In a free-electron laser an electron beam interacts with a periodic electromagnetic structure to amplify electromagnetic waves. A short history of the device, a summary of major accomplishments, and a tutorial single particle free-electron laser physics analysis are presented.

After an introduction to the main ideas of high order harmonic generation, a review of recent experimental results is presented. Calculations dealing with the effect of intermediate resonances and of the ion also are reviewed. Finally, the ideas behind inversionless amplification are briefly discussed.

CO₂-lasers are widely used in industrial applications. The increasing number of demanding applications as well as market needs are defining several trends in laser development. For welding and cutting of heavy steel sections as well as high-speed surface treatment, lasers with high output power are needed, well above 5 kW, which is the maximum for most applications now. Increasing the beam quality is a second trend which can be observed. High beam-quality systems are essential for sophisticated applications in the field of quality controlled and precision processing such as fine cutting, high speed cutting, and high speed welding. Along with these technical considerations the market development is demanding CO₂-lasers, which can be produced at much lower prices compared to conventional systems in order to be competitive worldwide and to open new applications and markets for CO₂-lasers. In order to meet the physical and technical requirements of these trends, a couple of basic investigations in the field of excitation and resonator physics as well as new technical designs are necessary. An overview on the respective activities of the ILT along with examples of technical designs are given in this paper.

This paper reports on a new type of Q-switched high power CO₂ laser constructed for applications in materials processing. The working principle is described and the experimental results for pulse repetition rates at f equals 11 kHz are discussed.

A new design for a rf-excited CO₂-laser uses coaxial electrodes. A high frequency blower located at one end of the electrode system generates a fast axial gas flow between the electrodes. The gas flow is reversed after passing the blower and moves back to the other end of the electrode system along the outer electrode, that carries a water-cooled heat-exchanger. So the gas reaches the other end of the electrode system with lower temperature and after a second reversal it again enters the cooling system. This kind of geometry then allows a very efficient cooling. Since the plasma has a hollow cylindrical shape, the extraction of radiation is not trivial. Several possibilities are available, e.g., a multipass-zig-zag-beam geometry. A theoretical estimation shows that this geometry, where the electrode system, gas flow and heat exchanger arrangement, and the resonator are integrated in a very rugged module, allows the user to obtain -- at least theoretically -- a beam power of 15 kW with a length of approximately 1 meter and an overall diameter of 70 centimeters.

A low pressure (23 mbar) CO₂ laser with fast mechanical Q-switch (f <EQ 20 kHz) provides pulses [t(0.5) approximately equals 200 ns] at several wavelengths simultaneously. The special design of a three-mirror cavity allows the oscillation of up to six arbitrarily selected wavelengths of the same vibrational band.

Laser-induced chemistry has received much attention in the past few years. The economics of such applications are dominated by the costs of photons and the quantum yield of the specific reaction. For a typical multiple-IR-photon process the quantum yield can be as low as 10^-4 which emphasizes the importance of reducing the cost of laser photons. Based on 1982 technology, CO₂ TEA laser operating costs were approximately $100/watt per year for a laser with an electrical efficiency of 6% and an average power of more than 100 kW. Capital costs dominated the energy cost as well as the maintenance and labor costs. At the South African Atomic Energy Corp. we have been involved in the development of high pulse frequency, high average power TEA-CO₂ lasers for the application in the field of laser-induced chemistry. Much of the attention, however, has been focused on the application to separate the isotopes of uranium via a multiwavelength infrared irradiation scheme. The progress that has been made towards the establishment of CO₂-lasers and laser chains for industrial use has been quite outstanding.

Performance characteristics of a high power Q-switched CO₂ laser based on a fast gas flow system have been investigated. A combination of a fast rotating chopper and an intracavity telescope as a Q-switching device has been applied to a fast axial flow, dc- discharged 1.5 kW continuous wave CO₂ laser. The laser delivered 480 W average output power at 12 kHz pulse repetition rate on a free-running basis. Peak power of each Q-switched pulse is approximately 100 kW, and the pulse duration is 250 nsec. An analytical model describing the unique characteristics of the CO₂ laser medium and of the rotating chopper Q-switching technique has been developed and applied to the experimental results. Good agreement was obtained between the theory and experiment for various pulse parameters such as peak power, pulse duration, and pulse buildup time. A method to improve the pulse characteristics using a fast axial gas flow system is proposed.

Microwave excitation of laser gases allows the deposition of high excitation energy densities into an active medium of relatively large diameter and volume. To take advantage of low cost commercially available generators a microwave frequency of (upsilon) equals 2.45 GHz was chosen. An oscillator amplifier system with TEM₀₀ beam quality and tunable wavelength is used in order to obtain a cw laser output of maximum P_L equals 7 kW. The output of a dc discharged, diffusion cooled CO₂ laser oscillator operating in the TEM₀₀ mode with an output power of 265 W (cw) and an average power of maximum 140 W in the pulsed mode is fed into an amplifier consisting of 24 modules, excited by two magnetrons of 2.7 kW power each. The discharge diameter of an amplifier module is d equals 0.051 m and the active length for one discharge module is 1 equals 0.3 m, yielding an active gain path of L_akt equals 7.2 m. In the pulsed mode the amplifier delivers a peak power of up to 1.9 MW at a repetition rate of 10 kHz.

We have developed a compact kW-class CO laser excited by transverse-rf-discharge. The discharge and output characteristics of the rf-excited CO laser were investigated experimentally. Six gas conditions (pressure, temperature, flow speed, CO-, N₂-, and O₂-concentration) are chosen as the experimental parameters. The optimum values of them to maximize the output power are 50 Torr, 200 K, 17 m/s, 5%, 12%, and 0.4%, respectively. The maximum output power obtained is 1.3 kW with the electrical conversion efficiency of 27% and the slope efficiency of 39%.

The use of radio frequency discharges for excitation of two types of carbon monoxide laser is discussed, with emphasis on the attainment of both good beam quality and operation at relatively high temperature. The first laser type uses fast axial flow in circular cross-section discharges, producing power output comparable to similarly dimensioned carbon dioxide lasers at cryogenic temperatures. Problems with obtaining good beam quality in this type of laser, related to the wavelength and high mass flow requirements, are identified. The second laser type uses stationary gas in a slab waveguide arrangement, and has been shown to operate with good efficiency nearer to ambient temperature than the fast flow device. Good beam quality is being sought by the use of waveguide-unstable resonators.

A gasdynamically cooled CO-laser with a dielectrically stabilized rf-discharge in the subsonic region is described, and laser performance data are presented. The laser is based on the well proven design realized several years ago in a 1-kW-laser at DLR Stuttgart. The dimensions have been scaled up to enable laser output powers up to 5 kW. The laser is run in a blow- down gas system which allows operation times of about 10 seconds. At present a maximum output power of 3.2 kW can be achieved.

A simple and compact computer code for the output beam mode analysis of the CO laser has been developed. The theoretical model based on the asymptotic expansion is converted to the computer model, which is divided into smaller structural modules. The program is described in C language in order to be compiled and executed by PCs, Mackintoshes, and workstations. The beam mode profiles under the multiline operation and misaligned cavity conditions are analyzed.

By insertion of a fast mechanical Q-switch into the resonator, continuous discharge CO₂ lasers can yield high peak power pulses at multi-kHz repetition rate. First experiments have been done with diffusion-cooled low-pressure CO₂ lasers. For this type of laser the pulse energy can only be increased by increasing the length of the active medium. A limit is given by the onset of uncontrolled self oscillation which prevents regular Q-switched operation. Single pulse energies can apparently not exceed 30 mJ at 250 ns pulse duration for this type of laser. Fast gas-flow convection-cooled laser discharges allow us to increase the stored energy by increasing diameter and pressure of the active medium as well as the electrical power density. We present the results of Q-switching of a 5 kW industrial laser. Our Q-switch is scalable in optical power. It is based on a fast chopper and a conical mirror. In some experiments we tuned the laser over a wide range by a diffraction grating. The influence of gas pressure and mixtures as well as discharge parameters has been studied. Single pulse energies of 100 mJ have been found, limited by the electrical input power density.

A laser kinetics code is developed to simulate the performance of a dc excited fast axial flow CO₂ laser module. Being a five temperature model, the simulation consists of a set of equations for the relevant vibrational mode temperatures, the intra cavity radiation intensity, and the ambient gas temperature. Input electron excitation rates are computed by solving the Boltzmann equation for a given laser mixture and corresponding reduced electrical field strength E/N (N: gas density). The latter parameter is obtained experimentally by means of floating Langmuir probes. The significant population densities of vibrationally excited states in CO₂ laser discharges necessitate taking into account the effect of superelastic collisions on the electron energy distribution function (EEDF). The calculated dependence of the laser output power on discharge current and pressure is in good agreement with the experiment.

