This PDF file contains the front matter associated with SPIE Proceedings Volume 10898, including the Title Page, Copyright information, Table of Contents, and Conference Committee listing.

The Laser Megajoule facility, developed by the CEA is based on 176 Nd:glass laser beams focused on a micro-target positioned inside a 10-meter diameter spherical chamber. The facility will deliver a total energy of 1.4MJ of UV light at 0.35 μm and a maximum power of 400 TW. To complete the experimental capabilities of LMJ, a PW beam, PETAL, has been added to the LMJ’s beams. PETAL offers a combination of a very high intensity beam, synchronized with the nanosecond beams of the LMJ. This combination expands the LMJ experimental field in High Energy Density Physics (HEDP). LMJ-PETAL is open to the academic communities for 20 to 30 % of the facility operating time. Since the operational commissioning of the LMJ in October 2014 (with the first bundle of 8 beams) and the second step with two bundles at the end of 2016, several experimental campaigns have been achieved. The installation of new bundles is continuing, simultaneously to plasma experiments. The first phase of nuclear commissioning of LMJ has been achieved to take into account high-energy particles created by PETAL and neutron production from D-D fusion reaction. A subsequent phase will take into account tritium targets. More than a third of the laser bundles are assembled, forty beams are operational, nine plasma diagnostics are commissioned (including 4 specific for PW physics). The next milestones for 2019 are: the first physics experiments using the 40 operational beams, the commissioning of two additional bundles, and the first experiments with neutrons production.

AWE’s Orion Laser Facility comprises ten 500J nanosecond (“long pulse”) beam lines (3ω) and two petawatt (“short pulse”) beam lines, each delivering 500J, 500fs pulses at 1054nm. One short pulse beam can operate at 2ω (at reduced aperture), producing ultrahigh contrast pulses. This paper reports on recent developments and planned future work. Static wavefront correctors have been implemented to mitigate prompt aberrations in the long pulse beams, which alter the onshot wavefront characteristics compared to the CW alignment beams. This mitigates aberration accumulation through the day, increasing the maximum number of shots in one shift. A TIM-mounted wavefront sensor/focal imager has been developed, which is better able to characterise the post-compression system aberrations, resulting in higher focal intensity. A diode-pumped, multi-joule rod amplifier has been prototyped. This is planned to replace the ageing, flashlamp-based ns-OPCPA pump laser, which constitutes a single point failure mode for our short pulse capability. Preliminary design work has commenced for a facility life-extension project, planned for ~2023. The infrared performance will be enhanced to ~1kJ per beam in 300fs, the additional bandwidth being supported by greater use of silicate glass. The two-grating, single-pass compressor systems will be replaced by four-grating compressors, retaining the extant vacuum vessels. The frequency doubling option will be retained. Since the greater near-field intensity inevitably over-drives the doublers, compressor detuning is necessary. We assess a novel, small compressor at the second harmonic. Simulations suggest that up to 500J in 150fs is possible in this configuration.

The OMEGA EP laser has been upgraded to provide a UV wavelength-tunable beam to support the study of wavelength detuning for the mitigation of cross-beam energy transfer in direct-drive inertial confinement fusion. The beamline delivers up to 0.5 TW in pulses up to 1-ns duration (0.1 TW up to 2.5 ns), to either the OMEGA or OMEGA EP target chambers with wavelength tunable from 350.2 to 353.4 nm. The upgrade leverages the existing optical parametric amplification (OPA) system in the short-pulse front end of OMEGA EP Beamline 1 for amplification of a new tunable, narrowband fiber front end over a broad spectral range. The tunable OPA output is spatially shaped to form a round OMEGA-like beam, which is amplified in the OMEGA EP beamline, then frequency tripled and characterized using the existing OMEGA EP long-pulse infrastructure. A new 3ω beam-transport system intercepts the tunable UV beam near the OMEGA EP target chamber and image relays it to the P9 port of the OMEGA target chamber for joint shots with the OMEGA 60-beam laser. Commissioning of the tunable UV capability has been completed, and four experimental campaigns have been supported with the tunable beam

