The paper describes the results of a direct experimental observation of the development of an electrohydrodynamic instability at an explosive-emission liquid-metal cathode along with a theoretical analysis of the growth and destruction of the protrusion formed on the cathode surface.

An experimental set-up allowing measurements of DC field emission currents in the range of 10^-18 A to 10^-6 A in ultra high vacuum (P < 10^-10 Torr) at room temperature has been developed. The detection of currents in the range of 10^-18 A to 10^-12 A was carried out with an electron multiplier located behind a grid anode. Higher currents were measured using a massive anode and a picoammeter. The whole system was designed to study niobium samples from sheets used to build electron superconducting accelerator cavities. Comparisons of the electron field emission from samples prepared by different surface treatments as chemical etching, electropolishing, oxide coating by anodization, were done. Some of the obtained results are presented below.

A fast-video imaging technique was used to observe the pulsed breakdown behavior of an alumina tube, having two concentric planar electrodes on its end-face. Voltage-pulses, typically of 5 kV amplitude (5 kV microsecond(s) ^-1 time rate-of-rise), were applied to the radial M-I-M insulator-electrode regime under ultra-high vacuum (pressure < 10^-8 Torr), with video recordings made at 1000 frames per second. Images of the observed breakdown phenomena are presented, viewing along the center-axis and also in the plane of the M-I-M structure. These images are discussed in relation to plasma-jets associated with vacuum arcs, and the nature of the ion species within such jets.

The results of the investigation of amplitude and spatial-time characteristics of the initial conduction current at pulse voltage of a microsecond duration in the centimeter gaps in commercial vacuum are presented and discussed.

The electrodes used in this investigation to experimentally evaluate the effect of diamond turning on electrical breakdown characteristics in vacuum were first roughly machined by turning, and then machined to a mirror finish by diamond turning. Residual stresses of the electrode surfaces were measured by an X-ray diffraction method before and after the diamond turning. Several electrodes were annealed in vacuum for one hour at 400 degree(s)C or 700 degree(s)C before the diamond turning. Vacuum breakdown experiments using impulse voltages with the wave form of 64/700 microsecond(s) revealed that the diamond turning improves breakdown strength, that the conditioning of vacuum gaps is achieved by only few tens of breakdowns, and that annealed electrodes have a better hold-off voltage capability.

Stability of output parameters of power high-voltage acuuin devices is connected with the probles of interelectrode vacuua gap electric strength and ectrode tihera1 kO&dS increac a WOli as dveiopent of tab1e forsi electrodes. One of the way to solve this problea is using graphite being a well as known in eIectronis material or swe other attrials on the base of carbon with their unique aechanical and thermophysical properties as soon as their essential defect, the law electricstrength is eliminated.

The paper deals with ectons, emission centers produced on the surface of a cathode as a result of heating and explosion of its material due to large specific energy accumulated in cathode microvolumes. The criteria for the initiation and self-sustaining of ectons are given and their parameters and types are discussed.

For production of ionized atom fluxes a method of evaporating materials in vacuum by means of non-self-maintained discharge was developed. The conditions of the initiation discharge were examined. The necessary conditions are attainment of fixed vapor or gas pressure between the electrodes and presence of pre-breakdown current. The characteristics of the discharge in the voltage range from 30 to 700 V with currents of 1 to 50 A were investigated. The discharge plasma was studied with the assistance of electrostatic probes. The considered discharge region occupies the transition region from glow to arc. The cathode temperature is one of the fundamental parameters which defines the type of discharge and its burning. The regulation of temperature and discharge power influences on the processes which proceed on the cathode and in the space between the electrodes. The operation in different regimes results in the ability to control of ionization degree from 2 to 100%.

Using three kinds of copper electrodes with different surface areas, experiments were performed to investigate the influence of electrode area on conditioning effect, which is a characteristic of dielectric breakdowns in vacuum gaps. The conditioning process varied with electrode area: the smaller the electrode area, the sooner conditioning ended. Breakdown voltages after completion of conditioning also depended on electrode area: the smaller the electrode area, the higher the breakdown voltage.

Results are presented of studies of pulsed electric strength and threshold current densities for niobium cathodes in the superconducting and normal states. The process of heating and failure of cathode microemitters under the action of high density field emission current is analyzed. Data of the work function, limiting current densities and tensile strength of YBa₂Cu₃O_7-x cathode tips, effect of emission current on the superconducting properties of YBa₂Cu₃O_7-x thin films and peculiarities of prebreakdown phenomena in vacuum gap with YBa₂Cu₃O_7-x electrodes are given.

The cause of pre-breakdown currents appearing between metal electrodes copper soldered with ceramics after bakeout in vacuum up to 200...440 degree(s)C was established. Quest for it was conducted from the viewpoint of a hypothesis suggested by an author earlier on vacuum insulation violation. `Triple point'--metal-insulator contact in vacuum--is referred to as a pore in the gas medium of which ionization process resulting in various kinds of vacuum insulation violation develop. Experimental results confirm the hypothesis suggested.

`The total voltage effect'--being a main characteristic of vacuum insulation is explained from an earlier offered mechanism of vacuum insulation deterioration point of view. When an external electric field sagging in the negative electrode cracks and pores in their volume ionization processes occur. Gas medium in the pores appears as a result of the resonance gas desorption from the pores surface. `The total current effect' is explained by the same processes.

The effect of injection processes on the pulse discharges propagation mechanism in solid dielectrics has been shown. It has been established that during a cathode discharge these processes are analogous to laser action on solids, while in the case of an anode discharge they are similar to detonations of explosives.

A discharge initiation system is developed which is intended for discharge gap breakdown in an electrothermal launcher. It consists of a high voltage transformer, a pulse sharpening discharge element and a nonlinear inductance or magnetic switch. The high voltage transformer with air as dielectric produces output voltage up to 130 kV. It permits to realize with certainty the breakdown of gaps with length up to 3 cm. The breakdown occurs along the surface of dielectric dividing the gap electrodes, polyethylene, fluoric plastic and ceramics being used as a material between electrodes.

A `transparent anode' imaging technique has been used to display the spatial distribution of prebreakdown emission sites and the location of subsequent breakdown events. Using a real time video recording technique, it has been possible to address the following issues: (1) whether or not a spatial correlation exists between a DC prebreakdown emission site and the location of a subsequent DC or pulsed-field (PF) breakdown event, and (2) the evolutionary character of both DC and PF breakdown processes.

In nanosecond vacuum arc discharges, the size of cathode plasmas has been found to have the order of 10 micrometers . This outcome has decisive consequences for particle density, current density, time constants, and radiation properties. The present paper explores the question whether there occur principal changes in longer discharges. Experimental methods were: imaging the spots of microsecond(s) -discharges by absorption and emission techniques, and monitoring the shape of spectral lines in the range of nanoseconds through milliseconds. It has been found that in the light of atomic or ionic lines the diameter of emission pictures have the order of 100 micrometers , irrespective of the discharge duration in the range 10 ns through 100 microsecond(s) . Pictures obtained by differential laser absorption in the time range 100 ns through 50 microsecond(s) yielded a size <EQ 30 micrometers . Time-integrated analysis of emitted spectral lines yielded broadened profiles for discharges of 800 ns duration with a FWHM (full width at half maximum) up to 0.5 nm. This indicates the effect of high densities and magnetic fields, possible only with small spots. The same has been found for discharge durations up to 100 ms. Thus, it is suggested that the spot size remains small up to discharge durations of milliseconds. This outcome is associated with a principal non-stationary spot nature.

The influence of the Buneman instability on the parameters of a current carrying plasma is investigated. By using the quasi-linear approximation the analytical expressions which connect the plasma parameter changes and the turbulent electrical conductivity with the energy of the unstable electrostatic field fluctuations were obtained. The latter is expressed by the plasma supercritically. The experimental data of the arc discharge in an extended gap are analyzed. It is shown that the changes of the electron temperature and potential drop can be explained by this instability.

In this paper the processes in the low arc current (less than 1 kA) on a cathode with moderate melting temperature are discussed. In this arc an anode typically plays a passive part, acting only as a collector of the microparticles, metal vapor, ion and electron flux, which are ejected into a gap from the cathode spot region.

In this paper we shall discuss the peculiarities of the arc discharge in large interelectrode gaps and the breakdown of the near-electrode plasma sheath depending on the surface condition.

The cathode spot vacuum produces a jet of highly ionized plasma plus a spray of liquid droplets, both consisting of cathode material. The droplets are filtered from the plasma by passing the plasma through a curved, magnetic duct. A radial magnetic field may be applied to the face of the cathode to rotate and distribute the cathode spots in order to obtain even erosion and avoid local overheating.

The paper discusses experimental data which indicate that type 2 cathode spots originate only on oxide-cleaned metal surfaces. The suggestion is made that an instant at which an ion resonance recharging starts to play an active role in a near-cathode region can be taken as a criterion of type 2 spot origination.

This paper reports studies of contact admixtures effects on recovery voltages of contact pairs when admixtures have a form of multiple inclusions of some metal oxides introduced into standard contact material CuCr 50/50. The breakdown and recovery voltage studies of these materials were carried out at a simulating circuit with load current up to 300 A and frequencies 50 kHz - 120 kHz. Grains of alumina, and hafnium, lithium and lanthan oxides were used as inclusions. CuCr 50/50 with small tungsten chips was used as reference material. Statistical characteristics of recovery voltages were obtained for these materials and some promising compositions were selected for future studies.

