US 3675063 A
A continuous dynode electron multiplier for detecting and multiplying charge flow in a vacuum. Continuous dynodes are deposited on thermally conductive substrates. Heat sinks attached to the substrates permit operation of the multiplier at high current levels without the threat of thermal runaway. Further, the method of making the continuous dynodes includes depositing from a composite source, whereby the dynode resistance can be accurately controlled.
Claims available in
Description (OCR text may contain errors)
United States Patent Spindt et al.
[451 July4, 1972  HIGH CURRENT CONTINUOUS DYNODE ELECTRON MULTIPLIER Stanford Research Institute, Menlo Park, Calif.
 Filed: Jan.2, 1970  Appl.No.: 360
 US. Cl ..313/104, 328/243  Int. Cl. ..H0lj 43/00  Field ofSearch ..313/103, 104, 105, 112,95, 313/107; 328/243  References Cited UNITED STATES PATENTS 2,620,287 12/1952 Bramley... ..313/103 3,128,408 4/1964 Goodrichm. ..313/103 3,387,137 4/1968 Adams ..313/103 3,345,533 10/1967 Washbum ..3l5/3.6
3,244,922 4/1966 Wolfgang ...3 1 3/103 3,458,745 7/1969 Shoulders ..313/104 OTHER PUBLICATIONS Rev. Sc]. Inst. Vol. 36 No.6 June 1965 pp. 775- 779 Article by C. A. Spindt and R. K. Shoulders entitled Electron Multipliers Primary Examiner-Eli Lieberman Att0rneyUrban H. Faubion and James Todorovic [5 7] ABSTRACT A continuous dynode electron multiplier for detecting and multiplying charge flow in a vacuum. Continuous dynodes are deposited on thermally conductive substrates. Heat sinks attached to the substrates permit operation of the multiplier at high current levels without the threat of thermal runaway. Further, the method of making the continuous dynodes in cludes depositing from a composite source, whereby the dynode resistance can be accurately controlled.
1 Claim, 7 Drawing Figures PATENTEDJUL 41972 3,675,063
sum 20F 2 :E"IIEl 3 IE-IIEIQA l 169mm 7." Foams do I NVE NTORS HIGH CURRENT CONTINUOUS DYNODE ELECTRON MULTIPLIER BACKGROUND OF THE INVENTION 1. Field of the Invention This invention pertains to the field of electron multipliers and more particularly to high-current continuous dynode electron multipliers.
2. Description of the Prior Art The rapid advances in ultrahigh vacuum techniques, scanning electron beam microscopy, and mass spectrometry have created a need for efficient, high-current electron multipliers capable of detecting charge flow in a vacuum. Electron multipliers have been used in the prior art for such applications, but such prior art electron multipliers have had limitations. The earliest kind of electron multipliers utilized discrete dynodes and required perpendicular magnetic and electric fields to direct electrons from dynode to dynode. Because of the additional complexity of requiring a magnetic field, electrostatic secondary-emission electron multipliers were soon developed. The electrostatic multipliers, however, were discrete dynode multipliers and consequently required a large number of vacuum feedthroughs, which unduly complicated the multipliers.
The next stage of development was continuous dynode electron multipliers. In these, as the name suggests, continuous rather than discrete dynodes are utilized, requiring a minimum number of vacuum feedthroughs. These have reasonably high gains and a simple no-magnet geometry and are stable to air exposure. A continuous strip dynode electron multiplier generally comprises a pair of opposed plane surfaces which are coated with a dynode material having a secondary electron emission ratio greater than 1. Typically, the dynode material is a composite molybdenum/aluminum-oxide film produced by simultaneous deposition of the two materials from separate electron-bombardment sources and is deposited on a fused silicon substrate. The two dynodes define a channel having an input end and an output end. Corresponding ends of the dynodes are electrically connected together and a voltage source applied between the two ends of the channel. This voltage creates in the channel an electric field which has equipotentials perpendicular to the dynode surfaces. The channel input is operated at a high negative potential relative to the channel output. Current to, be amplified is directed into the input end of thechannel with a velocity sufi'rcient to produce a secondary electron emission ratio greater than 1 upon striking either of the dynode surfaces. Secondary electrons are accelerated along the channel in a series of hops between the two dynode surfaces and the amplified current is collected by a collector at the channel output, which is operated at ground potential. A more detailed description of a typical continuous dynode electron multiplier may be found at C. A. Spindt and K. R. Shoulders, Stable, Distributed-dynode Electron Multiplier, Review of Scientific Instruments, Vol. 36, No. 6, pp. 775-779 (June 1965).
