GB2090657A

GB2090657A - Photoelectric Method and Apparatus for Detecting the Position of Edge-defined Structures

Info

Publication number: GB2090657A
Application number: GB8200345A
Authority: GB
Original assignee: Carl Zeiss Jena GmbH
Current assignee: Jenoptik AG
Priority date: 1981-01-07
Filing date: 1982-01-07
Publication date: 1982-07-14
Also published as: DD156290A1; FR2497568A1; DE3143948A1

Abstract

The invention can be used in optical precision measuring instruments, such as coordinate measuring machines, or in processing machines, such as in apparata for photolithography, and also for the scanning of edge-defined and line- defined structures. With the aim of ensuring rapid, objective and contactless detection of measured quantities in the measurement of optically effective structures which are not necessarily stationary with respect to a fixed reference system and which do not have to have any particular orientation relative to the reference system, the invention provides a method and a receiver arrangement for two-dimensional orientation- independent detection of the position of the structure. The radiant flux determining the position of the structure is evaluated by an array of photodetectors 1', 11'-14' the outputs of which are combined to produce a signal having a predetermined value eg. zero, when the structure is in a predetermined position relative to a reference point O regardless of the orientation of the structure. Concentric circular and annular detectors may be used. <IMAGE>

Description

SPECIFICATION Photoelectric Method and Apparatus for Detecting the Position of Edge-defined Structures to be Measured The invention relates to a photoelectric method of, and an arrangement for, detecting the position of edge-defined (kantenförmiger) structures to be measured.

Many forms of photoelectric systems are used in the construction of photoelectric microscopes and displacement measuring systems, and also for detecting objects and image analysis. A large number of static and dynamically operating photoelectric methods and arrangements for detecting the position of edges, particularly for the detection of line-defined (strichfijrmigen) planar objects are described in the existing literature, particularly in the field of photolithography (for example, in Optik 21 (1964) page 605 ff. and Wissenschaftliche Zeitschrift Friedrich-Schiller University Jena, Math.-Nat. R., 25th Volume (1976) H. 5 page 683).

These methods are frequently very expensive or they do not entirely fulfil the demands made by co-ordinate measurement (for example, the necessity for defined marks on the objects to be measured and pre-orientation relative to the measuring system).

Furthermore, devices for photoelectrically determining the position of an object are known in which a television camera tube is used as a transducer (WP 1 28637 DD). The image is scanned once or twice in the x and y direction in the television camera tube by suitable control of the deflection unit.

Suitable electronic evaluation means ascertain the misorientation relative to an ideal mark in the x, y and I directions. The measuring time is < 100 ms.

Moreover, a method is known (Offenlegungsschrift 2405 102 DB) in which a line-defined or edge-defined structure is scanned by means of a mechanically deflected and circularly rotating point light proble. The measuring signals thereby produced in a photoreceiver are evaluated with reference to their phase relationship to the rotational movement for the purpose of determining the angular position of the structure. The analysis of the frequency of the measuring signals gives the deviation from the centre of rotation.

The two last-mentioned solutions are characterised by a dynamic measuring principle. The use of mechanical deflection means thereby limits the use of the method to quasi-stationary objects.

However, the use of non-mechanical deflection means (such as the use of television camera tubes) also involves problems owing to the limited local resolution. Not least, the aforegoing methods are likely to lead to considerable measurement errors in the case of non-ideal structures, since the location of the structure is not measured in accordance with any of the generally recognized local definitions.

Offenlegungsschrift 2102027 DB describes a method which renders it possible to measure the location of the edge, by definition, at the location of the maximum density gradient of an edge-shaped intensity profile.

In this connection, the two-fold derivation of the intensity profile is formed optically. However this method can only be used for one or two (see Offenlegungsschrift 201 7400 DB) privileged or specified directions of measurements.

The basic aim of the present invention is to detect measured variables in a rapid, objective and contactless manner when measuring optically effective structures which are not necessarily stationary with regard to a fixed reference system and which do not have to have a particular orientation relative thereto. The invention is to enable greater accuracy of positioning and speed of measurement by replacement of the visuai positioning method currently used, especially in the case of optical precision measuring instruments, and particularly coordinate measuring instruments having a digital displacement measuring system.

