GB2081048A

GB2081048A - Position Control Circuit

Info

Publication number: GB2081048A
Application number: GB8122073A
Authority: GB
Original assignee: Optimetrix Corp
Current assignee: Optimetrix Corp
Priority date: 1979-02-27
Filing date: 1980-02-21
Publication date: 1982-02-10
Also published as: DE3006395A1; FR2451064A1; FR2451065A1; GB2046955B; GB2046955A; GB2081048B; FR2451064B1; FR2451065B1

Abstract

A position control signal is produced in response to the phase difference between a reference signal f1-f2 and an input signal f1- f2+/-2 DELTA fL1. The latter is produced by the "interferometer" system used in the orthogonally controlled stage of the parent specification. In that system a control signal operates a servo motor to position the stage until a predetermined number of cycles of a difference (doppler) frequency 2 DELTA fL1 (produced by the stage movement) have been counted. The present invention is then brought into use to position to a part (one eight) of a cycle. This is effected by phase shifting the reference frequency by circuit 118 to produce at 126 eight signals, all of the same reference frequency but phase shifted by 45 DEG . The desired arc is selected at 128 and used for the phase comparison at 130. An identical system is used in an orthogonal plane and the sum and difference of the two systems produce the ultimate control signals at 116, 116'. <IMAGE>

Description

1

GB 2 081 048 A 1

SPECIFICATION Position Control Circuit

This invention relates generally to interferometrically controlled stages movable 5 along X and Y axes for positioning or aligning a first object, such as a photomask or a semiconductive wafer, with respect to a second object, such as a reticle, or an image thereof, and more specifically to position control circuits for an 10' interferometrically controlled stage movable along precisely orthogonal X and Y axes for successively positioning or aligning different . regions of a photomask or a semiconductive wafer with respect to the same reticle or image 15 thereof.

The invention is related to the invention disclosed and claimed in the specification of our copending UK patent application number 8005832.

20 In the semiconductor industry interferometrically controlled stages movable along X and Y axes are employed both in the fabrication of photomasks and in the processing of semiconductive wafers to form integrated circuits 25 and the like. A high (submicron) resolution photomask is typically fabricated by employing such an interferometrically controlled stage to position, successively, different regions of the photomask with respect to a reticle, or an image 30 of a reticle, representing a level of microcircuitry to be printed on the photomask at each of those regions. This step-and-repeat printing operation forms an array of adjacent regions of microcircuitry of one level on the photomask in 35 rows and columns parallel to the X and Y axes of motion of the interferometrically controlled stage. A set of such photomasks, each bearing an array of microcircuitry of a different level is typically employed in the fabrication of integrated circuits 40 or the like from a semiconductive wafer. In the course of this fabrication, the semiconductive wafer is sequentially aligned with each photomask of the set and the level of ; microcircuitry printed on the photomask is in turn 45 printed on the semiconductive wafer. However, it is also possible to eliminate the operation of , forming a set of such photomasks by employing the interferometrically controlled stage to successively align different regions of the 50 semiconductive wafer with each of the reticles employed in fabricating the set of photomasks so that the level of microcircuitry represented by each of those reticles may be printed directly on the semiconductive water at each of those 55 regions during separate step-and-repeat printing operations.

In order to facilitate the precise positioning or alignment of an array of adjacent regions of one level of microcircuitry being printed on a 60 photomask, or on a semiconductive wafer, relative to each array of adjacent regions of microcircuitry of another level previously printed or yet to be printed on the other photomasks of the same set, or relative to each array of adjacent regions of microcircuitry of another level previously printed or yet to be printed on the semiconductive wafer, it would be highly desirable to employ an interferometrically controlled stage having precisely orthogonal X and Y axes of motion for step-and-repeat printing operations such as those described above. Unfortunately, however, conventional interferometrically controlled stages do not have precisely orthogonal X and Y axes of motion. Moreover, the degree of nonorthogonality of the X and Y axes of motion of such stages is normally different from stage to stage so that different stages have different frames of reference and cannot therefore be employed interchangeably in printing different levels of microcircuitry on different photomasks of the same set or on the same semiconductive wafer or batch of semiconductive wafers.

Conventional interferometrically controlled stages typically employ a separate interferometer system for each axis of motion of the stage with a first movable mirror of the interferometer system for the X axis of motion being mounted on the stage in a vertical plane normal to the X axis of motion and with a second movable mirror for the Y axis of motion being mounted on the stage in a vertical plane normal of the Y axis of motion, as • shown in British Patent No. 1,196,281 entitled INTERFEROMETRICALLY CONTROLLED POSITIONING APPARATUS.

Since these mirrors must be disposed in vertical planes precisely orthogonal to one another for the stage to have precisely orthogonal X and Y axes of motion, special measurement equipment and procedures involving considerable effort and expense are employed to mount and maintain these mirrors in vertical planes as closely orthogonal to one another as possible. However, since even the best measurement equipment has a finite accuracy, it is in fact not possible to mount and maintain these mirrors in precisely orthogonal vertical planes. As a result the stage does not have precisely orthogonal X and Y axes of motion.

According to the illustrated preferred embodiment disclosed in the specification and drawings of our copending U.K. patent application number 8005832, a stage comprises a platform movable along X and Y axes in a horizontal plane, and has fixedly mounted first and second movable plane mirrors of first and second interferometer systems, respectively, on the platform in vertical planes intersecting one another at the Y axis with the first and second movable mirrors symmetrically disposed about the Y axis. First and second stationary plane mirrors are fixedly mounted above the platform on a housing of a projection lens or some other such utilization device and are disposed parallel to the first and second movable mirrors, respectively. The first interferometer system has a first measurement path normal to the first movable mirror and a first reference path normal to the first stationary mirror. As the platform is moved along either the X or the Y axis, the first interferometer system

65

70

75

80

85

90

95

100

105

110

115

120

125

2

GB 2 081 048 A 2

produces a first measurement signal indicative of the velocity of the first movable mirror while it is being moved (relative to the first stationary mirror) along the first measurement path. 5 Similarly, the second interferometer system has a second measurement path normal to the second movable mirror and a second reference path normal to the second stationary mirror. As the platform is moved along either the X or the Y axis, 10 the second interferometer system produces a second measurement signal frrdicative of the velocity of the second movable mirror while it is being moved (relative to the second stationary mirror) along the second measurement path. In 15 response to sums and differences of these first and second measurement signals, first and second position control circuits move the platform along precisely orthogonal X and Y axes with the Y axis bisecting the angle between the first and 20 second movable mirrors. Thus, the stage is provided with precisely orthogonal X and Y axis of motion without requiring the first and second movable mirrors to be mounted in precisely orthogonal vertical planes and without requiring 25 any other such unattainable relationship between those mirrors or oth§r parts of the stage. This eliminates the principal source of degradation in the orthogonality of the X and Y axes of motion of' the stage. By comparison, other sources of 30 degradation, such as unevenness of the first and second movable mirrors, are insignificant and are therefore disregarded for purposes of this application.

