US3579031A - Zero arc drop thyratron - Google Patents

Abstract

A thyratron has a tubular cathode, preferably made of rhenium and circumscribed by a grid which in turn is circumscribed by an anode tube. The thyratron cavity is filled, e.g., with vapor pressure regulated cesium. The operating parameters can be chosen to establish essentially zero voltage drop across the thyratron. The cathode cavity is contiguous with a cavity of a thermionic converter.

Description

United States Patent lnventor William J. Keams Arcadia, Calif. Appl. No. 644,146 Filed June 7, 1967 Patented May 18, 1971 Assignee Xerox Corporation ZERO ARC DROP THYRATRON 18 Claims, 7 Drawing Figs.

U.S. Cl 315/363, 313/40, 313/180, 313/186, 313/211, 313/218, 313/227 Int. Cl. ..l-l0lj 17/08, l-lOlj 17/20 Field of Search 313/218,

Primary Examiner.lames W. Lawrence Assistant ExaminerPalmer C. Demeo AttorneySmyth, Roston & Pavitt ABSTRACT: A thyratron has a tubular cathode, preferably made of rhenium and circumscribed by a grid which in turn is circumscribed by an anode tube. The thyratron cavity is filled, e.g., with vapor pressure regulated cesium. The operating parameters can be chosen to establish essentially zero voltage drop across the thyratron. The cathode cavity is contiguous with a cavity of a thermionic converter.

74' erna/fire yy fade fn/t/F/tC/fd/t 3 Sheets-Sheet z Patented May 18, 1971 ZERO ARC DROP THYRATRON The present invention relates to a thyratron for high temperature operation and is adapted for zero voltage drop between anode and cathode electrodes. One of the principle operating features of thyratrons is a high thermionic emission of the cathode, but low thermionic emission of the anode. The high thermionic emission of the cathode can best be obtained by operating the cathode at a high temperature and by selecting a material which the so-called work function is very low. High temperature and low work function ensures high thermionic emission of any particular material where Richardsons law is applicable.

A low cathode work function and a high anode work function makes it inherently impossible to operate a thyratron with zero voltage drop between cathode and anode electrodes because the polarity of the difference of the work function is then such that the internal voltage drops in the thyratron have the same sign and thus add to the work function difference, In order to obtain zero voltage drop between anode and cathode of a thyratron it is, therefore, essential to have the reverse relationship of work functions, i.e., the cathode work function must be high and the anode work function must be low. From the standpoint of thermionic emission this appears to be an undesirable condition for thyratron operation. Of course, a suitable temperature differential between anode and cathode can still produce a high thermionic emission of the cathode and a low thermionic emission of the anode even though the work functions are selected that they do not aid in this relationship.

It has been found now that thermionic emission of a material with a high work function can be increased (for any given temperature) if the gas in the discharge chamber of the thyratron has a very low work function and a low ionization voltage and, therefore, adheres to the cathode even at cathode temperatures well above the boiling point of the gas. As a result of this adhesion the effective thermionic emission of the cathode is materially increased, In such a situation the thermionic emissions of cathode and anode are not determined any more in the simple manner following the Richardsons Law, but the gas or vapor modifies the thermionic emission of anode and cathode to a substantial degree and in proportion to the adhesibility of the gas or vapor molecules, which characteristics is in turn predominantly determined by the work function difference of the cathode material on one hand, and of the gas orvapor, on the other hand.

In the following it appears to be convenient to speak of the work function of a gas or vapor even though such terminology is not exactly correct. What is meant then is the work function of that material when in the liquid or solid state. Even though vapor and gas molecules adhere to a surface such as a cathode surface having a temperature well above the boiling point of that gas or vapor material, no true liquid or solid state of that gas or vapor material is established at the hot surface, but the adhering molecules define a surface strata which modifies the thermionic emission of the substrata and the work function thereof. Only when the surface has obtained substantial dimensions at rather low temperatures, then the work function of the strata itself is substantially that of the vapor or gas material when in the liquid or solid state.

If the anode is kept relatively cool its effective work function will be superseded by the bulk work function of the vapor material allowed to precipitate to a material extent on the anode. If, in addition, the vapor has a low ionization voltage, the total internal voltage drop in the thyratron can in fact be reduced to the effective work function difference between anode and cathode. And the result is a zero or substantial zero voltage drop measured externally between anode and cathode of the thyratron.

In particular the internal voltage drop in the thyratron depends on the plasma drop and the voltage drops in anode and cathode sheaths. These latter sheath drops in turn depend on the ionization voltage of the vapor which has to be rather low. Cesium and rubidium are highly beneficial for this purpose, but all alkaline vapors can actually be used, at least to the extent that the voltage drop across the thyratron can be made rather low.

The plasma drop depends primarily on the gas pressure and the electrode spacing in accordance with Paschens pd law for established plasmas. A separate component of the plasma drop is the grid drop, which is usually considered separately. This grip drop results from the constriction of the electric current path between anode and cathode, by the grid of the thyratron. It has been found possible to select the aperture of the grid wide enough without losing control so that that grid voltage is indeed also small. An important aspect is that for establishing zero voltage conditions, the cathode work function itself, and even the difference between cathode and anode work functions is the largest single potential difference in the entire anode-plasma-cathode system and enters into the consideration as modified by the adhesion of the plasma atoms such as cesium atoms and thus becomes subject to control and selection in accordance with existing operating conditions. It follows that the cathode work function must be relatively high. A high work function is a work function having value comparable with the work functions of rhenium, molybdenum, tantalum, etc. Cesium and rubidium have typically low work functions.

A large temperature differential between anode and cathode is beneficial to the operation of the thyratron. A high cathode temperature can, e.g., be obtained when the cathode is a cylindrical tube with grid and anode being concentrically disposed about the cathode tube. The interior of the cathode tube fonns a radiation cavity which is exposed, e.g., to thermal radiation. In the preferred form of practicing the invention, this cavity is an extension of a cavity defined by thermionic converters which convert thermal radiation energy into the electrical energy used for powering the thyratron or several thereof. This kind of arrangement is, e.g., highly useful as a power supply source in a space vehicle in which reflectors direct solar radiation into the above defined cavities for operating the thermionic converters and thermally biasing and energizing the thyratrons. These thyratrons may then pertain to an inverter which converts DC electric energy as provided by the thermionic generators into a suitable AC voltage and current.

While this specification concludes with claims particularly pointing out and distinctly claiming the subject matter which is regarded as the invention, it is believed that the invention, the objects and features of the invention and further objects, features and advantages thereof will be better understood from the following description taken in connection with the accompanying drawings in which:

FIG. 1 illustrates somewhat schematically a power supply system;

FIG. 2 illustrates in perspective view, partially broken open with section view of a thyratron;

FIGS. 3a, 3b and 3c illustrate relevant characteristics for the thyratron shown in FIG. 2;

FIG. 4 illustrates a top view of the thyratron shown in FIG. 2; and

FIG. 5 illustrates a section view of a different thyratron in accordance with the present invention.

