In cooperation with the SDIO/IST goals of generating phase-locked microwave and millimeter wave power at the hundreds of megawatt to gigawatt power level, a program has been in existence at the Naval Research Laboratory to generate such sources by enhancing the performance of gyrotron oscillators. The NRL program has emphasized the higher frequency portion of the spectrum, feeling that low frequecy phase-locked sources already exist in, for instance, SLAC klystrons. Accordingly, phase-locked sources are being developed at frequencies between about 10 and 85 GHz.

A self consistent and time dependent particle code has been used to simulate the transient effects of a high current gyrotron. At high beam currents, the output wave exhibits the effects of self-modulated oscillation and period doubling. In the regime of high injected signal powers, the locking bandwidth and time constant of high-power gyrotron phase locking simulations are observed to be respectively larger and smaller than the prediction from a lumped circuit theory. The multimode simulation of the risetime wave generation processes also demonstrates some nonlinear processes such as transient mode co-existence and mode suppression which appear to be of general interest and importance for short-pulse gyrotrons.

We present preliminary results of a longer pulse (200 ns) relativistic magnetron experiment. This experiment demon-strated the excitation of pr. mode besides the usual 27r mode through voltage variation and through magnetic field tuning, and reconfirmed the importance of voltage stability in achieving the mode stability and power output stability required by phase locking conditions. The experiment also indicated the possibility of using velvet as the cathode material for achieving long microwave pulses (200 ns). The long pulse behavior of graphite cathodes with varying cathode locations was found to be similar to the short pulse case. These experimental results aid in the understanding of the high-current magnetron under long pulse operation which is crucial for conducting the long pulse phase locking experiment. We also describe the design and construction of our long pulse, high repetition rate (1 ps, 4 pps) relativistic magnetron system (700 kV, 770 A) with the stability and pulse duration suitable for studying phase locking.

An experiment is currently underway at PI to investigate the phase-locking of relativistic magnetrons. Phase-Locking has been achieved at power levels of - 1/2 GW at 3.8 GHz.

The dispersion relation of the diocotron instability in the relativistic electron flow generated inside a cylindrical magnetron is derived within the framework of a linearized macroscopic fluid model. A massless guiding-center approximation is made for the theoretical calculation to be analytically tractable. In addition, a sharp-boundary density approximation is made, which ensures a strong surface perturbation. The closed algebraic dispersion relation obtained in this article is fully relativistic and fully electromagnetic. Influence of the anode resonator effects are also fully incorporated. The dispersion relation is a transcendental function which can be numerically solved to investigate the diocotron stability properties for a broad range of magnetron system parameters.

The design of GW level plasma Cherenkov masers is discussed. Using linear and particle simulation computer codes, it is shown that extrapolation of the 0.1 GW level Russian PCM to GW levels can result in decreased microwave generation efficiency. Techniques to increase the efficiency are discussed. Higher energy and thinner beams are shown to increase the efficiency. The result is that a high efficiency ( ~10%), GW level PCM is possible.

The Aurora electron-beam generator driving a reflex diode has produced the largest single microwave pulse measured in the laboratory to date (early 1988). In this experiment up to 374 J per pulse is extracted from the device into two radial waveguides. The maximum peak power is 22 GW obtained in a single narrow spike; the peak average power is 9 GW over 20 ns. The power flux through the 0.061-m² area of the two WR975 waveguides is 145 GW/m2, four times the flux once thought to be present in a previous experiment'. In many aspects the experimental results agree with simulation predictions. This paper discusses the recent measurements and indicates directions for further developments with the potential for terawatt peak RF power pulses.

We have carried out theoretical and experimental study to optimize the efficiency of microwave production of the reditron. In the optimal configuration, we have achieved the production of 3.3 GW of microwave radiation at 10.0% efficiency with a very narrow spectrum centered at 2.15 GHz. This is roughly a factor of 3.5 increase in efficiency and about 3 in bandwidth narrowing over conventional vircators. In additon, we found that the use of cavities can achieved bandwidth narrowing, stability of frequency during repeated operation, improvement of mode selectivity.

Preliminary experimental and theoretical results are reported on priming and phase locking of high power cavity vircators. Experimentally, we examine the selection criteria and performance of rectangular and cylindrical resonant cavity vircators. Theoretically, we conduct phenomenological modeling of a magnetron driven cavity vircator. In addition, we are undertaking basic vircator instability analysis to determine the linear growth rate and non-linear saturation physics.