Comparative characteristics of the cathode processes in the discharge with electrodes of pure metal and of CuCr composition are presented. The effect of the gap, electrode shape and other geometric factors on the cathode spots dynamics is investigated. The accomplished investigations allowed the elaborate methods to influence the cathode spot dynamics by geometric factors.

Plasma sheath near the substrate plays an important role in Vacuum Arc Deposition Technology (VADT). To analyze the sheath characteristics of the substrate in VADT, the unique ejection pattern of the cathode region must be taken into account. This paper describes a steady-state, two-dimension sheath model in which ions distribute specially according to a cosinoidal law. The model equations can be numerical soluted under specified conditions, to describe the sheath behavior of the substrate in the process of titanium film deposition using VADT. The model can be used for calculation of ion flux and heat flux flowing to the substrate in VADT, or in some other cases in which biased conductors were in vacuum arcs with cathode spots.

The paper presents a 2D hydrodynamic model of the plasma jet running out of the cathode spot of a vacuum arc, which is an outgrowth of the 1D problem that we have solved earlier. Also given are refined equations of state for a matter in a transitory phase, which conceivably might allow a through numerical simulation of a cathode spot from cold metal to hot plasma.

Hollow-cathode glow discharge, which operates at gaa pressures as low as those corresponding to a vacuum regime with \ d, where ) is the free path leugth of electron, d is the Bize of a cavity, 18used in electron and ion beam sources with plasma emitter.' However, a signilIcant difference in value between igniting and operating discharge voltages motivates the use of special systems for discharge ignition. Such systems, operating under increased preaurcs as compared to the preesure in the hollow cathode or using a high voltage ignition pulses, complicate source design and its electric power supplies.

A theoretical model is presented for transport of vacuum arc generated metal vapor plasma through a magnetized quarter-tours duct used for filtering out macroparticles in order to deposit high quality thin films. The model utilizes a two fluid approximation which takes into account collisions among the plasma particles. It is found that centrifugal forces must lead to a charge separation generated field, that determines plasma drift in the centrifugal force direction to the duct wall and give rise to ion loss. Another cause for plasma is the plasma pressure gradient. The plasma output flux is an increasing function of the magnetic field strength. The plasma flux in the output plane is asymmetrically skewed to favor the outside half. A further asymmetry in the flux distribution in the direction of the torroidal axis of symmetry is introduced if ions of different charge states are present in the plasma.

Multifluid equations describing the plasma of vacuum arcs expanding anisotropically from a cathode spot are given and discussed. Some first and preliminary results from an approximate analytical integration of such a system of equations are presented. Though the dependence of the plasma parameters on the direction is weak, the inclusion of angular terms and angular equations changes some results significantly, compared with the experiences from usual 1D models.

The present paper discusses the appearance of arc cathode spots observed in emission. Photographs yield a size of about 100 micrometers when using line radiation, whereas imaging by absorption techniques or emission pictures with continuum radiation result in values of about 10 micrometers , comparable to the crater traces left at the surface. Besides the size, the slope of the spot profiles is considered. Experimentally, it is found that the light intensity decreases with the distance r from the spot center proportional to r^-2. This slope is much less than the expected dependence proportional to r^-(beta), (beta) > 4, i.e. the observed spot edge is much less sharp than expected. The behavior of the radiation is theoretically studied by modeling the radial dependence of particle densities and temperature. It is shown that the line radiation stems from a region which is greatly influenced by the plasma expansion in competition with the finite transition probability of excited levels. It is concluded that emission pictures of cathode spot yield considerably higher spot diameters than corresponding to the spot core.

Voltage-current and other characteristics of the poorly explored type of discharge has been investigated and was found out to be quite different from the characteristics of other wide- known types of vacuum and gas discharges.

In this work the kinetics of macroparticle charging in the rarified part of the arc plasma jet are studied. The sheath in the vicinity of the macroparticle is collisionless and the problem with different Debye length to macroparticle radius ratio is analyzed. Maxwellian velocity distribution functions with different temperatures for the electrons and ions in an arbitrary ratio are allowed in the model. The influence of the plasma drift velocity on the macroparticle charging is discussed.

The paper deals with modeling processes at the electrode regions of pseudospark devices. We have analyzed the physical processes that could result in the initiation and development of explosive electron emission.

The results of investigation of high-current vacuum arc with Ti cathode filled by deuterium till x >= 1 (at.D/at.Ti) are presented. Electric and erosion characteristics were measured. The investigations proved this discharge to be very differing from the arc with degassed electrodes. The results obtained indicate the gas-filled electrodes have good prospects for some technical applications.

The paper presents the numerical implementation of the state-variable model of the high current arc interruption.

While one is trying to create the model of high speed charged participle flows from the cathode spot in vacuum arcs, attention should be paid to the non-contradictory choice of boundary conditions system.

Heat transfer to a thermally isolated graphite anode in a long duration vacuum arc was investigated. Anode bulk temperatures were measured as a function of time using two high temperature thermocouples. The anode surface temperature was optically determined. Surface temperatures of 2300 K were obtained in a 340 A arc. A 1D non-linear heat flow model for the anode was developed. A solution was obtained using a dynamic numerical method and the effective anode potential was determined to be approximately 6 V.

With the experiments presented in this paper applications of a retarding field analyzer for the measurement of ion potentials U_i in a vacuum arc plasma are discussed. The examined plasma was produced by a sinusoidal half-wave vacuum arc current. The experiments were concentrated on evaluating the plasma parameters at the last three milliseconds before current zero. In a current range from 300 A_rms to 10 kA_rms the ion potential distributions and their peak values were evaluated. With the increase of the arc current a decrease of the ion potential was found. By additional investigations of the angular distribution of the ion energies, a transition from a collision dominated interelectrode plasma to a freely expanding plasma was observed, depending on the arc current.

The use of reduced gaps in applications of high-current vacuum arc devices presents a number of interesting challenges. Specifically, standard contact designs have been developed over several decades to achieve controlled motion of high-current ac arcs in vacuum interrupters. For medium-voltage applications, the optimal maximum contact gap can typically range from about 6 mm to about 2 cm. However, the influence of the contact design may be gap dependent, so additional research may be appropriate if the contacts are to be used at smaller gaps. For example, the current through spiral contacts produces a magnetic field perpendicular to the arc column, but this will force the arc to move outward and run along the periphery of the petals only if a threshold separation is achieved. In this investigation, a framing camera was used to record the appearance and motion of drawn vacuum arcs between spiral-petal contacts with final gaps of 2 to 3 mm. After the rupture of the molten bridge, a high-pressure arc column formed and expanded across the width of the spiral arm. At the reduced gap, an intense anode spot formed if the peak current exceeded approximately 15 kA. Compared to results previously obtained at larger gaps, the arc motion was greatly reduced, and severe contact damage was observed at lower currents.

In this paper, the high current vacuum arc behavior with horse shoe electrode has been described. These are dynamic vacuum arc distribution, the maximum current density related to the interrupting limit and factors which influences the interrupting limits. Then based on the above analyses, several vacuum interrupters within (phi) 70 mm outside diameter have been tested. From the results, it shows that the small size vacuum interrupters can be made with the oblique, ellipse shaped horseshoe electrode. The interrupting ability of such type vacuum interrupters is further increased.

This paper analyzes the effect of the gap of electrodes on the state of vacuum arc by experiment and theory. And the model of vacuum arc is set up. The optimal gap can be deduced from controlling the vacuum arc to be diffusion state, in order to get the optimal moving characteristics of electrodes.

The main parameters and dimensions of cathode spots were under discussion for years. To solve these current questions, a new system was especially designs. The image converting High Speed Framing Camera, which combines a microscopical resolution of 5 micrometers with a nanosecond time resolution and a very high optical sensitivity. This camera was used to study the microscopical behavior of vacuum arc cathode spots in a pulsed high current arc discharge on copper. The direct observation of these spots with high resolution revealed the conclusions that one single cathode spot, as normally observed by optical means consists of a number of simultaneously existing microscopical sub-spots, each of them with a diameter of about 15 micrometers and a mean distance of 30...50 micrometers between them. The mean existence time of these sub-spots on copper was found to be about 3.2 microsecond(s) , where the position of a sub-spot remains unchanged (with an upper limit of about 2...3 micrometers ) during its existence time. An upper limit of the crater surface temperature was estimated by a comparison between the brightness of a cathode spot and of a black body radiation lamp to about 3000 K.

Micron-size macroparticles can be removed from vacuum arc plasmas using magnetic filters, permitting the deposition of high-quality thin films of the cathode material. The principle of magnetic filtering is explained, and ways of improving filter efficiency using additional electric and magnetic fields are explored. Under optimum conditions, about 25% of all ions entering the filter can be used for deposition. Other problems such as down or upscaling of deposition facilities are briefly discussed.

The present paper deals with the creation of a new mathematical model for 2D-computer simulation of the axisymmetrical radiative plasma magnitohydrodynamics (MHD) flows under vacuum spark discharge. In difference with the earlier works we use not only one but also two temperature approximation and take into consideration not only electrical but also magnetic fields in equation's system of the mathematical model. Created MHD-model is based on implicit full conservative difference scheme in combined Eulerian and Lagrangian approach. The system of difference equations is solved by using of Newton-like iteration procedure. Groups of equations are selected in accordance with the similar character of physical processes and the energy balance calculation is carried out for the convergence control.

