The efficiency and reliability of high power (multi-kW) fiber lasers has revolutionized material processing and other applications. However, diode lasers offer the promise of even higher efficiency, yet scaling to high power while maintaining good beam quality remains a significant challenge. We discuss the challenges, advances, and potential future of this technology.
We recently demonstrated a seed modulation format that is effective in suppressing SBS in kW-class fiber amplifiers. By applying synchronous amplitude and frequency modulation, a broad low power seed spectrum can be spectrally narrowed via self-phase modulation as it is amplified, thus reducing the SBS gain and enabling narrower linewidths. Since the approach leverages the fiber Kerr nonlinearity, SBS suppression becomes more effective as fiber powers and lengths increase. We will present updated results extending this approach toward higher modulation depths, higher powers, and narrower linewidths.
State-of-the-art diffraction-limited fiber lasers are presently capable of producing kilowatts of power. Power levels
produced by single elements are gradually increasing but beam combining techniques are attractive for rapidly scaling
fiber laser systems to much higher power levels. We discuss both coherent and spectral beam combining techniques for
scaling fiber laser systems to high brightness and high power. Recent results demonstrating beam combination of 500-W
commercial fiber laser amplifiers will be presented.
Thulium-doped fiber lasers (TFLs) emitting retina-safe 2-μm wavelengths offer substantial power-scaling advantages over ytterbium-doped fiber lasers for narrow linewidth, single-mode operation. This article reviews the design and performance of a pump-limited, 600 W, single-mode, single-frequency TFL amplifier chain that balances thermal limitations against those arising from stimulated Brillouin scattering (SBS). A simple analysis of thermal and SBS limits is anchored with measurements on kilowatt class Tm and Yb fiber lasers to highlight the scaling advantage of Tm for narrow linewidth operation. We also report recent results on active phase-locking of a TFL amplifier to an optical reference as a precursor to further parallel scaling via coherent beam combining.
The power record for near-diffraction limited output from a single fiber laser amplifier is now 10 KW. However, a
single fiber appears unlikely to approach powers greater than 100 kW, which is needed for some applications.
Therefore, there is great interest in methods for combination of many high power fiber beams that maintain aggregate
beam quality. A number of methods have been proposed, including active and passive phasing, and spectral
combination. These methods have varying implementation and performance advantages. The limitations of these
methods, and approaches to address them, are discussed as applied toward combination of kilowatt-class fiber
amplifiers.
A four-stage, Tm-doped fiber amplifier chain emitted 608 W of single-frequency (SF) output power with 53 dB gain,
54% slope efficiency, and M2 = 1.05 beam quality. The output power was limited by available pump power. The final
amplifier stage preserved the input <5-MHz linewidth and imposed negligible phase noise above 3 kHz. SBS limits at
the 2040-nm operating wavelength were measured by splicing different lengths of passive fiber to the amplifier exit.
Thermal limits of the fiber were explored analytically and are consistent with the measured power performance.
Comparison of the SBS and thermal limits suggests a maximum SF power of ~750 W from this fiber configuration, with
further potential to scale past 1 kW with different fiber parameters. To our knowledge, this is the highest power reported
to date from any single-frequency, single-mode fiber laser.
Theoretical and numerical analyses are presented of the passive coherent phasing of an array of fiber lasers that are
combined in a single laser cavity by an Nx1 coupler. In an initial linear analysis it is found that the brightness gain of the
passive coherent array grows linearly for a small number of fibers, however, for practical parameters, the coherent
brightness gain saturates at ~ 8 - 12 for large arrays. An intensity dependent index (Kerr) nonlinearity is then introduced
and it is shown that the expected maximum improvement is modest, with the coherent brightness gain saturating at 10 - 14, depending on the strength of the nonlinearity. These results are compared with recent experiments.
