US 3863094 A
The invention includes a channel-type electron multiplier incorporating a glass plate having holes therethrough. The plate is treated so that the holes will support secondary emission. When primary electrons from a photocathode enter the holes, an increased current density is produced at the output of the holes. When the output is directed onto a phosphor screen, an intensified image of the scene being viewed is produced. Night scenes may thus be brightened for use in military reconnaissance or the like. However, channel-type electron multipliers have what is known as a saturation level of operation beyond which an increase in input current produces little or no increase in output current. The device of the present invention alleviates this problem by increasing electrical resistance near the output. This construction makes it possible to produce a special electric field distribution in the holes. This field distribution then causes the gain of the multiplier to saturate at substantially higher input current levels. In addition, a multiplier of this type will exhibit a nonlinear input versus output relationship such that, if incorporated in an image tube (direct view or TV pickup), it imparts to this tube highly desirable contrast enhancing characteristics.
Claims available in
Description (OCR text may contain errors)
l nited States Patent 1 Orthuber 1 IMAGE lNTENSlFlER AND METHOD OF MAKING AN ELECTRON MULTlPLIER THEREFOR Richard K. Orthuber, Sepulveda, -Calif.
 Assignee: lnterntional Telephone and Telegraph Corporation, New York, NY.
22 Filed: Apr. 8, 1969 211 App]. No.: 815,519
3,260,876 7/1966 Manley et a1. 313/68 3,487,258 12/1969 Manley et a1. 315/11 3,497,759 2/1970 Manley 315/11 OTHER PUBLICATIONS Primary Examiner-Maynard R. Wilbur Assistant Examiner.l. M. Potenza Attorney, Agent, or Firm-John T. OHalloran; Menotti J. LainbardiflrfjEdward Goldberg 1 Jan. 28, 1975  7 ABSTRACT The invention includes a channel-type electron multiplier incorporating a glass plate having holes therethrough. The plate is treated so that the holes will support secondary emission. When primary electrons from a photocathode enter the holes, an increased current density is produced at the output of the holes. When the output is directed onto a phosphor screen. an intensified image of the scene being viewed is produced. Night scenes may thus be brightened for use in military reconnaissance or the like. However, channeltype electron multipliers have what is known as a saturation level of operation beyond which an increase in input current produces little or no increase in output current. The device of the present invention alleviates this problem by increasing electrical resistance near the output. This construction makes it possible to produce a special electric field distribution in the holes. This field distribution then causes the gain of the multiplier to saturate at substantially higher input current levels. 1n addition, a multiplier of this type will exhibit a nonlinear input versus output relationship such that, if incorporated in an image tube (direct view or TV pickup), it imparts to this tube highly desirable contrast enhancing characteristics.
surface resistivity variation.
12 Claims, 14 Drawing Figures I .4 7 z 7 I a I I 7 I l a PATENTEDJANZEIIHYS 3,863,094
' SHEET 1 UF 4 Fla 1].
INVENTOR E/CH/ZED K Off/V0565 ATTOZ/WEV IMAGE INTENSIFIER AND METHOD OF MAKING AN ELECTRON MULTIPLIER THEREFOR BACKGROUND OF THE INVENTION This invention relates to electron multipliers and, more particularly, to a channel-type electron multiplier having an unusually high saturation level.
In the past, image intensifiers have incorporated a transparent, evacuated envelope having a phosphor screen on an output side of a channel-type electron multiplier and a photocathode on the input side thereof. A scene illuminating the photocathode causes primary electrons to enter holes of channels in the multiplier. The multiplier includes a perforated glass plate which has been heated in a hydrogen atmosphere to treat the channel surfaces in a manner to cause them to support secondary emission.
When the primary electrons bombard the channel surfaces, secondary electrons are released at a ratio greater than unity. The secondaries are accelerated part way down a channel and bombard it again. This process is then repeated over and over again so that the current density at the output of the multiplier is much greater than that at its input. A bright image of the scenes being viewed is then displayed on the phosphor screen. Night scenes may, thus, be displayed with substantial brightness for military reconnaissance or other purposes.
Conventional channel-type electron multipliers have evaporated conductive input and output coatings on the input and output sides, respectively, of the dielectric plates. These coatings have holes therethrough in registration with the glass plate holes. The coatings act as electrodes. The output coating or electrode is maintained positive with respect to the input electrode. These electrodes at these relative potentials, thus, create a field in the channels and produce an electron current at the output of the multiplier. Electrons also flow from the negative electrode in the channel surfaces to balance positive wall charges caused by the emission of secondaries.
