Accelerators at the frontier of high energy physics (-10⁶ MeV for protons, -10⁵ MeV for electrons) have reached enormous proportions (tens of kilometers). The use of lasers to accelerate particles is attractive because of the extremely large electric fields they can produce (up to 10⁷ MeV/m). If such fields could be effectively coupled to charged particles ultra-high energies could be attained in a short distance, thereby miniaturizing a high energy accelerator. This paper reviews the status of several laser-driven particle acceler-ator schemes. These include (a) the inverse-free-electron-laser (IFEL), (b) laser and free-electron-laser driven microstructures such as minaturized linacs and gratings, and (c) laser-driven plasma space charge waves. Longitudinal accelerating fields of order 1 GeV/m have already been demonstrated in a recent laser accelerator experiment. New roles for lasers in particle acceleration continuously emerge. Examples of these include laser induced final focusing of a very high energy particle beam and laser-driven photocathodes for improved beam quality. Several novel accelerating schemes make use of picosecond laser pulses as fast switches to convert stored energy into high peak electric fields.

Several laser plasma accelerator schemes are reviewed, with emphasis on the Plasma Beat Wave Accelerator (PBWA). Theory indicates that a very high acceleration gradient, of order 1 GeV/m, can exist in the plasma wave driven by the beating lasers. Experimental results obtained on the PBWA experiment at UCLA confirms this. Parameters related to the PBWA as an accelerator system are derived, among them issues concerning the efficiency and the laser power and energy requirements are detailly discussed.

In order to accelerate electrons, large amplitude plasma waves are to be set up. It is shown that using laser wiggler beating instead of beating between two lasers reduces the necessary laser intensity by several orders of magnitude. Saturation mechanisms are investigated. A comparison is made between the two kinds of beating.

The differences in the plasma waves generated either in the laser wake plasmon concept^l or with long laser pulse accelerator concepts are pointed out. An emphasis is also given to the concept that uses the stimulated forward Raman scattering effect by seeding a second wave, such as in the beat wave concept but with a much smaller amplitude. In various cases, stochastic heating is observed. Conditions for an efficient accelerator and considerations on the effective accelerating gradient are also given.

In this paper, the possibility of amplifying two frequencies simultaneously in a free electron laser is discussed. Stability conditions for electron motion are derived, and numerical calculations reported. The advantages offered by using a D.C. Pelletron beam to power such a device, with application to the plasma beatwave accelerator - high efficiency, high average power, radiation wavelength tunability, and the possibility of phase locking several independent optical cavities - are evaluated. Related research efforts, and their impact on the design of this device are discussed.

The requirements on laser power sources for advanced accelerator concepts are formidable. These requirements are driven by the need to deliver 5 TeV particles at luminosities of 1033 -- 1034 cm-2 sec-1. Given that optical power can be transferred efficiently to the particles these accelerator parameters translate into single pulse laser output energies of several kilojoules and rep rates of 1-10 kHz. The average laser output power is then 10-20 MW. Larger average powers will be needed if efficient transfer proves not to be possible. A laser plant of this magnitude underscores the importance of high wall plug efficiency and reasonable cost in $/Watt. The interface between the laser output pulse format and the accelerator structure is another area that drives the laser requirements. Laser accelerators break up into two general architectures depending on the strength of the laser coupling (ratio of particle accelerating field to the laser optical electric field). For strong coupling mechanism (beat wave and grating linac), the architecture requires many "small" lasers powering the accelerator in a staged arrangement. For the weak coupling mechanisms (inverse free electron laser and inverse Cherenkov), the architecture must feature a single large laser system whose power must be transported along the entire accelerator length. Both of these arrangements have demanding optical constraints in terms of phase matching sequential stages, beam combining arrays of laser outputs and optimizing coupling of laser power in a single accelerating stage.

At the Lawrence Livermore National Laboratory, a pulsed electromagnetic wiggler has been coupled to the Experimental Test Accelerator forming the Electron Laser Facility. This is a single-pass Free-Electron Laser which, because the wiggler excitation can be varied, can operate over a wide frequency range. Efficiency of conversion of electron beam to micro-wave power is 7%. This new power source is being used in a collateral program for deve-loping and testing structures suitable for a Two-Beam Accelerator. The Two-Beam Accelerator shows much promise for achieving the high average accelerating gradients, e.g., >250 MV/m, required in such next-generation electron accelerators as 1 TeV on 1 TeV linear colliders. In this paper, the results of recent tests characterizing the Free-Electron Laser are summarized. Also, progress in the fabrication and testing of key Two-Beam Accelerator hardware components are reviewed. Finally, Two-Beam Accelerator problem areas are discussed.