Proceeding now to a detailed description of the drawing in FIG. 1 thereof, there is shown the general layout of the novel conversion system, particularly the high temperature section thereof. As was stated above, this system may serve as a power supply system for a space vehicle using solar energy. Radiation enters the system, in the drawing, from the left. This radiation may have been focused by a large concave reflector (not shown) which reflector is oriented towards the sun to receive solar radiation. That reflector may be positioned to the right of the system as illustrated in FIG. 1. A second smaller reflector, also not shown, may be disposed in the focal area of the larger and first mentioned reflector to direct light now as in dicated as radiation particularly towards and into a cavity 15 of a thermionic conversion system 10.

The cavity 15 of the system is coaxial with a second cavity pertaining to two coaxially positioned, ring- shaped thyratrons 20 and 30. The thyratrons pertain to an electric inverter which includes a transformer 40. The secondary winding 41 of the transformer leads to what can be described as a high temperature-low temperature interface 50 separating the high temperature unit 10-20-30-40 from other elements placed at a sufficient distance therefrom to be maintained safely at lower temperatures.

The low temperature unit may include particularly a rectifier 51 and all circuit elements driven by the output of the rectifier 51. The rectifier completes the system as a DC-AC-DC conversion system. The circuit elements connected to the rectifier 51 are in particular all instruments and electrically powered components in the space vehicle to which this unit pertains. These other elements are summarily denoted with numeral 52 and do not pertain to the invention proper. The low temperature unit includes also a circuit network 60 for driving the grids of the thyratrons. This circuit network 60 includes a voltage regulator 61 connected to the rectifier 51 and driving an astable multivibrator 62 which in turn drives a switching circuit such as a bistable multivibrator 63. lndividual output amplifier stages 64 and 65 lead from the multivibrator 63 to the grids of the thyratrons 20 and 30 through suitable connections. Since the grid current can be kept quite low the circuit connections between the control circuits and the grids of the thyratrons do not impose any weight problem nor are there considerable losses. In addition these wire connections can be included in the series impedance for the grid circuit which is usually necessary for the thyratron grids.

The thermionic converters are not themselves subject matter of the present invention, but they are the principal electric power source and shall thus be described briefly. There are four units 11, l2, l3 and 14, (not shown), each having an emitter and the four emitters are arranged so that their outer contour defines the cavity 15. Thus, the four converter units are radially disposed around the cavity and their respective collectors face in outward direction.

The converter units are electrically connected in series which connection includes the connectors 16 and 17 illustrated extending between the collector of converter 11 and emitter of converter 12 (connector 16), and between collector of converter 12 and emitter of converter 13 (connector 17). The radiation boils electrons from the surface of the emitters in the interior of the converters; the emitters are made of, e.g., rhenium. These electrons are collected by the respective collectors, lowering the potential thereof. Each converter unit will yield a voltage of about 0.04 to 0.7 volts so that the total yields of the four units when connected in series may be 1.5 to 3 volts at about 50 amperes. This is in the order of 10 watts. The temperature of the emitter electrodes and thus of the cavity 15 is raised up to 1700 C. or above. That temperature is needed to operate the thermionic converters at sufficient efficiency. Of course, the cavity 15 communicates with the vacuum of outer space. The system has two output bus bars, one of them is connector 18 and leads from the collector of converter 13 to the center tap of the primary winding 42 of transformer 40. The other output bus bar is not shown, but it leads from the four converter (14) to the cathode of the thyratrons 20 and 30.

These thermionic converters yield only a rather low voltage at high current. ln'order to make efficient use of this electrical energy it is essential that it be converted into a high voltage at correspondingly lower current in order to avoid long and thick connectors for a high current. Such connectors could be heavy, or, if thin, they would be lossy. Both features are undesirable in space vehicles where weight must be kept low and electrical energy must not be wasted. Thus, the electric conversion DC- AC unit 20, and 40 should be very closely positioned to the thermionic conversion unit 10 producing the high current-low voltage DC in order to avoid long output connectors for the thermionic unit. It follows, therefore, that the electrical conversion unit must be operated at high temperatures.

FIG. 2 illustrates a first embodiment for the two thyratrons designed to operate at high temperatures, whereby the high operating temperature is of advantage for efficient thyratron operation. The two thyratrons are constructed similarly so that details are shown only for one thyratron of this particular embodiment.

The thyratron 20 is comprised of a cylindrical or cathode tube 21 from which extend two supporting rings or annuli 211 and 213 having respectively two ring-shaped grooves to provide heat chokes 212 and 214. The two rings 21] and 213 serve as mounting media for the other electrodes, and the heat chokes 212 and 214 provide some thennal insulation between the cathode proper and the portions of rings 211 and 214 used for mounting the other electrodes.

The cathode tube with annuli could be made of molybdenum or tungsten, tantalum or niobium; preferably, however, it is made of rhenium. The cathode tube 21 with annuli 211 and 213 defines an annulus having a double U-shaped cross section wherein the entire interior of the U is a ring-shaped discharge chamber of the thyratron which, by itself, is an annulus. The principal cylindrically-shaped surface of the cathode emitting electrons is denoted with reference numeral 215. The emission occurs thus principally in radial, outward direction.

The anode of the thyratron is formed by an annulus 22 having a T-shaped cross section whereby the crossbar of the T defines the tubular-shaped anode proper, 221. The stem of the T alternates in length around the circumference of the annulus 22 to define a plurality of, e.g., three radiator vanes 222, 223 and 224. These vanes are the single cooling means for the anode.

The structure should be mounted to the space vehicle so that the radiators are exposed directly to the vacuum of outer space and so that other parts of the vehicles are not exposed to the thermal radiation emanating from the radiators nor should conduction be possible to the low temperature section of the vehicle. Cooling occurs thus solely by the emission of radiation. It will be appreciated that this provides for the hardest conceivable environment for operation of the thyratron. In a different environment a more efficient cooling may be available so that the operation can be improved to that extent.

The anode is made of any of the materials mentioned above for use as cathodes. For reasons below the anode material is of lesser importance. Two flat supporting rings or annuli 225 and 226 are hermetically joined to the crossbar of the anode T and extend parallel to the stem thereof as well as to the radiators 222, etc., and form sealing surfaces for joining the anode to the insulator rings 252, 253.

