods
still require significant heating of the substrate that induce residual stress, redeposit coverage, material flow, and
compromise the final surface finish and damage threshold. While very reactive fluorinated gases are capable to
reduce treatment temperatures even further, they are also inherently toxic and not readily transferable to large
processing facilities. In this report, we look at whether a short-lived gas plasma could provide the safe and effective
etchant sought, while still reducing the thermal load on the surface. We test this approach using a YAG laserinduced
gas plasma to act as a source of the etchant for fused silica, a common optical material. The configuration
and orientation of the beam and optical apparatus with respect to the surface was critical in preventing surface
damage while etching the surface. Results with N2 and air gas plasmas are shown, along with a description of the
various experimental implementations attempted.
We present results from a study to determine an acceptable CO2 laser-based non-evaporative mitigation protocol for use
on surface damage sites in fused-silica optics. A promising protocol is identified and evaluated on a set of surface
damage sites created under ICF-type laser conditions. Mitigation protocol acceptability criteria for damage re-initiation
and growth, downstream intensification, and residual stress are discussed. In previous work, we found that a power
ramp at the end of the protocol effectively minimizes the residual stress (⪅25 MPa) left in the substrate. However, the
biggest difficulty in determining an acceptable protocol was balancing between low re-initiation and problematic
downstream intensification. Typical growing surface damage sites mitigated with a candidate CO2 laser-based
mitigation protocol all survived 351 nm, 5 ns damage testing to fluences ⪆12.5 J/cm2. The downstream intensification
arising from the mitigated sites is evaluated, and all but one of the sites has 100% passing downstream damage
expectation values. We demonstrate, for the first time, a successful non-evaporative 10.6 m CO2 laser mitigation
protocol applicable to fused-silica optics used on fusion-class lasers like the National Ignition Facility (NIF).
We compare force fields (FF's) that have been used in molecular dynamic (MD) simulations of silica in
order to assess their applicability for use in simulating IR-laser damage mitigation. Although pairwise
FF's obtained by fitting quantum mechanical calculations such as the BKS and CHIK potentials have
been shown to reproduce many of the properties of silica including the stability of silica polymorphs and
the densification of the liquid, we show that melting temperatures and fictive temperatures are much too
high. Softer empirical force fields give liquid and glass properties at experimental temperatures but may
not predict all properties important to laser mitigation experiments.
The advantage of using mid-infrared (IR) 4.6 μm lasers, versus
far-infrared 10.6 μm lasers, for mitigating damage
growth on fused silica is investigated. In contrast to fused silica's high absorption at 10.6 μm, silica absorption at 4.6 μm
is two orders of magnitude less. The much reduced absorption at 4.6 μm enables deep heat penetration into fused silica
when it is heated using the mid-IR laser, which in turn leads to more effective mitigation of damage sites with deep
cracks. The advantage of using mid-IR versus far-IR laser for damage growth mitigation under non-evaporative
condition is quantified by defining a figure of merit (FOM) that relates the crack healing depth to laser power required.
Based on our FOM, we show that for damage cracks up to at least 500 μm in depth, mitigation using a 4.6 μm mid-IR
laser is more efficient than mitigation using a 10.6 μm far-IR laser.
Knowing the ultimate surface morphology resulting from CO2 laser mitigation of induced laser damage is important both
for determining adequate treatment protocols, and for preventing deleterious intensification upon subsequent
illumination of downstream optics. Physical effects such as evaporation, viscous flow and densification can strongly
affect the final morphology of the treated site. Evaporation is a strong function of temperature and will play a leading
role in determining pit shapes when the evaporation rate is large, both because of material loss and redeposition. Viscous
motion of the hot molten material during heating and cooling can redistribute material due to surface tension gradients
(Marangoni effect) and vapor recoil pressure effects. Less well known, perhaps, is that silica can densify as a result of
structural relaxation, to a degree depending on the local thermal history. The specific volume shrinkage due to structural
relaxation can be mistaken for material loss due to evaporation. Unlike evaporation, however, local density change can
be reversed by post annealing. All of these effects must be taken into account to adequately describe the final
morphology and optical properties of single and multiple-pass mitigation protocols. We have investigated,
experimentally and theoretically, the significance of such densification on residual stress and under what circumstances
it can compete with evaporation in determining the ultimate post treatment surface shape. In general, understanding final
surface configurations requires taking all these factors including local structural relaxation densification, and therefore
the thermal history, into account. We find that surface depressions due to densification can dominate surface
morphology in the non-evaporative regime when peak temperatures are below 2100K.
Localized damage repair and polishing of silica-based optics using mid- and far-IR CO2 lasers has been shown to be
an effective method for increasing optical damage threshold in the UV. However, it is known that CO2 laser heating
of silicate surfaces can lead to a level of residual stress capable of causing critical fracture either during or after laser
treatment. Sufficient control of the surface temperature as a function of time and position is therefore required to
limit this residual stress to an acceptable level to avoid critical fracture. In this work we present the results of 351
nm, 3ns Gaussian damage growth experiments within regions of varying residual stress caused by prior CO2 laser
exposures. Thermally stressed regions were non-destructively characterized using polarimetry and confocal Raman
microscopy to measure the stress induced birefringence and fictive temperature respectively. For 1~40s square
pulse CO2 laser exposures created over 0.5-1.25kW/cm2 with a 1-3mm 1/e2 diameter beam (Tmax~1500-3000K), the
critical damage site size leading to fracture increases weakly with peak temperature, but shows a stronger
dependence on cooling rate, as predicted by finite element hydrodynamics simulations. Confocal micro-Raman was
used to probe structural changes to the glass over different thermal histories and indicated a maximum fictive
temperature of 1900K for Tmax≥2000K. The effect of cooling rate on fictive temperature caused by CO2 laser
heating are consistent with finite element calculations based on a Tool-Narayanaswamy relaxation model.