The invention seeks to provide a photoelectric method and a receiver arrangement which are suitable for the detection, independently of orientation, of optically effective, line- or edge-defined structures to be measured. A suitable control signal for a peripheral electronic measurement means, such as the position output of a displacement measurement system, is to be derivable upon coincidence of the location of the structure sighted in accordance with the photometric centre definition.

In accordance with the present invention, there is provided a static photoelectric method which, by two-dimensional photometric evaluation of the radiant fluxes emitted by characteristic structural elements, renders it possible to determine the position of the structure, irrespective of its orientation, in a fixed reference system (co-ordinate system) with respect to a co-ordinate origin.In accordance with this method, the minimum and the maximum of the radiant flux in the immediate vicinity of the edge transition is optically and integrally detected two-dimensionally, the sum of these integral radiant fluxes being either formed optically with a peripheral diaphragm surface (such as a circular ring) and a proportional electrical signal then being derived, or electrical summation of individual signal values is effected, the individual signal values being proportional to the component radiant fluxes passing through a plurality of peripheral diaphragm component surfaces, the integral radiant flux of the edge transition being entirely received by a central diaphragm surface and a proportional electrical signal again being obtained. The resultant electrical signal values are subsequently subjected to difference formation.The signals may be evaluated by preamplification in conformity with the ratio of the peripheral diaphragm surface to the central diaphragm surface. It will be appreciated, however, that preamplification can be omitted if the surfaces are of equal size.

Preferably, the zero passage occurring in the difference signal is used to indicate a defined relative position of the structure to be measured and the co-ordinate origin.

The radiant fluxes emitted by the central and the peripheral diaphragm surfaces are either conducted to two photoelectric receivers by suitable beam-guiding elements, or they are combined and projected onto a receiver by way of a beam interrupter (chopper method).

It is also within the scope of the invention to form the diaphragm surfaces by correspondingly designed and arranged receiver surfaces of photoelectric receivers, wherein the latter can be integrated on a chip or can be manufactured as a hybrid structure. The photoelectric receiver arrangement can include at least one centrally arranged receiver surface and at least one peripheral receiver surface surrounding the central surface. The receiver surfaces are electrically insulated from one another, and the signals of the individual surfaces are freely switchable.The central receiver surface is preferably of circular or square configuration, and the peripheral receiver surrace can be annular or can comprise four separate component surfaces of the same shape, each component surface having the same area as the central surface, or the total peripheral surface being equal to the central surface.

In addition to achieving the method of scanning, independently of orientation, edge-defined structures to be measured by the free switchability of all the individual receiver surfaces, the receiver arrangement in accordance with the invention also renders it possible to achieve the known two-cell differential method for scanning line-defined objects. Since the method used constitutes, in principle, a static method of measurement, the arrangement can be used to particular advantage for the purpose of measuring moving (non-stationary) structures to be measured.

The invention is used in, for example, optical precision measuring instruments (co-ordinate measuring machines) for the purpose of establishing the positioning operation in the co-ordinate directions x, y and I, it being possible to automatically measure moving structures, that is to say, structures which are not stationary. A further field of application of the invention relates to the detection of the position of simple, geometrical structures on templates and semi-conductor chips in photolithography.

The invention is described further hereinafter, by way of example only, with reference to the accompanying drawings, in which: Fig. 1 a shows a receiver arrangement comprising five integrated square receiver diodes of the same dimensions; Fig. 1 b shows a receiver signal in the case of edge capture by translation (I=0 or 450); Fig. 1 c shows a receiver signal in the case of edge capture by rotation; Fig. 1 d shows a receiver signal in the case of line capture by translation; Fig. 2a shows a receiver arrangement comprising five integrated rectangular receiver diodes of differing geometry; Fig. 2b shows a receiver signs in the case of edge capture by translation;; Fig. 3a shows a receiver arrangement comprising two circular, concentric receiver surfaces having the surface ratio 1:1; Fig. 3b shows a receiver signal in the case of edge capture; Fig. 3c shows a receiver arrangement in accordance with Fig. 2a having a signal evaluator for line capture within a desired range; Fig. 3d shows a receiver arrangement comprising three concentric receiver surfaces; Fig. 3e shows a receiver signal in the case of line capture; Fig. 4 shows a receiver arrangement in accordance with Fig. 2a, each of the two circular surfaces being divided into four sectors of equal size.