After interruption of a high-frequency (hf) current by a vacuum arc, two distinct types of reignition can be observed. The first type follows immediately after interruption, the second type tends to allow a currentless pause of a few tenths of a microsecond. The post-arc current that flows after hf-current interruption has been measured and has a peak of several A and a decay time of a several hundreds of ns. A dynamic sheath model for the decaying plasma after hf current interruption has been used successfully to model the measured post-arc current waveshapes. The first type of reignition may be attributed to thermal stress during the period in which no ion sheath has been formed yet, the second type to high electric fields, either due to the ion sheath, or due to the TRV, the weight of each depending on gaplength and di/dt.

The cathode spot formation within first 22 ns was investigated by laser absorption photography and ps-pulse interferometry. The discharge was initiated between W-, Ag-, AuNi-, Pd- electrodes with cathode-anode distance below 100 micrometers , the arc duration was some milliseconds and arc current 5 A. A ps-pulse holographic interferometer and momentary absorption photography enabled us to determinate spatial-temporal density distributions in the ignition phase of the cathode spot. An absolute electron density value of order of 3 - 10²⁶ m^-3 has been determined indicating high conductivity values of the metal vapor plasma. Present measurements show that cathode spot plasma is essentially non- ideal and verify theoretical calculations resulting in an ionization potential decrease in dense cathode plasmas.

Computer calculation of pure metal film thickness distribution, using vacuum arc deposition technique, is presented in this paper. For random arc with a lower arc current, a model of single cathode spot with ion beam flux taking into account the cosine function spatial distribution is deduced. For arc steered by an external magnetic field, which is parallel to the cathode surface, a multiple cathode spots model is developed. And, in the case of random arc, calculation comparison between static substrate and rotating substrate is made. Results show that film thickness distribution is non-uniform when the arc is not controlled and tends to be uniform when the arc is steered by external parallel magnetic field with cathode geometry and substrate location being well chosen. Also, film thickness distribution is more uniform on a rotating substrate than on a static substrate.

Characterization of the cathode spot, which connect to a vacuum dc arc on a consumable cathode as a source of material for deposition, is studied using a high speed photograph technique. For a titanium cathode, external magnetic field effects on the behavior of the cathode spot is given with respect to number, size, and motion. Lastly, some discussion to the experimental results are presented.

The shield current distribution in a vacuum arc between CuCr40 contacts was studied by means of three cylindrical shields arrangement. It was found that the total shield current is affected by the arc mode. For a high-voltage oscillation sequence a significant effect on shield current distribution has an anode plasma jet. Then a great increase of the shield current is observed and the ratio of shield current to arc current was even more than 20%.

This report reviews research and development efforts within the last years for vacuum electron tubes, in particular power grid tubes for industrial applications. Physical and chemical effects are discussed that determine the performance of todays devices. Due to the progress made in the fundamental understanding of materials and newly developed processes the reliability and reproducibility of power grid tubes could be improved considerably. Modern computer controlled manufacturing methods ensure a high reproducibility of production and continuous quality certification according to ISO 9001 guarantees future high quality standards. Some typical applications of these tubes are given as an example.

Linear electric circuit with constant parameters becomes a nonlinear one with parameters variable in time after vacuum discharge gap is connected to it. Change of discharge parameters occurs by cycle. Full vacuum discharge cycle includes predischarge, spark, arc phases and current break phase (Fig. 1). Discharge development by incomplete cycle as well as development of one and the same discharge by several parallel channels with different cycles are possible. Every moment discharge current I is limited by one ofthe factors and namely - cathode emission ability L, plasma emission ability I, vacuum gap conductivity P. outer circuit resistance R -but it can not be limited by simultaneous infidence of several these factors. Fragment of the scheme (Fig.2) reflecting succession of turning on of current limiting factors may present how the process ofevery discharge goes on.

Switching characteristics of the triggered vacuum interrupter (TVI) combining the properties of vacuum interrupter and triggered vacuum switch are studied. In order to reduce the chopping current, electrode insertions made of a highly arcing-proof composite material are used. Along with the expected low chopping current less than 1 A the TVI has good switching capabilities both at anode and cathode modes of operation with a trigger current I_t > 100 A and trigger voltage U_t > 500 V. The effect of the load on the switching capability of the interrupter at the anode mode of operation is investigated. The obtained results make it possible to specify the requirements for trigger source output parameters with the aim to provide a reliable switching for different polarities of the main electrodes. Model tests of TVI in power frequency networks showed the possibility of its utilization under controlled switching with an accuracy to microsecond units and keeping all the properties of the vacuum interrupters.

The small-size sealed-off triggered vacuum switch TVS-7 designed to switch current pulse to 10 kA at high repetition rate to 1 kHz and maximum voltage to 25 kV is described. TVS switching characteristics are studied systematically versus the trigger current parameters for different polarities of the main electrodes. It is shown the TVS ensures a monotonous current rise of submicrosecond duration with a rate of current rise up to 10¹¹ A/s and a minimum switch-on delay time of 50 ns at a trigger current amplitude of I_t >= 100 A. In the cathode mode of operation the TVS is reliably triggered with reducing of trigger current up to I_t >= 1 A. But in this case current and voltage oscillations occur at an initial discharge development stage. This unstable stage of discharge burning is subjected to investigation. A minimum trigger energy, of about 10 mJ is required to ensure stable conditions of discharge burning with any polarity at the main electrodes. Under the condition power losses at switch-on stage of the TVS are substantially reduced in comparison with those of traditional vacuum switches. This became possible because of the use of a two-stage trigger system of the erosion type and a small main vacuum gap.

The paper considers pulsed power repetitive systems using plasma opening switches. The solutions of a row of technical problems connected with both repetitive operation of accelerator having high peak power and x-ray radiation converter are proposed. These machines are to be prospective for advanced radiation technologies.

We made some experiments on "Angara-5-l" (2-3 MA, 100 ns) [1], using low linear mass liner (10-40 iukg/cm ) as a part of double Z-pinch. The low mass liner surrounded coaxially a cylindrical load of higher mass (100-200 mkg/cin). Current instability and poor compression of the low mass liner were specific features of that inhoinogeneous Z-pinch. We registered '-0.5 TW pulse of soft X-ray radiation from inner load in the instability moment. We treat the effect as a Plasma Opening Switch (P05) process in external current shell of double pinch.

The current switching by means of a gas-filled electron switch has been studied experimentally and theoretically. Under deep volume charge compensation conditions the switch showed complete controllability. The switched current density was more than current density of a similar vacuum switch by an order.

In this report the results of the optimization of the multi-mode mcs-scale PEOS-based generator parameter's are presented. The PEOS parameters influence on the energy transfer efficiency was studied, the PEOS being considered as the part of the magnetically self- insulated line.

The vacuum state of vacuum switching elements after production is checked normally by Penning- or Magnetron methods (combined electrical and magnetic field). Vacuum in the range of 10^-1 - 10^-4 Pa can be measured in this way. After assembly into circuit breakers however, these methods are not applicable. HF interruption performance during making operation has been proposed earlier as a possible alternative. Further investigations show that differences in the number of HF prestrike current loops can be found in the pressure range of 10^-1 - 10⁵ Pa. Current chopping of DC arcs between 5 and 30 A during opening operation, may be another option for determination of the pressure range by measuring the lifetime of the arc, but the resolution in the vacuum range below 10^-1 Pa is too poor.

Series resonance inverters based on semi-conductor switching elements are well-known and have a wide range of application, mainly for lower voltages. For high voltage application many switching elements have to be put in series to obtain sufficient blocking voltage. Voltage grinding and multiple gate control elements are needed. There is much experience with the triggered vacuum gaps as high voltage/high current single shot elements, for example in reignition circuits for synthetic circuit breaker tests. These elements have a blocking voltage of 50 - 100 kV and are triggerable by a light fiber control device. A prototype inverter has been developed that generates 0.1 Hz, 30 kV AC voltages with a flat top for tests on cables and capacitors of many micro farads fed from a low voltage supply of about 600 V. Only two TVG elements are needed to switch the resonant circuit alternatively on the positive or negative supply. The resonant circuit itself consists of the capacitance of the testobject and a high quality inductor that determines the frequency and the peak current of the voltage reversing process.

The studies, carried out on high inductive energy transfer, has lead the Electrotechnical Laboratory to test switch device. Our works on trigger vacuum switches has shown switch off capacities. In preliminary studies, we have shown possibilities to pseudospark discharge for zero current switching. Our purpose here is to define electrical characterization of this kind of discharge. An axial magnetic field, pressure of gas (air), different shapes of geometrical electrode was the network constant of the pseudospark discharge. The surface observation of the cathode and the anode show us that a pseudospark discharge diffuse all over the electrode surface.

A high current surface discharge across the surface of an insulator in high vacuum is investigated. The mass erosion of the insulator and its surface voltage holdoff recovery are the two most important parameters of this investigation. Typically, the discharge current reaches approximately 350 kA peak with a pulse length of approximately 60 microsecond(s) (approximately 5 periods of a damped sinusoidal pulse). The vacuum level is maintained at approximately 10^-6 Torr. The insulator materials tested include a variety of polymeric (i.e., polyethylene, nylon, epoxy-fiberglass composites, and polyurethane) and ceramic (alumina, silicon nitride, and zirconia) insulators. Insulator mass erosion and surface voltage holdoff recovery versus electrode material has also been investigated. The electrode materials used include stainless steel, molybdenum, copper, copper-tungsten, brass, aluminum, and lead. Insulator materials that have low mass erosion and good surface voltage holdoff recovery have potential applications in high power vacuum switches. Breakdown voltage histories and mass erosion data were obtained for the plastics, but only breakdown voltages were obtained for the ceramics.

