US 3864597 A
The present invention relates to systems which utilise a very thin electron beam (diameter one-tenth of a micron or less) to reach a point upon a given surface or to write-in an information there or to read one information therefrom. The object of the invention is to achieve a relative accuracy in the order of 10<->5 to 10<->6.
Claims available in
Description (OCR text may contain errors)
tlnited States Patent 1191 Trotel DEVICE FOR INCREASING THE ACCURACY OF ADDRESSING AN ELECTRON BEAM STRIKING A TARGET  Inventor: Jacques Trotel, Paris, France  Assignee: Thomson-CSF, Paris, France  Filed: Apr. 13, 1973  Appl. No.: 350,719
 Foreign Application Priority Data Baldwin et al. 250/396 1451 Feb. 4, 1975 Primary Examiner-T. H. Tubbesing Assistant Examiner-J. M. Potenza Attorney, Agent, or Firm-Oblon, Fisher, Spivak, McClelland & Maier  ABSTRACT The present invention relates to systems which utilise a very thin electron beam (diameter one-tenth of a micron or less) to reach a point upon a given surface or to write-in an information there or to read one information therefrom. The object of the invention is to achieve a relative accuracy in the order of 10 to 10*.
Two scales are marked at two adjacent sides of a square or rectangular target. They comprise studs or spots of metal having a high secondary electron emission coefficient, these electrons being detected by an auxiliary electrode. These studs define a cross-ruled arrangement on the target, which serves for coarse guidance or location. Fine adjustment is obtained by interpolation within a small square of the cross-ruled arrangement.
10 Claims, 6 Drawing Figures PATENTED 4W5 3.864.597
SHEET 3 OF 3 ADDER' ADDER 60 y 605 94-. y
INTEGRATOI} DEVICE FOR INCREASING THE ACCURACY OF ADDRESSING AN ELECTRON BEAM STRIKING A TARGET The present invention relates to systems comprising an electron-gun and a target, located inside an evacuated enclosure. The object of the invention is to provide a suitable device in order to improve the accuracy of guidance and location of the electron beam striking said target. The term addressing will be used here to describe the operation which makes it possible to attain a given point on the target (guidance) and, to determine the coordinates of the point of impact of the electron beam. (location).
Those skilled in the art will be aware that in certain applications such as the production of masks for integrated circuits and the execution of write-in/read-out functions in relation to data stores, recourse is bad to extremely thin electron beams of the order l/lOth to l/lOOth of a micron. For a scanned width of cm, there is thus a ratio of 2.10 to 2.10 between the dimensions of the point and those of the scanned line. An accuracy of l/lOth of a micron in absolute value in a position of the point of impact, corresponds to a relative accuracy in the order of IO" to This kind of accuracy is incompatible with the unpredictable deviations of the electron beam which occur under the influence of parasitic magnetic fields around the cathode ray tube. To protect it against parasitic magnetic fields, a screening made of high-permeability magnetic material, generally mu-metal, is used. However, a difficulty still remains: the accuracy of the deflection system must be improved and, especially, the means for controlling the deflection must make it possible to locate the point of impact with the desired accuracy.
The object of the present invention is an addressing device which satisfies this requirement.
The device in accordance with the invention comprises:
two levelling-rods which define two rectangular coordinate axes, perpendicular to one another and located in the plane of the target;
graduations carried by said target levelling rods and taking the form of a substance which, under the impact of the electron beam, produces a secondary effect which differentiates said graduations from the remainder of the target and from those parts of said target levelling-rods which are located between said graduations;
means for detecting said secondary effect;
means for directing the beam, said beam successively impinging upon an element of each levelling-rod.
The invention will be better understood and other of its features rendered apparent, from a consideration of the ensuing description and the attached drawings in which:
FIG. 1 is a schematic section of an electronic masking apparatus comprising an addressing mask-maker type, equipped with the device in accordance with the invention.
FIG. 2 is a perspective view of the essential elements of the apparatus shown in FIG. 1.
FIGS. 3, 4 and 5 are explanatory diagrams pertaining to the operation of said apparatus.
FIG. 6 is a block diagram of part of the device in accordance with the invention.