Preliminary results are reported on the design and performance of a high power travelling wave tube amplifier. The amplifier is designed to work in the TMoi mode of a cylindrical waveguide and to operate in X band. The signal from a 250 kW magnetron is used as the input source for the amplifier. Details are presented of the amplifier design, the input coupler, and of the performance of the system when an 800 kV, 1.4 kA, POO nsec. duration pencil electron beam is used as the primary power source. Initial results indicate that the amplifier has a gain of 29dB at a frequency of 9.09 GHz, and an output power in excess of 100 MW.

Relativistic backward wave oscillators (BWOs) have proven to be relatively efficient, high-power microwave sources in the centimeter wavelength range. Recent Soviet results have indicated that the use of a background plasma in such a device can serve to increase the space charge limiting current and allow operation of the BWO at higher injected beam current levels, with accompanying higher microwave power outputs. Studies currently being initiated at the University of Maryland include studies of BWO performance with fill plasmas produced by both electron beam ionization of a low pressure neutral gas background and by an independently controllable plasma source used to inject plasma directly into the BWO structure. Results of the first configuration will be reported and compared with theoretical expectations.

The physics of modulation of an intense relativistic electron beam by an external microwave source and its usefulness for klystron-like devices are studied in this paper via experiment, theory, and simulation. It is found that the self fields of the electron beam, in general, intensify the current modulation produced by the external source. In the modulating gap, electron bunches may be generated instantaneously without the necessity of propagating the beam through a long drift tube. The excellent amplitude stability and the phase locking characteristics ( < 2°) of the modulated current, demonstrated in experiments, open new areas of research in high power microwave generation.

Cusptrons produce microwave radiation at the fundamental and harmonic electron cyclotron frequencies by the negative mass instability. A low-energy, axis-rotating beam of 28-30 kV, 0.8-3.5 A, 4 us, and 60 pps is used to generate a single RF-mode both in a circular cavity and in a six-vane circuit. In fundamental and second harmonic frequency generation with a circular circuit, the independently excited modes are TEll's and TE21's with radiation power of more than 1.8 kW and an electronic efficiency of approximately 7.5%. In high-harmonic frequency generation with a six-vane circuit, microwave radiation at 6.0 GHz (sixth harmonic) and 3.9 GHz (fourth harmonic) is also independently generated with more than 10.4 kW and 4.0 kW, respectively. In this case, corresponding electronic efficiencies are approximately 10.0 and 9.5%.

One of the important problems of strategic missile defense is discrimination between decoys and reentry vehicles (RVs). During midcourse flight, a ground-based HPM system could modify (by a physical damage mechanism) the radar cross section and/or the velocity of decoys (and decoy-camouflaged RVs), characteristics that are easily observable by microwave and optical radars.

New calculations confirm the ability of a random apriori initial-phase proton beam to drive a simple, single stage microwave cavity maser or transit-time oscillator (TTO) to saturation conversion efficiencies of = 11%. The required initial TE011 mode field can be provided from beam ramp-up bandwidth or excitation to a low level from an external source. A saturation field of 45 tesla, output power of .2TW (power density 2.2x10¹³ W cm 2) is calculated using an electron insulation field of 10 tesla and a 3 Mev, 400 Ka/cm² beam. (Addition of a "catcher" cavity can increase overall efficiency to = 15%.) Results are compared to those for an electron beam of the same energy and geometry, and it is shown that proton beams can potentially provide a three order of magnitude increase in over-all microwave power production density over that obtainable from electron beam TTO's.

The power handling capacity of a single planar-grating resonator is evaluated. When driven by an intense relativisitic electron beam, the predicted output levels are in the 0.1-1 GW range. In addition, the planar resonators may be stacked and cross-coupled in such a way that volume scaling will apply , and in this case the total power would be limited by the available beam power.

The magneto-orotron is conceived as a hybrid of the orotron and the uniform magnetic field free-electron lasers. Because of its hybrid nature, this device has four slow-wave structures, of which two are original (one due to the orotron and the other due to the uniform magnetic field) and two are dynamically generated. Here, a semiclassical analysis of radiation associated with original and dynamical slow-wave structures is briefly described; it is aimed at eventually giving a complete description of the device capable of generating radiation in the far infrared/microwave spectral region.