The new kinetic model of CO laser is developed. The basis of the model is multiquantum vibrational exchange rate constants given by Billing. The full Billing model of multiquantum VV exchange gives rise to satisfactory coincidence with experimental data on vibrational distribution function without any fitted parameters. Computer models of vibrational kinetics for CO containing mixtures are currently widely used in the analysis of problems of highly nonequilibrium vibrational excitation and in investigations of CO lasers. Until now only the single quantum VV exchange models were considered. For single quantum processes CO(v) + CO(u) yields CO(v - 1) + CO(u + 1) the rate constants (RC) are usually calculated using expressions, based on the first order perturbation theory assumptions. The parameters of well known theoretical expressions are fitted thereafter to get the magnitude and vibrational quantum number dependence of RC, measured experimentally. The RC, extended in this way to VV exchange of highly excited molecules, grow rapidly with v,u and exceed gas kinetical RC at rather low v,u > 7 for quasi-resonance exchange. Thus the validity of the first order perturbation theory expressions breaks and vibrational kinetics models using these RC become doubtful for relatively low levels. Nevertheless, these RC were widely used in practical calculations of CO laser kinetics. Their justification was in satisfactory agreement with measurements of vibrational distribution function (VDF), available in the literature.

Our center has developed a new type of commercially available high-power (up to 5 kW) multichannel diffusion cooled waveguide industrial CO₂-lasers with ac-pumping (MTL). The physical principles and the results of scientific investigations of these lasers have been published. Such a laser has a parallel bunch of gas-discharge laser tubes placed between two flat mirrors. The laser generates a large-aperture multibeam. The components of this multibeam may be non-synchronized, but there exist the optical methods of their synchronization and aperture-filling. The main advantages of the MTL are: relative simplicity of their construction, low cost of production and maintenance, high level of compactness. For example, the MTL-4 model (4 kW rated power) has a 60 kg, 2 X 0, 2 X O, 2 m laser head, 3 nl/h He consumption and 12% plug efficiency.

In the optical resonators of industrial CO₂-lasers there exist stagnation zones with unexcited working mixture. In these regions the beam is partially absorbed by carbon dioxide; this results in variation of the laser beam energy and spectral characteristics. In this connection, of practical interest is the study of the unexcited mixture heating stability under various procedures of heat removal: gas mixture pumping, heat conductivity, or natural convection. For the optical thermal instability under consideration, there is an effect of thermal runaway known in chemical burning theory and emerging under self-acceleration heat release, when it cannot be compensated by heat removal.

A one-dimensional theoretical model has been developed to simulate the amplification and oscillation processes in a transverse electrically excited CO₂ laser with gas flow longitudinal to the optical axis. It allows us to describe the complete beam generation process and it is able to solve specific problems of gas discharge or amplification performance. In this paper the model is used as a support for the understanding of special properties of longitudinal flow CO₂ lasers that strongly affects the power performances. Particularly the uncertainties in the discussion of experimentally achievable coefficients of small signal gain are shown.

Fast axial flow rf-excited discharges ((nu) _rf equals 13.5 MHz) in tubes with rectangular cross sections and plane electrodes have been investigated. Probing the tubes on-axis the dependence of small signal gain as well as saturation intensity on the input power density is identical for all the different cross sections. Under equal conditions the small signal gain profiles revealed significant decrease in the vicinity of the side walls with increasing tube dimensions. This is attributed to the change of discharge behavior with increasing interelectrode discharge length. It was found that the critical pressure for the onset of filamentation is reciprocal to this dimension. With this geometrical dependency the laser power of axial flow rf-excited discharges is not volumetric scalable in a straightforward way.

Axial-fast-flow CO₂ laser was investigated numerically taking into account turbulent and convective cooling of the gas medium. The model developed includes a description of the vibrational kinetics and the gas motion. The calculations were made for two real 500 W and 1200 W CO₂ lasers, respectively. The results obtained show a good agreement between the experimental and theoretical data.

An output power of 2 kW is attained with a compact coaxial slow-flow CO₂ laser. The resonator consists of two toric copper mirrors, which are tilted a small amount. These mirrors form an unstable resonator in azimuthal direction, from which two laser beams are extracted with high efficiency through a coupling aperture in one of the mirrors. The beam divergence of a single beam is nearly diffraction limited in the radial and azimuthal direction.

In order to achieve a compact industrial fast axial flow CO₂ laser delivering an output power of a few kilowatts with a high beam quality, the authors have investigated the behavior of an amplifying medium excited by an ac discharge also called `silent discharge.' The external electrodes discharge tube and this ac power supply have been developed. The power supply was working in a resonant mode within a frequency range of 95 kHz to 125 kHz and showed that the power transmitted to the load can be modulated by frequency modulation. The dependence of the capacitive coupling between the power source and the discharge load on the electrode geometry are studied. The dependence of the gas temperature on the electrical characteristics of the source such as voltage, current, phase, and injected power are displayed. In this paper we show that for a gas temperature of 200 degree(s)C the maximum input power density was limited to 11 W/cm³. In an attempt at a laser application, an efficiency of over 15% was obtained with a laser output power of 800 W for an electric power of 4,800 W.

A fast axial flow high power CO₂ laser excited by dc discharge has been developed. An aerodynamic technique was developed for achieving homogeneous discharge of power density up to 6 W/cm³ in 5.5 cm diameter tubes. An output power of 10 kW was obtained with an unstable two-folded resonator with a magnification of 2. The laser is made of 12 active mediums of 5.5 cm diameter and 60 cm length and the electrical discharge efficiency is 17%.

A multikilowatt operation of a multijoule transversely excited atmospheric CO₂ laser driven by an all-solid-state exciter has been demonstrated. The exciter consists of a high- voltage pulse generator employing a newly developed semiconductor stack switch, 48-series and 2-parallel insulated gate bipolar transistors, and a two-stage magnetic pulse compressor. The maximum average laser power of 4.6 kW was obtained with an overall efficiency of 10.5% at a repetition rate of 550 pps. The maximum repetition rate of 1100 pps was attained with an average power of 3.4 kW and an overall efficiency of 7.4%.

For a dc self-sustained discharge adapted for excitation of a transverse-flow 1.2 kW CO₂ laser stable operating condition in a CO₂/N₂/He equals 1/9/15 mixture at static pressure of 30 - 40 Torr and by flow velocity up to 70 m/s were obtained in the excited volume of 2.1 dm³ by the dissipated electric power density up to 9.5 W/cm³. It has been confirmed that the passivation and conditioning of electrodes influences strongly the long-term output characteristics.

The kinetics of chemical processes in the laser gas mixture is theoretically studied to explain its effect on power extraction of the transverse-flow 1.2 kW CO₂ laser system. The reactions of chemical species with excited electronic states, ions as well as heterogeneous recombination, are included into the model. The chemical equilibrium strongly depends on the excitation conditions.

The small signal gain distribution along the flow is taken into account in the diffraction type analysis of the output properties of a stable multipass resonator designed to operate with transverse-flow cw CO₂ laser. Experimentally obtained gain profiles for different laser excitation parameters are introduced to the numerical model of the resonator. The laser output characteristics are analyzed and compared with the experimental data.

The conditions for fundamental mode operation of the transverse flow CO₂ laser designed for industrial applications have been investigated experimentally. The effect of the field limiting apertures on the laser output characteristics has been studied for three stable resonator configurations described by the different value of Frensel number, G-parameter, and active medium length. The results are compared with the predictions of the theoretical analysis concerning the diffraction properties of the stable resonator.

The strikingly small affect of the mixture degradation on the power extraction observed experimentally in the MLT-1200 laser has led to the examination of the time characteristics of laser output as a function of the gas mixture composition. Processes competitive to the working mixture degradation were considered and we found that the optimum concentrations of CO₂ and N₂ lie in the region 2 <EQ [CO₂] <EQ 3% and 30 <EQ [N₂] <EQ 40%, respectively.

In this work, a new grid was added to a grid corona discharge TEA CO₂ laser. The new grid was mounted under the cathode and inside the laser tube. The effect of this new mesh on the uniformity of the glow discharge was investigated. Experiments showed that the new mesh extends the arcing limit of oxygen concentration up to 9.5%. In this study, the effect of the tube size and the electrode spacing was tested on the output and loading energy density in the medium aperture of 2.5 cm on the laser tube.

Plasma injection method has been used to stabilize the main electric discharge in the atmospheric high power cw CO₂ lasers. In this work, an analytical perturbative method is introduced for solving Boltzmann equation for a general plasma. This method is applied to a specific plasma injection CO₂-N₂-He laser. Small signal gain and output power of the laser are calculated in the steady state as a function of gas mixture and pressure. Dependence of small signal gain and output power on the main discharge current also were considered.

Transverse electric atmospheric (TEA), or multi atmospheric (TEMA) lasers deliver intense short laser pulses of considerable energies. Recurrent high repetition rate pulse trains afford substantial average power levels. In a high rep-rate operation the gas flows across the cavity and is externally cooled to maintain a reasonably low temperature. The gas flow gear and heat exchanger are bulky and costly. In this work we present a repetitively pulsed TEA or TEMA laser that combines energy and peak power features in an individual pulse with the substantial average power levels of a pulse train in a thin layer of gas. Excess heat is disposed of, by conduction through the gas, to cooled enclosing walls. The gas does not flow. The method applies to vibrational transition molecular lasers in the infrared, where elevated temperatures are deleterious to the laser operation. The gist of the method draws on the law that heat conductivity in gases does not depend on their pressure. The fact lends unique operational flexibility and compactness, desirable for industrial and research purposes.