We find Mg illuminated in vacuum with picosecond UV pulses to be a robust and useful photocathode even at roughing pressures. Photocurrent densities as great as 180 A/cm2 occur at moderate collecting field strengths E of 0.5 to 2 kV/cm. In pulses briefer than the electron transit time across a diode, space charge causes the collected charge to increase linearly with anode voltage. The observed Mg quantum efficiency of 2.5 x 10^-6 electrons per 300 nm photon was doubled by "red assist" when intense 600 nm pulses were simultaneously present. We give a criterion for E_min for complete charge collection from trains of picosecond optical pulses.

In Inverse Cherenkov laser accelerators, a passive gas is used to match the phase velocity of a suitably shaped laser field pattern to the electron velocity. Electron collisions with the gas molecules cause energy loss and have also a beam scattering effect. The former is small at high energies (multi Gev), even for pressures in the one atmosphere range, if the proper gas is used. Elastic collision scattering, however, is important because it can displace the electrons to less favorable positions within the field pattern. Computer simulation trajectory studies were carried out using the average of many electrons started at the same initial conditions and moving through the scattering medium while interacting with the laser fields. Both longitudinal and transverse forces were modelled, and the increase in electron mass with energy was included. The results show there is a transverse focusing force due to the laser fields, very effective for confinement over a wide range of laser phase angles within the acceleration region, so a small electron beam cross-section can be maintained even near the maximum acceleration condition and over extended distances. Individual electron trajectories display the transverse oscillation behavior expected from an approximately linear restoring force combined with random angular perturbations. The acceleration gradient differential caused by the transverse field variation is thus smoothed over the oscillation period. The above data confirm the suitability of this method of acceleration for the production of high-energy electron beams and give quantitative expression to the resulting beam properties.

Net absorption of a laser light by relativistic electrons through net inverse bremsstrahlung in a longitudinal electrostatic wave was previously found to give rise to a very large net dc force. The predicted incident energy dependence of electron acceleration by co-propagating laser waves and the weaker longitudinal electrostatic waves is in qualitative agreement with the preliminary results of a recent Canadian laser-plasma electron acceleration experiment in which phase matching was not attempted. Furthermore, an analysis of the preliminary data indicates the existence of a non-Lorentzian force, which is the only force responsible for the net electron acceleration.

The experimental results on multicharged ion formation at nonlinear atom ionization are analyzed. A presence of data referring to a collective character of interaction of the atomic electrons with radiation field is clarified.

The energy spectrum of electrons produced in the multiphoton ionization of rare gases in the 10¹²-10¹⁵ W.cm^-2 range displays interesting features. Firstly, the energy spectrum does not correspond to a single electron peak as would be expected in an N-photon ionization described by the lowest order perturbation theory. It generally consists of a series of peaks evenly spaced by an amount equal to the photon energy. The number of peaks strongly depends on the laser wavelength. For example with Xe, only one additional peak is observed at short wavelengths, while about ten and even tens of peaks are observed at longer wavelengths such as 1064 nm. These absorption processes can be described in terms of continuum-continuum transitions, or better still, in terms of the absorption of photons by an electron in the field of the ion to which it was originally bound. Furthermore, as soon as the electron-ion pair is formed, the electron acquires a quiver energy A in the presence of the e.m. field. The absorption of additional photons, corresponding to an energy less than A, is made energetically impossible. This leads to the suppression of the first peaks of the electron energy distribution. The disappearance of a certain number of electron peaks strongly depends on laser wavelength and laser intensity. The disappearance of nearly 30 peaks has been observed for He at 1064 nm and 10¹⁵ w.cm^-2. Finally, additional effects, such as electron angular distributions and space charge effects can change the relative amplitude of the first electron peaks. These effects must be taken into account before any comparisons can be made between theory and experiment.

At three-photon barium and strontium atom ionization the intermediate two-photon resonances with the triplet states were discovered. The ion yield maxima corresponding to two-photon singlet-singlet and singlet-triplet resonances have approximately the same amplitude. Large probability of two-photon excitation of the triplet states sharply differed from the case for one-photon excitation. The possible reasons leading to large probabilities of two-photon excitation of the triplet states are discussed.

Multiphoton simple and double ionization of strontium was studied using picosecond laser light and electron spectroscopy techniques, under laser intensities ranging from 10¹¹ to a few 10¹² W.cm^-2. Simple ionization is shown to leave the ion in either the ground state or one of the lower-lying excited states. Two and three-photon resonances can occur in this process on intermediate states lying either below or above the first ionization limit. Double ionization is shown to be essentially a stepwise process.

