KR100360554B1 - Method of manufacturing a semiconductor device using the scanning exposure method and the scanning exposure method - Google Patents

Abstract

The present invention relates to a scanning exposure method, wherein a pattern of reticle is exposed to a shot area on a wafer in an exposure by a scanning exposure system, and to detect a focus position in the lead-ahead area before the exposure areas (16,116) with respect to the scanning direction. When lead-ahead, (1) for each shot region S10, S1-S4, S5, S23, S28, S29-S32 on the peripheral region of the wafer 15, exposure is performed by the scanning wafer 15 So that the exposed area 16, such as a slit, moves relatively from the inside of the wafer 15 to the outside, and (2) between the focus position of the lead-ahead area 135 and the focus position of the imaging plane 139 in the exposed area. When the absolute value for the difference of exceeds the allowable value, the height of the wafer 15 is fixed to the set height up to then to ignore the read head data.

Description

Scanning exposure method and method of manufacturing a semiconductor device using the scanning exposure method

The present invention relates to a scanning exposure method for sequentially exposing a mask pattern on a photosensitive substrate, for example, by a slit scanning or step-and-scan type apparatus, in particular when exposure is performed by scanning exposure in autofocusing and autoleveling. It relates to a scanning exposure method applied to.

In manufacturing a thin film magnetic disk by a semiconductor device, a liquid crystal display element, or a photolithography technique, a projection exposure apparatus is used, which transfers a pattern on a reticle (or photomask, etc.) to a wafer to which a photosensitive agent is applied. A conventional projection exposure apparatus which is frequently used is a step-and-repeat (stepper) reduction projection exposure apparatus, which sequentially moves each shot region on the wafer to an exposure field of an optical projection system, and moves a reticle pattern image to each shot of ground. It exposes to an area.

Such stepper projection exposure apparatus is provided with an autofocusing mechanism and an autoleveling mechanism, both of which align each shot region on the wafer with the imaging plane of the optical projection system. This autofocusing and autorailing mechanism measures the focus position (or tilt) at a predetermined measurement point (or measurement area) in the exposure field of the optical projection system, for example, by the servo system based on the measurement results. Correct the focus position (or tilt). In this case, since the wafer is stopped during exposure, there is no particular inconvenience even if the autofocusing and autoleveling mechanisms have a low response speed.

However, it is necessary to increase the resolution of the optical projection system as the pattern becomes more and more fine in recent semiconductor devices. Methods for improving the resolution include shortening the wavelength of exposure light and increasing the numerical aperture of the optical projection system. In both cases, when it is desirable to fix an exposure field similar to the prior art, it becomes increasingly difficult to maintain imaging performance (warping, bending of a field, etc.) with a predetermined precision over the entire exposure field. Therefore, it is preferable to use a projection exposure apparatus of a scanning exposure type apparatus such as a so-called slit scanning system or a step-and-scan type system (hereinafter referred to as a "scanning exposure type system").

Such a projection exposure apparatus of the scanning exposure type exposes the pattern of the reticle to the wafer, while being relatively relative to a plurality of trapezoidal radiation regions (hereinafter referred to as "slit-shaped irradiation regions") of rectangular, arc or two-dimensional form. Scanning wafers and reticles synchronously. Therefore, when the pattern having the same area as the stepper-type area is exposed on the wafer, the scanning exposure type device can reduce the irradiation area of the optical projection system when compared with the stepper type device, thereby enabling improved imaging in the irradiation area. Done.

Also, conventionally, the size of the reticle is 6 inches, and the mainstream of the projection magnification by the optical projection system is 1/5 times. However, as the circuit pattern area for such a semiconductor device becomes larger and larger, the magnification of 1/5 or less for a 6-inch reticle becomes insufficient. Therefore, it is necessary to design a projection exposure apparatus having an optical projection system of variable magnification such as 1/4 times. The scanning exposure type is advantageous for performing an area increase of the transferred pattern.

However, such scanning projection exposure apparatus also requires a mechanism for aligning the imaging plane with each shot region during exposure. However, in a scanning exposure type similar to the stepper type, although the focus position (inclination) on the wafer is measured in the actual exposure area and the calibration is performed based on the measurement result, the wafer is scanned so that the actual exposure area is aligned with the imaging plane. It is difficult, and has the disadvantage that the response of the autofocusing (or autoleveling mechanism) is a predetermined value.

More and more methods are provided for detecting the focus position, in consideration of the response speed, in advance reading the height, inclination, etc. of the wafer at the measuring point of the lead ahead area with respect to the scanning direction with respect to the scanning direction, There is a so-called lead-ahead system that corrects the focus position (or tilt) of the exposure area based on the head result.

Even when the focus position is detected by such a lead-ahead system, Figure 9 (a) of Japanese Patent Laid-Open Publication No. 4-196513 (US Patent No. 068,101 filed by the assignee of the same application on May 28, 1993) As shown in Fig. 9 (c), when wafer exposure is performed on all shot regions in a sequential manner simply from one end to the other as in a conventional step-and-scan type apparatus, When the shoot region includes the periphery of the wafer, the value of the focus position detected in the lead-ahead region can vary considerably according to the scanning direction, making precise position control impossible, and the inclination of the wafer is inclined in the imaging plane. It takes too much time to align with, or the calibration is too large in the autofocusing mechanism (or autoleveling mechanism), so the actual exposure area There is a problem that can not follow.

Moreover, in the step-and-scanning projection exposure apparatus, the exposure is performed on each shot region of the wafer by the scanning exposure type, and the wafer stepping is performed while being exposed to the shot region. In this case, conventionally, the positioning of the shoot area of the wafer to be exposed in the non-scanning direction (the direction perpendicular to the wafer scanning direction) must be completed before exposure starts on the shoot area.

