Breakdown Voltage Measurements Research Articles

Effect of thermionic emitting heater on breakdown voltage in high pressure mercury vapor

HID lamps cannot be reignited instantly after they are extinguished, since the breakdown voltage is too high. As the result of investigation about the mothod that lowers the breakdown voltage at restart, we found that it is effective to set a heater as a preheated cathode in the discharge tube, and to utilize the thermionic emission from it. In order to investigate this effect of the heater more precisely, the discharge tube which had a heater at the end portion of the tube, and which was filled only with mercury, was prepared. Mercury vapor density in the tube was controlled in a furnace and the measurement of breakdown voltage was made in the heater temperature range up to 2,100°C. In the case of mercury density of more than 4.2×1018m-3, the results of the measurement have shown that the breakdown voltage was almost constant until the heater temperature reached about 1,700°C, and then it decreased greatly, while it reached a minimum voltage at about 2,000°C.

Performance of electrodes at the first voltage breakdown in liquid helium

A cryogenic system for the investigation of the dielectric properties of liquid helium is described. Results of first breakdown voltage measurements made with a great number of fresh (unsparked) sphere gaps are presented. Examples for typical conditioning effects resulting from repeated sparking of one gap are given and discussed. A main objective of this work has been the study of the influence of electrode material on the first voltage breakdown in freshly prepared sphere gaps.

Orientation dependence of electrical and optical properties of ZnSiP2 crystals grown from melt

AbstractUndoped and Ga‐doped ZnSiP2 crystals grown by spontaneous crystallization from nonstoichiometric melt show a pronounced dependence of electrical properties (conductivity, carrier concentration) and cathodoluminescence on the orientation of various crystal faces. Breakdown voltage measurements and Hall measurements yielded that the (112)B face is always low‐resistivity n‐type and the (112)A face is always semi‐insulating. The spectral position of the broad luminescence band observed at 80 K differs in a characteristical manner for the various crystal faces. The luminescence behaviour correlates partly with the electrical properties. The results are explained by an orientation dependence of the impurity incorporation which is reflected also in the bulk properties of the crystals.

Electrical phenomena at the surface of electrically stressed metal cathodes. I. Electroluminescence and breakdown phenomena with medium gap spacings (2-8 mm)

Spectral intensity measurements have been made of the light emitted from discrete points on the surface of electrically stressed metal cathodes. DC and pulsed breakdown voltage measurements have also been made immediately following surface arcing treatment. These show that the significant differences occur between different electrode materials. It is shown that the phenomena of electroluminescence, electron emission and vacuum breakdown, are all consistent with a model based on the production of conducting filamentary channels, electroformed through surface oxide or oxide type impurities by the action of the applied field.

Semiautomatic measurements of thin-film breakdown voltages

A semiautomatic setup is described which provides the destruction-free measurement of breakdown voltages of thin-film structures for applications like process control and quality prediction. The method of measuring the breakdown voltage consists in feeding a constant current to the thin-film device and analyzing the voltage buildup across the device. This method has been described in a recent paper. The setup provides the successive analysis of breakdown behavior of a given matrix of wafer dots by means of a positioning table, an electropneumatic prober, and a microprocessor-controlled measurement unit via handshake commands. The results are given in a histogram form.

Investigations of the occurrence of erratic low breakdown voltages in compressed air

Previous investigations of the influence of electrode surface irregularities on the breakdown voltage in compressed gases were continued in order to get a better understanding of the reasons which cause the erratic low breakdown voltages in pressurized gas insulations. This paper contains the results of breakdown voltage investigations of smooth or roughened electrodes in air with pressures up to 30 bar. Measurements were carried out with + d.c. as well as 50 Hz a.c. using uniform and coaxial fields with gap lengths from 2.0 to 5.0 mm. The a.c. breakdown voltage measurements displayed that the lowest spark voltages of polished or sandblasted electrodes are in the same order of magnitude. With- d.c. an irreversible prebreakdown current behaviour of sand-blasted electrodes was found in air at pressures higher than about 5 bar, respectively at field gradients greater than about 80 kV/cm. Scanning electron microscope pictures of a single electrode surface protrusion showed the removal of the top of the irregularity by irreversible prebreakdown current pulses at -d.c.. It is assumed that field electron emission accompanied by space charges and ionization causes a heating as well as a removal of sharp electrode irregularities and without doubt this heating process facilitates the occurence of erratic low breakdown voltages.