Position control circuits may be employed, for 35 example, to control the position of an interferometrically controlled stage as described hereinafter.

Conventional position control circuits for controlling the position of an interferometrically 40 controlled stage typically employ a reversible or up-down counter to provide an indication of the actual position of the stage as described in U.S. Patent No. 3,458,259 entitled INTERFEROMETRIC SYSTEM and issued on July 45 28,1969. The resolution of such position control circuits is therefore typically limited by the ambiguity of the last or least significant digit indicated by the counter.

The present invention provides a position 50 control circuit comprising first means for producing a difference in phase between a reference signal and a related input signal as determined by a control signal, and second means, coupled to the first means, for producing a 55 position control signal proportional to the difference in phase between the reference signal and the input signal.

In a circuit as set forth in the last preceding paragraph, it is preferred that said input signal has 60 a frequency related to the frequency of the reference signal.

In a circuit as set forth in the last preceding paragraph, it is preferred that said first means is responsive to the reference signal and to the 65 control signal for producing an output signal having the same frequency as the reference signal and having a phase determined by the control signal, and said second means is coupled to the first means and responsive to the output signal from the first means and to the input signal for producing a position control signal proportional to ifte difference in phase between the input signal fcftd the output signal from the first means.

In a circuit as set forth in any one of the last three immediately preceding paragraphs, it is preferred that said first means comprises a ' variable phase shifter, and said second means comprises a phase detector.

In a circuit as set forth in the last preceding paragraph, it is preferred that said variable phase shifter comprises a voltage controlled oscillator for producing an output signal having a frequency N times greater than the frequency of the reference signal, a division circuit, coupled to the voltage controlled oscillator, for producing an output signal having a frequency equal to the frequency of the output signal from the voltage controlled oscillator divided by N, another phase detector, coupled to the division circuit and to the voltage controlled oscillator and responsive to the output signal from the division circuit and to the reference signal, for driving the voltage controlled oscillator to produce an output signal having a frequency N times greater than the frequency of the reference signal, a shift register, coupled to the voltage controlled oscillator and to the last-mentioned phase detector, for producing N output signals of difference phase, and a data selector, coupled to the shift register and to the first-mentioned phase detector and responsive to the control signal, for applying a selected one of the output signals from the shift register as determined by the control signal to the first-mentioned phase detector.

A circuit as set forth in the last preceding paragraph but one, and for positioning a utilization device, may comprise counter means, responsive to the reference signal and the related input signal, for producing an actual position signal indicative of the actual position of the j utilization device, register means for receiving a desired position signal indicative of a desired position of the utilization device, comparator, means, for producing a comparison signal equal to the difference between the actual and desired position signals, control means, coupled to the comparator means, for producing a velocity control signal, and drive means, responsive to the velocity and position control signals, for moving the utilization device to the desired position.

In a circuit as set forth in the last preceding paragraph, it is preferred that said counter means is also responsive to another input signal for producing the actual position signal, said other input signal is also related to the reference signal, said circuit comprises combining means, coupled to the phase detector, for combining the position control signal with another position control signal that is proportional to a difference in phase between the reference signal and said other input

70

75

80

85

90

95

100

105

110

115

120

125

130

3

GB 2 081 048 A 3

signal to produce a combined position control signal, said drive means comprises a motor for positioning the utilization device, and servo drive means, coupled to the motor, for driving the 5 motor, said control means comprises register means for receiving desired velocity signals, tachometer means for producing an actual velocity signal proportional to the actual velocity of the motor, and comparator means, coupled to 10 the last-mentioned register means and the ■ tachometer means, for producing velocity control signals proportional to the difference between each desired velocity signal and the actual ? velocity signal, and said drive means further 15 comprises selector means, coupled to the last-mentioned comparator means and the combining means, for selectively applying the velocity control signals and the combined position control signal to the servo drive means so as to drive the 20 motor for moving the utilization device to the desired position.

The present invention further provides a method of processing a reference signal of a first frequency and a displacement input signal of a 25 second frequency related to the frequency of the reference signal to provide a position control signal of extended resolution, said method comprising the steps of producing a difference in phase between the reference signal and the 30 displacement input signal, and detecting the difference in phase between the reference signal and the displacement input signal to generate a position control signal proportional to that difference in phase.

35 A method as set forth in the last preceding paragraph may further comprise the step of combining the position control signal with another position control signal that is proportional to a difference in phase between the reference 40 signal and another displacement input signal to generate a combined position control signal equal to the sum or the difference of those position control signals.

The illustrated preferred embodiment of the 45 present invention comprises a position control circuit having a variable phase shifter responsive to an input reference signal and to an input . control signal for producing an output signal of the same frequency as the reference signal but 50 shifted in phase as determined by the input control signal, and by employing a phase detector responsive to the output signal from the variable phase shifter and to an input measurement signal of a frequency related to the frequency of the 55 input reference signal for producing a position control signal extending the resolution of the position control circuit. The variable phase shifter comprises another phase detector responsive to the input reference signal and to an output signal 60 from a divide-by-N circuit for driving a voltage controlled oscillator to supply the divide-by-N circuit with an output signal having a frequency N times greater than the frequency of the reference signal. A shift register is responsive to the output 65 signals from both the voltage controlled oscillator and the divide-by-N-circuit for supplying N output signals.of different phase to a data selector. The data selector is responsive to the input control signal for supplying a selected one of these N output signals to the first-mentioned phase detector as determined by the input control signal.