The discharge effects on gas flow lasers have been treated from both aspects of the activating features to the laser performance and of demerits by electrical heating and disturbances in the laser medium. The approach has been from two types of studies, namely on the continuous discharge in CO₂ supersonic mixing electric discharge laser (EDL) and on the pulsed discharge in excimer laser. For CO₂ supersonic mixing EDL, small scale experiments have been performed on the small signal gain coefficient and on the laser power, together with the numerical analysis by using the quasi-one-dimensional and vibrationally nonequilibrium equation system. Discharge effect is included by solving the Boltzmann equation for electron energy distribution function, and power extraction analysis also is carried out. By comparing the experimental results, various characteristics have been clarified on this type of supersonic discharge laser. As for the pulsed discharge, flow visualization experiments have been conducted in the model cavity of excimer laser, along with a numerical calculation on the one- dimensional Euler equation system by TVD approach. There have been three types of waves in the laser cavity, and Mach numbers of horizontally propagating main waves have been discussed from both numerical and experimental aspects.

In previous works a numerical study of the unsteady two dimensional flow in the cavity of an excimer laser has been presented. In this paper, the numerical method based on a finite difference scheme, associated with a flux transport algorithm (SHASTA-FCT method), is developed for computing the flow in a pulsed chemical laser. The gas mixture contains SF6 and the evolution of the ratio of specific heats (gamma) as a function of temperature must be included in the numerical approach. The elliptic shape of the lower electrode leads to a convergent divergent laser cavity and allows a better damping of pressure waves due to the non-uniformity of the energy deposition. From a comparison between numerical and experimental results on the evolution of the wall pressure at a given position it is possible to come up to the real shape of the energy deposition. In addition, the numerical simulation of the flow is carried out over two interpulse time to get the description of the flow field in a multidischarge configuration.

Theoretical and experimental investigations on the propagation, reflection, and attenuation of shock waves as they occur in excimer lasers have been performed. The numerical simulations have been carried out using a two-dimensional, unsteady finite difference scheme. The experimental setup is a piston driven shock tube with a rectangular cross section working in air at atmospheric pressure. The shocks were detected interferometrically as well as by means of pressure transducers. This shock tube allows us to investigate basic phenomena of shock diffraction which can be used to confirm the computational results in the range of weak shock waves. In particular, the influence of the shape of the wall contour on the reflection of shock waves has been investigated theoretically. The decay time of pressure and density perturbations differs for various wall configurations in such a way that short electrodes accelerate the attenuation as well as does a strong area increase in the vicinity of them. After each laser pulse there is a shock travelling into the laser channel. Experiments have been carried out on the reflection of this shock at a specially formed bend that is able to focus the shock into a muffling element.

A theoretical analysis is made to describe the influence of an active disturbing medium on the far-field diffraction pattern of an excimer laser beam. Results are in good agreement with experiences. The gas density perturbations induce amplitude and phase aberrations which are responsible for the degradation of the beam characteristics in the far-field.

The possibilities of a longitudinal N₂ laser pumped by BIW have been demonstrated. The developed laser gives us the possibility of exciting a nitrogen with specific stopping energy of about 0.5 J/cm³ and of producing laser action in N₂ with an output power up to 100 kW in pulse having a duration of 10 ns FWHM with excellent laser beam quality.

Depending on the mutual orientation of the output laser beam and direction of electric current in the discharge gap, two schemes of laser pumping by electric discharge are distinguished: longitudinal and transversal. Currently, most commercial pulsed lasers operate on the transversal scheme of excitation, with more powerful laser radiation due to higher stopping electrical power in the gas. The longitudinal scheme is used for producing a laser beam with more qualitative characteristics. In this case, due to the discharge gap length being longer than in the transversal scheme, the nanosecond electrical pulse with a higher amplitude is used. The value of the reduced electrical field strength E/P remains low for effective excitation and ionization of the gas. Typically the time of breakdown development and electric current growth is not enough for effective laser generation. We propose overcoming these shortcomings with the use of nanosecond gas discharge in the form of breakdown ionization wave.

The comprehension of the performance of pulsed gas lasers requires knowledge of several parameters which are time dependent and consequently difficult to measure. These parameters are mainly influenced by the two electric discharges which take place during the laser performance, namely by the ignition system and laser tube. The present work illustrates a new method to determine the most fundamental of these parameters, the current. This can be achieved exploiting only the voltage waveform on which much concealed information exists about all the time dependent parameters. The revelation of these parameters can be achieved by further elaboration of the waveforms of the high voltages. This elaboration, in our case, leads to finding the current and is described in this paper. The method is applied to a simple RLC circuit, to a doubling, and to a C-to-C circuit and it is delivered from experimental errors existing in other methods achieving the best accuracy to date.

Utilizing a pulsed beam of a Nd:YAG laser, hole-burning through the opaque cloud of products formed following the detonation of lead azide is demonstrated. The characteristics of the hole and of the expanding cloud are monitored in real time by a HeNe beam and by high- speed framing photography. The hole is carried with the cloud and propagates at a constant velocity of 0.5 - 2.8 km/s, depending on the time and location of burning. The hole-burning is a result of eliminating solid particles from the cloud. The reduction in the number and size of the particles is monitored by scanning electron microscopy of the deposits formed on a substrate following the detonation. The application of a laser to burn a hole in the detonation products from a solid explosive is demonstrated for the first time. This technique may serve as a method for flow visualization in an aerosol medium.

The effects of the supersonic nozzle shape on the performance of CO₂ gasdynamic laser are analyzed by solving time-averaged two-dimensional Navier-Stokes equations coupled with the vibrational relaxation equations for a CO₂ - N₂ system. The solver is based on a finite difference method with an implicit high accuracy TVD scheme. Numerical simulations are carried out for various types of supersonic nozzles. The performance is estimated with the indicator of the small signal gain coefficient. The results show possibility of improvement for small signal gain coefficient by nozzle shape modification and suggest a suitable nozzle shape.

It is a serious task to control the flow pattern in the cavity of a chemical oxygen iodine laser (COIL) in order to improve the performance of the facility. The major task is to mix the iodine with activated oxygen uniformly. We performed the computer calculation simulating the actual test plant. Results of the calculation show a uniform concentration rate of iodine corresponding to the actual power plant.

In order to increase high power closed cycle CO₂ laser performance, an experimental investigation of the velocity field in a cylindrical glow discharge tube was carried out. The axial velocity profile was measured (without electric discharge) using a laser Doppler anemometry technique at several sections of the discharge tube. The turbulent compressible flow has also been theoretically investigated by means of the PHOENICS code and comparisons of numerical results with experimental data show a reasonably good agreement.

Emission spectra from a high-current line plasma produced by means of a formed ferrite plasma source have been measured in the 1200 - 2900 angstrom spectral region. An observation of the plasma expansion with the formation and the propagation of a blast wave (1.8 km/s) is presented. The possibility of a peak intensity shift of VUV radiation to shorter wavelengths is discussed.

High repetition rate excimer lasers are expected for wide industrial application. The power of excimer laser, however, decreases rapidly in a higher repetition rate operation. Shock or acoustic waves, which are caused by the periodic pulse discharge, may limit the repetition rate of an excimer laser up to 2.5 kHz. Such waves cause inhomogeneity of gas density in the discharge region of the excimer laser. In high repetition rate operation this inhomogeneity remains at the next discharge. Arcing may be generated by this inhomogeneity and the homogeneous excitation of the laser gas is obstructed. Although these phenomena have been reported, the research for the effects of shock waves has remained insufficient. And the relation between these shock waves and discharge phenomena has not been clarified. To resolve this problem, we developed a scaling model chamber of a UV preionized excimer laser cavity with windows for flow visualization. We report the first result by using this model and Schlieren technique in a pure helium gas case. In our experiment three types of shock waves are found in the discharge cavity. Those shock waves are generated from the boundary of the main discharge area, from sparking pin gaps, and from the main electrode surfaces. In this study we focus on the effect of xenon gas on the generation and the propagation of shock waves. Components of the Xe-Cl excimer laser gas are helium, xenon, and hydrogen chloride. In those gases xenon has the largest molecular weight of 131.29. So we conclude xenon plays an important role in the shock wave propagation and in discharge phenomenon.

Sensitivity analysis has been applied to the description of an XeCl laser to analyze how the changes in the kinetic rate coefficients influence the species concentrations.

Supersonic expansion of an He fast discharge excited plasma is shown to produce the HeII 164.0 nm line by recombination. Results related to the spectrum, the time profile of the emitted radiation, and the effect of the backing pressure are presented. The study is of interest because of the possibility of getting amplification in the HeII line at 164.0 nm.

The overview summarizes past and present research on short-wavelength chemical lasers (SWCL) with special emphasis on recent advances in the field. Guidelines for developing SWCL and the most promising candidates are presented. It is concluded that premixed, pulsed chemical lasers may have some inherent advantages over their continuous-wave counterparts for operating a long sought after viable SWCL.

We demonstrated a high energy delivery of a pulsed wavelength-selected HF chemical laser by fluoride glass fibers (core/cladding equals 450/500 micrometers ). The optical energy of 19 mJ in a 540 ns pulse was successfully delivered with a peak intensity of 22 MW/cm² at the exit core surface. We have also theoretically investigated the operational characteristics of the pulsed chain first vibrational overtone HF chemical laser using a computer code. The model used can describe simultaneously both the fundamental and overtone oscillations. The higher overtone output energy of 2.73 J/l can be obtained with a gas mixture consisting of F₂/H₂/He equals 10/4/786 (Torr) by successfully suppressing the ASE.