Prior an multipliers such as described above have high gains, simple geometry and vacuum feature configurations, as well as being stable to air exposure and able to withstand repeated high-temperature vacuum bake-outs. The most significant limitation of prior art multipliers is their current-handling capability. The gain of the multiplier depends upon the re sistance of the dynodes and the voltage applied thereto. A typical multiplier has a resistance of perhaps ohms from end to end of the parallel pair of dynodes. For a multiplier having this resistance, the gain theoretically rises exponentially with an increase in supply voltage. Practically, however, this turns out not to be valid for higher output currents. The exponential rise is valid to the point where the collected signal output current becomes an appreciable fraction of the zero signal or quiescent dynode current. When this occurs, the normally uniform voltage gradient along the multiplier is progressively depressed toward the output end, and the overall gain is correspondingly smaller. For example, with the dynode resistance typically about 10 ohms, the quiescent dynode current is of the order of several micro-amperes. It has been found that under these conditions the exponential gain characteristic is valid for collector currents up to about 10 amperes, but further increases in output current result in a flattening of the gain curve. Higher output current capabilities would be obtained by using lower dynode resistance values to produce correspondingly higher quiescent currents at the same applied voltages. However, in common with other cermet materials, molybdenum/aluminum-oxide films have a high negative tem perature coefficient of resistivity, thus giving rise to the possibility of dynode thermal runaway at the higher power dissipation levels unless steps are taken to remove much of this heat.
SUMIVIARY OF THE INVENTION Accordingly, this invention is directed to a continuous dynode electron multiplier and a method for making the multiplier in which the heat-dissipation capabilities of the dynode structure are greatly improved. This permits using dynodes having a low resistance and higher current capability. More specifically, in accordance with the one embodiment of the invention, a relatively high conductivity material is used for the substrate and a heat sink is intimately attached to the substrate in order to improve the heat-dissipation capability thereof. Further, the dynodes are deposited on the substrates from a composite source in order to accurately control the dynode resistance.
The novel features which are believed characteristic of the invention are set forth with particularity in the appended claims.
The invention itself, however, both as to its organization and method of operation, together with further objects and advantages thereof, may best be understood by reference to the following description taken in connection with the accompanying drawings in which:
FIGS. 1 and 1A are diagrammatic illustrations of an electron multiplier according to this'invention and showing the manner in which the multiplier is electrically connected;
FIG. 2 is a side elevation of an electron multiplier constructed in accordance with the principles of this invention and showing the mechanical details thereof;
FIG. 3 is a cross-sectional view of an alternate multiplier and heat sink configuration;
FIGS. 4 and 4A are simplified electrical circuits of an electron multiplier; and
FIG. 5 is a diagrammatic illustration of apparatus for depositing the dynode material on the substrates.