As described above, it is also necessary to perform autofocusing continuously during scanning exposure even in the scanning exposure projection exposure apparatus. However, in the case of the scanning exposure type, since the wafer is scanned in the scanning direction predetermined for the projection optical system, the focusing position is measured only in the slit-shaped exposure area, and if the height of the wafer is adjusted based on the result of this measurement, the auto The response speed of the focusing mechanism has a predetermined maximum value. Thus, there is a possibility that a follow-up error occurs between the wafer surface and the imaging surface. Therefore, it is desirable to perform autofocusing even when the wafer is scanned so that a follow up error does not occur.

In addition, an appropriate amount is exposed to the photosensitive material such as photoresist on the wafer. In the scanning exposure type, when the irradiation intensity and the width of the slit exposure area are determined, the scanning speed for obtaining the appropriate amount of exposure is determined as the predetermined value. Therefore, when performing exposure on the wafer by the scanning exposure type, a predetermined acceleration portion (approach run portion) is necessary before the scanning speed of the wafer reaches a predetermined value. Preferably at least autofocusing is performed before the wafer passes through the accelerator.

Moreover, in the step-and-scanning projection exposure apparatus, there have been cases where the wafer is moved in the non-scanning direction until just before the exposure is performed by the scanning exposure type in a predetermined shoot area on the wafer. Therefore, if the wafer moves in the nonscanning direction, there is a possibility that an autofocusing follow up error will occur. Therefore, it is desirable that the follow-up error is not generated by being aligned.

Further, in the step-and-scan projection exposure apparatus, the wafer moving speed at the stepping time is considerably larger than the wafer moving speed (scanning speed) of the scanning exposure time. Therefore, there is a possibility that an unstable vibration occurs in the wafer due to the autofocusing operation at the stepping time.

It is an object of the present invention to provide a scanning exposure having excellent control accuracy in focus position or inclination with respect to the shoot area at the outer periphery of the wafer when the reticle pattern is exposed to each shoot area on the wafer in the scanning die.

It is still another object of the present invention to provide a scanning exposure that has excellent control accuracy at the focus position or inclination even in the shot region and does not waste the reticle.

It is another object of the present invention to provide an actual exposure area on a photosensitive substrate such as a wafer having an imaging plane of an optical projection system by exposure in a scanning exposure type without lowering the scanning speed, even if the height of the photosensitive substrate surface varies greatly. It is to provide a scanning method for precisely aligning and not reducing the follow-up accuracy of autofocusing or autoleveling as a whole.

It is another object of the present invention that it is possible to perform autofocusing with a small follow up error even if the exposure is performed by scanning exposure and the wafer is scanned when the autofocus is operated before the wafer passes through the accelerator. It is to provide a projection exposure method. It is also an object of the present invention to prevent the occurrence of autofocusing follow up errors caused by the movement of the wafer in the nonscanning direction.

It is still another object of the present invention to provide a projection exposure method for preventing unstable vibration in a wafer from occurring due to an autofocusing operation even at a time when wafer movement in stepping performs exposure at a high speed by a step and scanning type. .

A first scanning exposure method according to the present invention comprises the steps of: irradiating a mask (7) formed in a pattern transferred by a slit-shaped irradiation area, as shown in FIG. 10 and FIG. Projecting and exposing the mask 7 pattern of the irradiation area onto the photosensitive substrate 15; Moving each shot region S1-S32 on the substrate 15 to a scanning start position; Each shoot region on the substrate 15 is read sequentially when scanned in the predetermined or opposite direction with respect to the exposure region 16 coupled to the slit radiation region in synchronization with the scanning of the mask 7 in the predetermined or opposite direction. Detecting the height of the exposure surface of the substrate 15 before the exposure area 16 with respect to the scanning direction of the substrate 15 by an Ahead system; Adjusting the height of each shot region on the substrate 15 based on the lead head height (including the case where only the inclination degree is adjusted); And a step of sequentially exposing the pattern image of each mask 7 to a plurality of shot regions on the substrate 15.

In this scanning exposure method, the present invention provides that, when each shot region S1-S4 on the outer periphery of the substrate 15 (see FIG. 5) is exposed, the exposure region 16 moves from inside to outside (loci U1). (U3, U5, U7, U9 and U11) (see FIG. 10) move over this shoot area on a relatively outer periphery, while the height is detected by the lead-ahead system and based on the detection result of the substrate 15. The substrate 15 is scanned and exposed in a manner that corrects the height.

In this case, when the number (S1-S32) of the plurality of shooting regions on the substrate is even, the number of shooting regions (S1, S2, ...) in which the exposure region 16 is scanned in the same direction as the predetermined direction. And the exposure area 16 is made equal to the number of the shot areas S29, S30, ..... scanned in the opposite direction to the predetermined direction, so that the exposure of any shot area is started in the plurality of shooting areas. The exposure is preferably performed by alternately inverting the scanning directions for the respective shot regions.

On the other hand, when the number of the plurality of shooting regions on the substrate 15 is odd, the exposure starts from any one of the shooting regions in the predetermined direction in the scanning direction, and then alternately scanning directions for each shooting region. The number of the shot regions in which the exposure regions 16 are scanned in the predetermined direction so as to perform the exposure by inverting the number of the shot regions in which the exposure regions 16 are scanned one by one in the opposite direction to the predetermined direction in the plurality of shot regions. It is desirable to make it larger.

For an exposure area 16 opposite to the scanning direction (loci U3, U5, U7, and U9) of this area of the outer periphery, and a shooting area other than an inner shoot area on the outer periphery and an inner shoot area adjacent to the outer shoot area. The internal shoot regions (shoot regions scanned along loci U4, U6, U8 and U10) adjacent to the external shoot regions S1-S4 on the outer periphery of the substrate 15 with respect to the substrate 15, and the scanning direction of the substrate 15 ( It is possible to set so that the scanning directions of the shoot regions (shoot regions scanned along loci U2, U16, U15, U14, ...) arranged in the direction perpendicular to the X direction (Y direction) are alternately reversed. .