Design procedures and beam transport feasibility experiments are discussed for a novel, millimeter-wave free electron laser (FEL) concept employing short period magnetic undulators and a sheet electron beam. The advantages of this concept include: (1) lower beam voltage ("500 kV) compatible with thermionic Pierce gun technology and conventional power supplies (rather than electron accelerators), (2) a sheet electron beam geometry which enables high power operation without excessively high gun perveance, and (3) a streaming electron beam which is compatible with depressed collector beam energy recovery for enhancing overall system efficiency. Conceptual designs for pulsed and cw high power devices, and a cw low power device are presented.

A high power, 140 GHz, cyclotron autoresonance maser (or CARM) amplifier is under development at M.I.T. Theory, simulations, and the experimental design are discussed. The experiment will employ a high voltage (450-700 kV) Pierce-type beam and a helical wiggler to produce the required elec-tron beam. The initial experiment will be carried out using a 450 kV electron gun; the second phase of the experimental program will employ a 700 kV electron gun. The results of two-dimensional simulations are presented. The CARM nonlinear bandwidth and the sensitivity of the CARM to velocity spread are discussed. The stability of the CARM amplifier to absolute instabilities has also been investigated, and the criteria for stable CARM amplifier operation are presented.

A new space power concept incorporating earth-to-satellite microwave power beams coupled to onboard regenerative electrochemical energy storage is proposed for energizing defensive satellite constellations. The system addresses housekeeping, orbital maneuvering and burst mode power requirements, and offers an attractive alternative to the nuclear and solar space power systems currently envisioned for this application. This energy-conversion system incorporates six steps: (1) generate primary DC power at surface stations along the satellite ground-track, (2) convert to microwave (RF) frequencies, (3) transmit in a narrow beam to spacecraft using phased-array antennas which track and lock-on to satellite receivers as they pass in range during a fraction of their orbit, (4) receive the energy and convert to DC in space using lightweight and inexpensive rectennas; (5) store the energy onboard as chemical energy by electrolysis of water to oxygen and hydrogen and (6) recover free energy onboard the spacecraft during the balance of the orbit continuously or on demand as pulsed power with a high power-density fuel cell. Component and overall systems considerations of this scheme are discussed in comparison with alternatives, outstanding research problems are defined and preliminary analyses are described. These include orbital mechanics and ground tracks of satellites, accessibility of orbiters to microwave beams, transmission efficiencies, electronic and mechanical designs of the transmitter and rectenna, regenerative fuel cell energy storage, power conditioning and thermal management. The development of readily space-deployable rectennae, their supporting structures, and high specific power solid oxide monolithic fuels cells are the main pacing technologies leading to a wholly non-nuclear space power system capable of supporting all defensive satellite power requirements.

The generation of intense, high-brightness H- beams requires solutions of several physics and technological problems in the low-energy beam transport (LEBT) section of the accelerator to avoid particle losses and emittance growth. Existing accelerators and experiments use space-charge neutralization, usually in combination with magnetic lenses, to focus the beam. However, the demands for high beam brightness and intensity require a better understanding and control of gas focusing and associated instabilities or the use of alternate methods of transporting the beam. In this paper we present results of a new theoretical treatment of the charge-neutralized beam-plasma steady state. The second part of the paper presents computer results for a novel alternate focusing scheme using electrostatic quadrupole lenses in which no plasma buildup and charge neutralization occurs. Such quadrupole lenses provide efficient strong focusing in the LFRT line and, more importantly, good control of the beam matching both into the electrostatic column and into the RPO.

Conventional dc sources of H- are limited to current densities of the order of 50 mA/cm2 for sources with area larger than a few cm2. Early work at UCI and more recent work at the Lebedev Institute have shown that pulsed magnetical)sy insulated ion diodes can produce current densities larger by factors of the order 10 - 10'±. The large current density and space charge required considerable development of diagnostics for reliable measurements of current density; the methods involve a biased Faraday cup, Etch pit counting on CR-39 film, and nuclear reactions. The diagnostic development was done mainly with a coaxial diode because most of the previous data, particularly from the Lebedev Institute was obtained for this diode design. Since this diode is .not suitable for applications, several other types of magnetically insulated diodes have been developed and studied. For example, the annular and race-track diodes are suitable for an ipn gun. High current densities of H- are obtained only when the cathode plasma is at suitable density and temperature. In previous experiments the plasma was produced by flashover of a dielectric surface driven by the pulse line prepulse. The H- current density was quite sensitive to the magnitude and duration of the prepulse. The prepulse is a property of the machine design about which there is limited control in existing machines. We therefore developed a flash-board plasma source that is placed inside the cathode shank of the APEX generator and can be controlled independently of the pulse line. This paper will present the most recent results with this plasma source, and comparisons with the previous prepulse results.