The grid of the thyratron is composed of two separate annuli 23 and 24, each of them having a double L-shaped cross section. The long leg of each L is a flat ring structure, 234 and 244 respectively. From each ring there extend, in the same direction, three radiator vanes for cooling the grid. Thus, there are three radiator vanes 231, 232 and 233 respectively, extending from the ring 234 of the grid element 23 and three corresponding radiator vanes 241, 242 and 243, respectively, extend from ring 244 of the element 24. The short legs of the two L's define two coaxial grid tubes 235 and 245 respectively and each having narrow axial dimensions. The two axial grid tubes 235 and 245 are positioned to define a circumferential passage 25 between cathode surface 215 and anode surface 221.

The grid ring 234 is sandwiched in between two ceramic rings 251 and 252 by means of which this grid element 23 is mounted in between the two annuli 211 and 225, respectively, pertaining to cathode and anode structure. Analogously the grid ring 244 is sandwiched in between two ceramic rings 253 and 254, by means of which this second grid element 24 is mounted in between the annuli 213 and 226, respectively, pertaining to cathode and anode structure. The rings 25] to 254 can be made of any insulating material capable of withstanding the high operating temperatures contemplated; ceramic is the most suitable material here.

If space permits, a heat shield is interposed between the cathode and the grid. The heat shield is composed of two cylindrical or tubular elements 261 and 262. The aperture space defined between them may be somewhat narrower than the grid gap 25, to provide sufficient shielding for the grid and for the anode from the thermal radiation emanating particularly from the cathode. The shield elements can be made of any material capable of withstanding the high temperature encountered in the thyratron, and, further, they must be capable of bonding to the rings 211 and 213. In order to eliminate the latter problem, the shield elements can thus be made of the same material chosen for the cathode.

The interior space of the thyratron is filled with cesium vapor. There is provided a cesium reservoir 27 which communicates with the interior of the thyratron through a bore 271 in one of the anode heat dissipaters e.g., the radiator 224. The bore 271 may be a capillary so that the cesium in the liquid state cannot flow into the interior of the thyratron. Thus, the reservoir is of the O-g type. The reservoir 27 is located sufficiently far from any of the surfaces facing the interior of the thyratron, so that the temperature of the reservoir can be kept well below the boiling point of cesium, which is 670 C. for atmospheric pressure. The outer portions of the radiator vanes are sufficiently cool so that the reservoir can be kept at 300 C. for which a very low pressure equilibrium can be established in the interior of the thyratron.

Even though the thyratron chamber proper has rather hot walls, the pressure in the interior of the thyratron is determined by the coolest point in communication with the vapor and this is the reservoir. Such a low temperature is needed because the pressure in the thyratron is well below the atmospheric pressure under the desired operating conditions. For reasons below, the vapor pressure is in the range of 1 Torr or below. For such low vapor pressures, very little heat is transferred by and through the vapor. A quantitative analysis revealed that the vapor participates only very little in the heat transport from the cathode to the other electrodes so that the heat content of the vapor is also very low. The temperature of the reservoir is such that a heater 272 for the reservoir can be used to regulate the vapor pressure in the thyratron.

The thyratrons and 30 have the following electric circuit connections. The anode output lead is a bar 281 which is welded or brazed to one of the anode radiators, e.g., radiator 224. The bar 281 leads to one side of the primary 42 of the transformer 40. The cathode tube 21 of the thyratron is in intimate contact with or even integral with the cathode tube 31 of the second thyratron needed for the contemplates inverter operation. Thus, the two cathodes are connected to have common potential. As was said above, the two cathode tubes of the two thyratrons are connected to the one output bus from the thermionic converter, particularly the collector of one of the converter units thereof. Suitable high temperature connectors (not shown) provide further connection between the grids of the cathodes of the thyratrons on one hand, and the low temperature grid control device 60, on the other hand. Thyratron 30, which is constructed similar to the thyratron 20, has an anode output bus 381 which leads to the other side of the primary 42 of transformer 40.

As far as the operation of an individual thyratron is concerned, the high temperature environment sets the basic operating conditions which in turn affords the possibility of providing zero drop across anode and cathode electrodes. Turning now to FIG. 3a there is shown the voltage and postulated potential distribution between cathode and anode during operation, i.e., after firing of the thyratron. The individual voltages and potential differences or voltage drops shall be considered qualitatively at first; I is the work function of the cathode and is equal to the difference in the negative potential of an electron after having left the cathode and the Fermi level thereof. Since an electron loses energy when escaping the cathode the potential of the electron having escaped is negative with respect to the potential of the cathode at Fermi level. For purposes of reference the cathode potential can be considered equal to the Fermi level. V, is the voltage drop in the Langmuir or cathode sheath accelerating an electron and raising the potential thereof when passing through. The sheath drop depends on the ionization voltage of the vapor which presently is cesium.

V, is the plasma drop which is comparatively small and depends on the vapor pressure and the distance between cathode and anode in the thyratron. Somewhat larger is the voltage drop V across the grid aperture and resulting from the constriction of the discharge path between anode and cathode by the grid. The plasma drop actually occurs at both sides of the grid and thus is divided into two portions, but should be considered together as it is impractical to distinguish between an anode side and a cathode side plasma portion. Therefore, the total plasma drop V as shown in FIG. 30 includes the entire plasma at either side of the grid.

The plasma is essentially macroscopically neutral, i.e., it has not net space charge. Depending now upon the electric current drawn from the anode during operation, the anode surface may have a positive or a negative space charge sheath or the plasma may extend all the way to the anode as space charge free region. When the electric current actually flowing is high there will be a depletion of electrons near the anode resulting in a positive space charge and a corresponding increase in potential of the anode, as it is then necessary to pull electrons towards the anode and out of the plasma; this is normally observed in thyratrons.

When the electric current is low there will be also a negative space charge in the anode sheath because electrons resulting from the thermionic emission by the anode are added to the plasma electrons, and the required current does not deplete this space charge. This particular situation will be prevalent if, as here, the anode surface is large, particularly if it is larger than the cathode surface resulting in a lower current density for the anode than for the cathode. Thus, the sheath drops V and V may be of opposite polarity. In particular, electrons passing from the plasma to the anode may have to perform work to reach the anode through the anode sheath, and therefore, they lose potential in accordance with the value V,, providing the current through the thyratron is rather low. Finally, a l00 is the work function of the anode.

For establishing zero drop conditions the following relation is to be fulfilled: P '-I V,,+V,,+V V,,)=V approximately =0, wherein V, is the effective cathode-anode voltage drop for the general case. In this equation, absolute values have been assumed for the several voltages and the symbol i indicates that the anode sheath drop may have the same or the opposite direction as the other internal voltage drops in the thyratron combined in parenthesis. The problem is now to adjust the values of the several voltage drops in the equation so that V, is at least substantially equal to zero.