According to such a first scanning exposure method of the present invention, for example, as shown in FIG. 10, when the external shot area S1-S4 on the outer periphery of the substrate 15 is scanned by the scanning exposure system. The substrate 15 is scanned in such a manner that the exposure area 16 is moved from the inside of the substrate 15 to the outside, and the height is corrected by the lead-ahead system. Thus, even if any "sagging" appears around the periphery of the substrate 15, the exposure is not detected by the height of the substrate 15 without being sagging by the lead-ahead system, so that the exposure of the substrate 15 The height of the exposure surface is performed in the state of aligning the projected image of the mask 7. In addition, even if the exposure area 16 reaches the outer edge of the substrate 15 at the end of scanning and exposure and the height detected by the lead-ahead system fluctuates rapidly, the height adjustment mechanism remains in accordance with the predetermined time period. The height of (15) does not fluctuate so rapidly, and the height deviation is small.

Furthermore, in the case where two shot regions are successively exposed on the substrate 15, any idle movement is eliminated, and by scanning the mask 7 oppositely as compared to the slit-type exposure region 16. Yield is improved. This condition depends on whether the total number of shooting regions S1-S32 on the substrate 15 is even or odd. First, on the other hand, when the number is even, on the other hand, exposure equals the number of shot regions in which the exposure regions are scanned in the same predetermined direction and the number of shot regions in which the exposure regions are scanned in the opposite direction, and alternately reverses the scanning direction by the shot regions. Is performed. On the other hand, when the number of all shot regions on the substrate 15 is odd, the exposure is in the group of the shot regions arranged in the scanning direction, the number of shots in the group being larger than the number of other groups of the shot regions. By starting the exposure from the area and inverting the scanning directions by one shoot area alternately, the number of shoot areas in which the exposure areas are scanned in the same predetermined direction and the number of shoot areas in which the exposure areas are scanned in different opposite directions are made equal. Is performed.

Moreover, in the case where there are internal shoot regions (shoot regions scanned along loci U4, U6, U8 and U10) adjacent to the external shoot regions S1-S4 on the outer periphery of the substrate 15, The scanning direction for the exposure area 16 is set in a direction opposite to the scanning directions loci U3, U5, U7, and U9 for this external shoot area. With respect to such an external shot region of the outer periphery and a shot region other than this internal shot region, the scanning region is a shot region (loci U2, U16) which is disposed in a direction perpendicular to the scanning direction (X direction) on the substrate 15. , U15, U14, ...) are inverted alternately. This satisfies the conditions for reciprocally scanning the mask 7 scanning without any wasted movement.

The second scanning exposure method according to the present invention performs the same initial steps as the first scanning exposure method and, for example, when each shot region S1-S32 on the substrate 15 is exposed, the lead-ahead system It is determined whether or not the height detection region 35 or the exposure region 16 by means is completely included in the exposing region. The incomplete shot regions S1 and S4 in which the height detection region 35 or the exposure region 16 is not entirely included in the exposing region, compared with the shoot regions S1 and S4 in which the exposure is incomplete, the inside of the substrate 15 The exposure is performed by scanning the substrate 15 in a manner that is moved outward from the outside, the height is detected by the lead-ahead system, and the height of the substrate is corrected based on the detection result.

According to the second scanning exposure method of the present invention, the shot area S1 on the outer periphery of the wafer 15 in which the height detecting area 35 or the slit type exposure area 16 is not completely included by the lead-ahead system. For an incomplete shot region such as S4, the effect of the detected data obtained from the incomplete region is that the exposure region 16 is external from the inside of the substrate 15 along the locus U3 and U9. By scanning the substrate 15 in a relatively moving manner, it is reduced like the first scanning exposure method so that the height set value deviation with respect to the substrate 15 can be reduced as a whole.

The third scanning exposure method of the present invention includes, for example, irradiating an irradiation area in a predetermined form on a mask 7 formed of a transfer pattern, as shown in FIGS. Projecting the pattern in the irradiation area onto the photosensitive substrate 15 through an optical projection system PL; The mask 7 is scanned by scanning the substrate 15 in a predetermined direction (± Y direction) with respect to the exposure area 116 coupled to the irradiation area in synchronization with the scanning of the mask 7 in the predetermined direction with respect to the irradiation area. Pattern of the image is sequentially exposed on the substrate 15, where, for example, the height of the substrate 15 in the optical axis of the optical projection system, as shown in FIGS. 18 and 19, in the scanning direction. Read ahead at the measurement point 135D before the exposure area coupled to the irradiation area, and the lead head height is offset over the allowable area predetermined for the imaging plane of the optical projection system, and coupled to the irradiation area based on this lead head height. The height (focus position) of the substrate 15 is controlled in the exposed exposure area 116, so that the height of the substrate 15 is fixed to the height set up to this time.

In this case, the inclination of the substrate 15 is controlled (leveled) based on the lead head height, and when the lead head height is offset in a predetermined area with respect to the imaging plane of the optical projection system, then the substrate at that set It is preferable to fix the height and the inclination of (15).

Further, before the exposure area 116 coupled to the radiation area with respect to the scanning direction in parallel with measuring the height of the substrate 15 in the array of measuring points 135D arranged in the direction intersecting with the scanning direction, the substrate 15 The height of is also measured in the array of measuring points 135C arranged in the exposure area 15 in the direction crossing the scanning direction. The height of the substrate 15 is set so that the height measured in the array of measuring points 135 before the exposure area 116 with respect to the scanning direction is offset above a predetermined allowable area for the imaging plane of the optical projection system PL. Is set to.