A Hamiltonian-theoretic analysis of emittance growth during high current electron beam transport, coupled with an envelope equation analysis of the induced transverse beam oscillations, may be applied to develop scaling laws for the emittance growth suffered during such events as beam acceleration, propagation past irises or constrictions in the waveguide wall, beam aperturing, and axial variation in the magnetic guide field strength. Criteria may be developed for preventing excessive emittance growth by avoiding abrupt axial variations and providing sufficiently strong focusing forces. Designed variations in the waveguide wall shape and in the strength of the magnetic guide field may be introduced to greatly reduce emittance growth during events such as beam acceleration.

In the Plasma Wakefield Accelerator (PWFA) scheme a short bunch of high current (density nb), but low voltage electron beam is shot through a dense plasma. For a properly shaped bunch the background plasma electrons are adiabatically displaced from their initial positions and are subsequently released after the bunch has passed, resulting in a plasma wakefield. A trailing low current bunch can be accelerated by the plasma wakefield to energies on the order of.-1-11ln y0 mc2 when the driving bunch is appropriately shaped. The plasma thus acts like a transformer since it increases voltage at the expense of current. This scheme is thus in the general class of wakefield transformers, except that transformer ratios greater than 2 can be obtained even for axial co-propagating beams. In this paper we win describe recent work at UCLA on the PWFA.

The importance of producing accelerating gradients larger than those achievable with radiofrequency acceleration is well recognized. Large gradients will be necessary for the next generation electron-positron colliders and, in the more moderate energy range, for compact high brightness linacs suitable for applications as the injector for free-electron lasers. Among the alternative acceleration techniques of current interest are the laser-plasma, plasma wake-field and wake-field acceleration process. For those acceleration mechanisms involving plasmas, the periodicity of the structures is in general on the order of sub-millimeters so that accelerating gradients on the order of 1 GV/m is theoretically possible. Wake-field cavities are on the order of a few millimeters, thus will yield lower gradients on the order of 150 MeV/m. This gradient, if achieved empirically in particle acceleration, is sufficient for near term applications. Moreover, it probably is the most practical technique among those that were proposed recently from the point of view of fabrication and ease of phase matching in linac application. A program to investigate the feasibility of wake-field acceleration with an emphasis on the use of ellptical cavities is discussed here. A complete modal analysis was carried out which yields a full description of the wake-field acceleration mechanism. Experimental studies are in progress.

The spiral line recirculating induction accelerator (SLIA) is a compact, lightweight electron accelerator, designed to produce high current (> 1 kA), high energy 50 MeV) and moderate to high quality electron beams with macropulse widths up to several micro-seconds. The SLIA is an open-ended spiral configuration in which the electron beam recirculates in independent transport lines 20) passing through a common ferrite core accelerating section with high gain/pass (- 10 MeV), The concept differs from the spiral line accelerator studied at NBS by Wilson and Leiss in that a toroidal field threads the transport lines for space charge confinement and suppression of instabilities. A strong focussing field (2, = 2 stellarator) is used in the bends to provide tolerance to field errors and an energy bandwidth desirable with high gain/pass. The accelerator can be designed to produce effective accelerating gradients in the 50 MV/m range for compactness and yet be amenable to long pulse trains at high current. A modulated pulse train is applied to the accelerating cells to provide an accelerating field while the beam is present and to reset the cores during the recirculation period of the beam. Reuse of the core material significantly reduces system weight. The modulated pulse traip (tens of MHz) is made feasible by developments in branched magnetic switching by Birx'.. By using the ferromagnetic induction technology developed at LLNL for ATA, the SLIA can provide the necessary input power to rapidly accelerate high current beams using state-of-the-art technology for relatively near-term applications. An experimental (PSI) and theoretical (SAIC) program is underway to investigate key physics issues regarding energy bandwidth, control of emittance growth, and suppression of the growth of collective instabilities. A nominal 1 MeV, 1 kA, 100 ns beam is injected into a transport line with a 180° bend of 0.5 m radius. The program is intended to provide the information required for scaling to a multipass induction cavity experiment to study the beam breakup instability and complete a proof-of-principle demonstration.

