Search

Advances in optical parametric devices, in particular those requiring high conversion efficiency, rely on pump laser and gain medium properties. We describe and theoretically model the source of dephasing due to angular deviation from ideal phase matching in optical parametric amplification. Real laser beams have angular content, which is described by their spatial frequency spectrum. Such beams cannot be treated as single plane waves in nonlinear interactions. Our mathematical model is based on a plane wave decomposition of Gaussian and top-hat beams into their components in spatial frequencies. Several popular nonlinear materials (beta-barium borate, lithium borate, and potassium dihydrogen phosphate) are examined for phase matching angles and dephasing is rigorously calculated. The impact of the beam angular content on small signal gain and on conversion efficiency in the strongly depleted regime is evaluated numerically. In addition, a criterion is formulated for beam quality tolerance in optical parametric amplifiers, for critical and noncritical phase matching. The impact of initial conditions in optical parametric amplification is considered. Our calculations are intended primarily for devices pumped with long (nanosecond) pulses.

The parameter /spl xi/ (xi) is proposed as an alternative to the traditional solid-state laser media heating parameter, /spl chi/ (chi). /spl xi/ is the ratio of heat produced to energy absorbed, and /spl chi/ is the ratio of heat produced to the maximum stored energy in the upper laser level. The parameter /spl xi/ is particularly relevant to diode pumped systems. We demonstrate an experimental /spl xi/ characterization based on the determination of the steady state cooling rate (hence heating rate) of small sample crystals subjected to pump laser heating. Using measured fluorescent lifetimes of the samples and near zero doping (intrinsic) values, the doping independent (zero doping or zero quenching) parameters /spl chi//spl phi/ and /spl xi//spl phi/ are determined. The results for all samples are in excellent agreement with calculations based solely on energy defect and nonradiative quenching of the upper level. >

By using three 6-disk amplifiers of 10 cm clear aperture from the Nova laser as the active element in an unstable resonator of magnification M=3, we achieved a single shot output of 1.4 kJ in a beam which contains 90% of its energy within twice the theoretical beam divergence. We show that most of the beam quality degradation comes from about one wave of beam tilt during the pulse and that the tilt corrected beam quality corresponds to a Strehl ratio of /spl ap/0.6. >

A diode-pumped single Nd:YOS rod laser is operated at 110 Hz with 830-mJ free-lase pulses and 380-mJ, 13-ns (FWHM) Q-switched pulses. External doubling with an 8-mm-long KTP crystal results in 24 W at 0.537 μm after thermalization.

Detailed performance results for an efficient and low thermal impedance laser diode array heatsink are presented. High duty factor or CW operation of fully filled laser diode arrays is made possible at high average power. Low thermal impedance is achieved using a liquid coolant and laminar flow through microchannels. The microchannels are fabricated in silicon using an anisotropic chemical etching process. A modular rack-and-stack architecture is adopted for the heatsink design, allowing arbitrarily large two-dimensional arrays to be fabricated and easily maintained. The excellent thermal control of the microchannel cooled heatsinks is ideally suited to pump array requirements for high average power crystalline lasers. >

The authors have developed and tested stackable microchannel cooled laser bar diode pump packages suitable for direct pumping of slab lasers at high duty factor. A stack of 41 diode packages gives a pump array of 13.5 cm/sup 2/ and produces a peak power of 4000 W and an average power of 1000 W for an average irradiance of 75 W/cm/sup 2/. A high average power, total internal reflection face pumped Nd:YAG laser using 80 diode packages has been constructed. Preliminary testing of the slab laser using a 20% subset of diode packages arranged to pump a 4 mm*4 mm*80 mm volume of the slab has been completed. Seventy watts average power is obtained at 2.5 kHz pulse repetition rates and 100 mu s pulse widths. >

Flash-lamp-pumped laser operation of Nd3+:Y2SiO5 by use of the Nd3+ 4F3/2−4F11/2 transition at 1.074 μm is obtained and compared with that from an identically pumped standard-doped Nd3+:YAG rod. Higher free-laser performance in Nd3+:Y2SiO5 is attributed to its higher doping. Approximately 2.5 times more energy in Q-switched pulses is generated in the Nd3+:Y2SiO5 rod than in the prelase-limited Nd:YAG rod.