Furthermore, as shown, for example, in FIG. 21, the height of the substrate 15 is measured at a plurality of measuring points AF21-AF29 arranged in a direction crossing the scanning direction in front of the exposure area 16 with respect to the scanning direction. do. When the height measured at the measuring points AF21-AF29 is offset beyond the allowable range predetermined for the image plane of the optical projection system PL (e.g., spacing ΔY1, ΔY3), the height above the allowable range Except for the data, it is desirable to control the height of the substrate 15.

According to this scanning exposure method of the present invention, the height of the substrate 15 is determined on the substrate 15 when the pattern image of the mask 7 is projected onto the substrate 15 by synchronously scanning the substrate and the reticle or mask. It is measured at measurement point 135D before the exposure area with respect to the scanning direction. Thereafter, when the area whose height is measured by the lead head system reaches the exposure area 116 of the pattern image of the mask 7, the height of the area is set based on the lead head height. This allows the exposure surface of the substrate 15 to accurately align the imaging plane of the optical projection system PL without reducing the scanning speed even in the scanning exposure system. That is, autofocusing is performed.

Further, as shown in FIG. 19 (a), if the height read in advance at the measurement point 135D is offset beyond the allowable area predetermined for the imaging plane 139, the measurement result is ignored. That is, by setting the height of the substrate 15 when the area to be measured reaches the exposure area 116 at the height set up to that time, it is possible to prevent the sudden autofocusing control which causes the performance to deteriorate as a whole. Moreover, the area on the substrate 15 which is significantly changed in height is the portion on the periphery of the substrate 15 which is not suitable for exposure, so the effect is minimized even if the measurement data in this area is ignored.

Then, the inclination of the scanning direction on the substrate 15 can be detected by reading in advance the height of the substrate at the predetermined measurement point, and in this way the substrate 15 is scanned. In addition, when there are a plurality of predetermined measurement points in the direction crossing the scanning direction, it is possible to detect the inclination of the nonscanning direction on the substrate 15. Thus, the inclination of the substrate 15 can be controlled (level) based on the detected inclination. In this case too, follow-up accuracy during overall autoleveling is achieved by fixing the inclination of the substrate 15 at a set inclination until the lead-ahead height is offset above a predetermined allowable area relative to the imaging plane of the optical projection system PL. Will not worsen overall.

In parallel with the height measurement of the substrate 15 in an array of measuring points 135D arranged in a direction crossing the scanning direction before the exposure area 116 coupled to the radiation area with respect to the scanning direction, the height of the substrate 15 is exposed to exposure. When measured in an array of measuring points 135D arranged in a direction crossing the scanning direction of the area 116, autofocusing is performed more accurately by using the measurement data and the lead head measurement data in the exposure area 116. . Even in this case, if the pre-read height is offset above a predetermined allowable area for the optical projection system PL, the follow-up accuracy of performing autofocusing is entirely determined by fixing the height of the substrate 15 to the set height by then. It doesn't get worse.

Further, the height of the substrate 15 is measured at the plurality of measuring points AF21-AF29 arranged in the direction crossing the scanning direction before the exposure area 116 with respect to the scanning direction, and measured at the plurality of measuring points AF21-AF29. If part of the height gained is offset beyond the allowable area predetermined for the optical projection system PL (e.g., gaps Y1, Y2), the measurement data of the height offset from the allowable area is on the edge which is not suitable for pattern exposure. Data about the same area as the area. Thus, follow up accuracy for autofocusing and autoleveling can be improved for other areas by eliminating this measurement data.

The projection exposure method according to the fourth embodiment has an optical projection system PL, which projects the pattern image on the mask 7 on the photosensitive substrate 15 in a predetermined irradiation optical, and the optical axis of such an optical projection system. In the projection exposure method for sequentially exposing the mask 7 pattern to a predetermined shoot area 234 on the substrate 15 by scanning in synchronization with the scanning of the mask 7 in a first direction perpendicular to the substrate 15, the substrate 15 The position of the optical projection system PL with respect to the light direction in the lead-ahead area 235B before the exposure area 216 with respect to the second direction by the optical projection system PL on the image is measured, based on the result of this measurement. Thus, in adjusting the position of the substrate 15 with respect to the optical axis direction, the lead head region 235B is disposed between the predetermined shoot region 234 and the exposure region 216 when the substrate 15 is moved to the scanning start position. Is placed on.

In this case, after scanning of the substrate starts exposing the mask 7 pattern to the substrate 15, the positioning of the substrate 15 in the second direction (X direction) and in the vertical direction (Y direction) is read-out. It is preferable that the head region 235B is completed before reaching the predetermined shot region 234.

Further, the optical axis position adjustment of the substrate 15 is stopped immediately after the exposure is completed to the predetermined shot region 234, and the substrate 15 is placed at the next scanning start position in the second direction for the exposure of the substrate to the next shot region. The position of the optical axis of the substrate is preferably started after the positioning of the substrate is finished in the direction perpendicular to the second direction.

According to the method of the third embodiment, the optical axis position (focusing position) of the optical projection system is detected in the lead-ahead area 235B before the exposure area 216 on the substrate 15 in the second direction (scanning direction). The focusing position of the substrate 15 in the exposure area 216 is adjusted based on the detection result. Therefore, the response speed of the autofocusing mechanism has a predetermined upper limit range, and high accuracy follow-up auto focusing is performed even if the substrate 15 is scanned. Moreover, autofocusing as well as autofocusing can be performed to align the inclination of the imaging plane with the inclination of the substrate 15 surface.