In FIG. 1 a schematic section of an electronic masking apparatus incorporating an embodiment of the device in accordance with the invention, has been shown. The masks concerned here are employed in planar semiconductor technology and monolithic integrated circuits on silicon substrates.
An evacuated enclosure comprises an inlet port 11, connected to a pumping system which has not been shown, a bottom 12 with a bell-jar l3 and inside the latter a hollow base 121 designed to support a target I. By way of example, the method of attachment of the target involves a flange 3 and springs 122 which load the target against the flange. The target is generally constituted by a small plate which we will assume in this case to be rectangularly shaped. Two levelling-rods 41 and 42 (the latter being visible in FIG. 2 only) border two adjacent sides of the target.
The bell-jar 13 contains a tank 14 communicating with the outside through an opening 141 and a tank 17 communicating with the outside through an opening 171.
These tanks are designed to act as cryostats and to this end are supplied respectively with liquid nitrogen 140 and liquid helium 170. The tank 14 surrounds the tank 17 which latter in turn surrounds a cylindrical space in which the electron gun 5 containing cathode 7, is located. This space is protected from external magnetic fields (in the case of the invention) by a screen 19 which is rendered superconductive when the wall 18 against which it is located is at the temperature of the liquid helium. The screen 19 is extended at the end adjacent the target 1, in the form of a frustoconical portion 181 surrounding a further frustoconical portion 16 which is itself the extension of the gun 5 and the cylinder 51 containing the focussing and deflection elements of the cathode ray tube. The space 'between the screen 19 and the frusto-conical portion 16, is narrow. This arrangement facilitates magnetic protection and the production of a high vacuum in the neighbourhood of the gun 5, because a cryogenic pumping effect is produced along the wall 18 at the temperature of the liquid helium.
The gun 5 will preferably comprise an electron source 7 of the thin point field-excited cold-emission type, or again an electron source of some other type (the field effect benefits from the cryogenic pumping action). The electrons pass through an electrostatic or electromagnetic lense 8 and then through two successive deflection systems 9 and 10. Also, above the levelling-rod system 41 and 42, and angle piece 20, having respective sides parallel to the rods and constituting an electrode for picking up secondary electrons, is located.
In perspective, in FIG. 2, the electron beam 50 emitted by the source 7 and passing successively through the lense 8, the devices 9 and 10 which deflect the beam so that for example it strikes the levelling-rod 41 (direction OX of deflection), can be seen. These devices are deflection units which have been illustrated, by way of example,.in the form of double pairs of deflection plates. The plates 91 and 101 are OY deflection plates perpendicular to OX. They have been respectively connected to the terminals and 105. Similarly, the plates 92 and 102 are OX deflection plates, respectively connected to the terminals 94 and 104.
The deflection device 9, which has a low deflection amplitude is designed to improve the accuracy of addressing carried out by the deflection device 10, the latter being the main, high-amplitude deflector device. The device 9 is designed so that the positional error in the point of impact of the beam upon either of the levelling-rods, is less than the width of a scale line and therefore much smaller than the gap between two scale lines. In the following, other conditions which have to be satisfied by the devices 9 and 10, will be described.
The levelling-rods 41 and 42 are connected in parallel to the terminal 401 whilst the electrode 20 is connected to a terminal 402.
In the following: I, is the current flowing from the terminal 401 to the levelling-rods, I, is the current through the electrode 20, flowing between it and the terminal 402; I; is the electron beam current. Taking account of the secondary emission effect, we have:
However, since in fact. at the terminals 401 and 402, we have the respective currents I and I,, we can express in these terms by rewritting equation (I), to give:
I 'l' I In FIG. 3, a part of the levelling-rod 41 has been shown, on which a rectangular graduation has been indicated in the form for example of a gold stud or spot deposited upon a chromium substrate which constitutes the remainder of the levelling-rod. The zone of impact of the electron beam has been marked F. The aim is to locate the position of the beam, in which the two aereas H and H of this zone of impact having the side of a graduation N, as their common limit, have respective surfaces equal to one another.
Calling K, and K the secondary emission coefficients of chromium and gold, we have:
In the case where F only covers chromium; and
I, k I,
in the case where F only covers gold.