The receiver arrangement illustrated in Fig. 1 a comprises five separate square receiver surfaces, the central surface 1" and four peripherally disposed surfaces 11 12', 13', 14' having side lengths of dimension a. This construction can be integrated on a chip, or it can be a hybrid construction. The separation strips d between the receiver surfaces establish the axes (x,y) of the co-ordinate system.

When this receiver is used in an optical co-ordinate measuring instrument, it has to be disposed in an intermediate image plane or, for example, on a projection screen such that its centre lies in the line of sight. The optical axis of the measuring instrument is usually chosen as the line of sight.

The illustrated type of receiver permits: 1. The derivation of a coincidence signal during orientation-independent positioning of an optically effective edge relative to the co-ordinate origin; 2. The derivation of a parallelism signal or overlap signal when coincidence has been achieved in accordance with 1 when rotating an edge relative to the co-ordinate axes; 3. The derivation of a coincidence signal during the orientation-independent positioning of the photometric centre of optically effective lines of the widths. Lines having a width b a+2d b 3a+2d are thereby positioned relative to the co-ordinate origin, and lines having a width d b 2a+d are positioned relative to a reference point on the co-ordinate axes at a distance 1/2 (a+d) from the coordinate origin (see Fig. 1 a); 4.The derivation of a coincidence signal with lines spaced apart at 1/2 (a+d) and extending parallel to the co-ordinate axes, during rotation of a line.

Re 1. In accordance with the invention, the positioning of the edges is effected by the receiver of Fig. 1a by forming:

'1-radiant flux onto the surfaces (11 12', 13', 14') 0 radiant flux onto the surface 1" (=O for coincidence) The radiant flux is quadrupled after the conversion by corresponding amplification of the signal s (r,l) proportional to the radiant flux.

S'1...4--signals from 11', 12', 13', 14' s"1-signal from 1" rhrco-ordinates of the location of the edge The signal pattern characteristic of the method ensues upon translation by the receiver of an edge-defined structure located in any optional angular position relative to the co-ordinate axes. The difference signals s (Rk,#) for 1= or (p2=45O are illustrated in Fig. 1 b.

The receiver arrangement illustrated in Fig. 2a operates on the same measurement principle. The equality of the total signal from the outer receiver surfaces 21', 22', 23', 24' and of the signal from the inner receiver surface 2" when all the receiver surfaces are fully illuminated is in the present case obtained by the equality of the total of the surfaces 21', 22', 23', 24' and of the surface 2".

Fig. 2b shows the characteristic signal patterns s (rk,#). Irrespective of the angular position of the edge relative to the co-ordinate axes, the difference signal of the two embodiments has a zero passage when the edge is coincident with the co-ordinate origin.

At this instant, half the total of the external receiver surfaces and half of the inner receiver surface are darkened by the edge, and the difference signals according to Fig. 1 b or

according to Fig. 2b are produced.

The capture range in the coordinate direction amounts to Bxy=2(a'1+d)+a" (The capture range B indicates the dimension of the region on the receiver arrangement for which s (rk,#)#0).

When the outer dimensions are equal (circumscribed circular surface), the sensitivity of the isohedral construction of Fig. 1 a is greater than that of Fig. 2a.

The sensitivity of the two receivers increases when the angle between the edge and the coordinate axes (separation strips) increases from 0 to 450.

Re 2. Moreover, the two receiver arrangements can be used in a simple manner to indicate the coincidence during rotation (edge parallel to the coordinate axis). For this purpose, the receiver arrangements 11' and 13' or 12' and 14' in the arrangement of Fig. 1 are switched to form a difference signal.

The difference signals s(x,90 )=s'12(x,90 )-s'14(X,90 ) or s(y,o)=s'1 1(y,o) s' ,3(y,o) become zero when the above-mentioned outstanding angular positions are obtained.

The reaching of the coincidence position relative to a co-ordinate direction requires positioning according to Item 1. The characteristic of the difference signal is shown in Fig. 1 c. The limiting angle of the capture range generally ensues according to

when

when h-o (coincidence) h--distance of the edge from the co-ordinate origin in the co-ordinate direction 9;rnax-limiti ng angle between the edge capture range and the co-ordinate direction.