In addition, an accelerator having a predetermined width Lac is required to accelerate the substrate 15 at a predetermined constant scanning speed, as shown in FIG. When the lead head region 235B is in the acceleration portion at the scanning start point, that is, when the lead head region 235B lies between the shot region 234 and the exposure region 216 on the substrate 15, the auto Focusing is precisely performed when the shot region 234 passes through the accelerator and enters the exposure region 216.

Next, in FIG. 25, for example, even when the substrate is stepped in the non-scanning direction (Y direction) and the lead-ahead area 235B enters the shoot area 234, the shoot area 234 is in the Y direction. When moving to, autofocusing cannot be performed correctly. This is because the information of the focusing position obtained in the lead head area 234 is no longer the information of the focusing position at the appropriate measuring point. In order to prevent this, positioning (stepping) of the substrate in the non-scanning direction (Y direction) must be completed before the lead-ahead region 235B enters the shot region 234.

Furthermore, when the focusing position adjustment (autofocusing) of the substrate 15 is stopped immediately after the exposure is completed on the predetermined shot region 234, the substrate 15 is stepped at a high speed, and the focusing position to be detected is detected. Even if the wave amount of is increased compared with the case of scanning exposure, it does not oscillate unstable. Subsequently, scanning of the substrate 15 in the scanning direction is started, and when the autofocusing is started after the stepping in the non-scanning direction is finished, the moving speed of the substrate 15 is equal to or slower than the scanning speed in the exposure time, and thus is stable. Autofocusing is performed in a manner.

An embodiment of the present invention will be described with reference to the following drawings. Since this embodiment is an application of the present invention for exposure by a step-and-scanning projection exposure apparatus, the operation principle of the step-and-scanning projection exposure apparatus will first be described with reference to FIGS.

1 shows a scanning exposure projection exposure apparatus proposed by the inventor in US Patent No. 08 / 301,991. In the drawing, exposure light IL from the light source system 1 including such as a light source, an optical integrator, the first relay lens 2, the reticle blind (variable field aperture) 3, the second relay lens 4, The rectangular radiating region 8 on the reticle 7 is irradiated uniformly through the mirror 5 and the main condenser lens 6. The mounting surface of the reticle blind 3 is combined with the patterning surface of the reticle 7, and the position and shape of the radiation area 8 on the reticle 7 is determined by the position and shape of the opening of the reticle blind 3. do. The light source used in the light source system 1 includes a very high pressure mercury lamp, an excimer laser source and a harmonic YAG laser generator.

The reticle 7 pattern image of the irradiation area 8 is projected and exposed through the optical projection system PL to the rectangular exposure area 16 on the wafer 15 on which the photoresist is coated. The Z axis is taken parallel to the optical axis of the optical projection system PL, the X axis is in a direction parallel to the sheet plane of FIG. 1 in a two-dimensional plane perpendicular to the optical axis AX, and the Y axis is perpendicular to the sheet plane of FIG. Direction. In this embodiment, the scanning direction of the reticle 7 and the wafer 15 when exposed in the scanning exposure type is parallel to the X axis.

On the other hand, the reticle 7 is fixed on the reticle end 9 supported on the reticle base 10 so as to be driven in the X direction at a predetermined speed by, for example, a linear motor. The reticle 7 coordinates in the X direction are continuously measured by a moving mirror 11 and an external laser interferometer 12 fixed on one end of the reticle end 9 in the X direction, The measured coordinate information is provided to the main control system 13 for controlling the operation of the entire apparatus. The main control system 13 controls the position and movement speed of the reticle 9 via the reticle drive system 14.

On the other hand, the wafer 15 is fixed to the wafer holder 17, which is mounted to the Z leveling end 19 through a supporting point composed of three extendable piezo elements or the like. The Z leveling stage 19 is installed on the XY stage 20, which is supported on the wafer base 21 which is slidable in two dimensions. The Z leveling stage 9 finely adjusts the vertical position or the Z direction position (focus position) of the wafer 15 on the wafer holder 17 through three support points, and finely inclines the exposure plane of the wafer 15. Adjust On the other hand, the Z leveling stage 19 roughly adjusts the wafer 15 in the Z direction. Moreover, the XY stage 20 sets the position of the Z leveling stage 19, the wafer holder 17 and the wafer 15 in the X and Y directions, and the wafer is parallel to the X axis at a predetermined scanning speed during scanning exposure. Scan (15).

The XY coordinates of the XY stage 20 are continuously monitored by the movable mirror 22 fixed to the XY stage 20 and the external interferometer 23, and the detected signal of the XY coordinates is sent to the main control system 13. Is provided. The main control system 13 controls the operation of the XY stage 20 and the Z leveling stage 19 via the wafer drive system 24. When performing exposure by scanning, the shot region to be exposed on the wafer 15 is positioned at the exposure start position, and the wafer 15 is radiated region 8 at a speed V _RO through the reticle end 9. At the speed Vex (= β.VRO) at the amplification degree β of the optical projection system PL through the XY stage 20 in synchronization with the scanning of the reticle 7 in the -X direction (or X direction) with respect to Scanned in the X direction (or -X direction) with respect to the exposure area 16. Thus, the pattern image of the reticle 7 is sequentially exposed in its shoot area on the wafer 15.

The operation of the focus position detection system (hereinafter referred to as "AF sensor") apparatus for detecting the position in the Z direction of the exposure plane of the wafer 15 is described. Such a focus position detection system is proposed in U.S. Patent Application No. 172,098, filed Dec. 23, 1993. Although nine AF sensors having the same apparatus are actually installed in the apparatus of FIG. 1, FIG. 1 shows only three AF sensors 25A2, 25B2, and 25C2. First, light that is non-photosensitive to the photoresist emitted from the light source 26A2 in the intermediate AF sensor 25A2 emits a slit pattern in the light transmitting slit plate 27A2, and the image of the slit pattern is the optical projection system PL. Is projected to the measuring point PA2 of the wafer 15 arranged at the center of the exposure area 16 at an angle to the optical axis AX. The light reflected from the measuring point PA2 is focused on the oscillation plate 30A2 through the condenser lens 29A2 in which the image of the slit pattern projected on the measuring point PA2 is imaged again.