Previous research has shown that temperature gradients along a fiber can broaden the Stimulated Brillouin Scattering
(SBS) gain profile and thereby increase the SBS threshold. However, within practical temperature ranges this method
has been limited to SBS thresholds of a few hundred Watts. It is also well known that strain gradients applied to a fiber
can broaden the SBS resonance. To suppress the SBS threshold to kW levels in fiber amplifiers of length ~5 m requires
broadening of the SBS resonance width to ~1 GHz, which can be achieved with a strain of 1 - 2%. Although tensile
strain is generally limited by fiber failure to less than ~1%, compressive strain has been employed to the level of many
percent in a number of applications in the tuning of fiber Bragg gratings. We demonstrate the effect of SBS gain
broadening and suppression by strain gradients at high power (~ 190 W) for the first time to our knowledge, and explore
scaling of this method to kW output levels.
The fiber Bragg grating (FBG) with a very high count, up to N=80, identical channels for dispersion compensators has been commercialized. The design utilizes the phase-only sampling, which requires η / N time higher refractive index modulation than the N times required by the amplitude sampling design. The phase sampling is implemented by inserting phase shifts in the FBG. Hence, the positive valued coupling coefficients of the FBG become complex-valued, which greatly increase the functionality and spectral performance of the FBG. In the fabrication of the phase-shifted FBG we find that the phase shifts in the phase mask for writing the FBG are not replicated into the FBG, as believed by many people, but are split into two half-magnitude phase-shifts due to the near field diffraction. We show the split of the phase shift by using the Finite Difference in Time Domain (FDTD) analysis and we show that this split of the phase shifts results in the FBG spectral asymmetry by a physical model of the phase-only sampling function distortion and by the FBG experiments. The new pre-compensating phase mask technique is introduced to remove the errors and therefore ensure the success of the high count multiple channel FBGs.
A phase shift in phase mask is split into two half-magnitude phase-shifts in near field, resulting in FBG spectral asymmetry. The errors are removed using new pre-compensating phase mask for binary-phase Dammann sampled multi-channel FBGs.
We have applied the diffractive optics and digital holography technologies to the design and fabrication of fiber Bragg gratings (FBG). We studied the phase mask near field diffraction using the numerical simulation. Our new understanding on the phase mask has critical importance for fabrication of complex FBGs, such as the sampled grating DBR laser, multi-channel dispersion compensator, and phase-shift band-pass filters. In the design of the FBGs we introduced the phase-only sampling multi-channel FBGs using the Dammann diffractive grating, such that the maximum reflective index modulation required for N channels is only √N times of that for the single channel. We apply the discrete layer-peeling algorithm directly design the multi-channel FBGs, whose seed grating contains abrupt phase jumps in order to correct the channel distortions in the sampled multi-channel complex gratings. In addition, we introduced the iterative layer-peeling algorithm as a generic approach for the design of the FBGs. Our research provided a foundation of the advanced multichannel complex FBGs technology, which produces FBGs of very high channel count (45 channels and higher).
We describe a nonlinearly chirped fiber Bragg grating (FBG) based multichannel device that can perform tunable chromatic dispersion compensation. Complex or phase only sampling allows a single grating to operate on multiple DWDM channels. Various key performance parameters are discussed.
The use of nonlinearly chirped fiber Bragg gratings (FBGs) for tunable dispersion compensation is described. It is experimentally demonstrated that wide dispersion tunability can be accomplished over multiple wavelength division multiplexed (WDM) channels simultaneously in a single device with an approximately 10 cm long nonlinearly chirped FBG. Multiple channels are generated in such a short grating by fabricating a periodic superstructure over (sampling) the FBG. Proper design of the sampling also allows one to achieve dispersion slope compensation over broad bandwidth in WDM communication systems.
We have begun building the 'Mercury' laser system as the first in a series of new generation diode-pumped solid-state lasers for inertial fusion research. Mercury will integrate three key technologies: diodes, crystals, and gas cooling, within a unique laser architecture that is scalable to kilojoule energy levels for fusion energy applications. The primary performance goals include 10 percent electrical efficiencies at 10 Hz and 100J with a 2-10 ns pulse length at 1.047 micrometers wavelength. When completed, Mercury will allow rep-rated target experiments with multiple target chambers for high energy density physics research.
