The paper presents a review on the high-power electron and ion beam accelerators intended for pumping gas lasers, which have been developed during the last 15 years at the Institute of High Current Electronics (Tomsk) and the Institute of Electrophysics (Sverdlovsk) under the author's supervision.

The Los Alamos National Laboratory inertial confinement fusion (ICF) program is developing the krypton-fluoride excimer laser for use as an ICF driver. The KrF laser has a number of inherent characteristics that make it a promising driver candidate, such as short wavelength (0.25 micrometers ), broad bandwidth to target (>100 cm^-1), pulse-shaping with high dynamic range, and the potential for high overall efficiency (>5%) and repetitive operation. The large KrF laser amplifiers needed for ICF drivers are electron-beam pumped. A key issue for all laser ICF drivers is cost, and a leading cost component of a KrF laser driver is associated with the pulsed power and electron diode. Therefore, the efficient generation of electron beams is a high priority. The Los Alamos ICF program is investigating pulsed-power and diode designs and technologies to further the development of affordable KrF laser ICF drivers.

Nuclear reactors offer very large energy sources to pump lasers without the need for external power supplies. The large energy deposition possible in nuclear reactor pumped lasers (NRPLs) requires a flowing gas system to control thermal effects. A NRPL amplifier pumped by charged particles originating in fuel films parallel to a flowing buffer gas is presented. A 10 ms full width half maximum Gaussian reactor pulse is specified and the flow velocity in the cavity is varied to investigate the amplifier behavior for hydrodynamic time scales shorter, comparable, and longer than the reactor time scale. The index of refraction aberration in the cavity is dominated by a tilt in the flow direction and a cylindrical focus transverse to the flow. Higher order aberations are also significant and exhibit complex behavior during the pulse as the hydrodynamic time scale becomes long compared with the time scale of the reactor pulse. The far field beam quality resulting from the combined 3rd and 4th order aberrations is found to correlate well with the maximum index aberration in the aperture and not with the average power deposition. This correlation breaks down when the 4th order aberrations become significant and the qualitative nature of the aberration changes.

Large-area electron guns are critical components in many high-energy gas laser systems. The secondary emission electron (SEE) gun offers an attractive option for pulsed laser applications. With this type of cold cathode gun, a dc voltage is applied to the cathode and the electron beam is generated by secondary emission due to ion bombardment processes. The gun is controlled by modulating the source of ions which resides at ground potential. This design greatly simplifies the electron gun power system. SEE-gun systems have been developed which provide 150-220 keV beams at current densities exceeding 25 m(Alpha)/cm² with current density uniformities of approximately ± 10% over areas of up to 5x150 cm². Pulse lengths have ranged from 30 microsecond(s) to 20 ms at repetition rates from single-pulse to 30 Hz. It is expected that the SEE-gun can be scaled to beam voltages of greater than 300 kV, beam areas greater than 1 m², peak current densities exceeding 1 (Alpha)/cm², time-averaged current densities >0.5 m(Alpha)/cm², pulse lengths of 0.1 microsecond(s) to dc, and pulse repetition rates >1 kHz with good uniformity, high reliability and long life. Furthermore, the inherent simplicity of the SEE-gun results in low cost and a compact, light-weight system.

To investigate the excitation technique of high pressure gas lasers by high power microwave discharges, a WR-650 waveguide circuit was assembled. The pulsed microwave source was a magnetron transmitter with a frequency of 1355 MHz, a nominal pulsed power of 2.5 MW and a pulse length of 4 microsecond(s) . The pulse repetition frequency was 10 Hz. A double ridge waveguide coupling structure was designed. The results from these experiments have shown the need for a high power broadband modulateable microwave source to achieve further knowledge important for the design of compact, high power microwave excited lasers. For these purposes, a former radar transmitter, based on an amplifier chain terminated by a klystron, was modified. The transmitter has a center frequency of 1400 MHz with an instantaneous bandwidth of 100 MHz, a pulsed power of 10 MW, a maximum pulse length of 6 microsecond(s) and a maximum repetition frequency of 450 Hz. The amplifier chain is driven by a dielectric resonance oscillator which is pulse modulated by very fast pin diodes. This design allows the generation of multiple pulses within the 6 microsecond(s) time window with microwave pulse rise times of the order of ten's of nanoseconds. The results obtained with this equipment will be presented.

The performance characteristics of a small-scale discharge pumped XeCl laser that is driven by a new type of nonlinear pulse forming network are described. The network contains nonlinear ferroelectric capacitors and is switched by a glass thyratron whose switching capabilities would not normally be fast enough to drive the laser efficiently. The network is capable of providing both a fast voltage rise-time across the discharge electrodes and a fast rate of rise of discharge current. This type of pulse forming network could provide the best method of achieving the optimum driving conditions for this type of laser, and may be of importance not only to excimer lasers but to a wide range of high power lasers which are pumped by pulsed electrical discharges.

The magnetic pulse compressor is intended for pumping the excixner discharge laser (EDL) with a mean radiation power of 1 kW and pulse repetition frequency as high as 200 Hz. The compressor is formed by the input stage, consisting of four parallel capasitor chains to double the voltage in Fitch sceme, commutated by thyratrons, and one magnetic compression stage. The saturating chouke of the stage has one turn on the oval core the external dimensions of which are 800 cm x 350 cm and cross-section is 120 cm². The core is made of a 25 μm thick metal amorphous tape 2HCP. The compression stage capacitance is of 120 nF, maximum thyratron charge voltage of 50 kV, voltage amplitude at the laser discharge gap of 60 kV arid voltage rise time of 120 ns. At present XeCl laser radiates energy pulses of 3 J at the frequence of 200 Hz. In future the switches having charging voltage of 70 kV are supposed to be installed and thus I kW radiation power to be obtained.