Even though the equation above has quite a number of components as, so to speak, variables, there are several operating factors for the thyratron which determine most of these components regardless of the desire to produce zero drop across the electrodes. For conventional thyratrons it is common to select a low cathode work function of a few volts so that in accordance with Richardsons Law the current density can be very high. On the other hand, the ionization voltage of inert gases used commonly for thyratrons, e.g., xenon or argon and other noble gases, is about 10 volts or even higher, as is the ionization voltage of mercury. The cathode sheath drop V, is always somewhat below the ionization voltage but still rather high for these gases commonly used for thyratrons. Thus, for conventional thyratrons D -V, is already a negative value. This precludes zero drop conditions. Moreover, for heavy duty, V,, is negative also and 1 may even be larger than 1 so that it is absolutely impossible to arrive at V =0.

The result is different if one uses an alkaline metal vapor, preferably of higher atomic weight, such as rubidium or cesium, with cesium offering best performance. The reason why this is so shall be developed later. The equation above can be interpreted in this manner. The work functions 1 and In, are

to be such that their difference at least approximately equals the total voltage drop inside of the thyratron and expressed by V,,+V,,,,,+V,V,,. This total voltage drop is to be made as small as possible, because the available spread of work functions is, in general, not very large to form an appreciable difference. Thus, the individual components shall be made as small as possible to establish a small total interior drop in the thyratron because the first three components all have the same sign. However, the anode sheath drop should be made to correspond to a negative space charge and thus have an opposite sign when compared with the cathode sheath drop V,.

The two components V, and V are, of course, closely related as they both depend on the material chosen for the vapor. Cesium has the lowest ionization voltage among the alkali metals and the cathode sheath drop, which is the dominating drop, is accordingly the lowest for cesium vapor. Cesium is thus the preferred choice. As stated above, the current through the anode determines the extent to which the plasma adjacent the anode is depleted from electrons. In view of the ring configuration of the electrodes the discharge path extends radially from the cathode so that the current density in the anode is lower than in the cathode, even though the plasma develops fully also adjacent the anode. In the embodiment shown in FIG. 2 the grid has no grid bars in axial direction; thus, the distinction between a constricted and a spread discharge need not be made in this embodiment as the discharge is necessarily a spread one. Therefore, the plasma is developed all around the anode and the anode participates fully in the current conduction.

Presently a low anode current density can be realized due to a spread discharge using an anode area which is actually larger than the cathode area, so that nowhere near the anode is the plasma depleted of electrons. It can thus be seen that anode and cathode sheath drops are, in fact, oppositely oriented, so that the total sheath drop V +V,, can actually be made smaller than the cathode sheath drop alone. For cesium the cathode sheath drop is between about 0.6 and 1.0 volts, and the total sheath drop under the outlined conditions can thus be made to be 1.5 volts. However, this may vary with the load ultimately connected to this power supply system. How this variation could be counteracted will be also described below.

The plasma drop V is very low, or better, can be made very low if the pd" law is observed. The are drop or plasma drop for a particular material is uniquely related to the product of the vapor pressure p and the anode-cathode distance d. Even though the minimum of the characteristics is not very pronounced, it strongly suggests to construct the thyratron so that the product ofa p and dis in a range of 10 and 10 Torr mils, to obtain a small arc drop, for example, of the order of about 0.1 volts. It should be pointed out, however, that in practice the plasma drop V, cannot be measured, per se, as measuring requires the introduction of probes inherently providing incorrect measuring results. In particular, absolute values for the sheath drops are not independently measurable from the plasma drop and vice versa. One knows, however, the slope of the plasma drop so that a numerical approximation of the several values is not pure speculation. However, the region of a minimum plasma drop as a relative value can be ascertained by measuring the plasma drop with probes, or even the electrodes themselves. For cesium, the minimum plasma drop is at about 60Torr mils. For obtaining the minimum plasma drop the values for pressure and cathodeanode spacing have to be paired accordingly, regardless of the actual value for the plasma drop.

For structural reasons there is a practical minimum for the electrode distance, even though miniaturization techniques are well developed in this field. There is, however, a very real limitation as to the vapor pressure. The deionization time of any vapor is a very important consideration for operating a thyratron as a valve, because the plasma must have decayed before the voltage reversal across anode and cathode reaches appreciable amplitude so that the arc cannot reignite (backfiring).

The permissible deionization time is, therefore, determined by the frequency of the voltage applied to the thyratron or by the frequency with which the thyratron inverter is operated as an inverter. For 400 c.p.s. the plasma will decay and deionize in between successive half waves if the vapor pressure is maintained below about 1 Torr. It will be developed more fully below that, on the other hand, the thermionic emission, i.e., the cathode current density can be increased with vapor pressure, so that two opposing conditions limit the choice of the vapor pressure. The pressure should be as high as possible to achieve maximum electron emission while still being low enough so that the deionization time is short enough to meet the operating frequency requirements. Thus, a pressure of about 1 Torr or of that order of magnitude can be regarded as a more or less fixed parameter.

The recovery time or plasma decay time depends also on the electrode spacing, i.e., it increases with spacing. This is a further condition for a small electrode spacing. Operating in, approximately at least, the minimum arc drop region in accordance with the pd law" is compatible with this condition. Particularly an electrode spacing of the order of 60 mils (at 1 Torr pressure) is consistent with the desire to establish a plasma decay time sufficiently short for inverter operation in a technically desired frequency range (for example, 400 c.p.s.

It has been proven difficult, however, to provide such a small distance when this anode-cathode space has to accommodate both grid and heat shield. This can be circumvented by lowering somewhat the vapor pressure and choosing a somewhat larger electrode spacing accordingly. If only the latter is being done, then the plasma drop might increase somewhat but increase is quite small. However, with reference to FIG. 5 it shall be discussed that a modified structure does not require a heat shield and, for that reason, may actually be preferred even though posing other problems which shall be discussed more fully below. In conclusion, it is quite possible to construct and to operate the thyratron to establish minimum plasma drop conditions.

Next, the grid drop V has to be determined. It is this a resistance which, as stated above, occurs actually in the plasma itself, but which is not regarded in the usual sense as part of the plasma drop because it results from a constriction of the plasma by the grip aperture in comparison with the cross section of an electric current path between anode and cathode which would be available in the absence of the grid. The constriction is necessary so that the grid can exercise control over the occurrence (or inhibition) of the thyratron discharge. This, of course, establishes a tendency towards selecting very small grid apertures. On the other hand, a small grid drop requires a large grip aperture. Certainty of control is, of course, the predominant factor here so that the control conditions must exist without marginal design. A 50 percent grid aperture, referenced against the anode area size or thereabouts, is a workable size for a grid aperture and for providing a grid drop in the plasma of about 0. l to 0.4 volt.