When the input current to a conventional channeltype multiplier is increased from zero, the output current rises approximately in proportion to the input current, depending upon the gain of the multiplier. However, above a predetermined input current, the gain of the multiplier tends to decrease with increasing input. This condition is called saturation. It is believed that this condition exists because the channel surfaces near the output ends thereof become depleted of electrons. This, in turn, is believed to cause an electric field distortion that reduces the multiplier gain at elevated input and output currents.
The saturation condition of a conventional channeltype multiplier is a strict limitation of the output current density of an image intensifier.
SUMMARY OF THE INVENTION In accordance with the device of the present invention, the above-described and other disadvantages of the prior art are overcome by providing a channel-type electron multiplier in which the channel wall resistance near the input end of a channel is less than that at the output end thereof. It is believed that the invention in creases the gain at elevated input and output currents because the specific wall resistance provided in accordance with the device of the present invention corrects the said electric field distortion. At any rate, the multiplier of the invention has a relatively low gain at moderate current inputs and a relatively high gain at elevated input currents. This makes it possible to cascade the multiplier of the invention with a conventional multiplier. The conventional multiplier can then be operated to produce an output current for the input to the multiplier of the invention. The conventional multiplier may, thus. have a high gain at a low input current, and the multiplier of the invention may have a high gain at its elevated input current. Saturation, thus, limits the maximum gain of cascaded multipliers in accordance with the invention to a lesser degree as was the case in the prior art.
The invention also includes a method of making the multiplier of the invention.
The above-described and other advantages of the invention, particularly the possibilities of contrast enhancement, will be better understood from the following description when considered in connection with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS In the drawings, which are to be regarded as merely illustrative:
FIG. I is a sectional view of a conventional image intensifier tube;
FIG. 2 is a greatly enlarged sectional view of a channel-type electron multiplier;
FIG. 3 is a graph of a current characteristic of a conventional multiplier;
FIG. 4 is a graph of the potential distribution along the axis of a conventional multiplier;
FIG. 5 is a graph that is an approximation of a potential distribution shown in FIG. 4;
FIG. 6 is a graph of an approximation of the potential distribution of a multiplier made in accordance with the present invention;
FIG. 7 is a graph of both current characteristics demonstrating the shift in saturation level with the multiplier of the invention;
FIG. 8 is a graph of a current characteristic indicating how contrast may be enhanced with the multiplier of the invention;
FIG. 9 is a sectional view of portions of a multiplier to be constructed in accordance with the present invention;
FIGS. 10, 11, and 12 are sectional views of apparatus which may be employed in the treatment of the glass plate of a multiplier in accordance with the invention;
FIG. 13 is a sectional view of a multiplier constructed in accordance with the invention; and
FIG. 14 is a sectional view of an image intensifier constructed in accordance with the invention.
DESCRIPTION OF THE PREFERRED EMBODIMENT In the drawings in FIG. I, a conventional image intensifier tube 10 is shown including a transparent, evacuated envelope 11. A photocathode 12 and a phosphor screen 13 are positioned within the envelope 1] on 0pposite sides of a channel-type electron multiplier I4.
Multiplier 14 is shown greatly enlarged in FIG. 2 including a perforate glass plate 15 having perforate, evaporated, conductive coatings l6 and 17 thereon. Plate 15 is conventionally treated by heating in a reducing atmosphere to make the internal surfaces of channels or holes 18 capable of supporting secondary emismen.
If photocathode I2 is positioned contiguous to input electrode 16 and screen 13 contiguous to output electrode 17, a brightened image of the scene being viewed will appear on screen 13. Typical operating potentials are: Input electrode 16, -l .000 volts; Output electrode 17, zero volts; Screen 13, +5,000 volts; Photocathode 12, -l 100 volts.
Conventional image intensifer tubes using channeltype electron multiplier plates for intensification show a relationship between output brightness and input brightness or output and input current densities j and jg which is described by a characteristic of the form shown in FIG. 3.
For low values of output density j this characteristic showns a linear increase of the output density j with j and, thus, constant gain G, tana. At higher levels, however, the characteristic bends into a horizontal direction and approaches a saturation level of output current density j This saturation phenomenon is caused by the following fact.
As the emission density of secondary electrons increases (particularly near the output end of the channel walls) and approaches the strip current, i.e., the conduction current along the semiconductive channel walls, the potential within the channels ceases to increase uniformly towards the channel walls as it would in the absence of an electron input. However, close to the output end where the secondary emission density is highest, this potential drifts to a more positive than normal state due to the inability of the limited strip current to completely replace the charges lost be secondary emission.