Re 3. Advantageously, the desired position of a line can be indicated by means of the detector of Fig. 1 a (receiver component surfaces of equal size). The static two-cell difference method is used for this purpose. The receiver surfaces 11 12', 13', 14' can be optionally switched to form a difference signal with the surface 1" for, for example, utilizing the zero passage of the difference signal s(x)=s'12(x)-s1,,(x) s(y)=s'11(x)-s1,,(x) for small line widths (s < b < 2a+d) or s(x)=s'12(x)-s'14(x) x(y)=s'1 1(y)-s'13(y) for line widths a+2d < b < 3a+2d.

The zero passage is independent of the angular position for a specific pair of receiver surfaces in the range from 00140 < 450. In the case of larger angles, measurements must be made in the other coordinate direction respectively.

The characteristic of the signals is illustrated in Fig. Id as a function of the location of the photometric centre of the line.

Re 4. On the supposition that the line has been positioned in accordance with Item 3, it is possible to derive a coincidence signal upon attaining the co-ordinate directions by difference switching of the receiver component surfaces in accordance with Item 2.

A further receiver arrangement for orientation-independent edge scanning is illustrated in Fig. 3a.

This receiver arrangement is distinguished by a sensitivity which is independent of the orientation of the edge.

The arrangement comprises a circular receiver surface and a second, annular receiver surface disposed concentrically to the centre of the circle of the circular receiver surface. The two surfaces have to be wired such that the electrical signals supplied by them are proportional to the impinging luminous flux and are subsequently subtracted. Devices for this purpose are known (see, for example, W. G. Auslegeschrift 1957 161 DB) and do not require any further explanation.

In accordance with the present technique, the length of the radii has to be chosen such that the annular surface 31' is equal to the circular surface 3". This is the case when R1= ignoring the dimensions of the separation width d between the two receiver surfaces.

When, as is indicated in Fig. 3a, an edge-defined structure moves across the receiver surface, a difference signal having the characteristic shown in Fig. 3b ensues. It will be seen that, when the edge is coincident with the centre point of the receiver arrangement, a zero difference signal is produced and this can be used in an advantageous manner to activate a linear measuring system. In this circumstance, one half of the receiver surface is darkened and the other half is illuminated by the edgedefined structure, so that, assuming equal sensitivity of the receivers, the difference between the receiver signals becomes zero.

The analytical relation between the edge location rk and the difference signal s(x) ensues in the case of a separation width do

in which SD is the sensitivity of the receiver material, and E the intensity of illumination in the bright portion of the edge transition. (E=O has been assumed for the dark portion of the edge. With the values: E=1 500 lx, Sn=4.2 mA/1 m and r1=1 00 m, a value of

ensues for the receiver sensitivity of the arrangement in the vicinity of the zero passage (coincidence point).Since the dark current of the receiver material taken as a basis is approximately 0.07 nA during element operation, differences of less than 1 ium can be reliably detected in the case of ideal structures (maximum contrast).

The circular symmetry of the receiver arrangement makes the sensitivity of the arrangement entirely independent of direction. The sighting axis is determined by the centre of the circle. The arrangement can be used to particular advantage for measuring moving (non-stationary) edge-defined objects. In a development of the present technique, an arrangement is proposed which renders it possible to record a line-defined structure within a predetermined desired range.

For this purpose, with the use of the type of receiver proposed in Fig. 3a, the receiver signal originating from the circular ring has to be amplified electronically by a factor < 3x (for example, 2.5x) before recording the difference photocurrent (Fig. 3c). However, the same effect can also be obtained by connecting a further, second annular surface disposed peripherally outwardly of the first annular surface (Fig. 3d).The receivers have to be wired such that the resultant total signal sgeS is associated with the signals originating from the receiver component surfaces by means of the following equation: sge5(r,sD)=s"3(r,1)[(s'3(r,1)+s'32(r,1)] It is possible to change over to edge scanning by means of the switch s (switch position a).

a b S sir)=S(k) s(r)=S(rS) If the radius r2 (again ignoring the line of separation) is chosen to be somewhat smaller than twice the radius r0 of the circular receiver, the signal characteristic of Fig. 3e ensues upon a line-defined structure passing the receiver thus characterized. b, < b2 < b3 applies to the line width b.