It follows from the foregoing that the total drop V +V,, +V +V ranges between .5 and 1.60 volts. That total value is measurable and its value is, in summary, determined as follows: The anode-cathode spacing and the vapor pressure are chosen to permit sufficient deionization and to operate at or near the minimum voltage drop in accordance with the pd law for plasmas. The choice of the vapor is dictated by the requirement for a small ionization voltage. All this combined permits operation with a total internal voltage drop of 0.5 to 1.5 volts or thereabouts.

The cathode and anode work functions must now be selected so that their difference equals a value in that range. As stated above, the cathode should be made preferably of rhenium, having a rather high work function of about 4.9 volts.

A work function of 4.9 volts places rhenium actually into the class of the rather poor electron emitters, particularly if compared with other materials used as cathodes conventionally and when operated at comparably sir nilar temperatures. Moreover, the reason why rhenium is better in the present case than, e.g., molybdenum is that rhenium has a higher work function than molybdenum, which on its face appears to be a contradiction as far as cathode operation is concerned. However, the controlling factor for cathode emission of the thyratron in accordance with the present invention is the interaction between the cathode material or cathode substrate and the vapor in the thyratron cavity. The vapor has been chosen to have a very low work function and a very low ionization voltage for establishing low sheaths drops. That interaction between a high work function substrate and low work function-low ionization voltage vapor is crucial for the inventive thyratron and will be explained next.

The bulk work function of cesium is about 1.8 volts when in the solid state and somewhat less when in the liquid statev The cathode has a temperature well above the boiling point of cesium so that no precipitation of cesium on the cathode surface can be expected. n the other hand, the difference in work functions causes cesium molecules to adhere to the cathode surface. This adhesion, therefore, is the result of interaction between the individual cesium molecules and the particular cathode surface, and any cesium layer resulting therefrom is not the result of cohesion between cesium molecules. Of course, due to the high temperature of the cathode surface cesium atoms will boil away, but new cesium atoms will continuously be caught by the surface for deposition and adhesion thereon. Thus there is a dynamic equilibrium between the adhesion process and the vaporization resulting in a thin cesium layer on the cathode. The equilibrium conditions, i.e., the thickness of such layer is dependent on the vapor pressure and on the cathode temperature. The higher the pressure, the more molecules will adhere, but for increasing cathode temperatures the layer will become thinner. It has been found also that this adhesion is the stronger the larger the difference between the work functions of the two materials. Since, however, the work function of a gas has no direct meaning, it may be more accurate to say that the ionization voltage of the gas or vapor should be rather low, because the ionization voltage of a gas molecule and the work function of a solid or liquid both represent the ease with which an electron can separate from the material, i.e., from the individual molecule in case of a gas or from the bulk material when in the solid or liquid state. I

The adhesion, as described in the previous paragraph, in turn modifies the effective work function of the cathode. FIG. 3b shows this by way of example. The group of upwardly converging curves including, e.g., the curves A and B, represent Richardsons Law for different work functions, which is one of the parameters in the Richardson equation. The ordinate shows the logarithm of the emission current density and the abscissa shows the inverse of the temperature T in degrees Kelvin of the emitting surface, the scale being drawn to values of l000/T. Richardsons Law has a proportionality factor as a second parameter which is the same for all cathode materials of interest. However, this proportionality factor is different for cesium and the result thereof will be discussed more fully below.

Curve a now is a measured curve and represents the measured emission current density and dependence upon temperature for a particular vapor pressure and provided that the cathode material, i.e., the cathode substrate upon which cesium molecules may deposit is rhenium. This curve a, does not follow the relation given by Richardson's Law, i.e., it does not fit into the group of upwardly converging Richardson curves. Instead, it shows a different relationship qualitatively explained best as follows.

For very high temperatures (to the left of the abscissa) e.g., for 2500 Kelvin or thereabouts, very few cesium molecules will adhere to the cathode. Therefore, the cathode is, as far as electron emission is concerned, modified very little, and the current emission follows the Richardson curve for rhenium rather closely and according to which the current increases with increasing temperatures in a monotonic relationship.

Thus, curve a, approaches asymptotically the Richardson curve for rhenium. For decreasing cathode temperatures cesium molecules adhere to the cathode to an increasing extent, and they shield to some extent the otherwise exposed cathode surface, so that the potential barrier of the solid cathode normally retarding electron escapement from the cathode is lowered and the current density decreases with temperature at a rate less than in accordance with Richardsons Law.

For still lower temperatures, say between 2000 Kelvin and 1500 Kelvin the rate of adhesion of cesium goes up to such an extent that the emission current density now actually increases with decreasing temperature. Ultimately, for rather low cathode surface temperatures, adhesion of cesium establishes a thick layer of cesium which then determines the electron emission with little modification from the cathode substrate. Accordingly, the emission must approach asymptotically the Richardson curve for bulk cesium, i.e., for cesium in the liquid state. This is to be expected, particularly when the cathode temperatures approach sufficiently low values with corresponding low values for cesium vapor pressure where the surface coverage of the cathode by cesium is substantially complete. Such surface coverage is determined by the equation:

, where: 1 arrival rate of cesium atoms from the vapor n vapor density 6 average atom velocity in vapor, and

u. evaporation rate from substrate at given coverage and temperature conditions The influence of the cathode substrate material on the electron emission will then be negligible.

Thus, for very low temperatures the Richardson curve for bulk cesium is, in fact, the controlling characteristic for thermionic emission. That particular Richardson curve for cesium is, however, not one of the Richardson curves plotted directly in FIG. 3b because, as was already mentioned above, the Richardson emission curve for cesium has a lower proportionality factor than the high refractory cathode materials contemplated for use as a cathode substrate, so that in FIG. 3b the low temperature asymptotic line for the curve a is a curve precisely parallel to the plotted Richardson curve having the same work function parameter (1.8 volts) as cesium but being shifted down therefrom. In between the branch of curve a where the emission current 'density increases with decreasing temperature and the low temperature asymptotic portion of curve 0 there is a maximum.

For any given cathode temperature and for a particular vapor pressure there is now, in accordance with curve a a particular emission current density which is higher than for the substrate-cathode material (e.g., rhenium) if it had not the cesium layer. The reason for this modification is twofold. First, due to the cesium layer, the surface of the cathode itself is not exposed or is very little exposed directly to the environment so that the cesium molecules shield the cathode thereby lessening the potential barrier of the cathode substrate boundary. The magnitude of this shielding or reduction in effective surface work function is a function of the substrate temperature for a given cesium vapor pressure. Each point on a, therefore is the measured emission of the substrate-cesium combination. The intersection of a with the family of 'curves A through B yields the work function of the combination. The maximum in the curve indicates that there is an optimum coverage of the surface for maximum emission. By varying the substrate temperature, the emission can be varied anywhere between that for the bare substrate through the emission maximum to an emission level corresponding to bulk cesium. For example, in FIG. 3b a cathode temperature of 1277" C. the

current emission is such (in accordance with curve 11,) that it corresponds to a work function of 2.3 volts, which is the actual or effective work function of the cathode at that temperature.