A back-illuminating short-pulse photocathode using Cs2Sb films for use in a 45-MeV electron linac at the University of Texas at Arlington i under development. The photocathode is designed for an optimum simplicity without requiring an in-situ cathode preparation procedure and to achieve a sufficiently high current density for use in an wake-field acceleration test currently underway. Preliminary tests performed at 6 kV/cm indi1cates that a current of 2.5 mA/cm was satisfactorily achieved at a modest vacuum of 10 torr. However, fatigue effects were observed after a short period of operation in this test. Concurrently, a front-illumination metal photocathodes is also investigated for use in the wake-field test facility. Improvements and refinements are still in progress.

A promising approach to the generation of low-emittance e-beams for particle beam and FEL application employs a photoelectron cathode. IF such an e-beam source is to be viable at high power, a high-performance hard-UV illuminator is needed. Toward this end, experiments have been performed by illuminating a metal photocathode with the VUV radiation from a laser-guided gas-embedded high-density high-Z pinch. Such a VUV source is interesting because the plasma is created at high density and is optically thick. Thus, it is both a stable and an efficient radiator. Coupled with a copper photocathode it has generated a-beam current densities up to 60 A/sq-cm. The test device has been modified to utilize a pinch formed from a liquid jet in vacuum, rather than the laser-guided discharge in high-pressure gas. This is more suitable for rep-rate operation as it dispenses with the VUV-absorbing interposed gas, the channel-forming laser, and gas transport at high average power. A decane-jet device has been tested at 10 Hz with a peak pulse VUV power of 100 MW.

With multigap pseudo-spark chambers intense pulsed electron beams are generated by a low pressure discharge in different kinds of gases. The beams propagate in a self-focused manner through the low pressure gas. This depends on the influence of its own magnetic field and space charge neutralization. After a few millimeters the flight path of the beams hit metallic targets and drills holes of some tenths of millimeters in diameter. In principle the fact of drilling holes is not restricted to metallic materials. This is also possible into organic and inorganic insulators. The main applications for such beams therefore is high quality material processing. Furthermore it is shown to produce wideband soft X-rays. The option of postacceleration of such beams makes it attractive for future accelerators.

We have performed an experimental and theoretical study of a simple, axially directed Plasmoid which is neither charge or current neutral. This Plasmoid was produced in a novel virtual cathode ion diode, and subsequently propagated in vacuum. In this paper, we describe the simple theory underlying this type of Plasmoid, and the major results of the experiment.

The propagation of intense charged particle beams into vacuum without the aid of externally applied confining magnetic fields is under investigation both theoretically and experimentally. In the configuration under study an intense relativistic electron beam is injected through a localized plasma source into vacuum. Tons drawn into the vacuum region by the electron beam space charge provide the required neutralization for the effective propagation of beam electrons over significant distances. Fxperimental measurements indicate that under optimum conditions, both the ion and electron components of the propagating electron/ion beam are well focused radially. An equilibrium for this propagating electron/ion beam in which both species are in radial force balance has been postulated and matched to both electron and ion beam parameters at the source. Studies of the propagation of such an electron/ion beam across a transverse magnetic field are also reported.

The initial phase of research into plasmoids as relaxed, minimum energy states is described. A Lorentz invariant variational approach to finding plasmoid core equilibria has been developed. Three types of equilibria: nonrotating, weakly rotating and axially magnetized have been theoretically identified and verified by simulation using a fully electromagnetic PIC code, MAGIC. A rotational mode resembling that found in galaxies has been found for some plasmoid cores.

The possibility of reducing the rate of density decay in portions of an unconfined plasma by initial profile shaping is analyzed. For transit times of interest to the SDI program, such structuring may result in gains of several orders of magnitude in on-target plasma density.

The generalized covariant Virial Theorem can be used to provide both necessary conditions for plasmoid propagation equilibrium and accurate estimates of the disassembly time when such conditions are not met. Applied to long cylindrically symmetric geometries, it can be shown that radial confinement of the body of such a plasmoid requires continual transfer of momentum to the region external to the plasmoid; i.e., the plasmoid must have some external medium it can "lean" on. Such may be supplied by acceleration and loss of a sacrificial shell of material as proposed by A. Kadish and similarly by T. Lockner, or through coupling to an ambient ionized atmospheric component. This latter possibility has been explored and specific physical mechanisms identified and parameter ranges of validity established. The question of vacuum plasmoid propagation across a weak magnetic field has also been examined.