The vacuum energy concentrator has been designed on the 0.15 Ohm, 10 TW pulse power 8 module generator S-300. A vacuum 3D EC was constructed on the base of MITLs, connected parallelly at the central unit. For the total output inductance of 7.7 nH the concentrator provides currents of 5 - 7 MA. The calculations show that the transportation of efficiency electrical energy to the liner kinetic energy is about 30%.

Direct current breaker of a modular design is developed for the strong field tokamak power supply system. The power supply system comprises four 800 MW alternative current generators with 4 GJ flywheels, thyristor rectifiers providing inductive stores pumping by a current up to 100 kA for 1 - 4 sec. To form current pulses of various shapes in the tokamak windings current breakers are used with either pneumatic or explosive drive, at a current switching synchronously of not worse than 100 mks. Current breakers of these types require that the current conducting elements be replaced after each shot. For recent years vacuum arc quenching chambers with an axial magnetic field are successfully employed as repetitive performance current breakers, basically for currents up to 40 kA. In the report some results of researches of a vacuum switch modular are presented which we used as prototype switch for currents of the order of 100 kA.

Contacts design with amplified axial magnetic field for vacuum interrupters is described. Axial field specific induction reaches 10 mT/kA, that is the cause of time reduction for high current (> 10 kA) arc transition from contracted to quasi-diffuse modification. The arc movement on contacts surfaces is controlled by radial magnetic field up to arc transition from contracted to quasidiffuse modification. Both high current arc modifications control permits to decrease contacts separation velocity. During one pole tests 25 - 27 kA current was interrupted by contacts with diameter near 70 mm at 35 kV recovery voltage, 50 Hz, both at 1,1 m/s and 0,8 m/s separation velocity.

This paper studies the arcing and post-arcing medium-intensity recovery of vacuum arcs during short-circuit breaking of a vacuum interrupter by measuring the dynamic change waveforms of the relationship between shield potential, arc voltage and recovery voltage. On the basis of the experimental results, it makes mathematical derivation for the relationship between the vacuum interrupter shield potential and the voltage in contact gap. It also makes a theoretical analysis of changing regularity of the shield potential. It is proven that this method is feasible for deeper study of vacuum interrupter performance.

This paper studies post-arcing recovery characteristics of vacuum interrupters under the action of a high rising rate of recovery voltage by analyzing waveforms of recovery voltage and shield potential. It discovers that in the post-arcing recovery, there are two processes: electric recovery process and plasma recovery process. The two processes relate closely with the performance of a vacuum interrupter. The success of the breaking of a vacuum interrupter will rely on the plasma recovery process.

This paper measures and analyzes the waveforms of recovery voltage and shield potential. It discovers that in the shield potential waveform, when the breaking current is close to its limit, an irregular distortion happens to it. It is proved that not only the speed of post-arcing recovery process varies, but also non-stability exists in the process. This non-stability increase as the breaking current reaches its limit. The non-stability may cause post-arcing breakdown and breaking failure.

In this paper, the multi-channel array probes have been applied. And the computer measuring system has been adopted to deal with the measuring results to get the electron density and electron temperature at the point of the probe in vacuum chamber. And the electrode with self- axial magnetic field can be analyzed comparing with electrode with self-radial magnetic field by the distribution of electrode density in the chamber to study the effect of magnetic field on the diffusion ion.

Breaking capacity of vacuum circuit breakers is limited by dynamic voltage withstanding of their vacuum interrupters in a sense. Dynamic insulation in a vacuum interrupter can be represented externally by their dielectric recovery behavior and the stability of voltage withstanding. The internal feature of dynamic insulation includes two sides: one is the weak points under high electric stress, the other is macro-particles produced in the process of breaking high current. Some experiments were made to investigate these features. After analysis of the results, we given some measures to improve dynamic insulation of vacuum interrupters.

Electron-induced surface charging characteristics of sintered alumina were studied in relation to the polycrystalline features, such as (a) grain boundaries, (b) non-uniform distribution of grain size, and (c) orientation of grains, by adopting the scanning electron microscope `electrostatic mirror technique.' Significant variation in the charging characteristics of grains and grain boundary regions were not observed for bombardment with low energy (< 8 keV) electrons corresponding to secondary electron yield ((delta) ) > 1. Interestingly, for higher energies (> 8 keV) corresponding to (delta) < 1, the grain boundary regions were found to charge more negatively than the grains. The electrostatic mirror studies reveal that (1) the mirror image formation is not possible at high density defect zones, (2) the size of the mirror images, formed at one location varies significantly from that formed at another location, due primarily to the variation in the defect prone grain boundary volumes and, (3) the size of the image, formed on the grain, varies from that formed on another, due to variation in the crystallographic orientation of the grains. The results suggest that alumina undergoes inhomogeneous charging under electron bombardment.

Investigations have been carried out on right cylindrical teflon spacers in vacuum under DC stress to study the effect of cathode geometry on surface charging and flashover. Three cathode support arrangements were used. Type I involved a recess in the cathode into which the spacer was located. Type II was a simple flat plate cathode while type III had a raised insert on the cathode surface over which the spacer was located. In all cases the surface charge density and distribution is primarily dependent upon the magnitude of the electric area. Using the type I arrangement inception for charge deposition was always about 10 kV even when the spacer had been subjected to repeated flashovers and at applied voltages greater than this the density of the charge, which was fairly uniformly-distributed around the surface, was more or less proportional to the applied voltage. For the type II and III arrangements, a distinct, uniformly-distributed negative charge of up to 20 (mu) C/m² was always detected at low values of applied stress and at a fairly well defined transition voltage this gave way to a distribution which was substantially uniform in the case of type II but quite filamentary in type III and both involved the deposition of positive charge.

The high flashover sensitivity of high resistivity silicon in vacuum is discussed. Surface flashover fields of 5 - 30 kV/cm are common in silicon-vacuum systems, even though the intrinsic critical fields of silicon and vacuum are > 300 kV/cm. The influences of the material (bulk quality and surface processing), contact technology, contact geometry and electrode configuration on the preflashover and surface flashover characteristics of wafer samples in vacuum are analyzed.

Microwave surface flashover (MWSF) of dielectric articles--such as insulators, focusing lenses, waveguide windows, etc.--in vacuum is the primary factor limiting the applicable microwave intensities in physical and industrial installations. Besides this `negative' sense, MWSF also takes an applied significance. The present work is devoted to description of MWSF phenomenon itself and consists of two parts.

Alumina samples with different surface polishing (unpolished, 9 micrometers diamond-finished, and 9 micrometers SiC-finished samples) were used in our studies in vacuum, under pulse excitation, to understand the influence of surface preparation on the surface predischarge associated with microplasmas. The unpolished sample exhibited very active predischarge events, consisting of active localized plasmas on the surface, with associated large predischarge current and high intensity of luminosity, and low electrical breakdown strength. Compared to the SiC-finished sample, the diamond-finished sample showed more active predischarge events and lower conditioned and hold-off voltages. There seems to be a correlation between the predischarge activity and the surface preparation. Our experimental studies indicate that the phenomena of predischarge localized plasmas are related to the physical state of the dielectric surface.

A new method called the `Mirroir Method' for characterization of charging properties of insulators is presented. This method consists in the characterization of electrical potential formed after irradiation of a sample with 30 kV electrons in a scanning electron microscope. Analytical and numerical calculations allows to relate the electrical potential with intrinsic properties of materials such as the dielectric constant. Experimental results on various materials are well fitted by the models. The role of contamination layers is underlined. Relation between charging and mechanical properties are described owing to friction experiments.

Temporal change in electric field strength at the insulator-cathode-vacuum interface accompanied by charging of the insulator have been studied by using a 2D Monte Carlo simulation. The study employed cylindrical insulators and conical insulators. Each of the insulators was made of PMMA, polyimide or Al₂O₃. The field strength with cylindrical insulators show an abrupt increase, and with conical insulators considerable rapid reduction. Dependencies of the field change on the applied voltage height and material are well demonstrated. Flashovers occur after the simulated field change.

A method for preliminary treatment of electrodes by a microsecond low-energy intense electron beam is proposed. It has been demonstrated that such a beam melts off the electrode surface and cleans the surface layers from impurities and dissolved gases. In combination with subsequent conditioning of the vacuum gap by low-current pulsed discharges, high breakdown electric fields can be attained.

An ion beam sputtering technique was used to clean the electrode surfaces of vacuum gaps. Ions of the sputtering gas were irradiated by means of an ion gun in a vacuum chamber attached to a breakdown measurement chamber. By providing in situ ion-beam sputter cleaning, this system makes it possible to make measurements free from contamination due to exposure to the air. The sputtering gas was He or Ar, and the electrodes were made of oxygen-free copper (purity more than 99.96%). An impulse voltage with the wave form of 64/700 microsecond(s) was applied to the test gap, and the pressure in the breakdown measurement chamber at the beginning of breakdown tests was 1.3 X 10^-8 Pa. These experiments showed that ion-beam sputter cleaning results in higher breakdown fields after a repetitive breakdown conditioning procedure, and that He is more effective in improving hold- off voltages after the conditioning (under the same ion current density, the breakdown field was 300 MV/m for He sputtering and 200 MV/m for Ar sputtering). The breakdown fields at the first voltage application after the sputtering cleaning, on the other hand, were not improved.