There now follows a detailed description,

which is to be read with reference to the accompanying drawings, of position control circuits according to the invention, each of which is incorporated in an interferometrically controlled stage as disclosed and claimed in our aforesaid copending U.K. patent application; it is to be clearly understood that these position control circuits have been selected for description to illustrate the invention by way of example and not by way of limitation.

In the accompanying drawings:—

Figure 1 is a perspective rear view of an interferometrically controlled stage having precisely orthogonal X and Y axes of motion.

Figure 2 is a detailed schematic representation of one of the interferometer systems employed with the stage of Figure 1 ;

Figure 3 is a detailed block diagram of a pair of position control circuits employed for driving the stage of Figure 1 and constructed in accordance with the present invention;

Figure 4 is a detailed block diagram of another pair of position control circuits constructed in accordance with the present invention and employed, for example, in place of the position control circuits of Figure 3 to drive the interferometrically controlled stage of Figure 1; and

Figure 5 is a detailed block diagram of a pair of phase control circuits constructed in accordance with the present invention and employed in the position control circuits of Figure 4.

Referring now to Figure 1, there is shown an interferometrically controlled stage 10 for use in aligning a workpiece such as a semiconductive wafer 12 with an object such as a reticle 14 or a projected image of the reticle. The stage 10 comprises a lower platform 16 supported by air bearings on the flat upper surface of a granite block 18 for movement generally along an X axis of the stage, and an upper platform 20 supported by air bearings on the flat upper surface of the granite block 18 (through clearance openings in the lower platform 16) for movement generally along an orthogonal Y axis of the stage. In addition, the upper platform 20 is coupled to the lower platform 16 for movement therewith generally along the X axis of the stage 10. Thus, the upper platform 20 of the stage 10 may be moved in a horizontal plane generally along the orthogonal X and Y axes of the stage and, since such movements may occur simultaneously, may be moved along any straight line in that horizontal plane.

The semiconductive wafer 12 is held by a vacuum chuck 22 mounted on the upper platform 20 for movement therewith. Chuck 22 is

70

75

80

85

90

95

100

105

110

115

120

125

130

4

GB 2 081 048 A 4

positioned beneath a projection lens 26 or some other such utilization device for use in processing the semiconductive wafer 12. The reticle 14 is held by a vacuum holder 28 positioned directly 5 above the projection lens 26 and along an optical axis 24 thereof. Both the projection lens 26 and the holder 28 for the reticle 14 are mounted on a frame attached to the granite block 18. In the process of fabricating integrated circuits or the 10 like from the semiconductive wafer 12, the stage 10 is moved along the X and Y axes to successively align adjacent regions of one level of microcircuitry that may have already been formed on the semiconductive wafer with an image of 15 another level of microcircuitry contained on the reticle 14 and yet to be printed on the semiconductive wafer at each of those regions. This image is projected onto the semiconductive wafer 12 by the projection lens 26. 20 In order to provide the stage 10 with precisely orthogonal X and Y axes of motion, two elongated plane mirrors 30 and 32 are fixedly mounted on the upper platform 20 for movement therewith. These mirrors (hereinafter being referred to as the 25 first and second movable mirrors 30 and 32) are disposed symmetrically about the Y axis in respective first and second vertical planes intersecting one another at the Y axis at an angle of 26. No special measurement equipment or 30 critical measurement procedures are required in mounting the first and second movable mirrors 30 and 32 on the upper platform 20 of the stage 10 since, as hereinafter described, the stage is controlled so that the X and Y axes are precisely 35 orthogonal to one another with the Y axis , bisecting the angle 29 between the first and second movable mirrors. The first and second movable mirrors 30 and 32 may therefore be mounted in respective first and second vertical 40 planes intersecting one another at virtually any angle, and, for purposes of illustration, are shown as being fixedly mounted on the upper platform 20 of the stage 10 at a nominal right angle to one another by a carrier 34. First and second plane 45 mirrors 36 and 38 are fixedly mounted on a housing of the projection 26 above the carrier 34. These mirrors (hereinafter being referred to as the first and second stationary mirrors 36 and 38) correspond and are disposed parallel to the first 50 and second movable mirrors 30 and 32, respectively.

First and second interferometer systems 40 and 42 are employed to precisely measure the velocities of the first and second movable mirrors 55 30 and 32 (relative to the first and second stationary mirrors 36 and 38) while they are being moved along corresponding first (or AL,) and second (or AL2) measurement paths normal to the first and second movable mirrors, 60 respectively, as happens whenever the stage 10 is moved along either the X or the Y axis, and to produce measurement signals indicative of those velocities. Interferometer systems such as those manufactured and sold by Hewlett-Packard 65 Company and described in detail in Hewlett-

Packard Company's Application Note No. 197—2 for the 5501A laser transducer and in the aforementioned U.S. Patent No. 3,458,259 may be employed as the first and second interferometer systems 40 and 42. The interferometer systems 40 and 42 share a two frequency single mode laser transducer 44, such as the Hewlett-Packard 5501A laser transducer, for emitting a beam of light 46 including a first component having a frequency f, (hereinafter referred to as f, light) and a second parallel and overlapping component having a frequency^ (hereinafter referred to as f2 light). These parallel and overlapping components off, and f2 light? have linear horizontal and vertical polarizations (relative to the laser transducer 44), respectively. A beam bender 48 is employed to deflect the beam of light 46 from the laser transducer 44 upward to a beam splitter 50, which transmits fifty percent of the beam of light upward through an aperture 52 in a frame 54 for holding the block of granite 18. Beam splitter 50 also reflects fifty percent of the beam of light 46 from the laser transducer 44 to another beam bender 56 from which it is in turn deflected upward through an aperture 58 in frame 54.

The laser transducer 44, the beam benders 48 and 56, the beam splitter 50, and the various elements of the first and second interferometer systems 40 and 42 hereinafter described may all be mounted on a frame attached to the granite block 18 in the configuration shown. With the first and second movable mirrors 30 and 32 mounted on the carrier 34 at nominally forty-five degrees with respect to the Y axis as shown, the AL, and the AL2 measurement paths of the first and second interferometer systems 40 and 42, respectively, are rotated nominally forty-five degrees with respect to the Y axis. Thus, with the laser transducer 44 mounted along the X axis as shown, the laser transducer must also be rotated nominally forty-five degrees with respect to the Y axis as shown to orient the polarizations of the f, light and the f2 light at forty-five degrees with respect to the Y axis and hence parallel and = orthogonal to each of the first and second interferometer systems 40 and 42. This is essential since maximum output signal is obtained from the first and second interferometer systems when those polarizations are so oriented, whereas virtually no output signal can be obtained from the first and second interferometer systems when those polarizations qre oriented at forty-five degrees with respects to each of the first and second interferometer systems.