It will be seen that a zero passage of the signal voltage occurs twice symmetrically of the coincidence point (that is to say, the centre of the line is located on the sighting axis). The distance of these zero passages from the sighting axis is independent of the width of the line in the first approximation and is only determined by the radius r2 which has been chosen (or the amplication factor of the signal amplifier). The nearer the amplification factor approaches the value 3 (or the radii ratio the value rJrO=2), the nearer the zero passages approached the sighting axis, although the smaller become the negative signal components, so that reliable recording of the zero passages is no longer ensured.

Thus, it is possible to capture a line-defined structure in a prescribed desired range by means of the proposed arrangement. By way of example, a desiged range of +30 jum ensues when r0""75 m and r21 40 ym, and a desired range of +10 ym when ro#71 Mm and r2l 40 Mm.

Even when used in this way, it is readily possible to measure moving objects by realizing a static method of measurement.

The arrangement illustrated in Fig. 4 constitutes a further advantageous embodiment.

This receiver arrangement is particularly suitable for positioning the photometric centre of a line relative to a sighting point. For this purpose, the sectors are switched in pairs in difference, such that the receiver signal is formed in conformity with the equations -s(rl)=s4,+s42+s"4s+s"46(St43+s'44+Stt47+StB48) when 45 < 1P1 < 90 or s(r,I)=s'41+d'44+s"45+s"4,,-(s'42+s'43+s"46+s' 47) when 0o < I,iI < 450 The separation strips determine the axes of the co-ordinate system.

The zero passage of s(r,ç7) indicates coincidence of the photometric centre of the line with the centre of the detector. The same circuit is used to indicate the coincidence of an edge with the coordinate axes during rotation.

The coincidence of a line (photometric centre) with the co-ordinate axes during rotation is indicated by fulfilling the condition s'41(r,l)-s'42(r,l)=0 and s'42(r,l)=s'43 (r,l)=0.

The two last-mentioned applications require positioning relative to the co-ordinate origin. The arrangement can also be used directly for determining the angular position of a line relative to the coordinate direction. By way of example, the receiver surfaces 41' and 44; and the receivers 42' and 43' are switched in difference for this purpose upon movement of the object in the x direction, the zero passages of the two difference signals are recorded and the angle is calculated from the corresponding displacement co-ordinates.

2(r,+d)+r,--r,-d l=arc tan Ax An edge can be positioned by using a principle which corresponds to that of the receiver arrangement of Fig. 3. Angular positions (o=OO or l=90 must then be avoided owing to the separating line (smaller accuracy of measurement).

Claims

1. A photoelectric method of detecting the position of edge-defined structures to be measured, by means of two-dimensional photometric evaluation of the radiant fluxes emitted by characteristic structural elements, wherein the minimum and the maximum of the radiant flux in the immediate vicinity of the edge transition is optically and integrally detected two-dimensionally in a coordinate system irrespective of the orientation of the structure, the sum of these integral radiant fluxes being either formed optically with a peripheral diaphragm surface and a proportional electrical signal then being derived, or electrical summation of individual signal values being effected, with the individual signal values being proportional to the component radiant fluxes passing through a plurality of peripheral diaphragm component surfaces and the entire radiant flux influenced by the change in the intensity of the edge transition being fully received by a central diaphragm surface, a proportional electrical signal again being obtained, and the difference of the resulting electrical signal values then being formed.

2. A method as claimed in claim 1, wherein the diaphragm surfaces are formed by appropriately designed and arranged receiver surfaces of photoelectric receivers, and wherein the characteristic of the electrical signals emitted by the receivers indicates the existence of a specific relative position of the structure to be measured or its coincidence with the coordinate origin.

3. A method as claimed in claim 1, wherein the radiant fluxes emitted by the central and the peripheral diaphragm surfaces are conducted to two photoelectric receivers by suitable beam-guiding elements.

4. A method as claimed in claim 3, wherein the two radiant fluxes combined by beam-guiding elements are projected onto a photoelectric receiver by way of a beam interrupter, and the value of the electrical alternating signal emitted by the receiver indicates the existence of a specific relative position between the structure to be measured and the coordinate origin.