Multiphoton ionisation of Mg was studied with an excimer laser (XeCl, X=308nm). The laser intensity dependence of the number of ions indicates that the lowest order two photon process that have enough energy to reach the first ionisation limit, failed to dominate the ionisation of magnesium for some unknown reason. A 3 photon resonance with some autoionising states was proposed to explain this anormal phenomena.

Both experiments and quantum calculations have revealed some validity of classical predictions for the classically stochastic or "diffusive" many-photon excitation and ionization of hydrogen atoms with principal quantum number n near 60 in an intense microwave field. These processes partially arise from the relatively dense number of excited atomic states present near the continuum, and from the large radiative couplings between these states. The classical scaling relations are supplemented to predict approximate threshold conditions for "diffusion" and ionization. Diffusion occurs when an electron orbit changes during one orbit time, while ionization requires adequate energy absorption from the the external field. Indicated are interesting experiments at ten micron wavelengths, intensities of tens of gigawatts per square cm, and bound electrons such as in the helium ion with n=6. These values are similar to those predicted for n=1 electrons in quantum well structures, where the situation is somewhat different because of the inverted anharmonicity of the ladders of field-free atom energy levels.

The cross-sections for free-free processes in electron-argon scattering in the presence of a pulsed CO2 laser have been measured as a function of laser flux. This represents the first attempt at an experimental investigation of the theoretical predictions contained in the Kroll-Watson equation. The dependence of the cross-sections on the various experimental parameters is discussed and we stress the advantages of using moderate laser powers (10⁷W/cm ²) to carry out these tests of the fundamental interactions between free electrons and photons in the presence of an atomic potential.

The expressions for tunnel ionization probability of the complex atoms and atomic ions from arbitrary states in electromagnetic field are obtained. These expressions correctly describe the experimental data on rare gases atoms ionization in an infrared electromagnetic field.

Multimillion-degree plasmas are bright (intense, small, short-pulsed) sources of x-radiation. They can be heated by high-power electrical discharges or lasers. Plasmas generated by absorption of sub-microsecond laser pulses are intensely studied for scientific and technological reasons.

Continuum and line x-ray radiations are efficiently generated in high density plasmas produced by focusing high intensity laser radiation on solid targets. The short pulse x-ray that we obtain from the laser-produced plasmas are very useful not only in ICF research but also in practical applications such as x-ray lithography and x-ray microscopy. We discuss on the generation mechanism of x-ray emission in laser-produced plasmas. Experimental results concerning the x-ray conversion efficiency at various experimental conditions are presented.

A range of experimental techniques and applications is presented, for laser-produced plasma X-ray sources generated by the two high power laser systems at the SERC Rutherford Appleton Laboratory, U.K. Using continuum sources, EXAFS spectra have been recorded in both time-integrated and time-resolved modes, and a reflection geometry has been used to record grazing incidence reflection EXAFS (reflEXAFS) spectra. High intensity line emission sources have also been used for time resolved X-ray diffraction studies, and for high resolution X-ray contact imaging of fresh unfixed biological material.

Absorption spectra and photoionization cross-section measurements of multiply charged ions have been performed by using two laser-produced plasmas, one acting as background continuum radiation source the other as absorbing medium. The basic principles of the experiment are discussed and results for Bell, III, IV ions, in the grazing incidence spectral region, are reported. Measurements of photoionization cross-section are derived by extrapolating to the continuum the known oscillator strength of suitably chosen optically thin lines.

Laser produced X-ray spectra are studied both theoretically and experimentally in the range 0.1-5 keV for laser wavelengths λ = 0.26 μm and λ = 0.53 μm. The X-ray conversion efficiencies in various spectral ranges are presented. It is possible to obtain from a laser plasma source a well characterized X-ray spectra. As an example of application, the exposure of P.B.S. resist is studied both in the XUV and soft X-ray ranges.

Laser produced plasmas have been shown to be extremely bright sources of extreme ultra-violet and soft x-ray (XUV) radiation; however, certain practical difficulties have hindered the development of this source as a routinely usable laboratory device. To explore solutions to these difficulties, Sandia has constructed an XUV laser plasma source (LASPS) with the intention of developing an instrument that can be used for experiments requiring intense XUV radiation from 50-300 eV. The driving laser for this source is a KrF excimer with a wavelength of 248 nm, divergence of 200 jtrad, pulse width of 23 ns at 20 Hz and typical pulH energy of 500 mJ which allows for good energy coupling to the plasma at mgderate (10¹² W/cm²) power densities. This source has been pulsed approximately 2 x 10⁵ times, demonstrating good tolerance to plasma debris. The source radiates from the visible to well above 1000 eV, however, to date attention has been concentrated on the 50-300 eV region. In this paper, spectral data and plasma images for both stainless steel and gold targets are presented with the gold target yielding a 200 μ, plasma and reradiating 3.9% of the pump energy into 15-73 eV band, a flux of 1.22 x 10 photons/pulse/eV into 2n sr. Further efforts will expand these measurements to rare earth targets and to higher spectralenergies. A special high throughput wide angle XUV (50-300 eV) monochromator and associated optics is being concurrently developed to collect the plasma radiation, perform energy dispersion and focus the radiation onto the experimental area.