DESCRIPTION OF THE PREFERRED EMBODIMENT Referring now to FIG. 1, there is diagrammatically shown an electron multiplier according to this invention, showing the manner in which the multiplier is electrically connected. The multiplier comprises spaced, parallel substrates l0 and 11 having inner surfaces 12 and 13. The inner surfaces 12 and 13 are formed of a dynode material deposited on substrates 10 and 11 to form a channel therebetween having an input end generally designated by reference numeral 14 and an output end generally designated by reference number 15. Electrodes 16 and 17 are attached to the input end 14 of substrates 11 and 10, respectively, and are electrically connected together. Similarly, electrodes 18 and 19 are affixed to the output end 15 of substrates 11 and 10, respectively, and are electrically connected together. A voltage source 20 is connected between input 14 and output 15 of the channel and establishes an electric field therein parallel to the channel. A collector 21 is situated at the output end 15 of the channel and is at a positive potential with respect to the output end 15 of the channel due to voltage source 22 which is connected through a sensing device 23 such as a meter, cathode ray tube, integrated'circuit designed to drive further circuitry, etc. to the collector 21. Collector 21 serves to collect the amplified output current of the electron multiplier and sensing device 23 provides an indication of this collected output current. The voltage source 22 attracts the electrons coming from between the dynode plates. It is also quite feasible, as shown in FIG. 1A, to replace the voltage source 22 and the sensing device 23 by an impedance 23a of value Z across which the signal voltage is developed to drive other circuitry. The collector 21 is still at an attractive potential relative to the electrons coming from within the dynode strips. This forms a simple configuration which efficiently drives an extremely high frequency capability transmission line also of characteristic impedance 2,.
Turning now to consideration of FIG. 2, there is shown an electron multiplier constructed in accordance with the principles of this invention and showing in more detail the mechanical elements thereof with like elements to FIG. 1 being indicated by like reference numerals. In FIG. 2 substrates l and 11 are spaced, parallel and co-extensive. THE substrates l0 and 11 are made of insulator material which has a high thermal conductivity. For example, sapphire, polycrystalline alumina, or beryllia have proved to be very satisfactory substrate materials. Substrates l0 and 11 are coated with a dynode material to form continuous dynodes l2 and 13 which define a channel having an input end 14 and an output end 15. The dynodes 12 and 13 are deposited from a single composite source containing a mixture of molybdenum and aluminum oxide and the process of forming the dynodes is more fully discussed hereinafter in connection with FIG. 5. Electrodes l6 and 17 at the input end 14 facilitate an electrical contact to the input end of dynodes 12 and 13 and electrodes 18 and 19 at the output end 15 facilitate electrical connections to the output ends of dynodes 12 and 13. A heat sink 24 is associated with substrate and a heat sink 25 is associated with substrate 11. Inasmuch as the electron multiplier operates in a vacuum where heat transfer cannot take place through convection or conduction in air, some means must be provided for assuring intimate thermal contact between the heat sinks 24, 25, and theirrespective substrates. Cementing the heat sinks to their respective substrates with conductive epoxy has been found to be a very satisfactory method of assuring such contact, as has brazing the heat sinks to the substrates.
With the construction and materials described above, good heat dissipation results. The close match in thermal expansion coefficients of sapphire or alumina, of which the substrate is constructed, and the molybdenum-aluminum oxide mixture of which the dynodes are constructed, and the forming of intimate thermal bonds between substrates l0 and 11 and large area heat sinks 24 and 25, enable the heat which is developed in dynodes 12 and 13 to be very efficiently conducted to heat sinks 24 and 25.
In operation, electrons from a source of electrons (not shown) enter the input end 14 of the channel-defined by dynodes 12 and 13. The material of which dynodes 12 and 13 are constructed has a secondary electron emission ratio larger than 1. Therefore, when the entering electrons strike either of the dynode surfaces, secondary electrons are emitted. Due to the electrical field existing within the channel, the secondary electrons are accelerated in the channel towards the output end in a series of hops between the two dynode surfaces 12 and 13. The amplified current is collected at the output end 15 of the channel by a collector 21.