The paper presents electrical strength estimation of vacuum insulating system after long period of storage, when voltage was not applied to the system. The test objects were commercially manufactured, high voltage extinguishing chambers of vacuum interrupter. The method of chamber preparation with conditioning process was described. Maximum charge of microdischarges and also microdischarges onset (inception) voltage were the parameters taken as the criteria to evaluate the electroinsulating state of vacuum system. It was found that there is a relation between breakdown voltage and the charge of microdischarges. The measurements were made after long-lasting period of time, when the chambers were stored without any voltage applied to them. The investigations proved that the increase of microdischarge charge resulted in the breakdown voltage decrease. On the other hand, microdischarge charge decreasing was accomplished by the breakdown voltage increase. Limiting of microdischarge intensity involves the increase of electric strength of a tested insulating system. Microdischarges maximum charge in vacuum insulating systems may be chosen as the one of quantities used for defining the electrical state of a chamber. It is important in a case when the measurement of breakdown voltage is impossible to conduct, or should not be done, because it may change the state of an investigated system.

We observed the discharge type under non-uniform field in vacuum for air and He gas. In order to quantitatively discuss the change of the discharge type, we introduced an image processing method. In the first place, we newly proposed 4 shape parameters to characterize the discharge pattern: the area SH, the flatness rate H/V, the location G of the center of gravity for the luminous area, and the length L of the positive column. The analysis with the image processing for air revealed that SH, H/V and G continuously increased as the pressure decreased. Consequently, the discharge type in air was successfully classified into 3 regions with two boundary pressures 200 and 2000 Pa over the pressure range from 1.3 X 10⁴ Pa. We also analyzed the discharge type in He gas in the same way. We newly introduced another parameter: effective current density Je which was defined as a ratio of the discharge current to the luminous area viewed from the vertical direction. Je proved to be independent of the discharge current Id, so that we suggested the possibility that Je can be a universal parameter to classify the discharge type for non-uniform electric field in vacuum.

The paper presents the results of investigations of electric strength at AC voltage (50 Hz) of vacuum insulation systems which were not subject to conditioning. Plane electrodes with rounded-off edges of Rogowski's profile were used in the investigations. The electrodes, having diameters of 50 mm, were made of OFHC copper. The paper shows the influence of fundamental factors defining the conditions of a vacuum insulation system on its electric strength, such as: the value of pressure (within the scope of approximately 1 mPa - 1 Pa) and at constant value pressure approximately 1 mPa, the length of vacuum gap, and the pressure of cobalt-molybdenum alloy layer coating the surfaces of the electrodes.

In this contribution we discuss a number of practical implications from recent studies on high- voltage design concepts used in microwave tube technology. We show how the insights gained can be used in the design of, for example, insulators and cables, and how conditioning procedures and operating conditions (operating pressure, insulator charging) should be reflected in the design.

This report deals with electrical insulation investigation of quasi-stationary high-current relativistic E-beam (REB) injectors (electron energy up to 400 keV, current up to 250 A, and pulse lengths of 250 msec). They were accomplished during REB injectors development and exploitation including regime of REB releasing into the atmosphere.

The paper presents the experimental results on generating and transforming the ribbon (3.5 X 130 cm) beam with the electron energy 1 MeV, beam current 70 kA and pulse duration 8 microsecond(s) . The 70% efficiency of the energy transfer from a capacitor storage to the ribbon beam has been achieved. The close value of the efficiency has been obtained for transforming the ribbon beam to the compressed circular one at the optimal conditions. According to the measurements the angular spread of the compressed beam is less than 10⁰.

The traditional way of relativistic electron beam (REB) injection in the experiments on REB- plasma interaction is the injection through an anode foil that separates vacuum diode from plasma chamber. The presence of separating foil leads to the following: (1) replacement of destroyed foil is required after each shot, and (2) the beam angular characteristics making worse. A beam with low angular spread can be obtained from foilless diode placed into strong guiding magnetic field; the problem is how to avoid the diode shortening in the presence of a dense plasma from the interaction chamber. In the experiments on studying of Langmuir turbulence, carrying out on GOL-M device it becomes possible to avoid a separating foil and to obtain a foilless injection of REB into a dense (16 cm^-3) plasma, utilizing a special drift pipe as an anode of the foilless axially symmetric magnetized diode.

The cylindrical geometry of the superdense magnetron type discharge system version in the crossed electric and magnetic fields is attractive in that it allows a high azimuthal uniform plasma emission surface to be formed. This geometry offers no problem with screening the cathode from the magnetic field of the beam transportation region. There is a stabilizing effect of magnetic field on the discharge, which provides high value of discharge current in the diffuse (without cathode spots) phase of discharge operation. Thus, for a cathode surface area of 300 cm² and a pulse duration of 15 microsecond(s) , the maximum current of a discharge in He was 1.5 kA, in a steady-state mode is was more than 10 A. The processes of electron and ion emission have some peculiar features, and the experimental results of this investigation are also presented. In our device, for a 20 microsecond(s) pulse duration and energy of 200 keV it was possible to obtain an electron beam current of 600 A with a current density of 75 A/cm². The beam mean diameter and width were, respectively, 10 cm and 1 mm. The electron current in the steady-state mode was 2 A with a 10 keV energy. The ion beam current constitutes of 10% of the discharge current. The application fields of these sources are microwave generation, surface properties modification, etc.

The paper presents experiments aimed at focusing and unfocusing pulsed electron beams of energy from 0.2 to 1.5 MeV by 3 to 35 kA currents with a duration of up to 200 ns. A cathode construction has been found experimentally which in recording on an X-ray film permits obtaining a resolution of 5 lines per millimeter with a penetrability of up to 40 mm of lead. A pulsating voltage with an amplitude over 10 kV is shown to exist at the prepulse voltage across the cathode. An electric generator substitution circuit has been found that explains the occurrence of a voltage pulsation across the cathode. The presence of the aforementioned voltage leads to the high-current diode being filled with plasma before a major high-voltage pulse arises. Conditions have been found experimentally which enable electron beam focusing or unfocusing.

The formation of ion beams with a large current is provided by multiaperture ion-optical systems. Their characteristics are determined of properties of a single aperture of ion-optical system. A detailed experimental investigation of one-aperture ion-optical system was carried out1 with the aim of determining of optimal conditions which provide the maximum current density j in the beam at a minimum angle of divergence o. These requirements are produced to ion-optical systems of neutrals injectors for nuclear fusion setups.

Based on the state-of-the-art knowledge, the processes of formation and transport of intensive ion beams in the transverse magnetic field of a commercial electromagnetic separator are analyzed. For the first time, it is shown that suppressing the low-frequency noise oscillations in an arc discharge of an ion source allows for the formation of a highly stable ion beam with a minimum inner potential drop possible. At low pressures, the drop is determined by Coulomb's collisions between the beam particles and compensating electrons, the collisions being unremovable in principle.

The results of the computer simulations of electron beam emission and propagation in various spatially inhomogeneous system are presented. It is shown that with the appropriate choice of the geometrical parameters the beam may be created in its squeezed state.

In the recent years a number of publications on the electron emission from ferroelectric cathodes have appeared. The most recent reviews are those by Schachter et a!) and by Gundel.2 Ferrodielectrics are ceramics containing the compounds PbZrO3, LaO3, and PbTiO3. These ceramics are designated PLZT by the initial letters symbolising the metals involved. A cathode of this type is simple in construction. This is a PLZT ceramic plate of several milimeters in thickness with a solid silver layer fused into the plate from one side (a trigger electrode, TE). From the other side the plate is covered with a large number of silver strips parallel each other (a grid). The silver layer thickness is 1 m on both sides. The anode to which an acceleration voltage is applied is placed parallel to the grid. The grid is grounded. An up to 2 kV trigger pulse of duration iO s is applied to the TE. In the cathode-anode vacuum gap of the diode an electron current appears which is a hundred or more times higher than the Child-Langmuir current and shows a pronounced threshold depenthng on the voltage applied to the ceramics. Early in the pulse the electron current increases almost linearly with time; An investigation into the current structure has shown that it consists of individual bursts that merge into a single flow.2

The solution of the problem of a vacuum diodes development for a ribbon e-beams of nanosecond and subnanosecond time duration production meets some obstacles. They are defined both by electrotechnical problems and by difficulties arisen from e-beam formation. This paper is devoted to the investigation of the main causes of a current amplitude limitation of the short-time e-beams.

High-current ribbon e-beam vacuum diode was designed to operate with RADAN 303 compact pulsed power source for various laboratory research purposes and technological experiments. Maximum e-energy, 200 keV; peak e-beam current, 1.8 kA; beam cross section, 5 X 110 mm²; current pulse width, 4 ns; beam energy per pulse, 1.5 J; the average beam power (in air) is 40 W at reprate of 25 pps. The parameters are given for the 40 micrometers aluminum- beryllium foil window. The diode operates under a vacuum of 10^-2 to 10^-3 Torr. The metal-dielectric cathode and the foil life time exceeds more than 10⁶ pulses. The pulsed power source permits one to control the diode voltage and change electron's energy. The possibility of varying the cathode length and anode-cathode gap of the vacuum diode is also presented. Weight together with RADAN 303 is 31 kg.