2.8 A statistical approach to the analysis of dielectric breakdown strength of thin insulating films

Statistical models that are widely used in the study of high voltage insulating systems are applied to thin insulating films. By use of these methods it is in many cases possible to decide whether the electrical strength of a component is bulk-limited or electrode-limited. A discussion on which type of statistical distribution function can be expected for electrical breakdown in thin films is given. By analyzing measured breakdown voltages for more than 2000 capacitors it has been shown that a description of dielectric breakdown in thin films by means of the theory of the weakest link is fruitful. It is shown that the first selfhealing breakdown has its origin at weak spots randomly distributed over the capacitor area. The variation in the results is too large to determine whether the breakdowns start at the interface electrode dielectric or within the bulk of the film. These techniques have also been succesfully applied to previously published results, offering an explanation of results not previously understood.

Influence of the secondary electron emission coefficient of argon on Paschen breakdown curves in ac plasma panels for neon+0.1% argon mixture

A generalized expression has been derived for the effective secondary electron emission coefficient γ, in Penning mixtures which incorporates the effect of both the majority and minority ions. Specific calculations for ac panels having an Ne+0.1%Ar mixture show that even a relatively small value of γAr not only significantly lowers the breakdown voltages, but also moves the voltage minimum to higher p0d values and makes the right-hand side of the Paschen curve less steep. This sensitivity of the effective coefficient γ on γAr is sufficient to explain the critical dependence of the shape of the measured breakdown voltage curves in ac panels on minor changes in the surface properties of the deposited insulator surfaces.

Formation of Silica Films on Silicon Using Silane and Carbon Dioxide

The reaction of silane with carbon dioxide to form silica on silicon in the temperature range 700°–1100°C has been studied. The results of surface charge density and breakdown voltage measurements on the silica film are discussed. Etch rates, weight density, and infrared spectra obtained for these films are compared with those exhibited by oxide films thermally grown with steam. The dependence of the reaction yield on reactor design is briefly discussed.

Cryopower transmission studies in Europe

Abstract European studies on cryopower transmission are focused on semiflexible and totally flexible superconducting cables. Intense investigations of cable components are being performed. Nb/Cu and Nb/Al composite conductors have been developed, which exhibit 50 Hz surfaces losses well below the design limit of ac cables at 4–6 K and surface peak flux densities up to 0.1 T. Measurements of breakdown voltages and dissipation factors at 4 K indicate that wrapped tape electrical insulation, impregnated with helium, may be suitable for superconducting cables, the achievable dielectrical strength being about 10 kV mm −1 (rms). But further investigations are necessary within this topic. Different types of dc and ac cable joint and terminal have been designed or already constructed. A number of larger laboratory model tests with cable sections up to 50 m length have been performed or are planned for the near future with the aim of studying size effects, the interaction of different components and to demonstrate the feasibility and performance of superconducting cables.

Prediction of avalanche breakdown voltage in silicon step junctions

The results of two major theories that attempt to predict the electric field dependence of the ionization rates in silicon are compared with experimental data. For this purpose the published ionization rule measurements BUS supplemented by using the results of the two theories to predict the breakdown voltages of silicon step junctions. It is found that a. discrepancy between Baraff's analysis and the ionization rate data is also present in the comparison with breakdown voltage measurements. In contrast, a satisfactory agreement with both sets of results is found using the simple ballistic model first introduced by Shockley. The analysis of Shockley's model is re-examined with particular reference to certain discrepancies found by earlier workers. A revised expression for the ionization rate is obtained which removes the discrepancies and also remains consistent with both Shockley's earlier analysis and the experimental data.

Electrical breakdown of N2O, SF6 and N2O/SF6 mixtures

Measurements of static breakdown voltages were made for SF6 and N2O/SF6 mixtures containing between 1% and 50% SF6 by volume for values of Nds within the range 1023 < Nds < 1024 m−2 (where N is the gas number density and ds the sparking distance). These results, together with earlier data on N2O, show that the departures from Paschen's law become smaller as the percentage of SF6 in the mixture is increased over the range from 0 to 10% becoming negligible above about 6%. These results agree with predictions made on the basis of the ion-molecule reactions occurring in these gases. The dielectric strength of a 37% mixture was found to be about 83% of that for SF6 alone, and it is shown that, for this mixture, an increase in total pressure of about 20% gives the same dielectric strength as SF6 alone with a saving in cost of about 50%.

Electron and hole ionization rates in epitaxial silicon at high electric fields

Ionization rates for electrons and holes are extracted from photomultiplication measurements on silicon p + n mesa diodes for electric fields of 2·0 × 10 5−7·7 × 10 5 V/cm at temperatures of 22, 50, 100 and 150°C. These results are particularly pertinent to the analysis of high-frequency (∼ 100 GHz) silicon IMPATT diodes. The rates obtained here are in reasonable agreement with previously published data of van Overstraeten and DeMan, although slightly larger in magnitude. Calculated curves of breakdown voltage vs background doping level are presented using the room temperature ionization rates. Also a comparison is made to previously reported rates. The new rates provide a closer agreement between predicted and measured breakdown voltages for breakdown voltages less than 70 V.