Referring now to FIG. 3, there is shown a cross sectional view looking down the channel of an electron multiplier constructed in accordance with the principles of this invention and incorporating an alternate embodiment of a heat sink. More specifically, the electron multiplier of FIG. 3 comprises substrates l0 and 11 with dynodes l2 and 13, including the inner surfaces thereof. Heat sinks 26 and 27 are in thermal contact with substrates l0 and 1 1, respectively. In the embodiment of FIG. 3, however, this intimate thermal contact is achieved by clamping the heat sinks to the substrates rather than epoxying or brazing. Spacers 28 and 29 are placed between substrates 10 and l l to hold them apart and provide the proper channel spacing between dynodes l3 and 14. The heat sinks 26 and 27 are clamped to substrates 10 and 11 by screws 30 and 31 which thread into heat sink 26 and into springs 32 and 33, respectively, which bear against heat sink 27. This has proved to be a satisfactory intermediate arrangement for cooling the dynodes 12 and 13 without requiring epoxy or brazing between the heat sinks and substrates.
The advantages gained and problems overcome by providing heat sinks for a continuous dynode electron multiplier can best be understood by referring to FIG. 4. FIG. 4 is a diagrammatic representation of the electrical circuit of the electron multiplier of this invention. The dynodes 12 and 13 are, in ef fect, resistors as indicated by the dotted lines. The voltage source 20 connected across the paralleled dynodes creates an electrical field between the dynodes 12 and 13, which electrical field accelerates the secondary electrons created in the channel toward the output end and the collector 21. The voltage source 22 connected between voltage source 20 and collector 21 biases the collector positively with respect to the output end of the dynodes 12 and 13 and sensing device 23 gives an indication of the output current collected by collector 21. Again, an alternative output drive arrangement for driving a transmission line efficiently or as shown in FIG. 4A in which signal voltage is developed across an impedance 23a of value Z to drive, for example, a transmission line of characteristic impedance Z On the arrangement of either FIG. 4 or FIG. 4A, under conditions of no electron input to the multiplier a quiescent current flows through the dynodes 12 and 13 and no current is collected by collector 21. However, when there is an electron input to the channel between the dynodes l2 and 13, there is a current flow between the dynodes l2 and 13 and collector 21 in the form of secondary electrons which are emitted by dynodes 12 and 13. Theoretically, the gain of a multiplier rises exponentially with an increase in voltage applied across the dynodes. This exponential rise is, however, valid only to the point where the signal output current collected by the collector becomes an appreciable fraction (=10 percent) of the quiescent dynode current. When this occurs, the normally uniform voltage gradient along the electrical field established between the dynodes is progressively depressed toward the output end thereof, and the overall gain is correspondingly smaller. For example, with the dynode resistance typically about 10 ohms, the quiescent dynode current is of the order of several rnicroamperes. It has been found that under these conditions, the exponential gain characteristic is valid for collector currents up to about 10' amps, but further increases in output current result in a flattening or dropping off of the gain curve. Higher output current capabilities could be obtained by using lower dynode resistance values to produce correspondingly higher quiescent currents at the same applied voltages; that is, the collected signal output current would be a smaller percentage of the quiescent current. However, in common with other dynode materials, molybdenum/aluminumoxide films of which the dynodes in the present invention are preferably constructed, have a high negative coefficient of resistivity, thus giving rise to the possibility of dynode thermal runaway at the higher power dissipation levels unless steps are taken to remove much of this heat. In the continuous dynode electron multiplier of this invention lower values of dynode resistance (on the order of 10 -10 ohms) can be utilized because of the heat sinks which are associated with the substrates. Using heat sinks to cool the dynodes, the output cur rents can be increased several orders of magnitude. For example, a multiplier without a heat sink is limited to a maximum power dissipation of about 200 milliwatts and a corresponding maximum output current of the order of 10' amps, whereas a multiplier cooled with a heat sink in accordance with the principles of this invention can be operated at 10 to 20 watts with a corresponding increase in the maximum output current to about 10* amps. These magnitudes of current make feasible fast pulse (high frequency) output capability.
As discussed above, utilizing heat sinks the dynodes may be constructed to have a lower resistance whereby increased current capability results, but thermal runaway does not occur.