This phenomenon is discussed, and the above hypothesis for the cause of saturation is qualitatively substantiated in Review of Scientific Instruments, by D. S. Evans, Volume 36, Number 3, March 1965, Pages 375-382. This article illustrates the potential rise along a channel axis for a channel in the absence of an electron cascade (curve A-C) and for a loaded channel (curve ABC) as shown in FIG. 4.
The uniform field represented by the unloaded curve AC is seen to become distorted as the electronic load is applied in such a fashion that in a large section of the channel adjoining the input aperture, a slight increase in field results. In a short section near the output aperture, the field is drastically reduced and nearly disappears (horizontal section B-C).
It can be shown, using a simplified model of FIG. 4, that a transition from curve A-C to curve A-B-C results in a decrease of gain. This simplified model is shown in FIG. 5.
In FIG. 5, curve A-B-C of FIG. 4 is approximated two sections of uniform field, the section towards the input having a length M a L and being subjected to a potential difference BV, and the section containing the output having a length A, (l a)L and a potential difference l B)V where V, is the voltage applied to the entire channel of length L.
Using the expression G e(a 'b/E) for the gain of a channel section of length A (where a and b are constants involving channel radius and secondary emission properties, and E is the field gradient in the channel), the gain for the two sections in FIG. 5 becomes:
G =e e and 1-.. L a v.
The total gain:
This shows that for B 0 or B l, the total gain G tends toward zero, the case B 1 representing output saturation as illustrated by FIG. 4. An intermediate value must, therefore, represent maximum which is formed from dG/dB 0.
a (1-0:) f i 13: 2 3
Inspection of FIG. 5 shows immediately that B 01 implies a gradient uniform throughout the channel. The deviation B a induced by residual wall charges leads to saturation as stated before. The objective of this invention is an exploitation of the interdependence of gain and field uniformity for the purpose of contrast enhancement.
For this purpose, channels are used in which:
I. The longitudinal gradient is varying along the axis in the absence of electronic load;
2. Varies in the opposite sense as that in a conventional channel operating in or near saturation. This means in the unloaded state B a.
For such a channel, the potential diagram of FIG. 5 for a uniform field channel is changed to that shown in FIG. 6.
In FIG. 6, the potential distribution in the case of no or low electron input is qualitatively shown by the solid line B a. This distribution may be established by means of a channel plate in which the conductivity of the channel walls is higher in the input section than in the region close to the output end. According to the prior discussion herein, this channel would have less gain than an other wise similar one with uniform field distribution.
We consider now the electronic loading applied to increase until under the effect of non-neutralized wall charges the potential near a L has increased to the value B V =a V i.e. a B. In this state, a higher gain would be reached than with no or a low electronic load according to the preceding discussion.
We consider the load increased still further to make B 0:. Then, this channel will saturate as a conventional uniformly-conductive channel. The response characteristic for such a channel of the invention is, therefore, different from that shown in FIG. 3 for a conventional channel and shaped as shown in FIG. 7.
Whereas the conventional channel has a maximum and constant gain for all but very high levels, the channel of the present invention shows a variable slope of this response characteristic with a toe and shoulder somewhat similar to the H and D characteristic familiar in photography. The first with strict proportionality between input and output transmits contrasts unchanged except if driven to saturation. The second, however, has the capability of transforming low contrasts in the input image into high display contrasts, a capability which was so far restricted to photographic or TV-type electronic processing. The possibility of contrast enhancement is illustrated in FIG. 8 which shows a transfer characteristic of a channel of the present invention with the slope G signifying the total gain for an input j, and y, the maximum slope of the characteristic, which determines the differential gain, i.e., the ratio of an output brightness increment to the corresponding increment of input brightness.
A multiplier plate formed by an array of such channels of the invention and operated with an electron image having an input current density j, in the highlights and j, in the lowlights and, thus, an input contrast l 1 mu: ""jl mtnlji mu: l mu:
will then produce an intensified output image with contrast Thus, the contrast has been enhanced by a factor 7 /G which is I provided the low contrast input levels are adjusted to insure loading to the steepest slope of the brightness response characteristic. To accomplish very high degrees of contrast enhancement, it is possible to use two or more channel plates in cascade.
Adjustment of input'current densities to the most favorable level can be performed by:
l. Adjustment of the voltage across the channel plate of the present invention;
2. Adjustment of lens speed;
3. Gain adjustment of a conventional channel plate arranged ahead of the plate of the invention; and
4. Gain adjustment of a conventional linear intensifier coupled to the contrast enhancing intensifier.
It remains now to disclose ways to achieve the required non-uniform conductivity along the walls of the plate of the invention, i.e., a mosaic formed by many channels of the invention.