A compact, repetitively pulsed HF chemical laser, built with a gas recirculator loop to cool and process the gas mixture is described. A simple corona phototriggered discharge is used to produce vibrationally excited HF molecules at typical pressures of 90-150 Torr. This laser produced over 5 J per pulse in single- shot operation, and a maximum average power of 500 W. at a repetition rate of 110 Hz. The electrical efficiency is 3%.

The B yields X electronic transition of the iodine monofluoride (IF) molecule is a well known potential system for a chemical laser operating at visible wavelengths. By using the reaction between fluorine atoms and excited atomic iodine, IF(B) was produced in a supersonic flow. F and I atoms are produced in a combustor and excited I atoms are produced by mixing excited O₂(¹(Delta) ) to the expanded supersonic flow of combustion products. Chemiluminescence of IF was observed at a pressure ranging from 5 to 10 Torr. Experimental spectra have been recorded and compared to synthetic ones. A population density of IF(B) close to 4 X 10⁹ molecules/cm³ has been determined. No laser effect has been observed. A computer code has been developed and applied to the kinetic scheme. Calculated gain profiles, IF(B) population density, and laser power are in disagreement with the experimental values. Adding fluorine (F₂) to the flow increases the measured population density of IF(B) by a factor of 10. This value seems to confirm the existence of the step I^* + F₂ yields IF₂ in the production of IF(B).

Chemiluminescence was observed due to the reactions which take place in the supersonic flow of dissociated nitrogen trifluoride mixed with hydrogen. An intensive glow on the NF a yields X and b yields X transitions was identified. Concentrations measured of NF(a) and NF(b) states are high and permit hope for overcoming the lasing threshold in the supposed NF-IF laser.

A compact, repetitively pulsed HF chemical laser, built with a gas recirculator loop to cool and process the gas mixture, is described. A simple corona phototriggered discharge operating in SF₆/C₂H₆/Ne gas mixture is used to produce vibrationally excited HF molecules at typical pressures of 90 - 150 Torr. This laser produced over 5 J per pulse in single shot operation and a maximum average power of 500 W at a repetition rate of 110 Hz. The electrical efficiency is 3%.

The change of refractive coefficient of active gas medium in pulsed chain chemical HF-laser due to temporal variations of chemical composition and populations of vibrational states HF(v) and H₂(v)-molecules is analyzed. These changes are related to different polarizability of components of the chemical laser medium and the dependence of polarizability on vibrational state population for atomic and electronic constituents of polarizability. The atomic polarizability is characterized by sizable frequency dispersion. The change of reflective coefficient due to the change of vibrational excitation is a nonline effect, since the vibrational excitation depended on intensity of the laser beam.

Based on the measurements of the absolute emissivity of the thermally heated molecular chlorine in the shock wave flow, the redetermination of radiative characteristics of the Cl₂ B³(Pi) o+u yields X¹(Sigma) ⁺g and A¹(Pi) 1u yields X¹(Sigma) ⁺g electronic transitions was made with the aim of pre-laser studying a potential new laser active media. Reasons for errors in previous works are discussed.

The design, construction, and operational characteristics of an arc driven supersonic cw HF chemical laser of more than 1 kW power, which has been developed for laser-material investigations, are presented. Overtone lasing has also been obtained with a power of 220 W. The overtone chemical laser uses the same chemical reaction and produces the same excited populations as the conventional hydrogen fluoride laser does. It differs insofar as the optics employed suppress the lasing at 2.8 micrometers and allow lasing at 1.3 micrometers .

A wide area x-ray generator working at relatively low pulsed voltage was carried out in order to preionize an HF laser discharge. This x-ray source is based on the use of a corona plasma cathode. With this device, preionization densities of about 10⁸/cm³ in 1 bar argon and of approximately 10⁷/cm³ in the 110 Torr laser mixture were obtained for a 30 ns FWHM x-ray pulse duration (corresponding to approximately 4 J of input energy). The operating characteristics of this corona x-ray generator are presented, as are the preliminary results on the x-ray preionization of the HF chemical laser.

In the pulsed HF chemical laser the reactants, typically SF₆ and H₂ diluted in He, are premixed in the laser cavity. The laser pumping reaction, F + H₂ yields HF^* + H, occurs following the production of free neutral F atoms from the SF₆ host molecule. One method commonly employed in `cracking' the F atom host molecule is dissociative electron attachment during pulsed electrical discharge. Typically, this pulsed discharge is produced when a low-inductance storage capacitor, C<sub>1</sub>, is charged to high voltage, and discharged through a two electrode gap in the laser cavity with a low-inductance circuit. The introduction of a high-voltage switch into this circuit is necessary because the laser gap threshold breakdown voltage is usually too low to achieve appreciable efficiency. The efficiency of this process is improved by introducing an intermediate capacitor, C<sub>2</sub>, into the circuit. Generally this efficiency improvement is explained by the observation that the C<sub>2</sub>-laser circuit has a much lower inductance than the C<sub>1</sub>-laser circuit because the C<sub>1</sub>-laser circuit must include a high-voltage switch. Thus, in this capacity, C<sub>2</sub> is called a `speed-up' capacitor. In this paper we present data to show that for voltages above the discharge threshold region, and various C₁/C₂ ratios, there is a direct correlation between laser output energy, and the energy transferred to the C₂ capacitor.

From the viewpoint of theory, all of the approaches to the analysis of laser dynamics are based on a so-called Maxwell-Bloch system of equations, which has been the basic model of maser and laser dynamics since 1957. This model does not take into account the active medium motion as a whole, and so the influence of active medium motion on laser dynamics has been neglected. The present report is devoted to this problem.

A six-temperature vibrational relaxation model that takes into account the vibrational levels of O₂ and H₂O is proposed, based on which a simulation code has been developed with which analyses have been conducted on the effects brought on the performance of combustion- driven gasdynamic lasers by the presence of O₂ and H₂O in the medium. Inclusion of the vibrational levels of O₂ and H₂O in the vibrational relaxation model has proved to ensure good agreement of estimated results with experimental data over a wide range of gas composition (0 - 40% O₂, 0 - 10% H₂O). Both calculation and experiment have indicated that with the presence of 3% H₂O, an increase of 1% in O₂ concentration would lower the small signal gain by 1%. The simulation estimated that increasing concentration of H₂O in the laser medium accentuates the effect of O₂ on the lasing performance. This six-temperature vibrational relaxation model should permit reliable analysis to be performed on gasdynamic laser characteristics even when large amounts of O₂ and H₂O are present in the medium.

The vibrational relaxation of the CO₂ molecules in a mixture with Ar in the supersonic expansion of a coupled-modes gas-dynamic laser (GDL-CM) is studied in this paper. The measure of vibrational nonequilibrium is introduced and its relation to the available power of GDL is discussed. The effect of the initial temperature, pressure, and CO₂ concentration is examined.

Solutions of the binary master equations for the state populations in an expanding gas are facilitated by the use of a flow similarity variable. This similarity variable accounts for the density and velocity variation associated with a gas dynamic expansion and can be expressed as a function of the local Mach number, the stagnation conditions, and a nozzle geometry parameter. With the similarity variable, the master equations are transformed into a form which is independent of the nozzle shape and similar to those for a non-flow system. This simplification facilitates both numerical solutions and closed form approximations for populations in nozzle flow. With solutions in terms of the similarity variable the design conditions leading to a population inversion can be presented as a simple algorithm. An illustration is presented for a wedge nozzle expansion. The conditions for inversion are determined by the expansion Mach number at a stagnation pressure. With the algorithm these conditions for a range of nozzle sizes are presented as an inversion criteria curve.

We investigated performance of a gasdynamic laser (GDL) with combustion gas of a double base propellant theoretically. Small signal gain coefficients in the supersonic flow of the combustion gas were calculated for different stagnation conditions characterized as the operation region of conventional GDLs. Preliminary experiments show that less solid particles exist in the combustion gas flow than we expected. It was revealed from theoretical and experimental results that a conventional GDL which uses the doubled base propellant as a gas generator could be realized.

The basic challenge in the design of a pressure recovery system for a supersonic gas laser arises from the fact that the cavity pressure is quite low: 50 - 70 Torr for a CO₂ gasdynamic laser (GDL) and as low as 4 Torr for a chemical oxygen-iodine laser (COIL) system. The purpose of the pressure recovery system is to increase the pressure from its value in the supersonic cavity to one which allows the laser gas to be exhausted into the atmosphere. Two types of pressure recovery systems which are of primary interest for supersonic gas lasers are diffusers and ejectors. In this paper we derive some performance limits and examine the characteristics of both types of devices.

As the life time of the vibrational modes (001) and (100) of CO₂ are different, the population inversion in gasdynamic laser (GDL) is sensitive to the cooling rate of gas. If the cooling rate of gas is too fast, both levels will stay hot and so the gain decreases. However, if the cooling rate is too slow, two lasing levels cool down accordingly and again it lowers the gain. In turn, the slope of the supersonic nozzle affects the cooling rate of the levels and also the population inversion. Thus, the shape of a supersonic nozzle is an important factor in small signal gain of CO₂ gasdynamic laser. On the basis of mean value theorem, one would expect that for every family of supersonic nozzle shapes (shock free, wedged, logarithmic, and exponential), there will be a specific one having an optimum cooling and gain. Small signal gain of a GDL was calculated for the different nozzle shape of the different families. It was found that the exponential nozzle shapes would give an optimum population inversion.