Specification of visual defects (scratch and digs) on optics used in high-power laser facilities has always been a headache. Indeed, the wave degradation and the ensuing laser performances losses with regard to focal spot or downstream laser induced damage seem hard to predict. Indeed, one often has only partial information on each of the (often numerous) defects whereas the light behavior downstream strongly depends on the defect nature and morphology. So, determining general rules seem to be an impossible task. Borderline cases are then generally processed through timeconsuming optical profilometer measurements that are used in complex numerical laser propagation. We show in this paper that a simple analytic model can predict light intensification (responsible for some fratricide laser damage) with a high reliability. Defects are modeled by concentric quasi-circular rings of different radii, transmissions and phase shifts. The accuracy of these predictions will depend on the degree of knowledge of the model parameters set. In any case, upper bounds of intensifications can be provided as well as safe areas where intensification has decreased enough. These results allowed to specify defects dimensions and nature. We show good agreement between observed diffraction patterns downstream of real defects and model predictions, in terms of “hot spots” generation.

State-of-the-art physics experiments are pushing the development of lasers with ultra-high peak power pulses. 4 PW pulses have been produced with TiSa [1] and 10 PW with the same gain medium is scheduled at LULI (Apollon) and at ELI-NP. The other approach is to use Nd-doped glass as gain medium, whose interest is in its capability of delivering higher energy at the expense of a longer pulse duration. Based on this gain material combined with an OPCPA based front-end, a kJ-10 PW class laser has been designed and built. The front-end, consisting of picosecond OPCPA, temporal pulse cleaning and nanosecond OPCPA, delivers pulses with excess of 4 Joules at 5 Hz with a shaped spectrum to pre-compensate for gain distortions in Nd:glass power amplifiers. Two liquid-cooled, mixed glass power amplifiers, namely PA1 and PA2, are used for further amplification. Up to now, they have been activated demonstrating 70 J at 1 shot a minute after PA1 and 1 kJ at 1 shot every 7 minutes for PA2. The Fourier limit of the spectrum is 150 fs meaning 6 PW capability after compression. This energy level has been obtained with only 3 Joules seed energy, from the OPCPA and partial activation of PA2. Scaling of this result suggests that more than 1.7 kJ should be obtained leading to 10 PW after compression while the output spectrum will remain compatible with 150 Fs thanks to the OPCPA spectral tailoring capability.

3ω laser damage of fused silica optics is the bottleneck of high power laser systems for ICF. Excellent beam quality plays an important role in improving the anti-damage capability of final optics system. We have developed a new optical field measurement technology based on computational optical imaging. With the high power laser prototype of SGII-UP facility, damage resistance of final optics was experimentally studied. The near filed of laser beam was measured with a high resolution to study the effects of modulation and propagation on laser damage. The near field improvement of high power laser beam are reported and the influence of near filed quality on damage performance of final optics are discussed. The development of the defect detection techniques of final optics are introduced. Finally, we present the development perspective of final optics system for ICF laser driver. At present, the damage resistance capability of final optics assembly is 6J/cm² at normal operation, we will continue to improve the ability in the next step of work.

AWE’s Orion Laser facility contains ten nanosecond beamlines, each employing a series of flash-lamp pumped disk amplifiers capable of generating up to 750 J in 1 ns at 1053 nm. Discharge through the flash-lamps, however, causes unwanted disk heating which induces wavefront aberrations. This immediate effect, referred to as ‘prompt aberration’, alters the onshot wavefront compared to the wavefront of the alignment beams. In addition, a thermal load remains on the disks between shots, causing an evolution of both the alignment and on-shot aberrations over a typical day. The combination of the prompt aberrations and wavefront evolution has limited performance. After approximately six shots the alignment aberrations became so severe that further alignment was impossible. Operations would then cease to allow the disks to cool and the wavefronts to return to normal for the start of the following day. This paper reports on the development and implementation of static wavefront correctors. By mitigating the effect of the prompt aberrations, the alignment beam wavefront better matches the on-shot aberrations, and no longer limits operations, allowing more shots to be fired in a day. The wavefront analysis is discussed and shows the prompt aberration to comprise mainly of ~1 μm of defocus and astigmatism, averaged across many shots and all beamlines. Compensating static correctors are shown to reduce the effect of the prompt aberration to ~0.2 μm. The outcome indicates the possibility of firing more shots in a single day’s operation.