A ground-state-depleted laser is demonstrated in the form of a Q-switched oscillator operating at 912 nm. By using Nd(3+) as the active ion and Y(2)SiO(5) as the host material, the laser transition is from the lowest-lying Stark level of the Nd(3+4)F(?) level to a Stark level 355 cm(-1) above the lowest-lying one in the (4)I(9/2) manifold. The necessity of depleting the ground (4)I(9/2) manifold is evident for this level scheme as transparency requires a 10% inversion. To achieve the high excitation levels required for the efficient operation of this laser, bleach-wave pumping using an alexandrite laser at 745 nm has been employed. With KNbO(3), noncritical phase matching is possible at 140 degrees C using d(32) and is demonstrated.

We describe an electro-optically Q-switched, diode-pumped Nd:YAG slab laser oscillator operating at an average power of greater than 250 watts. More than 100 watts of frequency doubled light has been demonstrated.

We demonstrated a high-pulse energy, femtosecond-pulse source based on optical parametric chirped-pulse amplification. We successfully amplified 1-microm broadband oscillator pulses to 31 mJ and recompressed them to 310-fs pulse duration, at a 10-Hz repetition rate. The gain in our system is 6 x 10(7), achieved by the single passing of only 40 mm of gain material pumped by a commercial Q-switched Nd:YAG laser. This relatively simple system replaces a more complex Ti:sapphire regenerative-amplifier-based chirped-pulse amplification system. Numerous features in design and performance of optical parametric chirped-pulse amplifiers make them a preferred alternative to regenerative amplifiers based on Ti:sapphire in the front end of high-peak-power lasers.

Optical parametric chirped pulse amplification (OPCPA) is a scalable technology for ultrashort pulse amplification. Its major advantages include design simplicity, broad bandwidth, tunability, low B-integral, high contrast, and high beam quality. OPCPA is suitable both for scaling to high peak power as well as high average power. We describe the amplification of stretched 100 fs oscillator pulses in a three-stage OPCPA system pumped by a commercial, single- longitudinal-mode, Q-switched Nd:YAG laser. The stretched pulses were centered around 1054 nm with a FWHm bandwidth of 16.5 nm and had an energy of 0.5nJ. Using our OPCPA system, we obtained an amplified pulse energy of up to 31 mJ at a 10 Hz repetition rate. The overall conversion efficiency from pump to signal is 6%, which is the highest efficiency obtained with a commercial tabletop pump laser to date. The overall conversion efficiency is limited due to the finite temporal overlap of the seed (3 ns) with respect to the duration of the pump (8.5 ns). Within the temporal window of the seed pulse the pump to signal conversion efficiency exceeds 20%. Recompression of the amplified signal was demonstrated to 310 fs, limited by the aberrations initially present in the low energy seed imparted by the pulse stretcher. The maximum gain in our OPCPA system is 6x107, obtained through single passing of 40 mm of beta- barium borate. We present data on the beam quality obtained from our system (M2=1.1). This relatively simple system replaces a significantly more complex Ti:sapphire regenerative amplifier-based CPA system used in the front end of a high energy short pulse laser. Future improvement will include obtaining shorter amplified pulses and higher average power.

The next generation of Petawatt class lasers will require the development of new laser technology. Optical parametric chirped pulse amplification (OPCPA) holds a potential to increase the peak power level to >10 PW with existing grating technology through ultrashort pulses. Furthermore, by utilizing a new type of front-end system based on optical parametric amplification, pulses can be produced with substantially higher contrast than with Ti:sapphire regenerative amplifier technology. We performed extensive study of OPCPA using a single crystal-based OPA. We developed a replacement for Ti:sapphire regenerative amplifier for high peak power lasers based on OPCPA, with an output of 30 mJ, at 10 Hz repetition rate and 16.5 nm spectral bandwidth. We developed a 3D numerical model for OPCPA and we performed a theoretical study of influences of pump laser beam quality on optical parametric amplification. Our results indicate that OPCPA represents a valid replacement for Ti:sapphire in the front end of high energy short pulse lasers.

Summary form only given. Optical parametric chirped pulse amplification (OPCPA) offers several advantages when compared to regenerative chirped pulse amplification using Ti:sapphire or Cr:LiSAF. These advantages include low cumulative nonlinear phase (B-integral), greater wavelength flexibility, and high gain achieved without the use of gated electro-optic modulators and multipass amplification. High-energy OPCPA systems pumped by large aperture, low repetition rate glass laser systems have produced up to 0.5 J of amplified stretched pulses, subsequently recompressed to 300 fs. However, OPCPA using commercially available Q-switched pump lasers have, to date, only produced 600 /spl mu/J of amplified signal with a pump-to-signal conversion efficiency near 0.3 %. This low efficiency limits the attractiveness of OPCPA to facilities with high energy, shorter pulse pump lasers. We report chirped pulse amplification of broadband 1054 nm pulses to 31 mJ by single passing only 40 mm of gain material pumped by a commercial Q-switched laser, with a pump to signal conversion efficiency of 6%.