Water is widely used as an insulator for high power forming lines. Its high breakdown voltage and high electrical permeance gives a possibility to produce high power density flux. But an optimal wave resistance for cylindrical water forming lines is several ohm. Consequently, the power addition schemes are necessary for low impedance loads. In greatest pulsed power installations "SATURN" and "PBFA-H" (SNL, USA) water power addition schemes are used. On the contrary, "ANGARA-5-1" (TRINITI,Russia) scheme of power addition uses eight separate MIll in vacuum. Some experimental results of the parallel current addition on a common load are described in this report. Ajitter of parallel modules switching is a significant effect of "Angara- 5-1" operation. Quantitatively it is a square mean of time delay of modules T1-T8, when voltage of every module gets 0.1 of its maximal value. These moments can be defined with +/- 2 us accuracy.

Results of experimental study of a plasma-filled diode are presented. The diode was filed by a plasma from plasma guns. Two different regimes were tested. In the first one the plasma cloud was localized at the anode region. In this case there were not found a stable low impedance mode of a diode operation. In the second regime the diode was filled by plasma completely before a high voltage pulse and an additional plasma cloud was also created in the drift region. In this regime a low impedance accelerating gap, stable during up to 5 microsecond(s) was formed.

The powerful beam-plasma amplifier operating under continuous conditions is described. The amplifier is designed on the base of new principles using hybrid plasma-cavity electrodynamic structures. The theory and foundation of calculating the slow-wave structures having plasma filled propagation channel and interaction of electron beam with hybrid plasma-cavity system are considered. Considerably increased power of microwave radiation and expansion of the operating pass-band when the propagation channel is filled with plasma is shown.

The mockup channel tests for Space Power Experiments Aboard Rocket (SPEAR) I, SPEAR II, and SPEAR III will be examined in this paper. Specifically, high voltage results of mockup chamber tests will be compared with results of the flight hardware chamber tests and flight experiments. The authors will analyze the results obtained through the different phases of each of these programs and determine the effectiveness of the simulations that were performed in the vacuum chambers.

In this present report we consider dust particle striking surface of a target. Swift impact is a reason of a whole sequence of mechanical and physical phenomena. It generates spolation and crater formation, emission of matter, plasma creation and radiation flash. We have considered also dynamics of this plasma cloud in an external electric field (charge separation and charge extraction). This consideration is important from an experimental point of view for impact detection and analysis.

Investigations of fault arc propagation on spacecraft cables under spacecraft specific conditions, e.g. vacuum, normal and oxygen enriched atmosphere have been performed. Typical patterns of fault arc extinction have been identified and their consequences with respect to causes of damages are discussed. The results indicate that for a given test current and test voltage the behavior of the arc and the damages caused in a cable bundle depend strongly on the kind of insulation material, cable size and design as well as on the environmental conditions. For some cables the arc propagation effects are increased at higher oxygen partial pressure and significantly increased under vacuum. In many cases vacuum has turned out to be an important worst-case parameter. It was found that the amount of damage assessed by post-test measurements increases substantially, if the value of the test current grows.

This paper presents recent concepts and analysis on vacuum insulation issues in the lunar environment, including dust initiated breakdown, thermal management, gas contamination, and solar radiation effects.

An analysis of the kinetics of the electron subsystem of a superconductor based on a model with a phonon thermostat has shown that the nonequilibrium state of a superconducting thin- film cathode that appears at field emission changes the superconducting properties of the cathode as compared to the equilibrium state. The dependence of the superconducting gap on the intensity of the emissive source has been obtained.

The appearance or initial existence of microtips on the cathode surface is one of the main reasons of prebreakdown field emission (FE) current. Thermal explosion of these tips lead to explosive electron emission and initiating of vacuum breakdown. In the last years the sufficient role of super small tips (10^-6 - 10^-7 cm) in prebreakdown phenomena and initiating vacuum breakdown was established. The amplification of electric fields on such tips leads to very high FE current densities (J > 10⁺⁹ A/cm²). So, the creation and decay of nanometer structures define the dynamics of the vacuum breakdown development. The investigations of FE from nanometer formations are very actual, as the emission properties and stability of such structures are connected with the problems of field cathodes fabrication and exploitation in the devices of vacuum micro- and nanoelectronics. In this report the results of preliminary investigation of high current densities FE from nano-emitters are presented.

Field evaporation of Ir field emitters at high temperatures T and electric fields F was studied by means of field electron, ion and desorption microscopy. An initial emission of Ir-ions was observed only from the angles of build up emitters at relatively low F. The rise of F in 1.5 times in comparison with build up stage led to formation of microprotrusions. These microprotrusions moved along the edges between the angles of Ir-single crystal and played role of the centers of nucleation of new crystal layers. The macrooutgrowths on the close- packed planes were produced by field crystal growth and there were microprotrusions always on the corners of macrooutgrowths. Field evaporation of macrooutgrowths took place at definite T and F. This evaporation was observed as `effect of collapsing rings'.

A new set up is described which allows one to carry out field electron microscopic and spectroscopic study of cathode surfaces in direct correlation with their atomic structure.

Processes of field electron emission, field ionization and field evaporation of atoms from the emitter surface in intensive electric fields (107_lOs V/cm) are determined, in particular, by the crystallographic structure of the emitting surface. For pure metals, wherein the nonuniformity of density distribution of field emission current is due to various values of factor 3/2/E, this was first shown by E.Muller' . In the case of emission from the surface of metallic alloys the effect of the crystallographic structure on the distribution electron and field ion emission current density becomes more complicated. Such additional factors as the nature of interatomic forces responsible for structural trinsformations, different phase composition of various areas of the surface, the degree of long-range order in relevant sublattices, etc. begin to play a substantial role in emission processes. The purpose of this investigation is to elucidate the effect of alloy atomic surface structure crystallography on the processes offield electron emission and field ionization in intensive electric fields (107_108 V/cm). Such experimental conditions exist in the field emission microscope, where the distribution of electron and field ion emission current density is influenced directly by the structure of an atom-clean surface.

It is evident now that the internal mechanism of explosive electron emission and vacuum discharges is determined to a considerable extent by the dynamics of tion of the liquid phase at the cathode surface. The liquid phase is always present, if only on the individual sections of solid cathode, in every type of high voltage vacuum discharge. What is nkre, it presence is fixed actually even on carbon cathodes, although the formation of the liquid phase on it is fundamentally inhibited, since at normal pressures carbon sublis by passing the iilting stage1 The interaction between the external fields and the liquid surface of the emitter leads to the ergency of different types of hydrodynamic instabilities. As a result, the surface iuicrostructures of different types are formed2. The character diinsions of structures produced have the values which provides for the creation of the fields required for the emission sustaining. As our experiments show2, the development of such instability in super-high frequency (SHF) fields results in a capillary wave swing, that, even with relatively low SHF-field strengths, may cause both field electron and field ion emission3. In this case, the high degree of reproducibility of small- scale irregularities of the microrelief of a liquid surface is, in our opinion, the reason such emitters are highly stable.

Models about arc roots need a good knowledge of physical constants characterizing contact material. With pure metal, all the constants are well known; but for alloys some are not known. In this paper we relate how we have measured work function of silver alloys.

The destabilization of an infinite plane surface of liquid metal by a normal electric field in a vacuum has been the subject of different theoretical articles: above a critical field peaks grow on the liquid interface. A wavelength can be attributed to the resulting pattern. The electric field at onset and the corresponding wavelength only depends of the density and the surface tension of the liquid. Above this critical field a whole range of wavelengths are excited, each growing with a different growth time: the fastest is the dominant wavelength. We shall present the influence of dynamics of the liquid viscosity and of the liquid bath depth: it can be shown that the dominant wavelength is smaller with higher fields but differs if the liquid film is thick or thin, inviscid, or viscous.

The problem of field evaporation (FE) of the high temperature Superconductors (HTS) is discussed. The experimental data were obtained by techniques of wide-angle atom-probe and probe-hole atom-probe. These data are analyzed with the aims to estimate the binding energy (lambda) of removing particles from the surface and to find some possible correlation between simultaneously detected species. Both results of these processing show that destroy of unit cell of HTS-crystal at the elementary act of FE is essentially nonequilibrium process, and that well developed FE-theory of metals is only partly valid for HTS-materials.

This paper is devoted to theoretical and experimental investigations of high power microwave Cherenkov oscillators using of high current nanosecond electron beams. The influence of non synchronous waves and additional (non Cherenkov) resonance on the process of microwave generation in 3 cm relativistic BWO are discussed. Experimental data presented in this paper show that the efficiency of 3 cm relativistic BWO can achieve 40% and pulsed microwave power--500 MW. The enhancement of microwave power and efficiency of microwave generation was reached by creating of the non uniformity of the synchronous wave phase velocity and that of the coupling coefficient between the electrons and the electromagnetic wave. We experimentally studied the oversize Cherenkov oscillators in mm wavelength. One of them was 8 mm 120 MW orotron with TM₁₃-mode of the circular waveguide. Another one was two stage two electron beams 8 mm Cherenkov oscillator. The maximum of microwave power in this experiment was 600 MW, pulse duration was 10 - 15 ns. All experiments were carried out by high current electron accelerators of type SINUS. The performance parameters of our accelerators are the following: electron energy from 100 to 2000 keV, beam current from 1 to 20 kA, pulse duration from 5 to 50 ns, repetition rate up to 200 - 300 pps.