Thermal effects in the active laser medium influence on the quality of the laser beam in high-power laser systems. Temperature fluctuations lead to refractive index modulation in the medium, thus the intensity of the radiation is significantly reduced. Historically, to solve this task an adaptive optics technique is used. It allows to compensate for the wavefront aberrations. Stacked-actuator deformable mirror is a traditional technology that is used in order to improve the quality of the incident wavefront. This type of wavefront correctors has one significant shortcoming – the impossibility of the replacement of broken actuators. We developed a stacked-actuator deformable mirror with aperture of 120 mm and 121 control actuators to correct for the high-power laser radiation. Actuator arrangement was hexagonal. In our design, the broken piezoactuators could be easily replaced.

Multibeam lasers often require an output beam balance that specifies the degree of simultaneity of the laser output energy, instantaneous power, or instantaneous irradiance (power per unit area). This work describes the general problem of balancing a multibeam laser. Specific techniques used to balance the output power of the 60-beam pulsed OMEGA Laser System are discussed along with a measured reduction of beam-to-beam imbalance. In particular, the square-pulse distortion induced by a simple saturating amplifier operating with its output at some fraction of its saturation fluence is derived, and a method to exchange gain between saturated amplifiers in a single beam that have different saturation fluences to adjust balance is described.

Direct-drive fusion implosion experiments using the 60-beam OMEGA Laser System require ~1% rms uniformity on target. Measurements from laser diagnostics indicate that beam-to-beam power variation has been sufficiently reduced such that the uniformity nearly meets this requirement; however, experimental results suggest otherwise. To better understand this discrepancy, a full-beam-in-tank diagnostic has been developed to characterize the on-shot, full-energy focal spot of a single beam inside the target chamber using a small sample of the beam from the final optic assembly. In this paper, we describe the diagnostic and present the results of commissioning experiments.

The newly developed full-beam-in-tank (FBIT) diagnostic has the capability to characterize multiple beamlines in the target chamber. In addition to measuring multiple beams, we can obtain measurements of the step-by-step changes to achieve smoothing by spectral dispersion (SSD), the SSD kernel, and SSD synchronization. Since other existing diagnostics are all located upstream of the target chamber, this diagnostic can be used to explore a propagating beam through the final optics assembly. In this work, we investigate current discrepancies between laser diagnostics and experimental results by comparing results of on-shot direct measurements using FBIT and the equivalent-target-plane diagnostic.

We will describe a large research effort at Lawrence Livermore National Laboratory (LLNL) aimed at using recent advances in deep learning, computational workflows, and computer architectures to develop an improved predictive model – the learned predictive model. Our goal is to first train these new models, typically cyclic generative adversarial networks, on simulation data to capture the theory implemented in advanced simulation codes. Later, we improve, or elevate, the trained models by incorporating experimental data. The training and elevation process both improves our predictive accuracy and provides a quantitative measure of uncertainty in such predictions. We will present work using inertial confinement fusion research and experiments at the National Ignition Facility (NIF) as a testbed for development. We will describe advances in machine learning architectures and methods necessary to handle the challenges of ICF science, including rich, multimodal data (images, scalars, time series) and strong nonlinearities. We will also cover state-of-the-art tools that we developed to manage our computational workflow. The tools manage a wide range of tasks, including developing enormous (PB-class) simulated training data sets, driving the training of learned models on simulation data, and elevating learned models through exposure to experiment. We will end by drawing ties to a broad range of scientific applications.