The operation characteristics of a picosecond high power KrF laser are reported with an emphasis on picosecond pulse generation at the KrF wavelength, wide aperture discharge amplifiers, a 200J-class main amplifier, the synchronism among all laser devices and the optical system.

A subpicosecond KrF* laser system with a nearly diffration-limited beam and peak power on the order of 50 GW (capable of focusing to > 10¹⁷ W/cm') has been developed. The means of producing ultrashort seed pulses for the KrF* amplifier system are described. It is shown that efficient energy extraction is possible on a subpicosecond time scale and that deleterious effects such as optical damage and self-focusing do not limit the scaling of such a device to producing terawatt pulses in the 1 J energy range.

A laser system that produces intense subpicosecond pulses of 248 nm light is under development. Ultrashort pulses are generated in the visible in a synchronously-pumped mode-locked dye oscillator, heterodyned into the uv by two KDP crystals, and amplified in a chain of KrF* amplifiers. Front end output of 5 pJ is amplified to 20 mJ and focused to peak intensities of order 10¹⁷ W cm-2. Additanal amplification is expected to permit experiments at intensities >10²⁰ W cm^-2.

The parameters important for high energy amplification of picosecond pulses are discussed. A saturation energy greater than 400 mJ/cm² is obtained. We also describe a laser system which produces picosecond (T > 2.5 ps) pulses with a peak energy of up to 250 ITO.

In intense fields, the Rabi frequency ^ω_R = pE/h can easily be of the order of rotational and vibrational energies of molecules. This means that rotations as well as vibrations become strongly perturbed so that perturbative methods no longer apply. We will show that nonperturbative methods can be derived from the concept of the dressed molecule. This leads to coupled equations which are used ko simulate numerically the multiphoton processes which will occur at intensities > 10⁸ W/cm². Furthermore, for multiphoton rotational tran-sitions, one can derive analytical models which help one understand the temporal behaviour of energy flow in a molecule in terms of its dressed spectrum, such as chaotic or regular (nonchaotic) behaviour. These results are of relevance to the manifestation of multiphoton coherences in a CO₂ spectrum at very high intensities (I % 10¹² W/cm²).

In recent years a strong interest has developed in the study of the basic coupling mechanisms governing high order multiphoton processes in atomic systems.¹ Such interest has been stimulated by the rapid advances in short pulse generation techniques which allow atomic interactions at an intense field strength of the order of atomic field to be studied. Furthermore, good understanding of the highly nonlinear photon interactions is necessary to evaluate the possibility of using multiphoton processes² as a means to couple energy into excited states emitting in the x-ray region.

Collisionless ionization of five of the rare gases has been observed using a CO₂ laser, at intensities in the range 10¹³ - 10¹⁴ W/cm², with a pulse duration of 1 ns. Calculations taking into account the space - and time - dependence of the laser pulse show that the ionization rate is a highly nonlinear function of intensity. It is shown that the nonlinearity is so strong that an "ON-OFF" hypothesis, stating that ionization is saturated as soon as it begins to appear, is sufficient to reproduce the experimental results.

Four- and five-photon ionization of krypton has been investigated in the autoionizing region between the P_1/2 and P_3/2 ionization limits with a high-powered pulsed tunable dye laser and a static ionization cell. For the four-photon case, the np' and nf' Rydberg series leading to the P1/2 limit were observed for n values 10-14 (p') and 7-18 (f'). The measured term values are compared with available earlier results. Quantum defects and ionization limits have been obtained. No new atomic structure was observed in the five-photon scans in krypton or in limited three-photon scans in xenon.

The influence of dc electric field on three-photon ionization of magnesium atoms via two-photon resonance with bond electron state is investigated experimentally.