We have generated approximately 100 watts of frequency doubled light from the output of an electro-optically Q-switched, diode-pumped Nd:YAG slab laser oscillator operating at an average power of 200watts (2.5 kHz repetition rate, 80 mJ/pulse, 25 ns pulsewidth). The Q-switch was a compensated z-axis propagation LiNbO3 electro-optic modulator, and the frequency conversion crystal was a thin slab ofKTP. In addition, Q-switched operation at an average power of approximately 250 watts with 26 ns pulsewidths has been demonstrated. 1. INTRODUCTION Compact and efficient frequency converted solid state lasers with multi-hundred watt average powersare highly desirable for a number of industrial and military applications. Previous demonstrations of 100 watt frequency doubled output from solid state laser systems have utilized either master- oscillator/power amplifier configurations with the unavoidable degree of complexity inherent to thatapproach,l2 or intracavity frequency conversion of acousto-optically Q-switched lasers which gives

High average power Nd:YAG lasers are increasingly interesting for industrial applications such as drilling and machining. Diode pumping of this solid state medium offers longer services intervals, reduced thermal optical distortions, higher system efficiency and more compact packaging than lamp pumping. The zigzag slab geometry is well suited for applications where the average power exceeds a few hundred watts and a good beam quality is desired, particularly if the laser pumping level is to be varied. We present the status of our latest upgrade to our (originally 300 watt 1 ) diode pumped slab laser. In what follows we first describe the diode pump source. We then discuss the zigzag slab laser design and its present performance.

A method based on polarization selectivity and three step laser photoionization is presented for separation of the odd isotopes of gadolinium. Measurements of the spectroscopic parameters needed to quantify the excitation pathway are discussed. Model results are presented for the efficiency of photoionization. The vapor properties of electron beam vaporized gadolinium are presented which show dramatic cooling during the expansion of the hot dense vapor into a vacuum. This results in a significant increase in the efficiency of conversion of natural feed into enriched product in the AVLIS process. Production of enriched gadolinium for use in commercial power reactors appears to be economically viable using technology in use at LLNL.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Although operational for over twenty years, diode pumped solid state lasers have, for most of their existence, been limited to individual diodes pumping a tiny volume of active medium in an end pumped configuration. More recent years have witnessed the appearance of diode bars, packing around 100 diodes in a 1 cm bar which have enabled end and side pumped small solid state lasers at the few watt level of output. This paper describes the subsequent development of how proper cooling and stacking of bars enables the fabrication of multi kW average power diode pump arrays with irradiances of 1 kW/cm peak and 250 W/cm 2 average pump power. Since typical conversion efficiencies from the diode light to the pumped laser output light are of order 30% or more, kW average power diode pumped solid state lasers now are possible.

We have designed a diode pumped Nd:YAG zigzag slab laser test bed. Our goal is to produce 200 to 300 watts of average optical power with good beam quality. It has been tested, to date, to 70 watts average power using a 20% subset of the design pump diodes.

A ground state depleted (GSD) 1,2 laser has been demonstrated in the form of a Q-switched oscillator operating at 912 nm. Using Nd 3+ as the active ion and Y2SiO5 as the host material, the laser transition is from the lowest lying stark level of the Nd 3+4 F 3/2 level to a stark level 355 cm -1 above the lowest lying one in the 4 I 9/2 manifold. The necessity of depleting the ground 4 I 9/2 manifold is evident for this level scheme as transparency requires a 10% inversion. To achieve the high excitation levels required for the efficient operation of this laser, bleach wave pumping using an alexandrite laser at 745 nm has been employed. The existence of a large absorption feature at 810 nm also allows for the possibility of A1GaAs laser diode pumping. Using KNbO 3 , noncritical phase matching is possible at 140°C using d 32 and has been demonstrated. The results of Q-switched laser performance and harmonic generation in KNbO 3 will be presented. Orthosilicate is a monoclinic biaxial crystal. An oriented spectroscopic evaluation consisting of a Judd-Ofelt analysis of oriented absorption spectra and the measurements of oriented emission spectra has been completed and will be presented. Results of modeling using these spectroscopically determined parameters will be compared with the actual laser performance. The performance of this laser at 911 nm which allows accessing Cs atomic resonance filters through harmonic doubling will also be discussed. Orthosilicate can be grown in large boules of excellent optical quality using a Czochralski technique. Because of the relatively small 912 nm emission cross section of 2-3 x 10 -20 cm 2 (orientation dependent) fluences of 10-20 J/cm 2 must be circulated in the laser cavity for the efficient extraction of stored energy. This necessitates very aggressive laser damage thresholds. Results from the Reptile laser damage facility at Lawrence Livermore National Laboratory (LLNL) will be presented showing Y 2 SiO 5 bulk and AR sol-gel coated surface damage thresholds of greater than 40 J/cm 2 for 10 nsec, 10 Hz, 1.06 μ pulses.