Light passing through the slit in the oscillation plate 30A2 is photoelectrically converted by the photodetector 31A2, and the photoelectrically converted signal is provided to the amplifier 32A2. The amplifier 32A2 detects a photoelectrically converted signal from the photodetector 31A2 with the drive signal of the oscillating slit plate 30A2, and amplifies the resulting signal to generate a focus signal, which is measured point PA2. And a focus signal is provided to the plane position calculation system 33. Similarly, another AF sensor 25B2 projects a slit pattern image onto the measuring point in the -X direction with respect to the measuring point PA2, and electrically converts light from the slit pattern image with the photodetector 31B2. It is provided to the amplifier 32B2. The amplifier 32B2 supplies a plane signal corresponding to the focus position of the measuring point PB2 to the plane position calculation system 33. Similarly, the AF sensor 25C2 projects the slit pattern image onto the measuring point PC2 in the X direction with respect to the measuring point PA2, and photoelectrically converts light from the slit pattern image with the photodetector 31C2. This is provided to the amplifier 32C2. The amplifier 32C2 provides the plane position calculation system 33 with a focus signal corresponding to the focus position of the measuring point PC2.

In this case, the focus signal obtained by the amplifiers 32A2-32C2 from the photoelectrically converted signal from the AF sensor 25A2-25C2 has the measuring point PA2-PC2 at the imaging plane by the optical projection system PL. Are each adjusted to zero when matched with. Thus, each focus signal corresponds to the deviation (defocus) of the focus position of each measurement point PA2-PC2 from the imaging plane.

FIG. 2 shows the measurement point distribution on the wafer of the FIG. 1 device. In the figure, the three measuring points PA1-PA3 are arranged in a straight line along the Y direction at the center of the rectangular exposure area 16 having the width D in the X direction. The lead-ahead areas 34 composed of the measuring points PB1-PB3 are arranged around the positions separated from the measuring points PA1-PA3 at intervals of d in the -X direction, respectively, and constitute the measuring points PC1-PC3. The lead head regions 35 are disposed around positions separated from the measuring points PA1-PA3 at intervals of d in the X direction, respectively. The measuring point PA2 is disposed at the center of the exposure area 16, and the focus positions of the nine measuring points are separately measured by an AF sensor having a device such as the AF sensor 25A2 of FIG. Then, the measured value of the focus signal in the lead-ahead area 34 before the exposure area 16 for the scanning direction is used when the wafer 15 is scanned in the X direction, and before the exposure area 16 for the scanning direction. The measured value of the focus signal in the lead ahead area 35 is used when the wafer 15 is scanned in the -X direction. However, there is a method of using the focus signal obtained at the measurement points PA1-PA3 of the exposure area 16 together with the focus signal of the lead head area 34 or 35.

In particular, it is assumed that the wafer 15 is scanned in the X direction or the reticle 7 is scanned in the -X direction, for example as shown in FIG. In this case, the focus position of the lead head area 34 before the exposure area 16 (or the measurement position of the measurement point in the exposure area 16 and the measurement point of the lead head area 34 in the scanning direction (X direction)). Are measured by the AF sensor, and the focus position of the shot region is corrected based on the measurement result. Thereafter, exposure has conventionally consisted of all the shot regions on the wafer 15 which will be described later sequentially.

FIG. 5 shows an example of an exposure procedure by the scanning exposure projection exposure apparatus shown in FIG. As shown in the figure, the exposure plane of the wafer 15 is formed with the shoot regions S1-S32 having predetermined pitches in the X and Y directions. Each shot region S1-S32 was formed in the same chip pattern as in the previous process when exposing the second layer or later to expose the wafer 15. First, assuming that the exposure starts from the shoot area S1 at the upper left corner of the wafer 15, the trajectory of scanning relative to the slit exposure area 16 is referred to as T1, T2, T3 is in order. In fact, since the wafer 15 is moved, the moving direction of the wafer 15 is the opposite direction of the arrow with respect to the trajectories T1-T32.

Further, conventionally, exposure starts in the order of slots S1, S2, S3 ... S32, and the trajectory of the relative scanning of the exposure area 16 with respect to each of the shooting areas S1-S32 is T1-T32. Is displayed. In this case, scanning is performed in the -X direction with respect to the first shoot region S1 of the first row X1 (see FIG. 5) at the outermost periphery above the wafer 15 with respect to the exposure region 16. do. That is, the wafer 15 is scanned as if the exposure area 16 is moved in the X direction along the locus T1 on the wafer 15. At that time, the focus position is read in advance in the lead head region 35 before the exposure region 16 with respect to the scanning direction, and the focus position of the wafer 15 and the inclination of the wafer 15 are based on the resulting focus position. Adjusted. In practice, it is possible to use the focus position obtained at the measuring point in the exposure area 16, which is applied later.

Thereafter, after the second shooting area S2 adjacent to the shooting area S1 in the Y direction is gradually driven to a stage by the wafer, the wafer 15 is moved to the scanning start position, and the wafer 15 is then exposed to the exposure area 16. ) Is scanned as it moves on the shoot area S2 in the -X direction along the trajectory T2. At that time, the focus position is read in advance in the lead head region 34 before the exposure region 16 with respect to the scanning direction, and the focus position and the inclination of the wafer 15 are adjusted based on the resulting focus position. Thereafter, the exposure is continued in the scanning exposure system up to the last shot area S4 of the first row, while the scanning directions are alternately reversed. Therefore, the direction of the trajectory T5 with respect to the exposure area 16 on the first shoot area S5 in the second row X ₂ is the exposure area (on the last shoot area S4 in the first row X ₁ ). 16 is the opposite direction of the trajectory (T4) for.