Silicon Surface Contamination: Polishing and Cleaning

A Rutherford backscattering study using 2 MeV He+ ions has shown that the principal contaminants heavier than Si remaining on silica‐sol polished silicon surfaces are , and a heavy metal in the range . Preoxidation cleaning using and was found to be erratic in the ability to yield surfaces free of Ca and Cu and always left heavy metal contaminants. Cleaning with and solutions always removed all elements heavier than chlorine to below the level of detectability. Both cleaning treatments left . Breakdown voltage measurements indicated that junction quality was not predominantly controlled by surface impurities present after preoxidation cleaning, although the diode quality on cleaned wafers was much higher than on as‐polished wafers.

The breakdown of argon at low pressure

Measurements of prebreakdown currents and breakdown voltages at low values of ND to the left of the Paschen minimum are given for argon. Analysis of the results shows that the model of breakdown previously applied to mercury vapour may also be used for argon. Values of M and γ are given at E/N values down to 2·8 × 10−18 V m2, where M is the total electron multiplication coefficient and γ is the usual secondary coefficient. Theoretical values of M are shown to be in good agreement with measured values.

Determination of Highly Stressed Volumes in Oil Dielectrics

The volume of oil under the highest stress has been shown to correlate well with the measured breakdown voltages of transformer oil in different electrode systems. The determination of the volume of oil that is most highly stressed in a given electrode system is not however an easy matter. It involves the location of equistress or equigradient surfaces followed by computation of the oil volume contained within the surfaces. This paper describes a differential probe and amplifier system, which has been developed to allow the direct plotting of equigradient lines on a conducting paper analog of any two-dimensional electrostatic field.

Surface breakdown in silicon planar diodes equipped with field plate

The distribution of the electrical potential around the edge of the field plate in a silicon planar diode has been solved by means of a computer. The breakdown voltage related to this region has been derived as a function of impurity doping in silicon and of oxide thickness. If the oxide layer is not thick enough, avalanche breakdown voltage is often considerably lower than the value allowed by the material. The effect of Q ss has been found to be negligible. Breakdown voltage vs. gate voltage measurement on gated diodes and avalanche light emission support an edge breakdown mechanism. Breakdown voltage on field plated diodes matches rather well with the calculated values. It normally drifts towards higher values and during this walkout a hot carrier current is collected by the field plate. The field plate current is fairly proportional to avalanche current and increases with bulk doping. A new planar diode structure suitable for a BV up to 900 V has been designed. The discharge voltage in air between field plate and equipotential ring (EQR) is shown to be correlated with the actual field at the edge.

Uniform-field breakdown-voltage measurements in sulphur hexafluoride

Uniform-field breakdown-voltage measurements in sulphur hexafluoride

AVALANCHE BREAKDOWN VOLTAGE OF GaAs p+-n-n+ DIODE STRUCTURES

The ionization rates of GaAs are first empirically determined from breakdown voltage measurements of p+-n junctions with thick n layers. The theoretical analysis is then extended to p+-n-n+ structures in which space-charge punch-through occurs before breakdown takes place. The avalanche breakdown voltages of GaAs p+-n-n+ diode structures are calculated as a function of the n layer thickness W as well as the doping density n. Comparison of the theoretical calculation with experimental data shows very good agreement.

The Effect of Electrode Temperature on Vacuum Electrical Breakdown between Plane-Parallel Copper Electrodes

Combined measurements of current-voltage, prebreakdown characteristics, and breakdown voltage have been made for plane-parallel, copper electrodes in ultrahigh vacuum as a function of both cathode initial temperature and anode initial temperature. These measurements have been carried out for an electrode separation of 0.1 cm over the temperature range from 313° to 913°K. From the experimental data the temperatures of the hottest point of the cathode surface, (TC)s, and of the anode surface, (TA)s, at breakdown have been determined as a function of electrode initial temperature. Over the whole range of the investigation it is found that (TA)s is greater than (TC)s. Moreover, the calculated values of (TA)s at breakdown are essentially independent of anode initial temperature and have a mean value of (1100±150)°K. In contrast, the values of (TC)s at breakdown increase with increasing values of cathode initial temperature. The results suggest that a thermal instability of a point on the anode surface leads to breakdown. A model for the breakdown mechanism is presented in terms of electron avalanche amplification of current in copper vapor. The vapor is generated by the detachment of an anode macroparticle and its subsequent evaporation during its transit to the cathode. The prebreakdown current plays an important role in the breakdown mechanism since it causes the initial thermal instability at the anode surface and, in addition, provides the necessary source of bombardment for evaporating the macroparticle.