The National Ignition Facility (NIF) baseline configuration for inertial confinement fusion requires phase modulation for two purposes. First, approximately 12 angstrom of frequency modulation (FM) bandwidth at low modulation frequency is required to suppress buildup of Stimulated Brioullin scattering in the large aperture laser optics. Also, approximately 3 angstrom or more bandwidth at high modulation frequency is required for smoothing of the speckle pattern illuminating the target by the smoothing by spectral dispersion method. Ideally, imposition of bandwidth by pure phase modulation does not affect the beam intensity. Ideally, imposition of bandwidth by pure phase modulation does not affect the beam intensity. However, as a result of a large number of effects, the FM converts to amplitude modulation (AM). In general this adversely affects the laser performance, e.g. by reducing the margin against damage to the optics. In particular, very large conversion of FM to AM has been observed in the NIF all-fiber master oscillator and distribution systems. The various mechanisms leading to AM are analyzed and approaches to minimizing their effects are discussed.
Polarization smoothing (PS) is the illumination of the target with two distinct and orthogonally polarized speckle patterns. Since these two polarizations do not interfere, the intensity patterns add incoherently and thus the contrast of the intensity nonuniformity can be reduced by a factor of (root) 2 in addition to any reduction achieved by temporal smoothing techniques. Smoothing by PS is completely effective on an instantaneous basis and is therefore of particular interest for the suppression of laser plasma instabilities, which have a very rapid response time. The various implementations of PS are considered, and their impact, in conjunction with temporal smoothing methods, on the spatial spectrum of the target illumination is analyzed.
The performance of the Beamlet laser with one dimensional smoothing by spectral dispersion implemented is investigated. Measurements of then ear field beam quality, nonlinear breakup, and transmission through spatial filter pinholes show a modest effect only at large SSD divergence. No measurable effect was found at the divergence level planned for indirect drive ignition experiments. The efficiency of conversion to the third harmonic was also measured with SSD present and found to be somewhat larger than expected form an ideal plane wave model.
We describe the Optical Pulse Generation (OPG) testbed, which is the integration of the MOD and Preamplifier Development Laboratories. We use this OPG testbed to develop and demonstrates the overall capabilities of the NIF laser system front end. We will present the measured energy and power output, temporal and spatial pulse shaping capability, FM bandwidth and dispersion for beam smoothing, and measurements of the pulse-to-pulse power variation o the OPG system and compare these results with the required system performance specifications. We will discus the models that are used to predict the system performance and how the OPG output requirements flowdown to the subordinate subsystems within the OPG system.
The smoothing of the spatial illumination of an inertial confinement fusion target is examined by its spatial frequency content. It is found that the smoothing by spectral dispersion method, although efficient for glass lasers, can yield poor smoothing at low spatial frequency. The dependence of the smoothed spatial spectrum on the characteristics of phase modulation and dispersion is examined for both sinusoidal and more general phase modulation. It is shown that smoothing with non-sinusoidal phase modulation can result in spatial spectra which are substantially identical to that obtained with the induced spatial incoherence or similar method where random phase plates are present in both methods and identical beam divergence is assumed.
The electro-optic deflector is analyzed and compared to smoothing by spectral dispersion for efficacy as a beam smoothing method for ICF. It is found that the electro-optic deflector is inherently somewhat less efficient when compared either on the basis of equal peak phase modulation or equal generated bandwidth.