An established photoionization branching-ratio measurement method, in which an excited level of an atom is populated by a pulsed laser from the ground level and repopulation of the ground level is measured by a second laser or lasers, is extended to the case with a finite lower-level lifetime. Two pulsed dye lasers populate the upper and the lower levels of the transition of interest. A third pulsed dye laser probes the time evolution of the lower level by a photoionization process. This measurement, coupled with the upper- and the lower-level lifetimes, permits the determination of the branching ratio and transition probability. Cerium branching ratios obtained are compared with those that we determined from the emission spectrum of cerium. The method is tested for the difficult case in which the lifetimes of the transition levels are nearly equivalent. The effects of magnetic sublevels are investigated in ytterbium.

We have used the 1-m Fourier-transform spectrometer at the National Solar Observatory, Kitt Peak, Arizona, to record the emission spectrum of 140Ce from an electrodeless discharge lamp in the 4000–27 600-cm−1 range. Branching ratios are determined from the intensity data for transitions originating from common upper levels. These measurements are combined with lifetimes determined by a delayed photoionization technique to yield transition probabilities or gA values for 30 transitions. A plot of the logarithm of intensity divided by gA versus upper energy was constructed; the excitation temperature of the source was determined from this plot. Also from this plot, absolute gA values were derived for 228 of the most intense observed lines of Ce i between 10 706 and 22184 cm−1. To our knowledge, the gA values obtained here are the first reported for neutral cerium.

We have used two methods to measure the oscillator strength of the transition between the ground and 17992 cm -1 level in 174Yb. The first technique involves exciting the transition with a laser pulse that is nearly time-bandwidth limited, of uniform intensity, and has a reproducible shape from shot to shot. The population left in the excited state after the pulse varies sinusoidally with a period that depends on the integral over time of the electric field amplitude and the transition oscillator strength. These are the Rabi oscillations that are predicted by application of the Schrodinger equation to the two-level atom. The excited-state population is probed using a two-step photoionization to the continuum. The electric field amplitude is determined from the temporal profile and the intensity of the laser pulse. The second method involves observation of the polarization rotation of a set of degenerate sublevels brought about by a time-bandwidth limited light-shift laser. One sublevel (mj = 0) of the J = 1 level at 17992 cm -1 is populated by a linearly polarized laser. A second copropagating light-shift laser, which is linearly polarized at an angle to the first laser, is tuned between 7.5 and 30 GHz off-resonance with the transition. The light-shift laser causes population to be promoted into the mi = ±1 levels by a virtual transition through the ground J = 0, = 0 level. Two linearly polarized photoionizing lasers photoionize the population only from the mi = ±1 levels. The photoion signal oscillates sinusoidally with a period that depends only on the integrated pulse intensity, the laser detuning, and the transition oscillator strength.

The output spectral characteristics of fully saturated pulsed dye amplifiers for multimode input waves having amplitude modulation (AM) are modeled and compared with experimental measurements. The AM input in our experiments was generated by combining the outputs of two single-mode dye lasers. The spectral output of the saturated dye amplifiers consisted of the input frequencies and several sidebands spaced by the input difference or beat frequency. For our experimental system with kiton red dye amplifiers, up to 20% of the output intensity was contained in these sidebands. Sideband production was observed to decrease as the beat frequency increased with a characteristic half-width of approximately 10 GHz.

Three-step laser excitation was used to populate and observe members of Rydberg series in neutral iron with high effective quantum numbers. These series converge to the ground and to the first excited state of singly ionized iron. The photoionization threshold was also observed. Analyses of the Rydberg series yield the value 63 737(1) cm−1 or 7.9024(1) eV as the ionization potential of neutral iron.