For the position in which H H we should obtain:
k (k k )/2 In FIG. 4, the variations in I, plotted as a function of the distance 1 measured algebraically along an axis OZ parallel to the levelling-rod 41 and having its origin at the left-hand edge of the graduation N as indicated in FIG. 3 (Z: distance from the centre of F to the axis 00), have been represented.
Taking the equations (2) and (6), at the point C where the graph of I, intersects the ordinates, we have:
This is the result which should be observed at the terminals 401 and 402 at the instant of passage of the centre of the zone F across the edge N,.
In fact, since it would be too difficult to manually verify the equation (7), recourse is had to a servo device which instantaneously measures the interval between the true position of F and the desired position (H H and translates the deviation into an error signal of such a sign that, through the deflector 9, it controls deflection of the beam 50 in a direction which shifts the beam 50 towards the desired position astride the edge of a graduation.
By way of example, in FIG. 6 a single-wire diagram of a servo-device comprising the following elements, has been shown:
A comparator-detector 601;
two two-pole two-position (I and Il) switches 606 and two integrators 602 and 603;
two adders 604 and 605, with controllable outputs.
It will be assumed first of all that the switches 606 and 607 are in the position I. The device 10 is then supplied in such a fashion (voltage V, at the terminal 104, voltage V, at terminal that the beam 50 impinges upon the levelling-rod 41 in the neighbourhood of the graduation N The comparator-detector 601 supplies the terminals 401 and 402 with a positive direct voltage and, as we have seen, currents I and I, are generated as a consequence. The detected error signal is applied to the input E of the adder 604 across the integrator 602. There are two other inputs to the adder. the purpose of which will be explained hereinafter. The sole output of the adder 604 is connected to the terminal 94 (OX deflection).
If we now assume that the switches 606 and 607 are both in the position II, the system functions as if V, had been replaced by V E, by E and the levelling-rod by the levelling rod 42.
Contrarily to that if the switch 606 is switched in position II and the switch 607 in position I, the connections between the elements 601, 602 and 603, are broken and the servomechanism disconnected.
The operation of the device will now be described by successively considering the two following problems:
first problem: guiding the beam and stopping it at the predetermined point P on the target 1; second problem, locating the position of the beam by measuring the coordinates of the point of impact. First problem:
In FIG. 5, a cross-ruled arrangement constructed on two rectangular axes OX and CY, has been shown which we will assume to be traced upon the plane of the target 1, this arrangement having been extended to contain the graduations on the levelling-rods 41 and 42, which are located along the axes OX and CY. Let Q, be a point having the coordinates N N such that for the point P (x, y), we have:
where x and y are less than unity (interval between two graduations of levelling-rod).
The method of utilising the overall device consists in adjusting the beam to the point Q (FIG. with a strict accuracy and then displacing the beam with a lesser accuracy to obtain an impact upon the point P.
The method described hereinafter assumes that the following conditions are satisfied:
a. Accuracy of control of the beam when shifted by the deflector 10, better than unity (interval between two scale lines);
b. Deflector has a scanning range which includes the target 1 and the levelling rods 41 and 42;
c. Accuracy of control of the beam when shifted by the deflector 9, better than the thickness of a scale line;
d. Range of scanning of the beam when shifted by the deflector 9, at least equal to unity;
e. Stability of the voltage supplies to the deflectors to be maintained during a time longer than a predetermined value, for example longer than that which is broadly required to execute a complete pattern in the case of a production of a mask.
Control of the beam to bring it to the point O is carried out in two stages. First of all, the switches 606 and 607 (FIG. 6) are switched in position I and the beam deflected in the OX direction. Then, the switches 606 and 607 are switched in position II and the beam deflected in the direction OY. Finally the servo-device is disconnected and the beam displaced simply under the control of the device 9, by means of the voltage u, and a applied respectively to the corresponding inputs of the adders 604 and 605. This presumes that the device 9 has been previously calibrated.
This method is the converse of the preceding one. The point of impact of the beam being given, but its coordinates being unknown, control is varied until the beam strikes a point of known coordinates, for example the point 0,, and from this the coordinate of the point of impact are determined from the variation u, and u, in the voltages across the terminals of the device 9.