The impact of smoothing method on the performance of a direct drive target is modeled and examined in terms of its l-mode spectrum. In particular, two classes of smoothing methods are compared, smoothing by spectral dispersion (SSD) and the induced spatial incoherence (ISI) method. It is found that SSD using sinusoidal phase modulation results in poor smoothing at low l-modes and therefore inferior target performance at both peak velocity and ignition. This disparity is most notable if the effective imprinting integration time of the target is small. However, using SSD with more generalized phase modulation can result in smoothing at low l-modes which is identical to that obtained with ISI. For either smoothing method, the calculations indicate that at peak velocity the surface perturbations are about 100 times larger than that which leads to nonlinear hydrodynamics. Modeling of the hydrodynamic nonlinearity shows that saturation can reduce the amplified nonuniformities to the level required to achieve ignition for either smoothing method. The low l-mode behavior at ignition is found to be strongly dependent on the induced divergence of the smoothing method. For the NIF parameters the target performance asymptotes for smoothing divergence larger than approximately 100 (mu) rad.
Recent simulations and experiments on Nova indicate that some level of smoothing may be required to suppress filamentation in plasmas on the National Ignition Facility, resulting in the addition of 1D smoothing capability to the current baseline design. Control of stimulated Brillouin scattering and filamentation is considered essential to the success of laser fusion because they affect the amount and location of laser energy delivered to the x-ray conversion region (holhraum wall) for indirect drive and to the absorptive region for direct drive. Smoothing by spectral dispersion (SSD), reduces these instabilities by reducing nonuniformities in the focal irradiance when averaged over a finite time interval. We have installed SSD on Nova to produce beam smoothing on all 10 beam lines. A single dispersion grating is located in a position common to all 10 beam lines early in the preamplifier chain. This location limits the 1(omega) bandwidth to 2.2 angstroms with sufficient dispersion to displace the speckle field of each frequency component at the target plane by one half speckle diameter. Several beam lines were modified to allow orientation of the dispersion on each arm relative to the holhraum wall. After conversion to the third harmonic the beam passes through a kinoform phase plate (KPP) designed to produce an elliptical spot at best focus. The KPPs produce a focal spot having an elliptical flat-top envelope with a superimposed speckle pattern. Over 93% of the energy is contained in the central 400 micrometers . Calculations indicate a 16% rms intensity variance will be reached after 330 ps for a single beam.
The performance of the Beamlet laser with 1D SSD implemented is investigated. Simulations indicate that the critical issue for laser performance is the amount of additional divergence owing to SSD in comparison to the size of the spatial filter pinholes. At the current +/- 200 (mu) rad pinholes used on Beamlet, simulations indicate that the levels of SSD divergence anticipated for the National Ignition Facility results in a very slight degradation to the near field beam quality. Experiments performed with the Beamlet front end show no degradation to the near field beam with up to 100 (mu) rad of SSD divergence. MEasurements of the smoothing of a far field speckle pattern generated by a phase plate show the expected improvement in contrast with increasing amounts of SSD divergence.
The National Ignition Facility (NIF) is a proposed 1.8 MJ laser facility for carrying out experiments in inertial confinement fusion, currently designed for indirect drive experiments. The direct drive approach is being pursued at the 30 kJ Omega facility at the University of Rochester. In this paper we discuss the modifications to the NIF laser that would be required for both indirect and direct drive experiments. A primary concern is the additional cost of adding direct drive capability to the facility.
The requirements for laser uniformity are discussed in terms of the l-mode spectrum. It is shown that the choice of smoothing methods can significantly alter this spectrum and that this choice should be made in the context of the target physics. Although two dimensional smoothing by spectral dispersion yields a high quality near field beam profile, it results in poor smoothing for low spatial frequency. The partially coherent light method (fiber smoothing) leads to superior smoothing at low spatial frequencies, but has very poor near field beam quality. As a result, it may be desirable to use partially coherent light during the drive pulse foot (at low intensity and when minimizing the laser imprint is critical) and smoothing by spectral dispersion during the main pulse.
Two dimensional smoothing by spectral dispersion is analyzed by using diffraction theory calculations. It is shown that by using standard frequency modulated light one can obtain bandwidth limited smoothing over integration times relevant to inertial confinement fusion (about 1 nsec) with modest induced beam divergence. At longer integration times one can obtain bandwidth limited smoothing by increasing the divergence and/or by using more advanced phase modulation methods.
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.