The propagation of plasmoids (neutralized ion beams) in a vacuum transverse magnetic field has been studied in the UCI laboratory for several years. These experiments have confirmed that the plasmoid propagates by the tcg drift in a low p and high p plasmoid beam (0.01 < β < 300) , where p is the ratio of beam kinetic energy to magnetic field energy. The polarization electric field, t, arises from the opposite deflection of the plasmoid ions and electrons, due to the Lorentz force, and allows the plasmoid to propagate undeflected at essentially the initial plasmoid velocity. In these experiments we used plasmoids (150 keV, 5 kA, 50-100 A/cm1, 1 'Is) injected into itifansverie fields of Bt = 0-400G. Anomalously fast penetration of the transverse magnetic field has been observed as in the "Porcupine" experiments. Our most recent experiments are aimed at studying the plasmoid propagation dynamics and losses in the presence of a background, magnetized plasma which is intended to short the induced polarization electric field and stop the beam. Background p2sma w4§ gensrated by TiH4 plasma guns fired along Bt to produce a plasma density, n = 10 - 10 cm . , Preliminary results indicate that the beam propagation losses 1.1crease with the background plasma density; compared to vacuum propagation, roughly a 50%,educOon in ion current density was noted 70 cm downstream from the anode for n, - 10 cm '. Principal diagnostics include: magnetically insulated Faraday cups, floating potential probes, calorimeters, microwave interferometer, and thermal-witness paper. Experiments in the near future will use an improved accelerator and transverse-field coil system which allows higher energy, 500 keV, higher current density plasmoids to be studied; this generator will improve the beam uniformity and angular divergence to allow beam propagation for up to 5 meters and permit the study of losses from surface erosion.

A variety of schemes have been proposed over the last two decades for delivering lethal amounts of energy and/or momentum to targets such as missiles and high speed aircraft. Techniques have ranged from high energy lasers and high voltage charged-particle accelerators to less exotic but still challenging devices such as electromagnetic railguns. One class of technology involves the use of high speed plasmas. The primary attraction of such technology is the possibility of utilizing relatively compact accelerators and electrical power systems that could allow highly mobile and agile operation from rocket or aircraft platforms, or in special ordnance. Three years ago, R & D Associates examined the possibility of plasma propagation for military applications and concluded that the only viable approach consisted of long dense plasma jets, contained in radial equilibrium by the atmosphere, while propagating at speeds of about 10 km/s. Without atmospheric confinement the plasma density would diminish too rapidly for adequate range and lethality. Propagation of atmospherically-confined jets at speeds much greater than 10 km/s required significant increases in power levels and/or operating altitudes to achieve useful ranges. The present research effort has been developing the experimental conditions necessary to achieve reasonable comparison with theoretical predictions for plasma jet propagation in the atmosphere. Time-resolved measurements have been made of high speed argon plasma jets penetrating a helium background (simulating xenon jets propagating into air). Basic radial confinement of the jet has been observed by photography and spectroscopy and structures in the flow field resemble those predicted by numerical calculations. Results from our successful initial experiments have been used to design improved diagnostic procedures and arcjet source characteristics for further experiments. In experiments with a modified arcjet source, radial confinement of the jet is again observed, with the 3 cm exit diameter of the jet preserved downstream for the duration of the quasi-steady flow. The modification of the arcjet consisted of the addition of a converging-diverging nozzle upstream of the 3 cm diam exit of the arcjet to expand the jet further before its entry into the background atmosphere. The jet penetrates for tens of diameters into the target atmosphere at a speed of 2 km/s, with an estimated jet flow speed of 5.3 km/s. The jet diameter is maintained for the range of the jet, which appears to equal 1 to 1.5 times the product of jet penetration speed and pulse duration (- 350 As in the present experiments), as expected from theoretical considerations. Further modification of the arcjet source (increasing throat diameter and electrode radius ratio) and extension of the current pulse time are expected to increase the jet range.