Hohlraums convert the laser energy at the National Ignition Facility (NIF) into X-ray energy to compress and implode a fusion capsule, creating fusion. The Static X-ray Imager (SXI) diagnostic collects time-integrated images of hohlraum wall X-ray illumination patterns viewed through the laser entrance hole (LEH). NIF image processing algorithms calculate the size and location of the LEH opening from the SXI images. Images obtained come from different experimental categories and camera setups and occasionally do not contain applicable or usable information. Unexpected experimental noise in the data can also occur where affected images should be removed and not run through the processing algorithms. Current approaches to try and identify these types of images are done manually and on a case-by-case basis, which can be prohibitively time-consuming. In addition, the diagnostic image data can be sparse (missing segments or pieces) and may lead to false analysis results. There exists, however, an abundant variety of image examples in the NIF database. Convolutional Neural Networks (CNNs) have been shown to work well with this type of data and under these conditions. The objective of this work was to apply transfer learning and fine tune a pre-trained CNN using a relatively small-scale dataset (~1500 images) and determine which instances contained useful image data. Experimental results are presented that show that CNNs can readily identify useful image data while filtering out undesirable images. The CNN filter is currently being used in production at the NIF.

We have previously shown that the ceramic Yb:YAG-based edge-pumped disk laser amplifier offers significant advantages over traditional face-pumped disk amplifiers. Such amplifiers may be used in laser drivers for inertial confinement fusion, laser acceleration, and other applications, which require a combination of high-pulse energy and high-average power. Unlike face pumping, the edge-pumping architecture beneficially allows for reduced Yb doping and enables a construction of very simple, compact, and completely modular amplifiers comprising identical and interchangeable gain modules. This paper reports on the development and early testing of a Ø5-cm aperture edge-pumped ceramic Yb:YAG disk amplifier module pumped by 100-kW diodes at up to 20 Hz and cooled by a high-velocity gas flow at near ambient temperature. In early testing, the amplifier module has demonstrated very uniform transverse gain and 37 J of stored energy. A laser oscillator operating in a quasi-cw mode with 1- ms pump pulses produced 43 kW of instantaneous laser power and 31 J of energy at a wavelength of 1029 nm. Experimental results compare well to model predictions.

The development at STFC’s Central Laser Facility (CLF) of multi-slab gas cooled amplifier technology based on ceramic Yb:YAG, known as DiPOLE, is well known within the field [1]. Previous successes using this laser architecture include a 10 J, 10 Hz prototype system and demonstration of the world’s first 100 J, 10 Hz, 10 ns system (DiPOLE100), supplied to and commissioned at the HiLASE facility[2]. The success of the first kW-class 100 J nanosecond pulsed laser led to development of a second system to be made available to users of the HED Instrument at the European XFEL in collaboration with HiBEF / HZDR. This system, DiPOLE-100X, will be used for fundamental research into the structure and compression of materials, such as present in extrasolar planets [3]. DIPOLE-100X will provide nanosecond, user defined, temporally shaped pulses that once frequency doubled and focussed on target, create the pressure and temperature states of matter relevant to exoplanetary interiors. These changes can then be probed using the bright x-ray pulses of the European XFEL. DIPOLE-100X is scheduled for completion at the end of summer, before shipment to the European XFEL later in the year. In this paper we will present the operational and commissioning results for DiPOLE-100X, highlighting design changes and upgrades over the first generation, along with future developments of DiPOLE technology at the CLF.

We present the experimental thermal study of a 88.6 W average power optical parametric chirped pulse amplifier (OPCPA) system operating at 100 kHz, delivering sub-20 fs pulses at a center wavelength of 800 nm. A 15 W pump laser at 1030 nm is used to derive the seed pulses via white-light continuum generation and to pre-amplify the seed pulses to a 1 W level. Further amplification takes place in two high power stages, that are pumped with a frequency doubled Yb:YAG InnoSlab amplifier. The system is a prototype for the future LCLS-II pump-probe laser system.