In this paper we intend to present some recent result obtained in our laboratory on excimer laser development and applications. Even if the activity has been initiated as long as 15 years ago after 1986, following the European Eurolaser Initiative, the programs were strengthened and enlarged to accommodate more ambitious goals, according to the increased attention to this field. Actually the activity includes the following: (1) development of source, (2) computer models, 3) special devices, and (4) applications. The description of our activity is presented according to the above mentioned order, with special attention to the most recent results.

Since 1986, five kinds of excimer laser technologies have been studied by five companies as part of the large-scale project `Advanced Material-Processing and Machining System,' which is one of the national R&D projects in Japan. The five kinds of excimer lasers being developed are as follows: (1) high power XeCl laser with an average power of 2 kW, (2) high repetition rate XeCl laser with a repetition rate of 5 kHz, (3) automatically power-stabilized 400 W average power XeCl laser with a stability of +/- 1% or less, (4) long-life ArF laser with a gas life of 10⁹ shots or more, and (5) high beam-quality ArF laser with an average power of 200 W. The project passed an interim evaluation at the end of fiscal 1990, and is now ongoing toward the achievement of final targets as of the end of fiscal 1993. The present status and future prospect of the technologies are reviewed.

Experiments performed with a terawatt-class laser system which is based on amplification in XeCl discharge amplifiers are described. Energy extraction in XeCl discharge amplifiers is examined experimentally and theoretically using 0.25-ps nearly transform limited pulses and both 16- and 75-ps strongly chirped pulses of the same bandwidth. We find that while the saturation fluence for the 16-ps pulse is comparable to that of the 0.25-ps pulse, stretching the pulse to a width of 75 ps increases the saturation fluence by 35%. A study of the interaction of the terawatt laser pulses with aluminum targets at irradiances exceeding 5 X 10¹⁸ W/cm² is also presented.

High power excimer lasers are required for industrial applications such as micromachining of printed circuit boards. Reliability and durability are as important as high output power. At the design of a new laser, the lifetime of gas and components must also be taken into consideration like output power and beam profile. Another point is the size of the laser. It should be transportable without being disassembled. The output power of excimer lasers can be written as pulse energy times repetition rate. There are various ways to reach 500 W output power. Opposite to the LPX805, which delivers a pulse energy of 3 J at a large beam cross section, we aspire to a pulse energy of approximately 800 - 1000 mJ with the system presented here. The maximum repetition rate of the laser is in the region of 500 - 600 Hz.

We have developed a high repetition rate, high average power excimer laser intended for industrial applications. The system operates in the phototriggered mode and is capable of continuous operation at repetition rates up to 700 Hz. A compact closed loop gas recirculation system, driven by two centrifugal blowers is employed, providing gas flow velocities up to 35 m/s. The laser delivers a maximum average output power of more than 500 W at 308 nm wavelength.

The VEL industrial XeCl laser has reached 1 kW average power at a pulse repetition rate between 80 and 100 Hz (12.5 J X 80 Hz or 10 J X 100 Hz) with an overall efficiency of about 2%. Single pulse operation gives a maximum energy per shot of 15 J (efficiency 2%) and a maximum efficiency of more than 2.5% at 10 J (150 ns FWHM); the beam size is 8 X 6 cm².

A high repetition rate excimer laser has been developed. The laser gas can be circulated at a flow rate of 120 m/s uniformly in the discharge region by a 4-stage axial blower. The recent results are presented on the XeCl laser operated stably at an ultra-high repetition rate of 4 kHz and an average power of 200 W in the condition of CR equals 3 with thinner He buffer gas mixtures. Above this repetition rate, the laser power saturates caused by the discharge constrictions. The residual charged particles probably induce this localized discharge instability, and Xe content has a harmful influence dominantly on the output power.

Improvement in efficiency and specific output energy on an XeCl long pulse laser ((lambda) equals 308 nm) and the first achievement of long pulse (150 ns FWHM) laser emission from KrCl (B - > X) at (lambda) equals 222 nm are reported. The system, which includes x- ray preionization and double discharge (pulser/sustainer) excitation, allows potentially high laser efficiency from a small volume active medium with relatively low pressure Ne/Xe/HCl, Ne/Xe/BCl₃, Ne/Kr/HCl and Ne/Kr/BCl₃ working mixtures.

To make clear the fundamental processes of a KrF laser pumped by electrical discharge, we measured the time history of the Kr atom density in an excited state and the electron density. The relative computer simulation was also put forward to explain the experimental results.

A complex model of XeCl selfsustained discharge laser and two simple models based on different approximations are used to study a small and a large volume device. It is found that both simple and complex models reproduce satisfactorily the experimental results obtained for the small volume device, while they give too long a laser pulse when applied to the larger one. A one-dimensional modeling is performed to study the possibility of a discharge contraction, and as an open problem the recombination kinetics of positive ions is discussed.

To reduce the load on switching devices, a new type of all-solid-state excitation circuit has been successfully developed, where only one GTO thyristor is utilized in the switching device and a saturable transformer is used as the magnetic pulse compressor. We have used a saturable transformer with a winding ratio of 1:7. The initial pulse of 5.5 kV, 1.7 kA, 1.0 kA/microsecond(s) was transferred to the output pulse of 32.7 kV, 8.7 kA, 282 kA/ns. Employing this circuit for XeCl-excimer laser, we have succeeded in lasing the output energy of approximately 30 mJ/pulse.

A dispersively compensated scheme for sum-frequency mixing has been developed to generate subpicosecond injection pulses at 193 nm for subsequent amplification in ArF excimer amplifier modules. Such a scheme is capable of generating 12 (mu) J at 193 nm with a spectral bandwidth of 0.22 nm corresponding to 250 fs pulse duration by mixing short pulse radiation at 266 nm and 707 nm in a 1 mm BBO crystal. This VUV source has been used to characterize the small-signal gain and the saturation energy density of a discharge pumped ArF excimer amplifier.

The `Mercury' KrF laser facility at Los Alamos, New Mexico, U.S.A. is being built with the benefit of lessons learned from the Aurora KrF laser. An increased understanding of KrF laser engineering, and the designed implementation of system flexibility, will permit Mercury to serve as a testbed for a variety of advanced KrF technology concepts.

This paper describes the design and operating characteristics of a KrF laser which operates at variable pulse repetition rates up to 100 pps.

A comprehensive kinetic model and numerical code were developed to simulate photolytic excitation and lasing in gas media pumped by intense, broadband optical sources. The kinetic model is generally applicable to any photolytically pumped lasing scheme. Initial simulations were performed for the XeF(C yields A) laser and computational results are compared with experiments using exploding wires (EWs) and segmented surface discharges (SSDs).

A high-speed wind tunnel, made by using a Ludwieg tube, has been successfully developed for a highly repetitive discharge-pumped excimer laser. This apparatus allows the gas flow of velocity approximately 204 m/s, pressure approximately 293 kPa, temperature approximately 254 K, and duration time approximately 48 ms. The rate constant for the recombination process of Xe⁺ + Cl^- + Ne yields XeCl^X + Ne is found to increase to a maximum of 4.2 X 10^-6 cm³/s at 180 K in a gas pressure of 294.2 kPa. The kinetic simulation of XeCl excimer laser using such a rate constant indicates the enhancement of the laser output in the lower gas temperatures.

The 1-D numerical code developed to model the performance of a high power excimer laser is described. The spectral method and the Rigrod equation are basic to the code, which is quite general for application to any two-mirror resonator geometry. The specific features of the code make it appropriate for analyzing laser resonators with high Fresnel numbers and accounting for gain medium nonhomogeneity and instability phenomena, as those arising in discharge pumped lasers.

This paper describes a simple intrinsic optical fiber current sensor, based on the Faraday effect, which has been specially designed for use in a high average power discharge-pumped excimer laser. The sensor built at Salford has a design bandwidth of 200 MHz, and has been demonstrated in action on the pulse-forming line of an excimer laser pulse power circuit. Its performance is compared with that of a low-inductance current-viewing resistor in the same circuit.

Acoustic phenomena may play a leading role in high average power excimer laser as soon as a high pulse repetition rate is achieved. In this paper we demonstrate both experimentally and theoretically that acoustic wave generation and evolution are strongly dependent on electric arcs appearing at the end of the discharge. These acoustic waves induce a strong decrease of laser power throughout an amplification phenomenon of pressure fluctuations at high PRF. Acoustic dampers allow us to reduce this phenomenon and to increase the laser output power up to 50%.

The requirements for obtaining extended discharge stability in an excimer laser system are summarized. We present a system working in auto prepulse that despite an intrinsic simplicity obtains a long pulse duration and a relatively high overall efficiency. The performances of the scheme have been tested on a small scale prototype. In these conditions computer modelling of the main electrical and optical features shows a good agreement with the experimental data, offering perspectives for scalability in energy and pulse duration.