Since the first the second intertarometer systems 40 and 42 are identical, the same reference numbers are generally employed for the same elements of both interferometer systems (with reference numbers for those of the second interferometer system being primed), and only the first interferometer system 40 is described in detail. Referring now also to Figure 2, the first interferometer system 40 includes a polarizing beam splitter 60 for reflecting fi light of linear

70

75

80

85

90

95

100

105

110

115

120

125

130

5

GB 2 081 048 A 5

horizontal polarization (represented by a double-headed arrow normal to the plane of the paper) passing through the aperture 52 of the frame 54 and for transmitting the f2 light of linear vertical 5 polarization (represented by a doubie-headed arrow in the plane of the paper) passing through the aperture 52 (an auxiliary arrowhead in the plane of the paper is associated with each double-headed arrow to indicate the direction of 10 propagation of the light). The f, light reflected by i the polarizing beam splitter 60 passes forward through a quarter wave plate 62 to the first movable mirror 30 along a first portion 64 of the • AL, measurement path which, as already 15 described, is normal to the first movable mirror. As the upper platform 20 of the stage 10 is moved along either the X or the Y axis, the corresponding movement of the first movable mirror 30 (relative to the first stationary mirror 20 36) along the AL, measurement path causes the f, light to under-go a frequency change of ±Af as it is reflected from the first movable mirror backward along the first portion 64 of the AL, measurement path and through the quarter wave 25 plate 62. the quarter wave plate 62 converts the polarization of the f, light passing forward therethrough to righthand circular polarization, which is in turn converted to left-hand circular polarization as the f, light is reflected from the 30 first movable mirror 30, and converts the polarisation of the f,± Af light reflected backward therethrough to linear vertical polarization. Thus, the f,±Af light is transmitted by the polarizing beam splitter 60 to an attached retro-reflector 66 35 from which it is reflected forward through the polarizing beam splitter and quarter wave plate 62 to the first movable mirror 30 along a second portion 68 of the AL, measurement path. The f,±Af light reflected from the first movable mirror 40 30 backward along the second portion 68 of the AL, measurement path undergoes another frequency change of ±Af as the upper platform 20 of the stage 10 is moved along either the X or the Y axis. In this instance the quarter wave plate 45. 62 converts the polarization of the f,±Af light passing forward therethrough to left-hand circular polarization, which is in turn converted to right-, hand circular polarization as the f,±Af light is reflected from the first movable mirror 30, and 50 converts the polarization of the f,+2Af light reflected backward therethrough to linear horizontal polarization. The f,±2Af light reflected backward along the second portion 68 of the AL, measurement path is therefore reflected by the 55 polarizing beam splitter 60 downward through a mixing polarizer 70 to a photoelectric detector 72.

In a similar manner, the f2 light transmitted by the polarizing beam splitter 60 passes forward through a quarter wave plate 74 to a beam 60 bender 76 from which it is reflected to the first stationary mirror 36 along a portion 78 of a AL, reference path which, as already described above, is normal to the first stationary mirror (at least from the beam bender 76 forward). This f2 light is 65 reflected from the first stationary mirror 36

backward along the first portion 78 of the AL, reference path to the beam bender 76 and then through the quarter wave plate 74. The quarter wave plate 74 converts the polarization of the f2 light passing forward therethrough along the first portion 78 of the AL, reference path to the left-hand circular polarization, which is in turn converted to right-hand circular polarization as the f2 light is reflected from the first stationary mirror 36, and converts the polarization of the f2 light reflected backward therethrough along the_ . first portion 78 of the AL, reference path to linear horizontal polarization. Thus, the f2 light reflected backward from the first stationary mirror 36 along the first portion 78 of the AL, reference path is reflected by the polarizing beam splitter 60 to the retroref lector 66 from which it is reflected back to the polarizing beam splitter where it is reflected through the quarter wave plate 74 and deflected by the beam bender 76 to the first stationary mirror along a second portion 80 of the AL, reference path. This f2 light is reflected again from the first stationary mirror 36 backward along the second portion 80 of the AL, reference path to the beam bender 76 and then through the quarter wave plate 74. The quarter wave plate 74 converts the polarization of the f2 light passing forward therethrough along the second portion 80 of the AL, reference path to right-hand circular polarization, which is in turn converted to left-hand circular polarization as the f2 light is reflected again from the first stationary mirror 36, and converts the polarization of the f2 light reflected backward therethrough along the second portion 80 of the AL, reference path to linear vertical polarization. The f2 light reflected backward along the second portion 80 of the AL, reference path is therefore transmitted by the polarizing beam splitter 60 downward through the mixing polarizer 70 to the photoelectric detector 72 with the parallel and overlapping" f,±2Af light from the second portion 68 of the AL, measurement path in an output beam of light 73. For simplicity of illustration the paths of the input light beam 46 entering the polarizing beam splitter 60 and the output light beam 73 entering the photoelectric detector 72, the first and second portions 64 and 68 of the AL, measurement path, the first and second portions 78 and 80 of the AL, reference path, and the retroreflector 66 have been represented as being spatially disposed in the plane of the paper in Figure 2, whereas they are actually spatially disposed in a plane normal to the plane of the paper as shown in the perspective view of Figure 1.

The mixing polarizer 70 mixes the f,±2Af light and the parallel and overlapping f2 light of the output light beam 73 passing therethrough to provide each of those components of the output light beam entering the photoelectric detector 72 with a component of similar polarization. These similarly polarized components are mixed by the photoelectric detector 72 to produce a first electrical measurement signal having a frequency f,—f2±2AfLl at the output of the photoelectric

70

75

80

85

90

95

100

105

110

115

120

125

130

6

GB 2 081 048 A 6

detector. A second electrical measurement signal having a frequency f,—f2±2 AfL2 is produced in the same manner as described above by the second interferometer system 42 at the output of the 5 photoelectric detector 72' of that system.