Small micrometer-sized roughness on optical surfaces, caused by laser damage and/or redeposition of laser ablated
material, can cause local electric field intensification which may lead to damage initiation both on the optics and/or
downstream. We examined the smoothing of etched periodic surface structures on SiO2 substrate with 10.6μm CO2
laser using atomic force microscopy. The characteristic surface tension driven mass flow of the glass under
different laser parameters were simulated using computational fluid dynamics and correlated with experimental
results. We found that during CO2 laser polishing the estimate viscosity of the silica glass appears to be higher than
typical literature values measured at a temperature similar to the laser heating conditions. This discrepancy can be
explained by the observation that at high temperature, a significant portion of the hydroxyl content in the layer of
heated silica glass can diffuse out resulting in a much stiffer glass.
In situ spatial and temporal surface temperature profiles of CO2 laser-heated silica were obtained using a
long wave infrared (LWIR) HgCdTe camera. Solutions to the linear diffusion equation with volumetric and surface
heating are shown to describe the temperature evolution for a range of beam powers, over which the peak surface
temperature scales linearly with power. These solutions were used with on-axis steady state and transient
experimental temperatures to extract thermal diffusivity and conductivity for a variety of materials, including silica,
spinel, sapphire, and lithium fluoride. Experimentally-derived thermal properties agreed well with reported values
and, for silica, thermal conductivity and diffusivity are shown to be approximately independent of temperature
between 300 and 2800K. While for silica our analysis based on a temperature independent thermal conductivity is
shown to be accurate, for other materials studied this treatment yields effective thermal properties that represent
reasonable approximations for laser heating. Implementation of a single-wavelength radiation measurement in the
semi-transparent regime is generally discussed, and estimates of the apparent temperature deviation from the actual
outer surface temperature are also presented. The experimental approach and the simple analysis presented yield
surface temperature measurements that can be used to validate more complex physical models, help discriminate
dominant heat transport mechanisms, and to predict temperature distribution and evolution during laser-based
material processing.
Mitigation of 351nm laser-induced damage sites on fused silica exit surfaces by selective CO2 treatment has been shown to effectively arrest the exponential growth responsible for limiting the lifetime of optics in high-fluence laser systems. However, the perturbation to the optical surface profile following the mitigation process introduces phase contrast to the beam, causing some amount of downstream intensification with the potential to damage downstream optics. Control of the laser treatment process and measurement of the associated phase modulation is essential to preventing downstream 'fratricide' in damage-mitigated optical systems. In this work we present measurements of the surface morphology,
intensification patterns and damage associated with various CO2 mitigation treatments on fused silica surfaces. Specifically, two components of intensification pattern, one on-axis and another off-axis can lead to damage of downstream optics and are related to rims around the ablation pit left from the mitigation
process. It is shown that control of the rim structure around the edge of typical mitigation sites is crucial in preventing damage to downstream optics.
Identification and spatial registration of laser-induced damage relative to incident fluence profiles is often
required to characterize the damage properties of laser optics near damage threshold. Of particular interest
in inertial confinement laser systems are large aperture beam damage tests (>1cm2) where the number of
initiated damage sites for Φ>14J/cm2 can approach 105-106, requiring automatic microscopy counting to
locate and register individual damage sites. However, as was shown for the case of bacteria counting in
biology decades ago, random overlapping or 'clumping' prevents accurate counting of Poisson-distributed
objects at high densities, and must be accounted for if the underlying statistics are to be understood. In this
work we analyze the effect of random clumping on damage initiation density estimates at fluences above
damage threshold. The parameter ψ=aρ= ρ/ρ0, where a=1/ ρ0 is the mean damage site area and ρ is the mean number density, is used to characterize the onset of clumping, and approximations based on a simple model are used to derive an expression for clumped damage density vs. fluence and damage site size. The influence of the uncorrected ρ vs. Φ curve on damage initiation probability predictions is also discussed.
We examine the effect of lattice temperature on the probability of surface damage initiation for 355nm, 7ns laser pulses for surface temperatures below the melting point to temperatures well above the melting point of fused silica. At sufficiently high surface temperatures, damage thresholds are dramatically reduced. Our results indicate a temperature activated absorption and support the idea of a lattice temperature threshold of surface damage. From these measurements, we estimate the temperature dependent absorption coefficient for intrinsic silica.
Growth of laser initiated damage plays a major role in determining optics lifetime in high power laser systems. Previous
measurements have established that the lateral diameter grows exponentially. Knowledge of the growth of the site in the
propagation direction is also important, especially so when considering techniques designed to mitigate damage growth,
where it is required to reach all the subsurface damage. In this work, we present data on both the diameter and the depth
of a growing exit surface damage sites in fused silica. Measured growth rates with both 351 nm illumination and with
combined 351 nm and 1054 nm illumination are discussed.
We examine the effect of pulse duration on both density and morphology of laser-induced damage in
KDP and SiO2. In both materials the density of damage sites scales with pulse duration to the ~ 0.4
power for 351-nm pulses between 1 and 10 ns. In SiO2 three types of damage sites are observed. The
sizes of the largest of these sites as well as the size of KDP damage sites scale approximately linearly
with pulse duration. Similarities of damage in very different materials points to properties of laser-induced
damage which are material independent and give insight to the underlying physics of laser-induced
damage.
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.