The multipactor and flashover phenomena of alumina rf windows used in a high-power klystron have been investigated. Multipactoring due to the high yield of secondary electron emission takes place during rf operation. A spectrum analysis of luminescence due to multipactoring shows that multipactor electron bombardment causes an F-center of alumina, thus leading to surface melting. From the results of a high-power examination of rf windows with several kinds of alumina ceramics, it was found that an alumina material with a crystallized grain-boundary and without any voids between the boundaries, thus having a low loss-tangent value, is not liable for F-centers, even under multipactoring. Flashovers in a tree- like pattern of alumina luminescence occasionally take place on a TiN-coated surface where multipactoring is suppressed. A sapphire window, whose surface was polished and having pre- existing F-centers, shows a lower flashover threshold. The annealing effect of polished window surfaces was also investigated.

The development status of a prebunched FEM is described. We are developing a 70 KeV FEM to allow high gain wideband operation and to enable variation of the degree of prebunching. We intend to investigate its operation as an amplifier and as an oscillator. Effects of prebunching, frequency variation, linear and nonlinear effects, will be investigated. The prebuncher consists of a Pierce e-gun followed by a beam modulating section. The prebunched beam is accelerated to 70 KeV and injected into a planar wiggler containing a waveguide. The results obtained to date will be presented. These include: characterization of the e-gun, e-beam transport to and through the wiggler, use of field modifying permanent magnets near the entrance and along the wiggler to obtain good e-beam transport through the wiggler, waveguide selection and characterization.

The experiments described used a Vircator operating in the frequency range 2.5 - 3.5 MHz and powered by a 10 ohm, 100 ns pulse generator operating at voltages up to 600 kV. The Vircator had a tunable resonant cavity and power was coupled out from one side. A series of experiments were carried out to elucidate various aspects of the Vircator's behavior. The first experiment conclusively demonstrated reflexing in which electrons from the virtual cathode region re-enter the anode/cathode gap and affect the diode impedance. The second experiment demonstrated the effect of the anode/cathode gap on the behavior of the reflexing electrons. The third experiment showed that the power output depended upon the thickness of the anode foil, and more surprisingly, on the thickness of the foil forming the adjustable rear wall of the cavity.

A detailed comparison of experimental results and computer simulation have been made for a number of simple axisymmetric MILO structures designed to operate at 1 GHz. The structures were built from a set of demountable components which enabled the number of cavities and their dimensions to be rapidly altered. Measurements were made of the fluctuating magnetic fields at the end of each cavity and of the applied voltage and current pulse. The amplitude and depth of RF modulation of the magnetic fields, although repeatable, changed drastically from one configuration to the next. These parameters were compared with predictions from VIPER, a 2D electromagnetic PIC code. Good quantitative agreement was obtained between experiment and simulation in most situations, however, late in the current pulse, after about 100 ns the level of RF began to decay; a phenomenon which became more pronounced as the applied voltage was increased. The effect was attributed to plasma formation on the cavity vanes and subsequent electron emission; this explanation was verified by computer modeling electron emission and by using vanes made from polished stainless steel in place of aluminum vanes.

The theory of electromagnetic radiation excitation by an electron flow is considered for a flat magnetically insulated transmission line with a comb anode. The dispersion relation describing the excitation of electromagnetic oscillations by the Brillouin flow in the slow-wave structure is derived. The numerical analysis of dispersion equation has shown that in the presence of the slow-wave structure, an essential increase in increments of unstable oscillations takes place. The conditions for optimum generation of microwave power are found. Efficiency estimates are given for the oscillators of the type considered.

Secondary Emission Discharge (SED) is one of the critical factors, which limit energy possibilities of the powerful HF-electronic devices with great magnetostatic fields (e.g., relativistic generator and amplifiers); it initiates a breakdown mechanism. The most dangerous, nonlocalized variety of the SED, which develops with the highest speed, is the one-sided resonance discharge. It is formed by electrons that oscillate near one of the Me surfaces and are driven back to this surface by the magnetic field H₀, which is directed in parallel or at some angle to this discharge surface. Characteristics of such secondary emission resonance discharge (SERD) have not been studied yet in the cases of arbitrary direction of the field H₀ to the discharge surface. However, near the walls of various devices (e.g., electrodynamic structures of generators) the angle of the direction of the field H₀ with respect to the surface can change in a wide range. The basic concepts of the theory of SERD starting regimes are studied in this report.

The specialized high-voltage installation for the investigation of cyclotron autoresonance masers as possible prototypes of HF source for fusion has been put in operation. The main parameters of the system are as follows: the maximal voltage 300 kV, the electron-beam current up to 300 A, the pulse duration 10 microsecond(s) , the pulse-repetition frequency up to 2 Hz. Experiments with the electron-optic system on the basis of the magnetron-injector gun being traditional for subrelativistic gyrotrons, are carried out. As an initial stage of the experiments, the relativistic gyrotron with the wavelength 4.4 mm and the operating mode TE_4,3 is studied.

The relativistic electron beam--electromagnetic field interaction near the high-frequency boundary of the TM₀₁-mode transmission band in overmoded uniform and sectioned structures with different slowing-down value has been studied experimentally and theoretically by means of the linear theory. The detailed comparison of the numerical simulation and the experimental data has been carried out for the three regimes: TWT-, TWT-BW()-, BWO- ones.

We considered the superradiance of ensembles of electrons rotating in the uniform magnetic field. It is demonstrated that such processes may be used for generation of powerful ultrashort microwave pulses.

It is proved that there is the possibility to use 2D distributed feedback to ensure powerful spatial-coherent microwave emission of sheet and tubular relativistic electron beam with transverse size exceeding wavelength by the factor 10² - 10³.

The peculiarities of the self-congruent interaction of surface fields of the superdimensional periodic waveguide and the relativistic hollow electron beam are considered with using methods of linear multimode and nonlinear nonstationary theories. The longitudinal and transversal electromagnetic field structure, the efficiency and the spectrum of radiation are determined.

The present work continues experimental investigations of high-power microwave oscillator driven by relativistic electron beam. We studied the possibility to design a microwave source of microsecond pulse duration, usually restricted by plasma generation due to a few reasons. It is shown that the breakdown is determined by microwave discharge, initiated, in turn, by electron beam destruction and bombardment of the slow-wave structure walls. To depress the process of plasma accumulation we propose to apply a gyrotron with axially-symmetrical TE- type mode.

Relativistic klystron amplifiers for the linear collider usually used a iris loaded waveguides as an output circuit. The behavior of electron and electromagnetic waves in traveling wave (TW) output structure of relativistic klystron with electron beams loading have been investigated. The linear interaction of space-charge waves of electron beam and electromagnetic waves of TW structure were studied and increment of radiation gain of TW's cell was determined. Longitudinal and transversal distributions of electron beam EM fields and their effect on TW circuit characteristics has been shown.

Experimentally studied microwave radiation from plasma waveguide when relativistic electron beam (600 keV, 2 divided by 4 kA, 100 ns) propagate axially the plasma waveguide. Spectrum of microwave radiation was measured by multi resonator spectrometer power was measured by broadband calorimeter. The calorimeter measured the total microwave energy emitted from output horn. Emitting microwave power strongly depended on plasma density, interaction length and gap between annular tube beam and annular tube plasma column. Spectrum of radiation was approximately 50%.

Physics processes in multiwave microwave devices are discussed. One of the important peculiarities of high-power microwave devices is multiwave interaction between electron and electromagnetic field. Multiwave interaction effects are illustrated for well-known microwave devices such as multiwave Cerenkov's and diffraction generators.

The description of an electron accelerator of microsecond pulse duration and the results of experiments with a relativistic carcinotron and a gyrotron, carried out on its base, is submitted in the report.

The paper presents a problem of radio-frequency interference (RFI) generated by complex high-voltage insulating system, i.e. extinguishing chamber of vacuum interrupter. A typical RFI measurement system with receiving antenna has been used to investigate prebreakdown phenomena in the extinguishing chamber. The experimental results allow to decide whether the vacuum chamber may be a source of RFI. Moreover, a strong relation between breakdown voltage and the RFI level in prebreakdown stage has been shown.

The subject of the project is to develop a version of a high-power pulse microwave amplifier similar to the conventional traveling wave tube on the basis of a relativistic electron beam emitted by a field emission electron gun. Anticipated parameters of the system are: electron beam energy 0.8 MeV, beam current 7 kA, radiation frequency 9.3 GHz, microwave output power up to 1 GW, microwave pulse duration 50 ns, gain about 40 dB.

Longterm tests with several electrode materials and discharge channel arrangements were performed. Two possible geometries were tested, a radial and a coaxial one. Optical fast shutter technique, and spectroscopic and interferometric methods were used to get more information about the discharge-character, the pinching of the discharge channels due to magnetic forces, and the erosion mechanism. The homogeneous ignition and the equal distribution of the current into the individual discharge channels was proved by fast photography and current measurements. At currents exceeding some kA the pseudospark transforms from a specific hollow cathode discharge into a metal-vapor-arc like behavior. Measurements of the forward voltage drop provide values of less than 100 V which are typical for metal-vapor-arcs. Also cathode spots and their traces on the electrode surface were observable. The performance of these switches in pulsed power devices will be reported.