New electromagnetic directed energy pulse train (EDEPT) solutions of Maxwell's equations have been obtained. One particular solution, the modified power spectrum (MPS) pulse, will be described in detail. This pulse can be tailored to give directed energy transfer in space in such a manner that theoretically it beats the diffraction limit. These EDEPT solutions are not physically pathological and can be reconstructed from causal Green's functions. They represent fields that recover their initial amplitudes along the direction of propagation out to extremely large distances from their initial location. Moreover, these fields appear to be launchable from finite aperture antennas.

In the analysis of focused radiation from large aperture systems, and especially for the generation of "bullet-like" strongly collimated fields under transient conditions, it is suggestive to employ field representations in terms of focused basis elements. Possible basis functions in the frequency domain include Gaussians and Hermite or Laguerre Gaussians, and, directly in the time domain, complex source pulsed beams and focus wave modes. Some difficulties concerning the excitability of the focus wave modes will be discussed. The basis elements are next embedded in a discretized phase space spanning the space-time and spatial-temporal frequency domains to yield a rigorous field representation. Using space-time Gaussians, the phase space representation has been implemented for radiation from a space-time truncated aperture distribution. The results reveal the effectiveness of this approach to charting the near to far zone evolution of the emitted pulse.

When an electromagnetic pulse (EMP) traveling along the inside of a waveguide reaches the open end, it will cause radiation. Can this EMP radiation be an electromagnetic missile,' i.e., a pulse with slowly decaying energy? This analysis of radiation from an open circular waveguide attempts to provide the answer. The problems of radiation of acoustic and electromagnetic waves from a cylindrical waveguide have been solved for many years.2-4 However, interest then was confined to the CW case in the low-frequency region, i.e., when the wavelength A is of the same order of magnitude as the radius a of the tube. It is interesting and significant to study the radiation of an EM pulse in which the wide high-frequency band (λa) plays an important role. The problem of radiation of an EMP is solved and examined first in the frequency domain. The analysis proceeds according to the following steps: 1) Specifying the incident currents for the TEn, modes; 2) Establishing the coupled integral equations for the currents; 3) Using the Wiener-Hopf technique to solve the equations for the currents; 4) Deriving the formula for the electric field along the axis; 5) Evaluating the asymptotic behavior of the radiated field; and 6) Transforming the radiated field into the time domain and calculating the Poynting vector.

The purpose of the experiment is to demonstrate missile-like electromagnetic pulses. More specifically, this is a program to build and test devices capable of launching electromagnetic missiles. I Because a missile-like electro-magnetic pulse (EMP) involves transients with rise times under 100 ps, the measurement of short pulses is essential. The program must measure the EMP accurately enough to confirm the slow decay of the energy and provide the means for improving the design of the EM-missile launcher. After adopting the newly designed transmitting antenna (V-conical antenna) and receiving antenna Pantenna), the entire system is found to work properly up to 10 GHz. The preliminary results show that the energy of an ENV launched from an EM-missile launcher does have the property of slow decay.

We consider a transient point dipole current turned on at time zero, with the time-derivative being discontinuous at zero. This point current source generates an electromagnetic field incident on a perfectly conducting smooth surface having a boundary. Surface current is induced in the conductor, producing a scattered field. The objective is to analyze the leading edge of the scattered field. Cases are shown in which the edge singularity is strong enough to generate an electromagnetic missile. This report includes a rigorous formulation of the problem in terms of Maxwell's equations and a review of the incident and scattered wave fronts associated with these equations. These fronts are used to extract information about the leading edge of the scattered field. A "tangent-plane" approximation is studied and applied to a case of a bounded paraboloidal reflector illuminated by a transient point dipole. This results in a missile with a radiated energy den-sity that drops off as slowly as the inverse logarithm of the distance. Steps toward the establishment of bounds on the error of the approximation are described.

The following paper reports on the possibility of using a plasma, as the nonlinear medium, for generating phase conjugate reflection of electromagnetic waves in the microwave region, by almost degenerate four wave mixing. The reflected beam can be significantly amplified by the plasma. We show that the phase conjugate reflection is of significant magnitude if the frequency and the wave vector difference of the signal wave, with respect to the pump waves, resonate with the frequency and the wave vector of the ion acoustic mode of the plasma. This resonance can be predicted from a fluid like description of the plasma, but the Vlasov description can provide more details of the behavior near the resonance. The main results of the theory developed are described, as well as the experimental set-up we are currently building to verify the phase conjugation. Possible applications of the emergent technology are discussed.