A prepulse-sustainer discharge circuit for excitation of a xenon chloride laser is described. A single thyratron switch is used to initiate an electrical sequence which preionizes, prepulses and sustains the electric discharge in the laser. Important parameters within each of the circuits are calculated and the methods of implementation are given.

The design criteria of pulsed laser power supply are examined. The components examined are the power supply, the pulse forming network, and the triggering circuit. In addition, the optimization and mathematical modeling of this system was achieved.

Two projects at the Los Alamos National Laboratory require lower jitter than available from standard commercial excimer lasers. The first experiment demanded an optical pulse width of only 5 ns. With the ± 5-ns jitter of the commercial lasers, it was impossible to achieve the desired timing accuracy. The second application uses commercial lasers as a diagnostic tool to determine the temporal profile in a large-aperture laser experiment. This arrangement uses three different lasers, all of which must be timed within a nanosecond in order to obtain accurate data. This report describes the standard laser performance before modifications, the modifications, and the resulting jitter reduction. In all four cases, jitter was reduced to less than 0.75-ns peak to peak. Details on the electrical and mechanical modifications are included.

Classically, electrolytic capacitors have been used for dc filter applications. However, the inherent advantages of 'electrolytics' such as self-healing and high energy density make them an attractive candidate for pulse circuitry. To this end, a new type of electrolytic capacitor has been developed specifically for high peak current applications. Preliminary results of an ongoing research effort to determine the operational characteristics of this new type of electrolytic are reported. Electrical characteristics such as equivalent series inductance (ESL) and equivalent series resistance (ESR) has been determined for the first generation test units. Along with the ESR determination, the adiabatic temperature rise per pulse was measured to assist in cooling system design as well as operation in environments where heat rejection is limited. The new electrolytic capacitors may facilitate the development of a new class of pulse generators based on high energy density components. Pulse generators for mass launchers, very low impedance loads, and space-based applications where volume and weight density are at a premium are ideal target applications for pulse electrolytics. High power flashlamp drivers for rod and slab lasers are of particular interest. With recent advances in fast solid state switching such as the dynistor, together the ultrahigh energy density capacitors can result in very compact pulsers. Compared to other energy storage alternatives, the electrolytic capacitor still offers the best source of Joules per dollar for traditional applications.

The glass laser fusion driver, GEKKO 12 beam system, at ILE, Osaka, has operated since 1983 to deliver 20 kJ output in 1 ns pulse with the second and third harmonic frequency converters and the pulse tailoring optics. The uniform irradiation of the spherical target in the implosion experiments is a key issue to get the higher density of final compressed core. The pulse power to drive 2000 flash lamps exciting the rod and disk amplifiers is a moderate system with the total energy of 21 MJ in 300 microsecond(s) duration. High reliability and reproducibility are required to keep the laser energy balance of 12 beams within 3%, and no misfiring of 88 high-voltage switches of the laser amplifiers and 14 Pockels optical switches with the total probability of less than 0.1% per shot. Power balance technology involves not only the power conditioning of the flash power supply but also the gain control of the amplifiers and the alignment of a phase angle of the frequency converters, both working very nonlinearly. The precise control is sustained by the precise measurements of the beam energies, power waveforms, and the optical components characteristics such as reflectivity and transmittance. This report includes the optical beam control to realize the uniform illumination with an incoherent light source and the optical phase control in a focusing optics.

The method of achieving a high energy discharge in a controlled manner with relative ease is to use a pulse forming network. The design of such networks generally uses a single inductor and capacitor. Although easy to design and implement, this practice does not provide optimum performance in all cases. For example, most laser welding applications require a uniform energy discharge within the pulse to produce a high quality weld. Analysis has shown that when the pulse forming network approximates a lumped constant transmission line a maximum flat pulse is produced with minimal degradation of the pulse rise time. This paper shows that a low pass filter ladder network composed of equal capacitors and inductors provides a flat pulse whose energy discharge becomes more uniform as the filter's order increases. This is highly favorable in achieving a producible design since all the capacitors and inductors will be of equal value. An algorithm is generated and used to calculate the coefficients for the Laplace transform of the networks. Design equations are derived for calculating the network's components for use with flashlamps. Then a computer's simulation of a network is compared to actual performance.

In this paper an attempt is made to investigate the effect of magnetic quantization on tbe pulse power output of spontaneous emission from II-VI and IV-VI lasers operated at relatively low temperatures, on the assumption that the radiative recombination occurs through the recombination centers which obey the well-known Shockley-Read statistics, by formulating the magneto-dispersion laws of the said materials. We have studied the pulse power out-put in the presence of a quantizing magnetic field by considering the combined influences of spin and broadening of Landau levels. It is found, taking CdS laser and PbTe laser as examples, that the pulse power output exhibits oscillatory dependence and increases with increasing carrier density in an oscillatory way. It is also shown that the power output at both low and relatively higher operating temperatures may pass through a maximum or may continuously decrease with increasing magnetic field at the quantum limit. The corresponding results for wide-gap lasers have been obtained as special cases of our generalized formulations.

The Pulse Systems Inc. Model LP-140 is a commercial pulsed CO₂ laser designed for marking and engraving. It is available with pulse energies in excess of 1 joule and repetition rates up to 7 Hz, thus making it a potential candidate for lidar applications. The authors document the characteristics of the LP-140 performance including power, temporal and spatial mode stability, chirp, and long term operational characteristics. The laser can be made to function as a coherent lidar only if modified to improve its inherent characteristics. This paper addresses work in progress on the following modifications and their effect on performance: gas flow, optical resonator configuration and discharge supply modifications.