Experimental investigations were carried out with Ar:Xe mixtures at 1 atm and with a wavelength of 1.7 micrometers using an existing CO₂ pulsed laser system without special adaptation. Laser pulse energies E_L <EQ 0.31 J (E_L^* equals 0.3 J/l(DOT)atm) were measured at efficiencies of (eta) <EQ 0.4% for various Ar:Xe mixtures. The electrical pulse energy supplied was E_d <EQ 70 J (E_d^* <EQ 5.83 J/l(DOT)atm) at a pulse duration of t_p equals 10 microsecond(s) . This limitation was a consequence of the mismatch between the electrical discharge modulator (PFN) with its impedance of 2 (Omega) and the plasma impedance of <EQ 0.2 (Omega) . The normalized field strength E/n equals 7.5(DOT)10^-18 Vcm² was determined for a power density of 1.16 kW/cm³. An improvised reduction of the PFN impedance to 0.20 (Omega) led to a c. 70% increase of the specific electrical energy supplied E_d^* and to a c. 35% increase of the laser pulse energy E_L. Typical characteristics such as discharge voltage and current over time, laser pulse power and plasma impedance are presented below. The specific data are given for scaling to a repetitive 1 kW laser design.

The Einstein coefficients and the stimulated emission cross sections have been calculated for far IR transitions of rare gas halides. The formation of inverse population has been predicted for certain vibrational-rotational levels of excimer molecules (EMs). The conclusion has been made about amplification of the IR radiation corresponding to the transitions between these levels.

The characteristics and output performances of a long pulse, e-beam driven, large volume KrF laser are presented. This laser, designed and realized at the IMFM, delivers pulses of 400 ns duration with maximum energy of about 200 J. It is now used in laser matter-interaction experiments at (lambda) equals 248 nm.

Short and continuously tunable (2 ns - 100 ns) UV laser pulsewidths have been obtained for the first time by applying the intracavity injection technique to a long pulse, x-ray preionized XeCl laser with remarkable stability and reproducibility characteristics.

Efficient wavelength shifting by means of stimulated Raman scattering (SRS) in hydrogen and liquid nitrogen of the blue-green XeF(C yields A) excimer laser was demonstrated. Energy conversion into the first Stokes line with an efficiency of 38% was achieved. Continuously tunable radiation from 523 - 579 nm and from 578 - 650 nm with pulse energies ranging from 100 mJ to 210 mJ was generated in liquid nitrogen and hydrogen, respectively. A peak power of 35 MW at 549.0 nm was obtained.

We present experimental results on the performance of a novel two-stage KrF excimer laser designed and constructed to drive a laser plasma XUV source, an application which requires high pulse energy and high beam quality in order to produce XUV radiation with high efficiency. The system consists of two high power laser modules that can be configured as power oscillator-power amplifier (popa), or as an injection seeded system. In the first case the pulse energy and beam divergence amount to 1.5 J and 0.15 X 0.12 mrad (FWHM), respectively, in the second case 0.95 J and 0.07 X 0.13 mrad. At 50 Hz the power amounts to 67 and 42 W. Progress in excimer laser engineering has led to many improvements that give more reliability at higher pulse energies and higher average power, while maintaining beam properties essential for the application under study. The aim of this work is to develop the laser plasma for x-ray lithography, i.e., to construct a prototype source with a high XUV flux as an alternative to electron storage rings.

An overview of the COIL history and corresponding technical milestones is given and major scientific contributions of laboratories all over the world are discussed, briefly. This paper reviews basic COIL technology for the subsonic and supersonic system including in particular new generator concepts. It highlights important achievements and analyzes the remaining problems which are primarily related to still open fundamental questions in the area of liquid and gas phase kinetics. It also attempts an outlook of the forthcoming technology. Finally, the COIL potential for future applications is addressed with special emphasis on the unique high power, high brightness performance capabilities of this laser.

The basic technology and performance of the chemically pumped oxygen-iodine laser is reviewed. The performance is discussed in terms of the operation of the chemical oxygen generator, the kinetics of energy transfer from oxygen to iodine, and the extraction of power by the optical resonator. Techniques for generation of excited oxygen and iodine are reviewed. In addition advanced concepts for switching the laser on and off, switching the polarization, and frequency shifting are discussed.

Rotational and vibrational populations in different electronic states of I₂ have been determined during the dissociation of I₂ by metastable oxygen in an oxygen + iodine flame. From these results, we propose a multi-step dissociation process which explains measured consumption rates of O₂(¹(Delta) _g in the dissociation reaction.

The injection and mixing of iodine vapor into a duct flow has been investigated experimentally and numerically. Laser-induced fluorescence has been used to visualize the progress of spreading of the iodine and its distribution homogeneity along the duct. Several injection methods, typical flow pressures for oxygen-iodine lasers, and various molar flux ratios have been examined to find the most favorable conditions for the mixing in supersonic lasers. A one-dimensional gasdynamic and kinetic computer model taking into account a zone of bulk type mixing with variable extension was used for estimating the influence of the mixing speed on the rate of dissociation and on the performance parameters of a supersonic oxygen-iodine laser.

A theoretical model for the amplifier of a pulsed chemical oxygen-iodine laser is presented. The calculation shows that the main point for obtaining high energy extraction efficiency is to let the oxygen-iodine laser beam emitted from an oscillator (1) have short pulse duration and high repetition rate, and (2) get multiple pass by mirrors through the excited medium within the oxygen-iodine amplifier.

A new method for small-signal gain measurements which uses a distributed feedback laser diode (DFB laser diode) as a probe laser has been developed. In this method, as the wavelength of the probe laser can be changed continuously, not only the small-signal gain but also the gain profile can be measured. Small-signal gain with the gain profile of a chemical oxygen iodine laser was measured by applying this method. Moreover, from the line width of the observed gain profile, the temperature of the laser medium was estimated.

Recently, modulation of the output from a chemical oxygen iodine laser (COIL) has been demonstrated by applying a pulsed magnetic field directly to the laser cavity. A peak-to- average power enhancement of 3.2 was obtained. The rate of change of the magnetic field in these experiments, however, was relatively slow, 0.5 gauss/microsecond(s) . In this article, we report the demonstration of a fast field switching technique (325 gauss/microsecond(s) ), in which switching occurs faster than the laser mode can build up. With this technique, we have achieved a peak- to-average power enhancement of 12.7 and achieved true gain-switched operation.

A simple, one-dimensional leak flow tube model was used to calculate the effect of mixing on the performance of chemical oxygen-iodine lasers (COIL). Both the maximum gain and the characteristic length of the iodine dissociation are shown to be nonmonotonic functions of the iodine flow rate, nI₂. The maximum nI₂ for which lasing is possible is less than 1 - 2% of the oxygen flow rate. This is in agreement with experimental data and is not explained by models assuming premixed flows. The present model was applied to calculations of the performance of supersonic COILs.

A theoretical model is developed for chemical generators producing O₂(¹(Delta) ) at high pressure. Such generators are especially important for supersonic chemical oxygen-iodine lasers. The model treats different types of generators, e.g., bubble column, film, aerosol, and jet generators. The main factor affecting the O₂(¹(Delta) ) yield under high pressure is liquid-phase quenching enhanced by depletion of HO₂^- ions near the gas/liquid interface. Simple analytical expressions are derived for the O₂(¹(Delta) ) yield at the exit of the generator. Output characteristics of different specific generators are calculated and compared with available experimental results. O₂(¹(Delta) ) yield > 0.5 can be achieved for oxygen pressure up to 50 Torr and flowrates of 3 mmol/cm²s. For equal velocities of the gas and the liquid the maximum flux of the energy carried by O₂(¹(Delta) ) for jet or aerosol generators reaches 200 W/cm². It can be increased by increasing the liquid velocity in the generator.

A magnetic quenching of generation in a chemical oxygen-iodine laser (COIL) has been studied experimentally. This work gives experimental data on a quenching threshold magnetic field in dependence on a resonator output coupling and an iodine concentration in the laser active zone, respectively.

Simple experimental criterion for finding the pyrolysis limit in optical pumping of the iodine photodissociation laser was suggested and tested for different experimental conditions. This method enables us to determine this limit for the considered laser system without knowledge of the absolute value of the absorbed UV pumping energy and of the laser mixture heat capacity.

A high power chemical oxygen iodine laser (COIL) has been developed at ONERA. A numerical model and numerous optical diagnostics are used for optimization. A maximum laser output of 600 Watts has been achieved. Laser power and efficiency are limited by an excessive water vapor content. A new water vapor trap has been successfully tested. A laser power of at least one kilowatt is expected.

The feasibility of a 100 kJ class chemically pumped iodine laser amplifier is studied by numerical calculations of the Maxwell-Bloch equations. The idea is based on a chemical iodine pulse laser utilizing a porous-pipe, high-pressure singlet oxygen generator. It is shown that at the high singlet oxygen pressure conditions, the energy transfer reaction from singlet oxygen to ground state iodine atom directly deposits energy on the pulse amplified. Therefore, it is possible to operate an amplifier with low iodine concentration, which ensures the slow deactivation of the stored energy sufficient to fill up a cell of 80 cm in diameter.

Intracavity second harmonic generation of chemical oxygen iodine laser utilizing LiB₃O₅ crystal has been studied. A chemical oxygen iodine laser of which the fundamental maximum output power is 3 W in TEM₀₀ mode with the Cl₂ flow rate of 300 mmol/min is used. Obtained total second harmonic power is a maximum of about 6 W. Therefore, we could estimate that the effective extraction efficiency is 200%. It is thought that the latter is above 100% due to the condition that the output coupling for the fundamental beam is not optimum. And applying the result of a fundamental laser power measurement, the internal conversion efficiency is estimated at 0.29%.