Another aspect to this invention is the method of making or depositing the dynodes on the substrates. In the prior art, dynodes of molybdenum/aluminum-oxide or similar materials have been deposited from two separate sources, one for the molybdenum and one for the aluminum-oxide. It has been felt that this was necessary in order to avoid a distillation process which would occur if the dynodes were deposited from a composite source consisting of a mixture of the two elements. In other words, it was felt that due to the differences in the rate of vaporization of molybdenum and aluminum-oxide, a nonuniform dynode would result. However, it has been found that this is not the case. In fact, dynodes can be formed by deposition from a composite source of molybdenum and aluminumoxide with very accurate control over the resistance of the dynode and a high yield of properly formed dynodes.
Refen'ing now to FIG. 5, there is shown a deposition arrangement in accordance with the principles of this invention for depositing dynode films from a composite molyb denum/aluminum-oxide source. A substrate 34 which is to be coated is held in position by a rotary substrate holder 35, a substrate furnace comprises an inner alumina shell 36 and an outer alumina shell 37. The inner shell 36 is shown as threaded as a coil form with a molybdenum filament wire 38 wound thereon. The substrate 34 is transported into the furnace by a manipulator and heated to 1 ,000 C. for cleaning and later for heat treating the deposited films. During deposition, the substrate 34 is held at 200 to 300 C. due to incident radiation thereon from the evaporator. By preheating and post heating the substrate 34 to this temperature, better adherence of the dynode and a more unifonn deposition thereof is achieved. Radiation shields generally indicated by reference numeral 39 surround the substrate furnace. An evaporator 40 is situated at some angle with respect to the normal plane of the substrate which in a preferred embodiment may be for example 30. The evaporator 40 is equipped with a shutter 41 which may be opened or closed independently of any other function. Prior to deposition the substrate 34 is heated to a temperature of x 1,000 C. by means of the substrate furnace. When this temperature has been achieved, the substrate 34 is lowered onto the rotary substrate holder 35, and the shutter 41 is opened, exposing the substrate 34 to the evaporation source 40. During deposition, the substrate holder 35 rotates, carrying substrate 34 with it. This rotation is necessary to assure a uniform deposition on substrate 34. If, for example, the dynode or film was sputtered on with a sputtering apparatus, rotation of the substrate would not be necessary. This because the relatively broad area source in sputtering would achieve uniform deposition on substrate 34 without rotation thereof.
While particular embodiments of the invention have been shown and described, it will of course be understood that the invention is not limited thereto, since many modifications in the electrical circuits, mechanical arrangements, and in the materials employed may be made. It is contemplated that the appended claims will cover any such modifications as fall within the true spirit and scope of the invention.
What is claimed is:
1. A continuous dynode electron multiplier comprising first and second opposed spaced substrates having inner surfaces which define a channel and having outer surfaces adapted to contact heat sink means, said inner surfaces of said substrates coated with a material having secondary electron yield greater than unity and forming continuous dynodes having input and output ends, a voltage source having a given voltage output connected between the input and output ends of said continuous dynodes for establishing an electrical field within said channel, said voltage source also establishing a quiescent current flow through said continuous dynodes, means for introducing input electrons into said channel adjacent the input ends of said continuous dynodes, collector means for collecting output electrons adjacent the output ends of said continuous dynodes, said continuous dynodes having a resistance substantially less than 10 ohms provided by evaporating a predetermined mixture of molybdenum and aluminum-oxide mm a single source whereby the quiescent current flow through the continuous dynodes is substantially larger than 10 microamperes to permit larger numbers of output electrons to be collected by said collector means while still maintaining an approximately exponential gain relationship between input and output electrons for any given voltage applied between the input and output ends of said continuous dynodes, and heat sink means contacting the outer surfaces of opposed spaced substrates to prevent thermal runaway of said continuous dynodes due to their having a resistance substantially less than 10 ohms.