Distortions which may be exhibited by the device 10 must be taken into account so that the addressing of the electron beam takes place in an identical way from one apparatus to another. To this end, a start is made by storing the corrections Au, and Au which have to be made to the voltages u, and u in order to bring the point 0 (N N,,) to the point 0,, which is the intersection of standard cross-ruling arrangement applied to the target 1. Calibration can be carried out, for example, by utilising the apparatus as a scanner-type electron microscope, in order to show the standard crossruling arrangement.
The voltages 14,, u, and the corrections Au, and Au are introduced into the servodevice of FIG. 6, at the auxiliary inputs of the adders 604 and 605.
Finally, the drift in the supply voltages over a period of time must not be neglected. The time T at the end of which this drift reaches the same order of magnitude as the resolving power of the apparatus, must be determined. The duration of the calibrating and addressing operations should be substantially less than the time T, if valid results are to be obtained.
By way of order of magnitude, a time of I second for the time T is indicated, and the operations can be carried out in less than 1 ms in the case of the servodevice described hereinbefore.
The invention incorporates variant embodiments in which the graduations of the levelling rods, under the effect of bombardment by the electrons of the beam are the location of secondary effects which differ from the secondary emission phenomenon, for example:
the emission of back-diffused electrons;
the emission of X-rays;
the emission of light;
the creation of charge carriers in the material.
The invention applies to numerous cases where use is made of a very thin electron beam which can be deflected in order to reach any point upon a given surface, either to deposit information there or to read out a piece of information.
Amongst these applications, the following can be listed:
television tubes: the information furnished by the electron beam constitutes an image which is rendered visible by a luminescent deposit;
television camera tubes: the information read by the beam is an optical image;
certain kinds of storage tubes for computers where the information is stored or read-out by electron beam respectively on or from appropriate data carriers;
peripheral equipment of computers, which displays results of calculations upon the screen of a cathode ray tube;
mask-making machines for the manufacture of semiconductor devices where the resist of which the mask is made is exposed to a very thin electron beam; certain machines for testing semiconductor devices in which the electronic state of certain points of the circuit can be measured by the effects produced there under the impact of an electron beam;
scanner-type electron microscope, microscopic sensing devices, etc.
What is claimed is:
1. A device for increasing the accuracy of addressing an electron beam striking a surface, comprising in an evacuated enclosure a cathode capable of emitting an electron beam in an electric field;
an electron beam deflection system;
an anode consisting of a target surface;
two levelling-rods perpendicular to one another said levelling rods respectively having graduations, said graduations being made of a substance responding to electron impact, in another manner than that of the remainder of the rods; means for detecting the response to said electron impact; means for adressing a predetermined point of said target, by scanning successively in the directions of one rod, and of the other, for impinging upon the corresponding graduation of the rods.
2. A device as claimed in claim I, wherein said response is secondary emission of electrons.
3. A device as claimed in claim 1, wherein said deflection system comprises a servodevice which causes the beam to displace to a position astride a predetermined edge of an arbitrary graduation on one or the other of said levelling-rods.
7 8 4. A device as claimed in claim 1, wherein said subval which is taken as unity; stance is gold, the remainder of said levelling-rods b. control of the deflection in order to bring the imbeing chromium. pact point of the beam to:
5. A device as claimed in claim 1, wherein said enclosure further comprises around said cathode a screen which is rendered superconductive by thermic contact Y with a cryogenic source consisting of a reservoir containing a liquid gas surrounding said screen.
6. A device as claimed in claim 1, wherein said dec. control of the deflection in order to bring the impact point of the beam to:
flection means comprise a first and a second unit; said 10 X 0 first unit serving to move the beam into the neighbour- Y N hood of a predetermined graduation, an said second unit serving to locate the beam astride one of the edges C t o f he eflection in Order to bring the imof said graduation. pact point the beam to the point:
7. A method of addressing an electron beam utilising N, a device as claimed m claim 1, comprising the following stages: u
dennficanon predeiermmed e. control of the deflection in order to achieve the get surface, by its coordinates final point (X Y) X NJr x 8. A cathode ray tube comprising a device as claimed in claim I. Y y 9. An electron microsco e incor oratin a device as a I v p p g where N and N,, represent the number ofmtervals beclaimed in claim 1. tween graduations on each of levelling-rods arranged 10. An electronic mask-making device comprising a on the reference axes of the system (XX), and the device as claimed in claim 1. quantities x and y represent fractions of such an inter-