Due to a progress in development of high power pulsed lasers the laboratory available intensities of electromagnetic wave reaching about IO¹⁸ W/cm² for neodymium and about IO¹⁶W/cm² for CO₂-lasers. This provides a possibility of experimental study of some effects in strong electromagnetic field. The interesting one is the electron-positron pair production by electromagnetic wave. Here a theory predicts two possibilities:(1) direct vacuum pair production /1,2/ and (2) indirect pair production in plasma by relativistic electrons, accelerated in electromagnetic field /3,4/. The calculations of the probability of vacuum pair production /I,2/ indicates that production of a detectable number of pair from this process will require the intensities (of the order of IO³⁰W/cm²) that many orders of magnitude greater than the contemporary lasers provide. The later effect, however, is expected to require only quite realistic intensity. In this paper we discuss the possibility of experimental observation of the pair production in plasma by electrons accelerated by focused laser beam, and suggest the simple method allowing for a small number of positron to detect in the presence of strong background from laser plasma.

A survey of research devoted to the achievement of a lasing action in the soft X-ray wavelength range is presented. General problems of laser produced plasmas considered as amplifying media are briefly described. The question of superradiance and of resonator for X-rays is considered. The most important mechanisms proposed to produce population inversions for XUV and some recent experimental results are described.

General conditions for generating high gain by expansion cooling will be discussed. Theoretical and experimental studies of the carbon fibre approach show good agreement. The development of scaling laws for this simple system allows optimisation and flexibility to be identified in the operating conditions. Seeding and coating the fibre with appropriate elements introduces further flexibility by additional radiation cooling. Fibre doping with elements from nitrogen to neon extends laser action from 182Å to 61Å. Shorter wavelength operation at 39Å may be generated from aluminium based targets.

This paper reports time-resolved measurements of gain at 105.7 Å in an aluminum plasma produced by laser. It is shown that the amplification occurs during the plasma cooling, at the end of the laser pulse. The observed time-variation of the gain agrees with the prediction of a polasma recombination model used for calculating the level populations of the lithium-like Al¹⁰⁺ ions.

Recombination and collisionally excited x-ray laser schemes in hot dense plasmas require ideally the existence of gradient-free uniform conditions for times long enough to generate high x-ray yield. The potential for imploding cylindrical plasmas to circumvent refraction effects associated with long plasmas generated from thin foils is discussed.

We will describe our optical laser pumped XUV Laser Program. To date, we have concentrated our efforts on exploding foil amplifier designs using Ne-like n=3p to 3s inversion schemes. We will describe our latest modeling results as well as measurements which demonstrate output power near the 1 MW level at 206 and 209 Å and lasing at wavelengths as short as 106 Å.

A series of exploratory experiments recently carried out at NRL were directed towards the possible utilization of laser-heated ultra-thin copper films for generating elongated plasmas of sufficient uniformity to produce amplification on 3s-3p transitions in neon-like copper ions. The NRL Pharos III laser operating at 140-335 J in 2-6 ns pulses was used as a driver. Variations were made in the plasma length, the laser energy and pulse shape, and the copper thickness in order to optimize the gain-medium conditions. A primary necessity was to assure on each attempt an accurately-aligned vuv grazing-incidence spectrograph. Using space-resolved x-ray crystal spectroscopy and pinhole photography as auxiliary diagnostics, axial homogeneity as well as front/rear symmetry were measured. The electron temperature in the Cu XX plasma, as estimated from intensity ratios of 2p-nd transition x-ray lines, was found to increase with copper thickness. Various explanations for the lack of measureable gain in these inital tests are discussed. A novel slotted copper foil (thicker) target design was also tested and showed similar characteristics to the thin copper film targets. Also, spectral features from a selenium target exposure are described.

The parameters are studied of active media (AM) of He:Xe ( λ =2.03; 2.65 µm)- and He:Ar ( λ =1.79 µm)-lasers pumped by C0₂-laser induced optical break-down (0B) in the mixture of these gates. The lasing is shown to occur as a result of action of UV radiation from a hot OB plasma kernel on the gas mixture under conditions of plasma shock wave (SW) compression capable of activating AM development.

The theory of electron and positron channeling and radiation production in crystals is extended to macroscopic scale systems. Systems that exhibit periodicity along the axis of beam propagation, specifically, periodic electrostatic fields are examined, where the dimensions of the field periodicity (1_L) are much greater than typical inter-atomic lattice spacing (10, e.g. lz >> 1_L. In addition, the limiting case of the flat plate (e.g. lz + + co) is studied and its relevance to Quasi-Channeling theory analyzed. Effects of beam properties on the proposed phenomenon of Quasi-Channeling are discussed. The feasibility of a laser based on this phenomenon is examined theoretically. Differences between the properties of the pro-posed Quasi-Channeling Radiation and Radiation produced by wigglers (e.g. Free Electron Lasers) are discussed.