Exposure by the scanning exposure system alternately inverts the scanning direction for each adjacent shoot area from the shoot areas S6-S10 of the second row to the shoot areas S29-S32 of the last row, and at the end of each row This is done by reversing the scanning direction between the shoot area and the first shoot area in the next row.

In the scanning exposure sequence as described above, the scanning direction is alternately inverted with respect to the exposure area 16 of the shooting regions S1-S4 at the outer periphery of the top row on the wafer 15 (ie, the shot adjacent to the outer periphery). A shot region partially overlapping the region or the outer periphery), the scanning direction is alternately reversed with respect to the shot regions S29-S32 at the outer periphery of the lowest row. However, as shown in FIG. 4, the end 15a of the surface of the wafer 15 is generally thinner from inside to outside due to sagging. Therefore, when the slot exposure area 16 is scanned from the outside to the inside for the shoot areas such as the shoot areas S2 and S4 of the outer periphery of the wafer 15 (the wafer 15 is scanned from the inside to the outside), Since the control mechanism and the tilt for the focus position detected in the lead-ahead area 34 are so large that they do not tend to be subject to variations in the focus position, the exposure plane of the wafer 15 is considerably deviated from the allowable area. There is a disadvantage of being exposed.

In addition, the exposure area 16 is moved from the outside to the inside along the trajectory T29 with respect to the shooting area such as the shooting area S29 at the lower right corner of FIG. 5, a part of which is lost by the outer peripheral end of the wafer 15. FIG. When scanned, the leveling begins in a state affected by the data for the focus position of such a loss. Thereafter, immediately after the scanning exposure is started, the exposure plane of the wafer 15 is exposed with a great inclination to the imaging plane, and even after the lead-ahead region with respect to the focus position is moved from the loss area to the wafer 15 Due to the response characteristics of the calibration mechanism, it takes time for the slope of the wafer 15 to align the slope of the imaging plane.

Moreover, since the scanning direction with respect to the slit-shaped or slit-shaped irradiation area 8 in the reticle 7 is a predetermined direction or the opposite direction, the most efficient movement such as the operation sequence of the reticle 7 is followed by the exposure from the shooting area. When moving to the shooting region, the scanning direction for the reticle 7 is always reversed. This means that when all the shot regions on the wafer 15 are exposed, the reticle 7 is reciprocated with respect to the irritated irradiation region 8 so that no idle return is required. Thus, even if an exposure order having good control accuracy with respect to the focus position and the inclination is required for the shot area at the outer periphery of the wafer, in the exposure order, the scanning direction is changed when the scanning direction is moved from one shoot area to the next area. It is required to be reversed, ie the scanning direction for the reticle is to be reversed.

In addition, as described above, if there is an area where the height of the wafer surface changes considerably in the periphery of the wafer, etc., the correction amount is increased in the autofocusing mechanism (or the autoleveling mechanism) so that the actual exposure area may not follow up the imaging plane. To avoid this, lowering the scanning speed is sufficient, but it worsens the productivity of the exposure processor.

A first embodiment of the scanning exposure method is described next with reference to FIGS. 1 to 5 and 6 to 11.

In FIG. 2, the distance from the end of the exposure area 16 to the center of each of the lead head regions 34 and 35 in the scanning direction is La. In this case, it is possible to use detection data of the focus position detected in the exposure area 16 together, but assuming that the next correction is performed based on the focus position detected in the lead head area 34 or 35. An explanation is given. First, when the exposure is performed in this embodiment, the reticle end 9 and the XY end 20 of the wafer 15 in FIG. 1 are described with reference to FIGS. 7 (a) and 7 (b).

FIG. 7 (a) shows the change of the moving speed V _R of the reticle stage 9 in the X direction as the scanning direction, and FIG. 7 (b) shows the moving speed Vwx of the XY stage 20 in the X direction. Fig. 7 (c) shows the change in the moving speed V _WY of the XY stage 20 in the Y direction perpendicular to the scanning direction. When the pattern image of the mask or reticle 7 is exposed on the wafer 15, the moving speed of the XY stage 20 in the X direction becomes V _ex . Moreover, since the projection magnification of the optical projection system PL is β from the reticle 7 to the wafer 15, the scanning speed of the reticle stage 9 in the X direction when the exposure is performed by the scanning exposure method is V _ex / β. Similarly, the acceleration of the reticle 7 (reticle stage 9) is equal to the acceleration of the XY stage 20 at the wafer 15 in the X direction such that the reticle 7 is synchronized with the wafer 15 from the acceleration start position. 1 / is the β. in other words, the 7 (a) Fig., the acceleration time if the acceleration of the reticle stage (9) in a _r t _a is

to be.

8 shows the positional relationship at the acceleration start time between the shot region Si exposed on the wafer 15 and the exposure region 16 in the slit or slit shape. In the figure, assuming that the shot region Si is scanned in the -X direction with respect to the exposure region 16, the distance Lc from the left edge of the shooting region Si and the right edge of the exposure region 16 is the acceleration of the XY stage 20. Is the acceleration distance on the wafer 15 required for the acceleration distance Lc,

to be.

In the case of FIG. 8, the lead head area 35 on the right side of the exposure area 16 is used as the lead head area for the focus position. The X-direction interval between the right edge of the exposure area 16 and the center of the lead-ahead area 35 to which the light for detecting the position (hereinafter referred to as "autofocusing beam") is irradiated is La (see FIG. 2). article). Therefore, in FIG. 8, the distance Lb of the acceleration start position between the center of the lead-ahead area 35 and the edge at which the exposure of the shoot area Si is started,

to be.