In order to better understand the strong chemiluminescence occurring during the chemical reaction between Cl₂, O₂(¹(Delta) _g) and Cu at 400 degree(s)C, we have investigated the electronic states of CuCl₂, recording Fourier transform spectra and measuring lifetimes of the ²(Pi) _u, ²(Delta) _g states. Implications on the likelihood of obtaining a chemical laser from this system are evoked.

Laser surface texturing of polymer films is of interest for the improvement of the tribological and adhesion properties of these films. The absorption of a UV laser pulse of high energy causes ablation and a simultaneous surface roughening may appear. For pulse energies approaching the ablation threshold semi-crystalline polymers are superficially amorphized over a depth that is controlled by the experimental conditions. In a new regime of irradiation that uses a polarized excimer laser beam of low fluence periodic modifications of submicron size aligned parallel to the electric field are formed on nominally smooth surfaces. These self- developing structures result from the interference of the incident and surface scattered waves. Various applications are anticipated owing to the high dose efficiency and controllability of the process.

In the early sixties, investigations in laser welding were started. Since then, parallel to the experimental work, physical models have been developed in order to get an enhanced understanding of the process. Especially since the time of industrial use, the growing interest in laser welding has led to an increasing number of models. In the first years, the models just treated the calculation of heat conduction. In those models solutions of the heat conduction equation, which had been found in the forties and fifties, were applied to laser welding. Some of these solutions had been found treating conventional and electron-beam welding. Duley was the first who put these models together and arranged them thematically in his book published in 1976. Some further review articles appeared later on with emphasis on different topics and with different structures. This article again is structured thematically in order to form a picture of the process -- how it is seen today.

The authors examine a contact free, optical machining process by illuminating ceramic substrates like alumina, silicon nitride, and silicon carbide with an excimer laser. Using the effect of sealing the surface by simultaneously smoothing and reducing of surface defects it is possible to generate a homogeneous, remelted surface layer. The influence of purity and pre- treatment on the wear-resistance against mechanical and dynamical stresses is reported. It can be shown that the applied laser parameters are strongly correlated to the results of the mechanical tests and the studies of the laser produced plasma concerning plasma formation, plasma propagation, and the local resolved spectroscopic measurements.

On the basis of previous experience in the field of theoretical and numerical characterization of the laser-plasma interaction at very high intensities, a computational scheme has been set up for the analysis of these processes and applied to the characterization of some of the most representative among them, with special emphasis in providing results directly suitable for experimental model validation and for use both in a predictive way for the a priori process design and with process control purposes.

Because of their specific properties, i.e., emission of UV-radiation and pulse length in the range of several 10 ns which results in high intensities, excimer-lasers enable surface modification and micro processing of brittle and thermoshock sensitive materials. Surface treatment near the removal threshold generates a planarization or an increasing roughness, depending on the fluence. The generation of thin conductive layers is possible when radiating AlN-ceramics, with higher fluence and adapted feed rate. The influence of process parameters on the resistance is pointed out in the first part of this paper. With increasing fluence, a homogeneous removal with minimum dimensions of a few microns and high contour sharpness can be achieved. The gas ambience of the process can avoid or induce chemical reactions. The second part of this paper deals with the effect of process gas on the removal of ceramic materials. Using Si₃N₄ as an example, the influence of reactive gas on the removal geometry and the plasma formation is represented.

Laser impulse space propulsion (LISP) has become an attractive concept, due to recent advances in gas laser technology, high-speed segmented mirrors, and improved coefficients for momentum coupling to targets in pulsed laser ablation. There are numerous specialized applications of the basic concept to space science -- ranging from far-future and high capital cost to the immediate and inexpensive, such as: LEO-LISP (launch of massive objects into low-Earth-orbit at dramatically improved cost-per-kg relative to present practice); LEGO-LISP (LEO to geosynchronous transfers); LO-LISP (periodic re-boost of decaying LEO orbits); and LISK (geosynchronous satellite station-keeping). It is unlikely that one type of laser will be best for all scenarios. In this paper, we discuss these most immediate applications, leaving LEO-LISP -- the application requiring the longest reach -- for another venue.

The two high-power industrial lasers (CO₂ and YAG) cannot be focussed on a very small spot. More efficient or new laser processings can be developed by using other high-power lasers, for instance CO, chemical HF, CO or HF overtone, oxygen-iodine, or copper vapor and dye, the beams of which can be focussed on a smaller spot. As a consequence of the higher absorptivity of the metals, the better transparency of the plasma and the higher power densities available, the processing efficiency increases whereas the processing width drops. A whole array of 1 kW power range, near-IR cw lasers as well as repetitively pulsed visible lasers is already (or will be soon) available at ISL.

The employment of laser systems for completely automated welding processes needs an analysis of physical mechanisms by means of mathematical models to correlate the characteristic quality indexes of welding to working parameters. In this paper we describe the application of neural networks methods in order to correlate, for different kinds of steel, the form factor and the penetration depth to working parameters and chemical and thermal properties of working materials. The results are quite satisfactory even a direct search of the optimal process conditions is not possible.

It's well known now that, under peculiar conditions, the laser-matter coupling is highly dependent on the plasma formation in front of the irradiated sample. The experiments with results presented here have been carried out over a range of intensity large enough to be able to discriminate the main stages of the interaction and particularly the effects of the plasma with the increasing intensity. The first part of this paper is devoted to the experimental conditions, whereas the results are discussed in a second part. Finally, a balance of the interaction is drawn up.

An investigation is made of the melt-pool formed during keyhole welding with a cw CO₂ laser on thin plate mild steel. The aim of the study is to analyze the dynamics of the melt-pool and keyhole in order to provide information on the causes of instabilities found during high speed welding. Such problems found during high speed welding include humping, keyhole failure, and surface tension driven melt-pool instabilities. The effects of varying laser power (2 to 4 kW), traverse speed, shroud gas, gas delivery angle, and plate thickness were studied. The methods used included various high speed camera techniques. Two high speed cameras are used, a high speed video camera at a frame rate of 1,000 frames per second and a high speed gated camera used in conjunction with a frame grabber capable of gate speeds as low as 25 ns and freeze frame multi-imaging. The high speed video system was used to gather information on the gross melt-pool characteristics, e.g., shape, length, width, and any other slow changes present (of the order of 100 Hz). It is hoped by correlating these results with theory that an insight into high speed behavior will be obtained.

This paper reports on a new arrangement for laser materials processing allowing the fast variation of beam parameters in the focus region, e.g., the polarization of the radiation and the diameter and position of the focus spot. Results of CO₂ laser cutting of mild steel with a dynamic polarization are discussed.

In the present paper, CO₂ laser coating of a cobalt base alloy on a 12% chromium steel with a view to applying it to the protection of steam turbine blades against wear was investigated in order to achieve coatings thicker than 1.5 mm and 20 mm wide and applying the process to test-blades for evaluating their abrasion/erosion resistance.

A set of numerical models for the characterization of laser processing applications is developed. The main physical and calculational features of these models are given along with some results on their comparison to experimental data and other well established theoretical models. Special emphasis is made on the suitability of the set of models for applications design and practical implementation.

It is well known, metallic plasmas created during welding with high power lasers are very important for the welding process. During recent years, we have been trying to characterize these kinds of plasmas, in order to get a better comprehension of the absorption of laser light by a steel target during welding with a CO₂ laser. A complete spectroscopic characterization (electron density and temperature) of these types of plasmas was achieved, in spite of the lack of spectroscopic data for metallic lines. Only one iron line was found to be well calibrated. The problem of electron density for iron plasmas has been solved with this line. This is not the case of titan.

In the present work, AlN has been sintered under nitrogen atmosphere using the cold isostatic pressing technique (3000 bar) for its shaping. Yttria has been used as a sintering aid and also samples of pure AlN without any sintering aids have been prepared. The excimer laser treatment of these samples has been investigated. The surface morphology (phases characterization, microcrystalline structures, roughness, surface layer orientation, grain size), the depth of the influenced layer and the quality of the laser treatment (presence of microcracks, sealing of porosity, homogeneity) have been studied in relation to the properties of the samples before the laser treatment. The electrical resistivity of the sample surfaces after laser treatment has also been measured. The proper conditions for excimer laser micromachining of AlN are also investigated and comparisons between two different laser wavelengths (193 nm, 248 nm) used are made.

Excimer lasers prove to be useful tools for precise structuring and uniform surface modification of thin surface layers of ceramics, glasses, and composites. They provide high energy laser pulses and a uniform intensity distribution. Processing quality is strongly affected by laser and process parameters. It is demonstrated that the integration of diagnostics in an excimer laser processing system results in an optimization of processing quality by appropriate choice of the process parameters.

The destruction process of PMMA under high power IR laser irradiation is a boiling process. In cast PMMA, bubbles remaining trapped at the end of the irradiation give a mat aspect to the burn pattern; on extruded PMMA this latter remains clear. This difference of behavior results from the wide difference of viscosity of these two qualities: the cast one is reticulated and its viscosity at the vicinity of vaporization temperature is still too high to be measured, whereas the extruded PMMA is rather fluid at high temperature. High speed video pictures show that a boiling process occurs for the two materials; however, at the end of the irradiation, the fast cooling of a high viscosity material traps the residual bubbles whereas the bubbles formed in the fluid extruded boiling PMMA can escape and the surface appears free of bubbles.