5. A photoelectric receiver arrangement for performing the method of claim 1, comprising at least one centrally disposed receiver surface and at least one peripheral receiver surfaces being disposed so as to surface, the receiver surface being disposed so as to be electrically insulated from one another, and the signals of the individual surfaces being freely interconnectible.

6. An arrangement as claimed in claim 5, wherein the arrangement is of monolithic construction.

7. An arrangement as claimed in claim 5, wherein the arrangement is of hybrid construction.

8. An arrangement as claimed in claim 5, 6 or 7, wherein the peripheral receiver surface comprises at least four separate component surfaces of the same shape, each component surface having the same area as the central surface.

9. An arrangement as claimed in any of claims 5 to 8, wherein the receiver arrangement comprises a square central surface (1") and four peripheral receiver surfaces 1', 12', 13', 14') identical to the central surface (1").

10. An arrangement as claimed in claim 9, wherein the measuring signal in accordance with claim 1 is formed by electrical summation of the signals S',1 14 produced by the peripheral receiver surfaces (11 12', 13', 14') and subsequently forming the difference between the total signal and the signal S", produced by the central surface and amplified by the factor 4.

11. An arrangement as claimed in claim 9, wherein the measuring signal is formed by using the signals originating from the respectively oppositely located peripheral receiver surfaces (12') and (14') or (11') and (13') in accordance with the equations s(r,l)=s'12(r,I)-s'14(r,l) and s(r,I)=s'11(r,l)-s'13(r,l), respectively.

12. An arrangement as claimed in claim 9, having the facility for differential connection of oppositely located peripheral receiver surfaces or the differential connection of a peripheral receiver surface to the central receiver (1")..

13. An arrangement as claimed in claims 5, 6 or 7, wherein the peripheral receiver surface comprises at least four separate component surfaces of the same shape whose total surface is equal to the central surface.

14. An arrangement as claimed in claim 13, wherein the receiver arrangement comprises a square central receiver surface (2') and four rectangular peripheral receiver surfaces (21'), (22'), (23'), (24'), the total area of these latter receiver surfaces being equal to that of tne receiver surface (2').

1 5. An arrangement as claimed in claim 14, wherein the obtaining of the measuring signal in accordance with claim 1 is effected by electrical summation of the signals S',, . . . 24 originating from the peripheral receiver surfaces (21'), (22'), (23'), (24') and subsequently forming the difference between the total signal and the signals S"2 produced by the central surface (2").

1 6. An arrangement as claimed in claim 14, wherein the measuring signal is obtained in accordance with claim 11.

1 7. An arrangement as claimed in claims 5, 6 or 7, wherein the receiver arrangement comprises a circular central receiver surface and an annular receiver surface surrounding the former surface.

1 8. An arrangement as claimed in claim 17, wherein the measuring signal in accordance with claim 1 is produced by differential connection of the two receiver surfaces.

1 9. An arrangement as claimed in claim 18, wherein the signal produced by the annular receiver surface is amplified by a factor 3.

20. An arrangement as claimed in claim 17, wherein the arrangement is supplemented by a second peripheral, annular receiver surface whose outer radius must be no more than twice that of the central circular receiver surface.

21. An arrangement as claimed in claim 20, wherein the measuring signal is produced by forming the difference between the total of the signals originating from the two annular receiver surfaces and the signal from the central receiver surface.

22. An arrangement as claimed in claim 17, wherein both the central receiver surface and the peripheral annular receiver surface are each divided into four sectorial receiver surfaces which are of equal size and are electrically insulated from one another.

23. An arrangement as claimed in claim 22, wherein the measuring signal is produced by combining the signals in accordance with claim 1 8.

24. An arrangement as claimed in claim 22, wherein the measuring signal is produced by forming the differences between the total signals originating from the sectorial receiver surfaces separated by separation strips.

25. An arrangement as claimed in claim 22, wherein two measuring signals are simultaneously obtained by differential connection of two respective sectors of the annular receiver surface.

26. A photoelectric method of detecting the position of edge-defined structures, substantially as hereinbefore described with reference to the accompanying drawings.

27. An arrangement for performing the method of claim 1, substantially as hereinbefore described with reference to and as illustrated in the accompanying drawings.