Referring now also to Figure 3, the first electrical measurement signal of frequency f,—f2±2AfLl is applied to a first input of a first dual output mixer 82 of the first interferometer system 10 40, and the second electrical measurement signal of frequency f,—f2+2AfL, is applied to a first input of a second dual output mixer 84 of the second interferometer system 42. An electrical reference signal of frequency f,—f2 produced by the laser 15 transducer 44 at an electrical output 86 thereof (see Figure 1) is applied to a second input of the first dual output mixer 82 and to a second input of the second dual output mixer 84. The first dual output mixer 82 combines the first measurement 20 signal and the reference signal to produce a first pulse train measurement signal (hereinafter simply referred to in this description as the first Dulse train signal) havina a reDetition rate of 2AfLf on an up or a down output thereof as 25 determined by whether the sign of the ±2AfLl component of the frequency of the first measurement signal is po'sitive or negative, respectively. The repetition rate of this first pulse train signal is proportional to the velocity of the 30 first movable mirror 30 while it is being moved (relative to the first stationary mirror 36) along the AL, measurement path of the first interferometer system 40, as happens whenever the upper platform 20 of the stage 10 is moved along either 35 the X or the Y axis of motion of the stage. Similarly, the second dual output mixer 84 combines the second measurement signal and the reference signal to produce a second pulse train measurement signal (hereinafter simply 40 referred to in this description as the second pulse train signal) having a repetition rate of 2AfLlon an up or a down output thereof as determined by whether the sign of the ±2AfLj component of the frequency of the second measurement signal 45 is positive or negative, respectively. The repetition rate of this second pulse train signal is proportional to the velocity of the second movable mirror 32 while it is being moved (relative to the second stationary mirror 38) along the AL2 50 measurement path of the second interferometer system 42, as also happens whenever the upper platform 20 of the stage 10 is moved along either the X or the Y axis of motion of the stage.

Pulses of the first and second pulse train 55 signals appearing on the up outputs of the first and second dual output mixers 82 and 84 are applied to a first pair of inputs of a first dual adder 88, which produces a pulse train representing the sum of those pulses on an up output of the first 60 dual adder. Similarly, pulses of the first and second pulse train signals appearing on the down outputs of the first and second dual output mixers 82 and 84 are applied to a second pair of inputs of the first dual adder 88, which produces a pulse 65 train representing the sum of those pulses on a down output of the first dual adder. The trains of pulses thereby produced on the up and down outputs of the first dual adder 88 represent the sum of the first and second pulse train signals. Pulses of the first pulse train signal appearing on the up output of the first dual output mixer 82 and pulses of the second pulse train signal appearing on the down output of the second dual output mixer 84 are applied to a first pair of inputs of a second dual adder 90, which produces the sum of those pulses on an up output of the second dual adder. Similarly, pulses of the first pulse train signal appearing on the down output of the second pulse train signal appearing on the up output of the second dual output mixer 84 are applied to a second pair of inputs of the second dual adder 90. The sums of pulses thereby produced on the up and down outputs of the second dual adder 90 represent the difference of the first and second pulse train signals.

In response to the difference and the sum of the first and second pulse train signals and to X and Y digital end point data signals received, for example, from a computer 92, X and Y axes position control circuits 94 and 96 move the upper platform 20 of the stage 10 along precisely orthogonal X and Y axes (with the Y axis bisecting the angle 29 between the first and second movable mirrors 30 and 32) to precisely position the upper platform 20 as specified by the X and Y digital end point data signals. These movements of the upper platform 20 along precisely orthogonal X and Y axes are effected by the X and Y axes position control circuit 94 and 96 in accordance with the following equations as hereinafter explained, where AL, and AL2 are the displacements of the first and second movable mirrors (relative to the first and second stationary mirrors) along the AL, and AL2 measurement paths of the first and second interferometer systems, respectively, as the upper platform is moved along either the X axis or the Y axis:

(1)

AK=KX(AL,-AL2),

where

Kx=1/2 cos 8; and

(2)

AY=Kv(AL,+AL2),

where

Ky=1/2 sin 8.

The orthogonality of the AX and AY movements of the upper platform 20 along the X and Y axes of motion of the stage 10 in accordance with equations (1) and (2) is substantiated by the fact that AX is a function of cosine 8, whereas AY is a functfon of sine 8, and by the fact that such cosine and sine terms always exist in quadrature.

Since the X and Y axes position control circuits 94 and 96 are identical, the same reference numbers are employed for the same elements of both position control circuits {with those or the X axis position circuit being primed), and only the Y axis position control circuit is described in detail.

70

75

80

85

90

95

100

105

110

115

120

125

7

GB 2 081 048 A 7

Pulses appearing on the up and down outputs of the first adder 88 are applied to an up down counter 98 for counting these pulses to produce a AY digital output signal proportion to the sum 5 (AL,+AL2) of the displacement AL, and AL2 of the first and second movable mirrors 30 and 32 (relative to the first and second stationary mirrors 36 and 38) along the AL, and AL2 measurement paths of the first and second interferometer 10 systems 40 and 42, respectively, as the upper ► platform 20 of the stage 10 is moved along either the X or the Y axes of the stage. In effect, the up-down counter 98 integrates the sum of the • velocities of the first and second movable mirrors 15 30 and 32 with respect to time as those velocities are measured by the first and second interferometer systems 40 and 42, respectively, to produce the AY digital output signal. This AY digital output signal is applied to one input of a 20 comparator 100, and the Y digital end point data signal from the computer 92 is stored in a register 102 and applied to the other input of the comparator. The comparator 100 produces a digital comparison signal equal to the 25 difference between the digital signals applied thereto and proportional to the distance the upper platform 20 must be moved along the Y axis to reach the Y axis position specified by the Y digital end point data signal. This 30 digital comparison signal is applied to a digital-to-anafog converter 104 which converts it to an analog voltage signal and applied it to one input of a summing circuit 106. Another analog voltage signal produced by a tachometer 108 as 35 hereinafter explained, is applied to the other input of the summing circuit 106. Thus, the summing circuit 106 produces an output voltage signal equal to the difference between the analog voltage signal from the analog-to-digital converter 40 104 to the tachometer 108. This output voltage signal is applied to a servo drive circuit 110 for driving a Y axis servo motor 112 mounted on the upper platform 20 and reacting against the lower platform 16 to move the upper platform along the 45 V axis to the Y axis position specified by the Y digital end point data signal. The tachometer 108 is coupled to the Y axis servo motor 112 for , producing an analog voltage signal proportional to the speed of the Y axis servo motor and 50 applying it to the summing circuit 106. This reduces the output voltage signal from the summing circuit 106 and therefore slows the Y axis servo motor 112 down as the upper platform 20 approaches the Y axis position specified by the 55 Y digital end point data signal so as to impede the upper platform from overshooting the specified Y axis position.