A computational analysis is performed to investigate unsteady shock wave (SW) formation from electrical foil explosion. The velocity of SW vs. generator power is investigated: results for momentary power release are compared with ones that are obtained for electrical circuit parameters calculation with effects in foil consideration. Physical effects in the foil have been considered in detail. The special case of SW profile formation is analyzed. Results of calculations compared to the experimental data.

This report deals with A.C. measurements of H.V. insulating arrangements on site and in laboratory. The inspection of H.V. insulation of electrical equipment is first of all to recognize defects of these insulations as early as to prevent breakdown of apparatus. The special difficulty of these inspection results form the large dimensions of this equipment. The PD- measurements consider the test-circuit transfer characteristic influenced by the large physical and geometrical dimensions, the individual conditions of the test objects and the disturbance levels found on several places. For state evaluations of electrical insulations different test arrangement and measuring devices are used. The report gives a comparison between such methods with regard to the impact on the PD-signal and its spectrum from the generation up to the evaluation.

The tester is a convenient and useful unit for complex damage-free test of current leakage of vacuum and ceramic insulation in a wide range of high voltage. PC-control and processing provide information about the leakage nature and characteristics of the electrodes' surfaces.

It is asserted that Impedance Collapse is due to a partial vanishing of space charge which limits current in an improperly designed field emitting device. Empirical support for this contention is presented.

Resistometry, X-ray and NGR methods have been used to investigate phase transformations long-range type in submillimeter layers of Fe₆₉Ni₃₁ and Pd₂AuFe alloys under ion bombardment. In situ measurements were made simultaneously resistivity and temperature on plate specimens 100 - 400 micrometers thick. The observed transformations are considered to be a new type of radiation-induced self-spreading phase transformation similar to combustion and detonation phenomena, propagating at a speed equal to or exceeding that of sound in the material. In accordance with the proposed model the transformations take place in metastable media at solitary wave fronts generated as a result of evolution of atom collision cascades. Equations are proposed describing the emergence of undamped solitary waves with self- regulating amplitudes which bring about such transformation.

The present report consists of three parts. The first one is a brief review of experimental works accomplished in MRT Institute and devoted to a study of nonlinear interaction of strong microwaves (in X-band at intensities of 0.01 - 10 MW/cm² and pulse duration from 50 - 100 nsec up to 1 - 10 microseconds) with a plasma flare, and of attendant physical processes. The small experimental plasma reactor bench is described in the second part. And the third part of our report reflects the first qualitative experiments on pulsed microwave deposition of thin carbon and metal films on various substrates. Obtained practical results and technique perspectives are discussed in detail.

The emerging capability to produce high average power (5 - 250 kW) pulsed ion beams at 0.2 - 2 MeV energies is enabling us to develop a new, commercial-scale thermal surface treatment technology called Ion Beam Surface Treatment (IBEST). This new technique uses high energy, pulsed (<EQ 100 ns) ion beams to directly deposit energy in the top 2 - 20 micrometers of the surface of any material. The depth of treatment is controllable by varying the ion energy and species. Deposition of the energy with short pulses in a thin surface layer allows melting of the layer with relatively small energies and allows rapid cooling of the melted layer by thermal diffusion into the underlying substrate. Typical cooling rates of this process (10⁹ - 10¹⁰ K/sec) cause rapid resolidification, resulting in the production of non-equilibrium microstructures (nano-crystalline and metastable phases) that have significantly improved corrosion, wear, and hardness properties. We have conducted IBEST feasibility experiments with results confirming surface hardening, nanocrystalline grain formation, metal surface polishing, controlled melt of ceramic surfaces, and surface cleaning.

Field ion microscopy techniques have been used to investigate the atomic structure of quenched FePd₂Au alloy before and after bombardment by Ar⁺ ions (ion energy E equals 20 keV, fluence D equals 2 (DOT) 10¹⁷ ions/cm²). It has been confirmed that ion bombardment induces a `disorder-order' phase transformation in a subsurface layer more than 20 micrometers deep.

An experimental circuit which can be used to measure the phenomena of HF (high frequency) vacuum arc is built. Numerous measurements are carried out and the number of half-cycles of the HF vacuum arc, which is considered important to the over voltage caused by multiple reignition of vacuum switch, is input into computer. A calculating method is developed to calculate the mean value of the number of half-cycles of vacuum arc at various fixed gaplengths and circuit parameters. The calculated results fit the experimental ones very well. It is concluded that the number of half-cycles of HF vacuum arc depends strongly to the frequency dependent decaying constant of the high HF circuit and the parasitic or lumped capacitor parallel to the vacuum interrupter.

Thermionic emission of new material compositions are studied. Combinations of rare earth materials and tungsten offer great potential as thermal electron emitter into vacuum. The thermal emission properties of these materials are studied and compared to thoriated tungsten as a well-known thermal emitter. The corresponding work functions and Richardson Dushman constants are evaluated. The chemistry involved and the emission mechanism are discussed.

The technology is based on the well-known process of hardening some organic liquid monomers and oligomers (resins and lacquers) under the action of ionizing radiation, i.e. electron, ion, ultra-violet, laser, and X-ray beams. The main mechanism of hardening is 3D polymerization of initial monomers induced by irradiation. First of all 1D polymer chains are created. And the next stage is cross-linking of them. Numerical attempts to apply such a process for plastic materials production met some earnest difficulties. Our decision to perform the whole processing in vacuum changed radically the main properties of radiation induced hardening technology. The inhibition of polymerization by reactive oxygen became unessential. The output window foil of accelerator became unnecessary. Application of super broad beams such as 1 sq.m became possible. The entire efficiency of grid electricity was increased to 60% and it was not the limit. One of the main advantages is that the processing carried out in vacuum may satisfy the highest ecological standards. The technology developed is contamination free and environment-saving.

Considerable interest has recently been generated in a new type of highvo1tage switches that depend on a low pressure gas discharge with a cold cathode, often named the pseudospark switches13. They are considered as an alternative to the hydrogen thyratrons in facility where the high level of currents and dud! are required. One of the problems for such switches is to increase the static breakdown voltage since this parameter is very important in itself and, besides, determines the ability for a switch to operate in high frequency mode. In order to solve the above mentioned problem Christiansen et al1'2 use the special design in which a positive potential is applied to additional blocking electrode from auxiliary power urnt.

Electrode systems with electrode areas exceeding a few square meters and interelectrode gap spasings reaching several tens of centimeters are used in physical experiments for the production of large-cross-section charged-particle beams as well as for the separation and deflection of the particles. Under commercial vacuum condition, the breakdown voltage is strongly affected by the electrode size and surface condition and the electrode separation. Thus, for a 1-mm gap, increasing the electrode area a hundred times reduces the breakdown voltage by about one half, ill other conditions being the same. Large and intricately shaped electrodes make the system degasing and conditic'ning more complicated. Prebreakdown currents produce an extra load on the power supply circuits and, in pulsed accelerators, they distort the beam currents waveform and often difurb the normal operation of the accelerator. This has stimulated the study aimed at refining the factors affecting the breakdown voltage, elucidating the mechanism for the prebreakdown conductance in the electrode gap, and developing the methods for suppressing prebreakdcwn currents.

A 1D nonstationary model of material melting and evaporation in anode spots of vacuum arc is developed. Calculations of melting and evaporation rates of copper and copper-chromium contacts were carried out and concentration of copper atoms was evaluated when the arc was burning and after its extinguish. The model of anode melting that accompanies its evaporation permits to assess the time of liquid metal film existence at the anode in the ablation approximation of the model.

This paper investigates the surface charge distribution of an alumina insulator surface after flashover tests by impulse voltage applications. The measurement of surface charge distribution and the voltage testing were both done in vacuum (in situ). The vacuum chamber for the voltage flashover testing experiments was maintained at the pressure of 1.3 X 10^-8 Pa. Impulse voltage was applied with positive and negative polarity. The surface charge was measured for both positive (flashover and unflashover) and negative (flashover and unflashover) polarity. Applying the positive impulse polarity produces a positive charge on the alumina surface, when the impulse polarity is reversed both the positive and negative charges were detected in case of flashover, and only negative charge was detected in case of unflashover.

1 .Introduction Electron field emission from microscopic metal cathodes subjected to high electric fields is observed at fields being much below the value predicted by the Fowler-Nordheim law for flab metallic surface.Un derstanding for this phenomenon is important for high field vacuum aplications [1J. This enhanced field emission (EFE) is known to originate from localized microparticles on a broad area cathode 12-51. The characteristic size L of the microtip was estimated to be of order of iO - 1O_8 cm. However,the physical mechanism of EFE is not clear up to now. The case of extremely small microtips is analised below Depending on the correlation of the microtipe size and lengt of elastic (l ) and in elastic ( 1 e electron scattering,three modes of emission may be realized: ballistic (1<> le Ji ).In the later case the principial solution may be obtained by means of hydrodynamic approximation [6] . Otherwise, in the case under consideration, the the analysis should be based upon the directsolution on the kinetic equation for the electron distribution function.For metall microcontacts such a programme in the single-particle approximation is relized in [7]. In this case the energy scattering length 2e Oi le )1/2 >> thus heat discharges in a region of volume 2 , exceeding considerably the region of formation of microtip tunneling resistance ('.3 ). According to [3], it is this particular phenomenon that causes the microtip thermal stability at greater values of the current density.