Some results on the experimental and theoretical investigation of high-pressure N₂ assisted CO₂-laser cutting of Al and stainless steel are presented. The quality of the laser cut is improved significantly at the expense of up to 50 percent decrease of the speed with respect to this for O₂ assisted cutting. The striations formed parallel to the surface of the work piece are observed. The influence of the waveguide attenuation of the focused laser beam on the quality of the kerf in depth is discussed.

High carbon steels for dies production need high surface hardness and core toughness. The feasibility of improving surface hardness by laser treatment has been studied, on samples with an improved core toughness, obtained by a former quenching and tempering treatment at higher temperatures. The behavior of three types of steel has been examined. Treated layers of high hardness and martensitic structure have been obtained in low chromium steels. Owing to the difficulties in the dissolution of chromium carbides, sufficient hardening effect cannot be obtained in the 13% chromium high carbon steel.

Laser beam penetrations have been carried out on URANUS 45 duplex stainless steel sheets, varying the beam power as well as the traverse speed. The microstructural results, in terms of bead geometry, microhardness, and austenite content, have been related to the working parameters, with the support of an analytical thermal model of laser beam welding.

Surface morphology of oxygen assisted laser cutting of low carbon steel sheets exhibits characteristic drag lines regularly spaced and more or less pronounced depending on process parameters and depth of observation. Examination of roughness profiles does not correspond directly to the periodicity of the drag lines, especially with cutting speed near to oxidation reaction front speed (about 2 m min^-1). In these conditions the two phenomena (laser cutting proper as opposed to oxygen cutting) overlap and the two types of accidents, superimposed over the cutting surface with different periodicity and amplitude, become confused. In this case the roughness profile can be considered as a sum of two populations with different mean and standard deviation, therefore, the height density probability distribution is bimodal and it can be represented as a mixture of two univariate distributions. In this paper a new algorithm is illustrated in order to establish the two parameters of each unimodal distribution, which is well modeled by a Beta-distribution, and the degree of mixing, useful for the evaluation of the relative influence of the two phenomena previously described.

The laser induced ablative decomposition of nitrocellulose and nylon sensitized with organic dopants (Stilbene 420, Coumarin 120, and Rhodamine 6G) has been studied. Ablation with a low power nitrogen laser is hereby reported for the first time. With the addition of dyes strongly absorbing at 337 nm the photoetching rate of the pure materials can be significantly increased. A two step photochemical mechanism considering the decomposition of the polymers from excited electronic states and the energy transfer process from the dye to the polymer are discussed.

Different materials like polymers, glasses, and ceramics have been machined by photoablation with excimer laser radiation (193 nm, 248 nm, 308 nm). Structuring with micrometer resolution can be achieved for these materials if the right wavelength, fluence, and pulse repetition rate is chosen. At high fluence or intensity the ablation process seems to be superimposed by some kind of plasma sputtering. The diameter of deep holes becomes significantly larger than the spot diameter on the sample surface. On the bottom of the holes a regular microstructure appears for several materials. The development of this structure depends on the fluence and the ambient atmosphere.

A large volume XeCl laser has been equipped with a confocal positive branch unstable resonator (CPBUR), obtaining a near diffraction-limited beam, with a brightness of > 10¹⁴W/(cm²sr). Focusing the radiation, an intensity of about 10¹²W/cm² is reached on a copper target, generating an x-ray emitting plasma in the region of 100 eV, with pulse length up to 100 ns.

This work concerns the lap welding of dissimilar (AISI 304 - 430, AISI 430 - 304, AISI 316 - 430, and AISI 430 - 316) stainless steels and the butt welding of dissimilar (AISI 304 - 316 and AISI 304 - 430) ones using a fast axial flow 2 kW cw CO₂ laser (B.O.C. Laser Ltd.). Two covering gases, He and N₂, are used fed coaxially to the laser beam through a 12 mm outlet diameter nozzle at a flow rate varying from 1.67 to 1.87 X 10-3)m³/s. The influence of the process parameters, such as power level, covering gas, and speed on the quality of the welded joints is examined. In the second part of the work laser lap and butt welds of dissimilar stainless steels are characterized by macro and micro graphic investigations to enable structural evaluations. Fillet weld morphology and quality of the same specimens is examined for the two different covering gases, He and N₂. Moreover, the different structural aspects of the melted zones obtained with laser welds of the same pair of stainless steels, but with the beam impinging first on the austenitic stainless steel (e.g., AISI 304 - 430) and then the ferritic one (e.g., AISI 430 - 304), or vice versa, are also studied and evaluated.

This work concerns a study on the integration of a laser system with a speech synthesis facility. The speech synthesis system uses a random access memory (RAM) and electrical contacts (NO and/or NC), controlled by an electronic circuit provided with a microprocessor, in order to take off the `spoken' information corresponding to the actual failure. Hence the laser system is able to `speak' to the operator and to keep him informed on the process conditions giving him step by step simple ON and OFF instructions (replacing the operator instruction manual) and keeping him informed about the actual state of the technological plants and giving him `the spoken messages' for maintenance with scheduled expiry (replacing the maintenance instruction manual). In other words, the future `speaking laser system' will be able to make a complete auto-diagnosis and to report in real-time the results to the operator.

The breakdown characteristics of a short-distance discharge gap in an atmosphere by TEA- CO₂ laser have been studied to control the lightning artificially. It is efficient to enhance the probability of electrical breakdown induced if the focal point is set behind the discharge gap axis or near the negative high-voltage electrode. The length of the optical-breakdown plasma channel is elongated by using the micro-particles diffused in an atmosphere. Using 3 micrometers -diam. aluminum particles, the optical-breakdown threshold is lowered to 15 MW/cm² compared to 0.5 GW/cm² in the absence of the micro-particles.

We present in this work a kinetic model which describes the evolution of charged particles produced by high-intensity, ultrashort laser pulses in nonresonant multiphoton ionization (MPI). The numerically predicted waveforms of the ion detector signal are compared to those observed in nonresonant MPI of Xe with 30 psec, 10¹³ W/cm², 1.064 micrometers laser pulses. The comparison highlights the relevance of space charge effects in typical nonresonant MPI conditions, and shows the effectiveness of our model in accounting for them.

A large amount of metals and alloys are successfully penetration welded by pulsed or cw laser. The laser welding process requires a gas flow (covering gas) to ensure adequate protection of the melt against atmospheric oxidation. The gas can be supplied in a variety of ways but in many cases the coaxial gas-laser geometry is used. This work is concerned with technological and structural investigations (using SEM and A.E.S. microanalysis) aimed at identifying any particular differences in the welds obtained using N₂ instead of He as the covering gas. The investigations conducted on INCONEL 600 and stainless steel appear to provide conclusive evidence that N₂ may be readily used as an alternative to He. Very recent works have studied some peculiarities (blanketing, transmission, and process efficiency) of N₂ as a covering gas, when used in 2 kW CO₂ laser welding. The influence of the covering gas in welding two stainless steels (AISI 304 and 430) and of a nickel alloy (INCONEL 600) using a 2 kW CO₂ laser is examined.

This study was employed to examine the interaction of a commercially used magnesium alloy AZ31B with an impulsional laser beam at 248 nm. The surface structure of the irradiated alloy was examined and the roughness factor was found to depend on the fluence of the incident pulse. The corrosion properties of the laser treated specimen were evaluated in 0.5 M NaCl and have been compared with that of the untreated alloy.

The irradiation of pure aluminum with a KrF excimer laser revealed that its surface morphology and roughness can be affected. The surface structure of laser treated aluminum was also found to be affected by the laser irradiation procedure. In addition, depending upon the lasing conditions, the corrosion resistance of aluminum in NaCl solution can be improved.

Emission spectroscopy studyworks permitted us to characterize plasma produced by a Q- switched neodymium-glass laser on a space and time scale (1.054 micrometers , 27 ns,), for incident power densities ranging between 2 10⁸ and 6 10⁹ W/cm², on solid targets (aluminum, graphite, SiC) under vacuum (ambient pressures of the order of 10^-3 Pa). For these experimental conditions the electronic temperature is a few eV and electronic density is around 10¹⁷ cm^-3.

An experimental study has been performed in order to characterize the metallic materials ablation by means of interaction between impulsional laser beam and matter for 248 nm, 308 nm, 532 nm, and 1.064 micrometers . The ablation is achieved on a static sample (oxidized stainless steel alloy), with one or several successive pulses and for different energy densities. It permits us to determinate the ablation fluence threshold according to each wavelength and the alteration of the surface as a function of the number of pulses. The aim of this application consists in measuring the superficial oxidized layer and quantifying the necessary fluence to ablate a part or all of it. In 248 nm-light, it has been determined that we need 28 J/cm² and 20 successive pulses to remove the totality of the 10 micrometers depth superficial oxidized layer. Other experiments have been realized under the gaseous cover of argon. We hope to avoid oxidation. The efficiency of ablation doesn't seem clearly better than in ambient air. Peripheral oxides redepositions exist as in ambient air.