The up-down counter 98' of the X axis position control circuit 94 similarly integrates the 60 difference of the velocities of the first and second movable mirrors 30 and 32, as those velocities are measured by the first and second interferometer systems 40 and 42, respectively, to produce a AX digital output signal proportional 65 to the difference (AL,—AL2) of the displacements

AL, and AL2 of the first and second movable mirrors (relative to'the first and second stationary mirrors 36 and 38) along the AL, and AL2 measurement paths of the first and second 70 interferometer systems, respectively, while the upper platform 20 of the stage 10 is moved along either the X or the Y axis of the stage. In response to this AX digital output signal and an X digital end point data signal stored in the register 102' 75 by the computer 92, the servo drive circuit 110' drives the X axis servo motor 112', which is mounted on the lower platform 16 of the stage 10 and which reacts against the granite block 18, to move the lower platform 16 and, hence, the 80 upper platform 20 to the X axis position specified by the X digital end point data signal.

Thus, it may be seen that the upper platform 20 is moved along precisely orthogonal X and Y axes in accordance with the difference (AL,—AL2) 85 and the sum (AL,+AL2) of the displacements of the first and second movable mirrors 30 and 32 (relative to the first and second stationary mirrors 36 and 38) along the AL, and AL2 measurement paths of the first and second interferometer 90 systems 40 and 42, respectively, as specified by the corresponding terms of equations (1) and (2) above. In actuality the constants Kx and Ky of those equations may be determined without the necessity of precisely measuring or knowing the 95 half angle 6 between the first and second movable mirrors 30 and 32. These constants can be determined in setting up the stage 10 by simply attaching a reference contact member to the upper platform 20; placing a gage block of, for 100 example, four inches in length on the upper platform along the Y axis and in abutment with the reference contact member; mounting a deflection type sensor of an electronic gage at a ' fixed position (with respect to the upper platform) 105 along the Y axis and in the path of the gage block and the reference contact member; moving the stage forward along the Y axis to bias the gage block against the sensor until the electronic gage is zeroed and then also zeroing the up down 110 counter 98 of the Y axis position control circuit 96; moving the upper platform backward along the Y axis and removing the gage block; moving the upper platform forward along the Y axis again to bias the reference contact member against the 115 sensor until the electronic gage is zeroed again; dividing the length of the gage block by the count thereupon registered in the up-down counter 98 to determine Kv in inches per count; and by repeating the same process for the X axis with the 120 same reference contact member, the same gage block, and the up-down counter 98' of the X axis position control circuit 94 to determine in inches per count. Since gage blocks are commonly calibrated by the National Bureau of 125 Standards to sub-microinch accuracies, this set up procedure permits the upper platform 20 of the stage 10 to be moved along the orthogonal X and Y axes with extremely high precision. The constants Kx and K,, along with other constants 130 such as might be employed to compensate for

8

GB 2 081 048 A 8

changes in atmospheric conditions etc., are stored in the computer 92 and utilized in determining a set of pairs of X and Y end point data signals required for a desired step-and-repeat operation. 5 As each pair of X and Y end point data signals is fed by the computer 92 to the registers 102' and 102 of the X and Y axes position control circuits 94 and 96, the upper platform 20 of the stage 10 is successively stepped along precisely 10 orthogonal X and Y axes to the position specified by that pair of X and Y end point data signals so as to align successively adjacent regions of microcircuitry of one level on the semiconductive wafer 12 with the projected image of the reticle 15 14. Since the upper platform 20 is stepped along precisely orthogonal X and Y axes, other such stages may, therefore, be employed interchangeably in printing different levels of microcircuitry on the same semiconductive wafer 20 12.

Referring now to Figure 4, there is shown another pair of X and Y axes position control circuits 94 and 96 that may be employed in place of those described above in connection with 25 Figure 3 to control the position of the interferometrically controlled stage of Figure 1. Since these X and Y axes position control circuits 94 and 96 are identical, the same reference numbers are again employed for the same 30 elements of both position control circuits (with those of the X axis position control circuit being primed), and only the Y axes position control circuit 96 is described in detail. Pulses appearing on the up and down outputs of the first dual adder 35 88 are again applied to an up-down counter 98 for counting those pulses to produce a AY digital output signal proportional to the sum (AL,+AL2) of the displacements AL, and AL2 of the first and second movable mirrors 30 and 32 (relative to the 40 first and second stationary mirrors 36 and 38) along the AL, and AL2 measurement paths of the first and second interferometer systems 40 and 42, respectively, as the upper platform 20 of the stage 10 is moved along either the X or the Y axis. 45 In effect, the up-down counter 98 again integrates the sum of the velocities of the first and second movable mirrors 30 and 32 with respect to time as those velocities are measured by the first and second interferometer systems 40 and 50 42, respectively, to produce the AY digital output signal. This AY digital output signal is again applied to one input of a comparator 100, and the Y digital end point data signal from the computer 92 is again stored in a register 102 and applied to 55 the other input of the comparator. The comparator 100 again produces a digital comparison signal equal to the difference between the digital signals applied thereto and proportional to the distance the upper platform 60 20 of the stage 10 must be moved along the Y axis to reach the Y axis position specified by the Y digital end point data signal. This digital comparison signal is applied to the computer 92, which in response to a nonzero comparison signal 65 sequentially stores each of a series of digital velocity signals in a register 103. These digital velocity signals and the durations they are stored in register 103 define an optimum profile of accelerating, maximum, and decelerating velocities, as determined in accordance with well known techniques, for the distance the upper platform 20 of the stage 10 is to be moved along the Y axis. Each digital velocity signal stored in the register 103 is applied to ji digital-to-analog converter 104 which converts it to an analog voltage signal and applies it to one input of a * summing circuit 106. Another analog voltage signal produced by a tachometer 108, as hereinafter explained, is applied to the other input of the summing circuit 106. Thus, the summing circuit 106 produces an output voltage signal equal to the difference between the analog voltage signal from the digital-to-analog converter 104 and the tachometer 108. In response to a nonzero comparison signal from the comparator 100, the computer 92 also activates a selected circuit 109 to apply the output voltage signal from the summing circuit 106 to a servo drive circuit 110 for driving a Y axis servo motor 112. This Y axis servo motor 112 is mounted on the upper platform 20 and reacts against the lower platform 16 of the stage 10 to move the upper platform along the Y axis towards the Y axis position specified by the Y digital end point data signal. The tachometer 108 is coupled to the Y axis servo motor 112 for producing an analog voltage signal proportional to the actual velocity of the Y axis servo motor and applying it to the summing circuit 106. This reduces the output voltage signal from the summing circuit 106 for the purpose of equalizing the actual velocity and the desired velocity of the Y axis servo motor 112.