The first approach to an acceptable contacting method for a high and low conductivity plate would require contact over the total areas of the contacting surfaces of two dielectric plate sections. In order to facilitate such a contact, the front plate on its output surface is ground or sagged to a very slightly spherical shape, whereas the back plate on its input surface is ground and polished flat, as shown in FIG. 9. The second plate may be significantly thinner than the first, since the break of field strength occurs close to the channel ends.
In FIG. 9, input and output electrodes 19 and 20, respectively, are conventional, vacuum-deposited, contiguous electrodes and connected to the power supply. On squeezing the two plates together by applying pressure to their rims, the second plate will wrap over the first plate, establishing a contact pressure more uniformly distributed over the area than would be the case if both plates were flat.
Omission of metallic electrodes between plates 21 and 22 may give rise to non-uniformity of contact resistance over the interface of the arrangement of FIG. 9. As a solution to this transition-resistance problem, low
resistivity is provided between the contacting surfaces,
while preventing lateral conductivity along the surfaces in the direction normal to the channel axis.
Metallic electrodes with this characteristic are formed by depositing mutually-insulated, conductive islands indicated at K on the contacting surfaces, either by evaporation through very fine mesh metal screens or by techniques known from the preparation of iconoscope targets e.g., by deposition of a contiguous silver-film which is then broken up into separate conductive islands exhibiting high resistance is a transverse direction. V. K. Zworykin and G. A. Morton, Television; Second Edition; Wiley Sons, I940, 1945; Pages 310-31 I.
Another way of preparing the multiplier of the invention starts out with a single structure and accomplishes non-uniform conduction by differential activation.
Conventional channel-type multiplier plates are usually provided with the required conductivity and secondary emission properties by subjecting the normally highly resistive base structure of a reducable glass to a hydrogen firing process. The duration and temperature at which the resulting reduction of the channel wall surface is carried out have a pronounced bearing on the conditionof the resulting semiconductive layer formed on the channel walls. For moderate duration of the process, e.g., less than 20 hours at around 400 centigrade, the resulting conductivity increases with the temperature of the glass during activation.
Non-uniform conductivity within multiplier channels can be accomplished by establishing, during activation, a temperature gradient through a channel plate in the direction of the channels. The following two procedures establish this temperature gradient during activation and may be used singly or in conjunction. The first is illustrated in FIGS. 10 and 11.
FIGS. 10 and 11 show a quartz cylinder 23 mounted within an oven which contains independentlyadjustable heater elements 24 and 25. Within the quartz cylinder 23, a baffling element 26 is mounted so that it runs parallel to the axis of quartz cylinder 23. The baffling elements 26 contains one or more apertures onto which one or more non-activated channel plates 27 are mounted. Hydrogen admitted from the bottom of FIG. 11 has to pass the channels of the channel plates in the baffie in order to leave through the outlet on top of FIG. 11.
This is all as in conventional channel plate activation. In order to obtain the channel plate of the invention, the outer heaters are energized to different levels, e.g., the right-hand heater 25 higher than the left-hand heater 24. In this case, the heat flow onto the righthand surface of the channel plate exceeds that onto the left-hand surface; and heat will flow between the two channel plate surfaces, establishing a temperature gradient through the plate so that the right-hand surface is hotter than the left-hand surface. Consequently, activation and conductivity build-up will proceed at a more rapid rate at the right-hand channel ends; and the end result will be channels with wall conductivity increasing from the left to the right, which results in a channel plate usable for contrast enhancement.
An alternate procedure is illustrated in FIG. I2. The heaters are arranged as in FIGS. 10 and 11. Differential heating by the heater elements may or may not be applied. Again, activation occurs in a quartz tube 28, which is again separated into two halves by a baffie 29 containing the channel plate 30 under activation. Gas
exchange between the two halves occurs only through the channel plates 30.
In contrast to FIGS. 10 and l l, differential heating of the plate occurs here by two separate hydrogen flows entering the two chambers of the activation vessel. The gases entering the chambers are preheated in heat exchangers 31 and 32 so that the gas entering the righthand chamber has a higher temperature than that entering the left chamber. This results again as in FIGS. 10 and 11 in a temperature gradient through the channel plate and thus non-uniform activation of the channels.