High-energy ultraviolet (UV) sources are required to probe hot dense plasmas from fusion experiments by using Thomson scattering resulting from the lower self-generated background from the plasma in the 180- to 230-nm spectral region. Although recently we demonstrated efficient, joule-class fifth-harmonic conversion of 1053-nm pulses using a cesium lithium borate (CLBO) crystal, larger crystals are necessary for increased UV energy. This paper presents an alternative approach for fifth-harmonic generation of a large-aperture neodymium laser using ammonium dihydrogen phosphate (ADP). An Nd:YLF laser was optimized to produce square pulses with a flattop, square beam profile (1053 nm, 1 to 2.8 ns, 12 × 12 mm, ≤ 1.5 J, ≤ 5 Hz ). Noncritical phase matching in ADP suitable for sum–frequency mixing the fourth harmonic with residual fundamental light was achieved by cooling the crystal (65 × 65 × 10 mm) in a two-chamber cryostat to 200 K. The crystal chamber used helium (1 atm) as the thermally conductive medium between the crystal and the crystal chamber, which in turn was surrounded by a high-vacuum chamber that contained a liquid nitrogen reservoir to cool the crystal chamber. Temperature variation of 0.4 K across the crystal aperture was obtained using two 50-W heaters with active feedback. The total conversion efficiency from the fundamental to the fifth harmonic, including surface losses and absorption, was 26%.

Independently managing the timing of individual beams so they all arrive at the target within 10 ps of the time specified by the principal investigator is crucial to the success of experiments on the OMEGA EP Laser System. A streak camera is used to observe the x rays emitted when the laser beams strike a gold target, while an optical streak camera is used to measure the UV pulses. Correlating the signal on the two instruments gives a timing accuracy of 10 ps for the short-pulse IR beams and 20 ps for the long-pulse UV beams.

We used a Monte Carlo method to generate error bars for deconvolved measurements from diagnostics on the National Ignition Facility (NIF). Through a process of masking and normalization of the diagnostic system’s known Impulse Response Function (IRF), we were able to diminish the deconvolved measurement error for all points of the waveform by a factor of < 2. This technique is generally applicable to deconvolutions with measured IRFs.

Beamlines at the National Ignition Facility (NIF) use large neodymium-doped glass slabs for amplification of pulsed beams with various temporal shapes and transverse dimensions _40 cm _ 40 cm. Currently, the Virtual Beam Line (VBL) simulator1 computes saturable amplification according to the approach of Frantz and Nodvik, modified to include drain of the lasing transition's lower level. Linearly chirped pulses are amplified by gain media parameterized by an emission cross section value referenced to the instantaneous beam wavelength. Expanding the capabilities of VBL to a family of waveforms that is more diverse in terms of spectral amplitude and phase calls for the adoption of an approach that is fundamentally dispersive. In this work, we describe an approach to computing broadband amplification in the time domain according to coupled equations that describe evolution of the population inversion and the associated resonant polarization. Considering the diversity of glass species with respect to various gain inhomogeneities, we explore various model extensions for capturing the non-Lorentzian emission cross section in the small-signal regime and how the underlying resonant susceptibility is deformed by gain saturation. The polarization envelope acts as transverse-spatial sources to (3+1)D spectral envelope propagation that fully accounts for linear-optical diffraction and dispersion in the host glass, and includes the usual instantaneous non-resonant third-order electronic response (optical Kerr effect).

We investigated the characteristics of a laser pulse width depended on an optical cavity length in a longitudinally excited CO₂ laser. The CO₂ laser was a simple device and had a discharge tube, an excitation circuit for a fast discharge, and an optical cavity. The discharge tube was made of a 45-cm-long alumina ceramic pipe with an inner diameter of 13 mm and metal electrodes at both ends of the tube. An excitation circuit was consisted of a low-voltage pulse power supply, a stepup transformer, a storage capacitance and a spark gap. The optical cavity was formed by a ZnSe output coupler with a reflectivity of 85% and an Au-coated mirror with a reflectivity of 99%. The length of the optical cavity was 65 cm to 185 cm. A short laser pulse with a spike pulse and a pulse tail was produced at a gas pressure of 2.6 kPa in a 1:1:2 mixture of CO₂/N₂/He gas. In the optical cavity length of 65 cm, the spike pulse width was 137 ns (FWHM), the pulse tail was 146μs, the energy of the whole laser pulse was 32.0 mJ, and the energy of the spike pulse part was estimated to be 0.86 mJ. The spike pulse width depended on the optical cavity length and was 137 ns, 195 ns, 274 ns, 304 ns and 332 ns at the length of 65 cm, 95 cm, 125 cm, 155 cm and 185 cm, respectively.