The up-down counter 98' of the X axis position control circuit 94 similarly integrates the difference of the velocities of the first and second movable mirrors 30 and 32 as those velocities are measured by the first and second interferometer system 40 and 42, respectively, to produce a AX digital output signal proportional to the difference (AL,—AL2) of the displacements AL, and AL2 of the first and second movable mirrors (relative to the first and second stationary mirrors 36 and 38) along the AL, and AL2 measurement paths of the first and second interferometer systems, respectively, while the upper platform 20 of the stage 10 is moved along either the X or the Y axis. In response to this AX digital output signal and an X digital end point data signal stored in the register 102' by the computer 92, the servo drive circuit 110' drives the X axis servo motor 112'. This X axis servo motor 112 is mounted on the lower platform 16 of the stage and reacts against the granite block 18, on which both the upper and lower platforms 20 and 16 are mounted, to move the lower platform and, hence, the upper platform, which is coupled to the lower platform for movement therewith along the X axis, towards the X axis position specified by the X digital end point data signal.

70

75

80

85

90

95

100

105

110

115

120

125

130

9

GB 2 081 048 A 9

Thus, it may be seen that the upper platform 20 is again moved along the orthogonal X and Y axes in accordance with the difference (AL,—AL2) and the sum (AL,+AL2) of the displacements of 5 the first and second movable mirrors 30 and 32 (relative to the first and second stationary mirrors 36 and 38) along the AL, and AL2 measurement paths of the first and second interferometer systems 40 and 42, respectively, as specified by 10 the corresponding terms of equations (1) and (2) - above. In actuality the constants Kx and Ky of those equations may be determined without the necessity of precisely measuring or knowing the * half angle 6 between the first and second 15 movable mirrors 30 and 32 as described above. The constants Kx and Ky, along with other constants such as might be employed to compensate for changes in atmospheric conditions etc., are stored in the computer 92 20 and utilized In determining a set of pairs of X and Y end point data signals successively fed by the computer 92 to the registers 102' and 102 of the X and Y axes position control circuits 94 and 96, as described above, to step successively the upper 25 platform 20 of tha stage 10 along the orthogonal X and Y axes to the position specified by those pairs of X and Y end point data signals.

The resolution of the X and Y axes position control circuits 94 and 96 of Figure 4 is extended 30 by providing the Y axis position control circuit 96 with phase control circuit 114, responsive to the reference signal of frequency f,—f2, the first measurement signal of frequency f,—f? 2 fLl,

and a three bit control or select code signal 35 supplied by the computer 92 in response to a zero comparison signal from the comparator 100, for producing a position control signal as hereinafter described. Similarly, the X axis position control circuit 94 is provided with a hase control circuit 40 114', responsive to the reference signal of frequency f,—f2, the second measurement signal of frequency f,—f2 2 fLj, and another three bit control or select code signal supplied by the computer 92 in response to a zero comparison signal from the 45 comparator 100', for producing another position control signal as hereinafter described. These position control signals are applied to a pair of inputs of a summing circuit 116 (in the Y axis position control circuit 96) for producing an 50 output voltage signal equal to the sum of the position control signals. They are also applied to a pair of inputs of a summing circuit 116' (in the X axis position control circuit 94) for producing an output voltage signal equal to the difference of 55 the position control signals. In response to zero comparison signals from the comparators 100 and 100' the computer 92 activates the selector circuits 109 and 109' to apply the output voltage signals from the summing circuits 116 and 116' 60 to the servo drive circuits 110 and 110'

respectively. This drives the Y and X axes servo motors 112 and 112' to move the upper platform 20 of the stage 10 to precisely the desired Y and X axes positions.

65 Referring now to Figure 5, there is shown a detailed block diagram of the phase control circuits 114 and 114' for the Y and X axes position control circuits 96 and 94, respectively, of Figure 4. Since these phase control circuits 114 and 114' are identical, the same reference numbers are employed for the same elements of both phase control circuits (with those of the phase control circuit 114' for the X axis position control circuit 94 of Figure 4 being primed), and only the phase control circuit 114 for the Y axis position control circuit 96 of Figure 4 is described in detail.

The phase control circuit 114 includes a variable phase shifter 118 for receiving the reference signal of frequency f,—f2 and for producing an output signal of the same frequency but shifted in phase as determined by the three bit select code from the computer 92. This phase shifter comprises a phase detector 120 having a first input at which the reference signal of frequency f,—f2 is applied and a second input at which an output signal from a divide-by-N circuit 122 is applied as hereinafter explained. In response to these input signals the phase detector 120 applies an output voltage signal to a voltage controlled oscillator 124 so as to drive the voltage controlled oscillator to produce an output signal having a frequency N times greater than the frequency f,—f2 of the reference signal. This output signal from the voltage controlled oscillator 124 is applied both to an input of the divide-by-N circuit 122 and to a clock input of a shift register 126. The divide-by-N circuit 122 divides this output signal by N, which for purposes of illustration is herein taken to have a value of eight, and applies the resultant output signal to the second input of the phase detector 120 and also to a data input of the shift register 126. In response to the applied output signals from the divide-by-N circuit 122 and the voltage controlled oscillator 124, the shift register 126 supplies N (or eight) output signals of different phase (each such output signal differing in phase from the preceding one by 360°/N or 45°) to a data selector 128. The data selector 128 supplies a selected one of these output signals from the shift register 126 to the output of the variable phase shifter 118 as determined by the three bit select code signal supplied by the computer 92 in response to a zero comparison signal from the comparator 100. As indicated above, the selected output signal has the same frequency f,—f2 as the reference signal.