If the pressure in both chambers is kept equal so that no net flow through the channels occurs, the hot and cold hydrogen will mix in the channels by diffusion so that a gradual temperature drop of the gas from right to left is established. The temperature profile within the channels can then be shifted by adjusting the pressures in the two halves at different levels, inducing a net flow in the channels. If the right-hand chamber pressure is increased above the pressure in the left chamber, a flow of hot gas through the channels will be superposed to the diffusion process. The highly-activated segments of the channels will expand more if the pressure excess of the hot gas chamber is higher. In this way, the method of FIG. 12 permits the establishment of controllable conductivity profile in the channels.
In accordance with the present invention, a channeltype electron multiplier 33 is provided as indicated in FIG. 13, including a glass plate 34 made of two layers according to FIG. 9 or by the other methods disclosed herein. Conventional input and output electrodes 35 and 36, respectively, may be provided. However, channel surfaces 37 of plate 34 have a resistance at the input ends thereof less than that at the output ends. Preferably, the length of the surfaces of greater resistance is substantially less than the length of the surfaces of less resistance to simulate the unloaded curve of FIG. 6.
According to the present invention, an image intensifier tube 38 is shown in FIG. 14 including an evacuated envelope 39, a photocathode 40, two channel-type electron multipliers 41 and 42, and a phosphor screen 43. Multipliers 4i and 42 are, thus, cascaded. Multiplier 42 is constructed in accordance with the present invention to have a current characteristic approximately the same as that shown in FIG. 8. Multiplier 41 is conventional and is operated to give an output current approximately equal to j, shown in FIG. 8, although the output current of multiplier 41 may be greater or less than j. This, then, makes it possible to operate multiplier 42 in the higher input currentmaximum gain portion of its current characteristic between and j or higher.
A source of potential 44 maintains the conductive portions of tube 38 inside envelope 39 at appropriate operating potentials.
It should be mentioned that the channel plate of the invention is not restricted to use in a direct view tube displaying an electron image on a phosphor. It is also advantageously applicable to tubes involving other processing steps for the output electron image. These may be, for example, image storing tubes, TV pickup tubes, or image dissectors.
What is claimed is:
1. An image intensifier comprising: an electron tube including an evacuated envelope; a main channel-type electron multiplier inside said envelope; first means to supply primary electrons to one side of said main multiplier; second means to receive the electron output of said main multiplier from the other side thereof, said main multiplier including a dielectric plate having holes 5 extending therethrough from said one side to said other side thereof, said main multiplier having internal surfaces defining said holes, said internal surfaces being adapted to support secondary emission, first portions of said surfaces adjacent said one side thereof, having a lower resistance than second portions adjacent said other side thereof, said main multiplier having conductive input and output layers on said one and said other side thereof, respectively, each layer having holes therethrough in registration with said plate holes; and third means to maintain conductive portions of said tube including said layers at predetermined operating potentials.
2. The invention as defined in claim I, wherein said third means includes conductive leads sealed through said envelope in electrical contact with said conductive layers.
3. The invention as defined in claim 1, wherein said input layer is in electrical contact with the ends of said first portions of said hole surfaces and said output layer is in electrical contact with the ends of said second portions of said hole surfaces, the mutually adjacent ends of said portions being in electrical contact with one another.
4. The invention as defined in claim 3, wherein the length of said first portions of said hole surfaces is substantially greater than the length of said second portions thereof.
5. The invention as defined in claim 4, wherein said third means includes means to maintain said output layer positive with respect to said input layer.
6. The invention as defined in claim 3, wherein said third means includes means to maintain said output layer positive with respect to said input layer.
7. The invention as defined in claim 3, wherein said tube includes an auxiliary channel-type multiplier fixed inside said envelope between said first means and said main multiplier, said auxiliary multiplier having first and second conductive electrodes, said second layer being disposed adjacent said input layer, said auxiliary multiplier also including a dielectric plate between said layers having holes therethrough, the surfaces defining said auxiliary multiplier plate holes having a uniform resistance throughout their lengths.
8. The invention as defined in claim 7, wherein said third means includes means to maintain said layers at potentials determined by the following conditions:
where E, is the potential of said first layer;
E is the potential of said second layer;
E, is the potential of said input layer; and
E is the potential of said output layer.
9. The invention as defined in claim 8, wherein said first portions of said surfaces defining said main multiplier plate holes are substantially longer than said second portions thereof.
10. A channel-type electron multiplier comprising: a dielectric plate having holes therethrough; and a conductive layer fixed in position on each side of said plate, each said layer having holes therethrough in registration with said plate holes, said plate holes having a surface resistivity lower toward one side of said plate than toward the other; and means for maintaining the layer at said one side at a lower potential than that at the other.
11. The invention as defined in claim 10 wherein the length of said surface of greater resistivity is substantially shorter than the length of said surface of lesser re-