The curves a and a; show similar characteristics respectively for higher and lower vapor pressures but still with rhenium as substrate and cesium being the vapor. The relationship between curves a and a verifies what was mentioned above, viz, that the emission current density can be increased with vapor pressure, and it will be recalled that the required deionization time limits the choice of too high a vapor pressure in the thyratron. It will be discussed more fully below how this afiects the choice of the operating parameter for the thyratron.

The curve b shows the characteristics for cesium vapor at the same temperature chosen in case for the curve 0,, but with a molybdenum cathode as substrate. One can see that higher emission current densities for similar cathode temperatures and vapor pressures can be obtained when rhenium rather than molybdenum is used as a cathode substrate. Nevertheless, the phenomenon of producing reduced effective work functions and higher thennionic emission is present in either case. In particular now we observe the inverse relationship for a considerable range of temperatures. Molybdenum actually has a lower work function and therefore its emission is higher than the emission of rhenium itself, but the higher work function of rhenium is instrumental in providing a larger attraction for the cesium vapor molecules than molybdenum does, so that for similar temperatures and vapor pressure the effective thermionic emission is ultimately larger for a rhenium substrate than for a molybdenum substrate over a wide range of temperatures. However, it should be repeated that the phenomenon of producing an increased thermionic emission is present in either case. In other words, using cesium vapor improves the thermionic emission of both rhenium and molybdenum but the improvement is better for rhenium than for molybdenum.

To summarize the effect of the vapor, each of the work function parameter curves such as a,, a b etc., has a high temperature value asymptotic line which is the Richardson curve for the substrate material of the cathode. The asymptotic line is, therefore, the same for the curves a and a i.e., it

. is the rhenium curve. For curve b the asymptotic line is the Richardson curve. for molybdenum. Each of these work function parameter curves 0,, a b etc., passes through a minimum, increases again towards a maximum for a particular temperature, and all these curves drop asymptotically towards the particular Richardson curve for bulk cesium, this relationship being explained here for declining temperatures. Thus, the low temperature asymptotic line is similar to all of these several curves and does not depend any more on the substrate. The degree of independence is temperature dependent but qualitatively the phenomenon is similar in all cases.

One can see, that over a wide range oftemperatures the current emission depends very little on the work function of the cathode substrate except that the work function controls the process of adhesion. The vapor temperature and, more important, a rather large work function difference between cathode substrate and vapor material are the dominating factors for the high current yield. The temperature regions between minima and maxima of the work function parameter curves 0,, a b etc., are the principal regions of interest here because there are available current densities at temperatures much below the temperature for which the cathode substrate material alone would produce comparable thermionic emission. The value of the cathode work function itself is of immediate importance for the efficiency of the thyratron as proper choice here permits the establishing of zero drop conditions. For thyratron efficiency however, the current density is also of some importance but only to the extent that too big a cathode surface, i.e., too big a thyratron construction is undesirable in general. In the usual case where adhesion does not modify the thermionic emission the work function of the chosen material for the cathode is a fixed parameter, and the current density can be varied only with the cathode temperature with the higher density requiring higher temperatures throughout. The presently observed interaction between vapor and cathode material modifies the simple Richardson relation, in that the work function becomes a variable operating parameter and both a high thermionic emission and a zero drop condition can be satisfied.

The selection of the operating parameters to be used yields a considerable variety of choices. From the equation above and from FIG. 3a, one can see that zero drop conditions require that the anode work function be below the cathode work function. By definition the anode electron emissivity for its operating temperature must be considerably below the corresponding value for the cathode at its temperature, otherwise no proper electric valve operation were possible. Normally, a lower work function means higher emissivity except for a drastic temperature difi'erential. However, it is beneficial that for low temperatures of a. surface exposed to cesium the emissivity is determined by the low temperature asymptotic line of the Richardson curve for cesium as stated. Even though the effective work function of such a surface will be rather low, one can see that this is not too critical because the choice of a rather low temperature for the anode then benefits from the fact that the proportionality factor for the cesium Richardson curve is lower than the proportionality factor of the higher refractory cathode materials and this, in turn, results in a very low anode emissivity in spite of its low work function. Thus, the effective anode work function will, in fact, be about 1.8 volts, and the anode must cooled to the extent that this work function and the corresponding low emissivity of such a cesiated anode can be realized which, in turn, makes the anode work function independent from the anode substrate and fixes it as a parameter in the equation written above for defining the condition to establish zero drop across the thyratron.

Here now we turn to the previous result, viz, that for zero drop conditions the anode-cathode work function differential D should be between 0.5 and 1.5 volts. This value was determined as a result of summing the various internal drops in the thyratron and it is the sum V,,+V,V +V For zero drop condition that sum has to be equal to the work function differential. In the previous paragraph we have determined the anode work function to be 1.8 volts as a result of the choice of cesium for the vapor material and the anode work function is substantially independent from the anode substrate. It follows, therefore, that the cathode work function has to be a particular value chosen from within the range of 2.3 and 3.3 volts in order to establish zero drop conditions and the particular choice, of course, depends on the ultimately resulting fixed value for the sum of the several internal drops.

The parameters to be considered now are, therefore, cathode work function, cathode temperature, density of the cathode emission and vapor pressure. The choice of particular values for any two of the four parameters fixes the values of the other two. The choice of the variables depends on the requirements and on the desired particulars. The choice is limited, however, by certain constraints, partially discussed above and briefly summarized as follows: The selection of the cathode work function, for zero drop conditions, is particularly critical. The cathode temperature will be as high as possible, as one can see from the general nature of the curves in FIG. 3b, but, of course, the temperature is limited by the amount of available thermal energy, available conditions for cooling and resistance of the various components including those in the environment against thermal destruction. The current density should be rather high for reasons of efficiency, i.e., a set of operating parameters resulting in a very low current density is of little practical value. The vapor pressure should be high, as follows from the curves of FIG. 3b because there is a direct relation between high current density and high vapor pressure. However, deionization time requirement is severe constraint on the vapor pressure, as is the observation of the pd law because too high a vapor pressure may result in impractically close electrode spacing.

For zero drop conditions the cathode work function is an' ternal drop of 0.5 volt can reasonably be expected. A cathode temperature of l320 C. and a current density of about 12 amp/cm are satisfactory values. The anode temperature being low as above-described, for example, 708 C. The resulting cesium pressure is above 2 Torrs at a reservoir temperature of 289 C. FIG. 30 illustrates the terminal voltage between anode and cathode as a function of interelectrode spacing for these parameters.