The phase control circuit 114 also includes a phase detector 130 having a first input at which the selected output signal (i.e., the output signal with the desired phase shift) of frequency f,—f2 from the variable phase shifter 118 is applied and a second input at which the first measurement signal of frequency f,—f2±2AfL is applied. In response to these signals the pnase detector 130 supplies a position control signal proportional to the difference in phase therebetween to an input of each of the summing circuits 116 and 116' as previously described. Similarly, the variable phase

70

75

80

85

90

95

100

105

110

115

120

125

130

10

GB 2 081 048 A 10

shifter 118' and the phase detector 130' of the phase control circuit 114' are responsive to the reference signal of frequency f,—f2, to the other three bit select code signal from the computer 92, 5 and to the second measurement signal of frequency f,—f2±2AfL for supplying another position control signal proportional to the difference in phase between the selected output signal from the variable phase shifter 118' and 10 the second measurement signal to the other input of each of the summing circuits 116 and 116'. The sum and difference of these position control signals are applied to the selector circuits 109 and 109' of the Y and X axes position control 15 circuits 96 and 94, respectively, of Figure 4 to extend the resolution of those position control circuits as previously described.

From the above description, it can be seen, inter alia, that the present invention provides an 20 improved position control circuit in which the ambiguity of the last or least significant digit indicated by the counter is eliminated and the resolution of the position control circuit is extended.

Claims

25 Claims

1. A position control circuit comprising first means for producing a difference in phase between a reference signal and a related input signal as determined by a control signal, and

30 second means, coupled to the first means, for producing a position control signal proportional to the difference in phase between the reference signal and the input signal.

2. A position control circuit according to claim 35 1 wherein said input signal has a frequency related to the frequency of the reference signal.

3. A position control circuit according to claim 2 wherein said first means is responsive to the reference signal and to the control signal for

40 producing an output signal having the same frequency as the reference signal and having a phase determined by the control signal, and said second means is coupled to the first means and responsive to the output signal from the first 45 means and to the input signal for producing a position control signal proportional to the difference in phase between the input signal and the output signal from the first means.

4. A position control circuit according to any 50 one of claims 1,2 and 3 wherein said first means comprises a variable phase shifter, and said second means comprises a phase detector.

5. A position control circuit according to claim 4 wherein said variable phase shifter comprises a

55 voltage controlled oscillator for producing an output signal having a frequency N times greater than the frequency of the reference signal, a division circuit, coupled to the voltage controlled oscillator, for producing an output signal having a 60 frequency equal to the frequency of the output signal from the voltage controlled oscillator divided by N, another phase detector, coupled to the division circuit and to the voltage controlled oscillator and responsive to the output signal from the division circuit and to the reference signal, for driving the voltage controlled oscillator to produce an output signal having a frequency N times greater than the frequency of the reference signal, a shift register, coupled to the voltage controlled oscillator and the the last-mentioned phase detector, forproducing N output signal of difference phase, and a data selector, coupled to the shift register and to the first-mentioned phase detector and responsive to the control signal, for applying a selected one of the output signals from the shift register as determined by the control -signal to the first-mentioned phase detector.

6. A position control circuit according to claim 4, for positioning a utilization device, said position control circuit comprising a counter means, responsive to the reference signal and the related input signal, for producing an actual position signal indicative of the actual position of the utilization device, register means for receiving a desired position signal indicative of a desired position of the utilization device, comparator means, coupled to the counter means and the register means for producing a comparison signal equal to the difference between the actual and desired position signals, control means, coupled to the comparator means, for producing a velocity control signal, and drive means, responsive to the velocity and position control signals, for moving the utilization device to the desired position.

7. A position control circuit according to claim 6 wherein said counter means is also responsive to another input signal for producing the actual position signal, said other input signal is also related to the reference signal, said circuit comprises combining means, coupled to the phase detector, for combining the position control signal with another position control signal that is proportional to a difference in phase between ths reference signal and said other input signal to produce a combined position control signal, said drive means comprises a motor for positioning the utilization device, and servo drive means, coypled to the motor, for driving the motor, said control means comprises register means for receiving desired velocity signals, tachometer means tor producing an actual velocity signal proportional to the actual velocity of the motor, and comparator means, coupled to the last-mentioned register means and the tachometer means, for producing velocity control signals proportional to the difference between each desired velocity signal and the actual velocity signal, and said drive means further comprises selector means, coupled to the last-mentioned comparator means and the combining means, for selectively applying the velocity control signals and the combined position control signal to the servo drive means so as to drive the motor for moving the utilization device to the desired position.

8. A position control circuit substantially as hereinbefore described with reference to Figure 3 of the accompanying drawings.

65

70

75

80

85

90

95

100

105

110

115

120

125

11

GB 2 081 048 A 11

9. A position control circuit substantially as hereinbefore described with reference to Figure 4 of the accompanying drawings.

10. A position control circuit substantially as

5 hereinbefore described with reference to Figure 5 of the accompanying drawings.

11. A position control circuit substantially as hereinbefore described with reference to Figures 4 and 5 of the accompanying drawings.

10

12. A method of processing a reference signal ► of a first frequency and a displacement input signal of a second frequency related to the frequency of the reference signal to provide a - position control signal of extended resolution, said 15 method comprising the steps of producing a difference in phase between the reference signal and the displacement input signal, and detecting the difference in phase between the reference signal and the displacement input signal to 20 generate a position control signal proportional to that difference in phase.

13. A method according to claim 12 further comprising the step of combining the position control signal with another position control signal 25 that is proportional to a difference in phase between the reference signal and another displacement input signal to generate a combined position control signal equal to the sum or the difference of those position control signals. 30

14. A method substantially as hereinbefore described with reference to the accompanying drawings.

Printed for Her Majesty's Stationery Office by the Courier Press. Leamington Spa, 1982. Published by the Patent Office, 25 Southampton Buildings, London, WC2A 1 AY, from which copies may be obtained.