For a value of 60 mils this voltage is, in fact, zero. For smaller distances the system actually generates voltage rather than providing loss. There is a second value, about 3.5 mils where the voltage drop is zero, but this is of little practical significance. The provision of a grid producing a voltage drop V, will, in effect, shift the curve in down direction and for the value of the' grid drop V,, leading to smaller electrode spacing values. Selection of a smaller vapor pressure permits an increase of the electrode spacing in accordance with the pd law. It is beneficial that a considerable drop in pressure is accompanied by only a modest drop in current density. Modifying somewhat the cathode temperature may lead to a different work function for more conveniently providing zero drop conditions, should the grid drop prove too large. A larger cathode work function operates in the diagram of FIG. as shift of the curve in up direction.

As a representative example, for a thyratron it was found highly suitable to use a current density for about l0 amperes per square centimeter. For about 50 percent grid aperture and a load current which does not deplete the plasma with electrons adjacent the anode, the internal drop, i.e., the required work function differential can be made as low as 0.5 volts. This results in a cathode work function of 2.3 volts and fixes the operating temperature at l277 for a cesium vapor pressure of about 1 Torr, and an electrode spacing of about 60 mils. The operating conditions thereby establish assured satisfactory performance and at a cathode temperature which is not too high and can be realized, particularly in the combination with a converter shown in FIG. 1.

Two other important considerations are to be made. First of all, the grid should also have a small thermionic emission. The construction illustrated in FIG. 2 shows that the grid is thermally insulated from the cathode through the shield and through the heat chokes. In addition, the grid structure is provided with cooling vanes to cool the grid down to a temperature comparable with the temperature of the anode. This, in turn, means that the low temperature branches of the several curves illustrated in FIG. 3b are valid also for the grid and the thermionic emission is similar for grid and for anode, which again depends on the fact that the thermionic emission in this cesium filled thyratron cavity is substantially independent from the material chosen for the grid.

A second point to be considered is that the internal voltage drop may vary with the load current passing through. On the other hand, one can see from FIG. 3!; that for a given cathode temperature the vapor pressure when varying for any reason also causes a resulting variation in the effective cathode work function. Therefore, by varying the electric current passing through the reservoir heater 272 the cathode work function can be adjusted. That control in turn may be carried out in dependence upon theload current passing through the anode to thereby establish a variable cathode work function which follows the variation of the anode sheath drop to maintain zero drop conditions in and across the thyratron.

FIG. 5 illustrates an alternative structure for the thyratron. The thyratron, as shown here, is a tubular cathode 121 with heat chokes (thin portions) 122 and 123 at the two ends of this cathode tube. The cathode can be made of any of the materials mentioned above, but again rhenium is the preferred choice. A pair of ring flanges 111 and 113 are used to mount the other electrodes to the cathode tube. In particular there are ceramic rings 112 and 114 for mounting two grid support annuli and 145 respectively to the flange rings 111 and 113. A plurality of grid bars are secured to the two supporting rings 135 and and define a grid cage concentric with and surrounding the cathode tube 121.

By means of two additional ceramic rings 116 and l18, t wo additional flange rings 125 and 126 are respective mounted to the two annuli 135 and 145 for supporting an anode tube 130 which in turn circumscribes the grid cage. Reference number denotes again the reservoir for cesium, and also for this embodiment there is provided a heating coil 172 to establish control conditions for the cesium atmosphere. The liquid cesium is kept sufficiently far from any internal wall facing and defining the thyratron cavity, so that the liquid cesium can be maintained at a temperature permitting cesium vapor' to be at equilibrium with its liquid state at a correspondingly low pressure.

The particular structure shown above differs from the one shown in FIG. 2 in several important aspects. First of all, the thyratron is longer in axial direction which enlarges the effec tive cathode emission surface, or, for the same area, a smaller diameter can be chosen. For a given anode-cathode spacing, this in turn means a relative enlargement of the anode surface in comparison with the embodiment shown in FIG. 3. The relative enlargement of the anode surface is instrumental in improving the anode sheath drop towards zero drop conditions.

Next, the cathode-anode spacing can be made smaller than in the case of FIG. 2 because no heat shield is interposed between the cathode and the other electrodes. This in turn means that the grid is maintained at floating heat potential meaning that it has a temperature in between the cathode and the anode temperature. This is a very important consideration as it introduces inherently a higher grid temperature into the i system. In order to counteract any higher thermionic emission from the grid the choice of the material for the grid is not open y t H a Looking at curve a,, for example, one can see that for a rhenium surface not too much cooler than the cathode, the thermionic emission is actually higher than for the cathode. For purposes of reference curve 0 in FIG. 3b represents the thermionic emission for a niobium substrate in a cesium atmosphere under the same pressure and temperature conditions for which the curve a for rhenium and b for molybdenum were developed. It can be seen that the current emission of niobium as modified by adhering cesium is well below the emission in accordance with curve a by about 2 orders of magnitude. Moreover, a rather hot grid is of advantage as in the contemplated range now an increase in the niobium surface temperature results in a decrease of the thermionic emissivity as long as one remains in the temperature range between the maximum and the minumum of the curve c,. In other words, the grid should either be as cool as the anode or, if the substrate for the grid is one exhibiting lessv adhesion than rhenium, then the grid temperature should actually be not too .xn qtbelq thes t d tsmp ata e v Quite obviously then the grid temperature can actually be rather close to the temperature of the cathode, which means that. the heat chokes 122 and 123 do not have to be very pronounced or do not have to be provided for at all. Moreover, one can see that little provision is required for any active cooling of the grid nor are any cooling surfaces for the grid structure provided. It is an interesting consideration that the low grid emission obtainable here permits the thyratron to be operated at low control powers, i.e., a high firing pulse can be made available at small current in the grid circuit.

The anode 130 has a rather large outer surface because the anode is here an elongated cylinder. This large outer surface therefor provides very active cooling.

Another important difference between th e structureslgwn is the possibility of a constricted discharge due to the bars 140., It has been found, however, that for theoperating conditions! expounded above the discharge is in fact not a constricted one. but a spread one. 7 V V,

The invention is not limited to the embodiments described above but all changes and modifications thereof not constituting departures from the spirit and scope of the invention are intended to be covered by the following claims.

I claim:

1. A thyratron comprising, a tubular cathode for direct exposure to a source of thermal energy and having a cylindrical outwardly directed emitting surface:

an annular anode having a cylindrical, inwardly directed operating surface circumscribing the tubular cathode, and being coaxial therewith, the anode-cathode spacing being exclusively determined by the difference in radii of said anode and cathode surfaces facing each other across the ring space between the cylindrical cathode surface and the cylindrical anode surface so that the electron flow, radially outwardly from cathode to anode, in-.

herently results in lower current density at the anode that first annular ceramic insulating means for mounting the,

anode in relation to the grid elements;

second annular means including insulating means for mounting the grid element in relation to the cathode;

means including at least some of the aforementioned elements to define a closed discharge chamber which contains the said cathode surface, the anode and the grid elements and 3 means for sustaining a particular vapor pressure in the discharge chamber, so that for said chosen difference in radii the plasma drop is about minimum.

2. A thyratron as set forth in claim 1, said second means including annular heat chokes integral with the cathode.

3. A thyratron as set forth in claim 1, at least one of said cathode, anode and grid elements being made of one of thematerials selected from the group which consists of molybdenum, rhenium, tantalum.

4. A thyratron as set forth in claim 1, said cathode being made of one of the materials selected from the group which consists of rhenium, molybdenum, tungsten, tantalum and nickel.

5. A thyratron as set forth in claim 1 including a plurality of radiator vanes connected for heat conduction respectively to said anode and said grid elements and extending outwardly therefrom.

6. A thyratron as set forth in claim 1, wherein the work function differential of the work functions of said cathode and of said anode is at least approximately equal to the voltage drop resulting from an arc between anode and cathode including the combined anode and cathode sheath drops and the drop in the grid space.

7. A thyratron as set forth in claim 1, said tubular cathode circumscribing a cavity for receiving thermal radiation.

8. A thyratron as set forth in claim 7 wherein a plurality of ,thermionic converters are provided, arranged'to define a cav ity for rebEWm radiation, said cavity being contiguous with the cavity as circumscribed by said tubular cathode, said tubular cathode beingheated by the radiation which also heats said thermionic converters, further including elt trEaLQrcuit k 5 nieansfforconnecting the thermionic converters to said anode,

and cathode of said thyratron.

9. A thyratron as set forth in claim 1, including a reservoir communicating with said chamber and containing a low boiling point material in the nonvaporized state.

10. A thyratron comprising, in combination:

a cathode having a particular cylindrical surface for thermionic emission comprising a high refractory material having a relatively high thermionic work function;

an anode having a cylindrical surface and disposed coaxial to the cathode for facing the cathode across a cylindrical ring space;

a grid mounted in spaced relation and coaxial to the anode and to the cathode to establish a controlled discharge path between cathode and anode through the grid;

the active surface of the cathode being spaced from the anode at a uniform distance across the cylindrical ring space between coaxial anode and cathode;

means including the anode and the cathode to provide a closed discharge chamber which includes said discharge P means including a reservoir for introducing a vapor into the means for providing a temperature differential between the cathode and the anode, including means for dissipating thermal energy from the anode, the anode receiving exclusively thermal energy from the cathode.

11. A thyratron as set forth in claim 10 wherein the vapor pressure is below 1 Torr.

12. A thyratron as set forth in claim 10 wherein the vapor is an alkali metal.

13. A thyratron as set forth in claim 10 wherein said cathode is rhenium or molybdenum, and the vapor is cesium or rubidium.

14. A thyratron as set forth in claim 13 wherein the grid is made of niobium.

15. A thyratron as set forth in claim 10, said reservoir being mounted to be at a temperature substantially below the cathode temperature.

0 16. A thyratron as set forth in claim 15 comprising in addition heating means for controlling the temperature of the reservoir.

17. A thyratron as set forth in-claim 10, wherein said grid assumes a temperature in between the cathode and anode temperatures and is selected of a high refractory material having a work function below the work function of the cathode material.

18. A thyratron as set forth in claim 10 wherein said anode and said cathode are made of the same material.

Claims (17)

1. 2. A thyratron as set forth in claim 1, said second means including annular heat chokes integral with the cathode.
2. 3. A thyratron as set forth in claim 1, at least one of said cathode, anode and grid elements being made of one of the materials selected from the group which consists of molybdenum, rhenium, tantalum.
3. 4. A thyratron as set forth in claim 1, said cathode being made of one of the materials selected from the group which consists of rhenium, molybdenum, tungsten, tantalum and nickel.
4. 5. A thyratron as set forth in claim 1 including a plurality of radiator vanes connected for heat conduction respectively to said anode and said grid elements and extending outwardly therefrom.
5. 6. A thyratron as set forth in claim 1, wherein the work function differential of the work functions of said cathode and of said anode is at least approximately equal to the voltage drop resulting from an arc between anode and cathode including the combined anode and cathode sheath drops and the drop in the grid space.
6. 7. A thyratron as set forth in claim 1, said tubular cathode circumscribing a cavity for receiving thermal radiation.
7. 8. A thyratron as set forth in claim 7 wherein a plurality of thermionic converters are provided, arranged to define a cavity for receiving radiation, said cavity being contiguous with the cavity as circumscribed by said tubular cathode, said tubular cathode being heated by the radiation which also heats said thermionic converters, further including electrical circuit means for connecting the thermionic converters to said anode and cathode of said thyratron.
8. 9. A thyratron as set forth in claim 1, including a reservoir communicating with said chamber and containing a low boiling point material in the nonvaporized state.
9. 10. A thyratron comprising, in combination: a cathode having a particular cylindrical surface for thermionic emission comprising a high refractory material having a relatively high thermionic work function; an anode having a cylindrical surface and disposed coaxial to the cathode for facing the cathode across a cylindrical ring space; a grid mounted in spaced relation and coaxial to the anode and to the cathode to establish a controlled discharge path between cathode and anode through the grid; the active surface of the cathode being spaced from the anode at a uniform distance across the cylindrical ring space between coaxial anode and cathode; means including the anode and the cathode to provide a closed discharge chamber which includes said discharge path; means including a reservoir for introducing a vapor into the closed chamber, said reservoir being positioned remote from the cathode the material of the vapor having a boiling point below the cathode temperature and a work function as well as an ionization voltage below the work function of the cathode, the vapor pressure being selected so that the plasma drop for a chosen small distance between anode and cathode is about minimum; and means for providing a temperature differential between the cathode and the anode, including means for dissipating thermal energy from the anode, the anode receiving exclusively thermal energy from the cathode.
10. 11. A thyratron as set forth in claim 10 wherein the vapor pressure is below 1 Torr.
11. 12. A thyratron as set forth in claim 10 wherein the vapor is an alkali metal.
12. 13. A thyratron as set forth in claim 10 wherein said cathode is rhenium or molybdenum, and the vapor is cesium or rubidium.
13. 14. A thyratron as set forth in claim 13 wherein the grid is made of niobium.
14. 15. A thyratron as set forth in claim 10, said reservoir being mounted to be at a temperature substantially below the cathode temperature.
15. 16. A thyratron as set forth in claim 15 comprising in addition heating means for controlling the temperature of the reservoir.
16. 17. A thyratron as set forth in claim 10, wherein said grid assumes a temperature in between the cathode and anode tempEratures and is selected of a high refractory material having a work function below the work function of the cathode material.
17. 18. A thyratron as set forth in claim 10 wherein said anode and said cathode are made of the same material.

Images