For example, the scanning speed V _ex at the exposure of the wafer is 50 mm / s, the amplification degree β of the optical projection system PL is 1/5 (reduced projection), and the reticle stage 9 at the start of scanning. When the acceleration a _r is 200 mm / sec ² and the interval La with respect to the lead-ahead area 35 of FIG. 8 is 10 mm, the acceleration time t _a , the acceleration distance L _c , and the interval ( L _b ) is represented by the formulas (1)-(3),

Becomes

In this case, since the lead-ahead area 35 is at a distance Lb from the front edge of the shot area Si to be exposed at the start of acceleration in FIG. 8, it is spaced from the front edge of the shoot area Si. The position at Lb is an area where an exposure area exists on the wafer 15 or the substrate to which the photoresist is applied so that the lead-ahead focus position is effective. In this case, the exposure area is simply defined by referring to the peripheral edge of the wafer 15. However, as described with reference to FIG. 4, since the outer periphery of the wafer 15 generally has "sagging" having a width of about Lb, the exposure area has a width Lp from the outer periphery of the wafer 15. It can be defined as an area except a ring-shaped area. In addition, the exposure area can be defined in any other area.

The relationship between the exposure area and the shoot area will be described with reference to FIG.

10 shows an example of the exposure procedure for this embodiment. As shown in the figure, the exposure plane of the wafer 15 is formed in the shoot regions S1-S32 at predetermined pitches in the X direction (scanning direction) and the Y direction. In order for the exposure to be performed on and after the second layer, each shot region S1-S32 is formed in the same chip pattern by the previous process.

As an example, assume that the exposure area is an area where the sagging of width Lp is removed from the periphery of the entire exposure plane of the wafer 15. In this case, since the shooting areas S1, S4, S5, S10, S23, S28, S29, and S32 itself protrude from the periphery of the wafer 15, each of them has a shooting area not entirely included in the exposure area ( Hereinafter referred to as the "loss region". Further, as described with reference to FIG. 8, it is necessary for the lead-ahead area 35 to be set at a position at a distance Lb from the front edge of the shoot area Si, which is included in the exposure area.

That is, in the exposure region in which the sagging of width Lp is removed at the outer periphery of the wafer 15 shown in FIG. 10, a band region having a width Lb in the X direction and the scanning direction from the contour region is further included. The removed area is generally an exposure area. In this case, since each shot region S2, S3, S30, S31 on the outer periphery of the wafer 15 of FIG. 10 has a portion which generally protrudes from the exposure region, the shot regions S2, S3, S30, S31. ) Is regarded as a lost shot region. In this embodiment, when the lossy shoot regions S1-S4, S5, S10, as well as S23, S28, S29-S32 on the wafer 15 shown in FIG. 10 are exposed by the scanning exposure system, the wafer 15 ) Is scanned so that the exposure area 16 moves from inside to outside with respect to the wafer 15.

However, if the widths Lp and Lb are narrow, the shoot regions S2, S3, S30, S31 can be included in the generally exposed areas. In this case, the scanning direction for the shooting regions S2, S3, S30, S31 may be the X direction or the -X direction, but the shot adjacent to the Y direction (non-scanning direction) so that the reticle can be moved oppositely. It is desirable to alternately reverse the scanning direction for the area. In this case, for example, the scanning direction is opposite to the shoot regions S2 and S3.

Here, the scanning direction is determined, for example, with respect to the shoot area in which the exposure area 16 is moved from the outside of the wafer 15 to the inside, the XY end 20 of the wafer is located near the lead-ahead area 34. Once driven stepwise to set the front edge of the shoot area on the wafer 15, autofocusing is performed from the focus position information. Then, while maintaining the focus position after the wafer 15 is moved to the acceleration start position for the scanning exposure, scanning starts while maintaining the focus position until the lead head region 34 enters the shot region. In this way, even when large sagging partially remains inside the region having the width Lp from the periphery of the wafer 15, the deviation of the focus position with respect to the shot region is not increased.

Thereafter, as shown in FIG. 7A, exposure is performed while the reticle and wafer movement speeds are maintained at constant scanning speeds at the exposure time t _ex after the acceleration time t _a , respectively. Then, the acceleration and deceleration of the reticle for the next exposure is performed at the acceleration or deceleration time t _A after the exposure time t _ex . However, this is an operation when the scanning direction for the reticle is inverted between the shoot regions that are continuously exposed. In addition, as shown in FIG. 7 (c), the moving speed VWY of the XY end 20 in the wafer in the Y direction is used to move the wafer 15 in the step form in the non-scanning direction (Y direction). Is changed within the travel time t _B during the acceleration or deceleration time t _A. The operations involving stepping in the non-scanning direction that must be performed between the acceleration or deceleration of such a reticle and the shoot regions continuously exposed are the following four operations.

(A) Acceleration and deceleration of the reticle 7

(B) Stepping the Wafer 15 in the Non-Scanning Direction (Y Direction)

(C) Stepping the Wafer 15 in the Scanning Direction (X Direction)

(D) Idle return of reticle (7)

The idle return of the reticle 7 in the last item is that the reticle 7 is scanned once without exposure to scan the reticle 7 in the same direction when the scanning direction is the same in the shoot area where the exposure direction is continuously exposed. It means action. If the time required for the four operations (A)-(D) is t _A , t _B , t _C and t _D , respectively, the relationship between these times is

to be.

Therefore, the time required from the time until the scanning exposure for the shot region on the wafer 15 exposed in advance at the time when the exposure is started for the next shot region is completed is shown in Figs. 9 (a) -9 (c). As shown in Fig. 2, the difference in the arrangement of the shoot regions and the difference in the scanning direction with respect to the reticle is varied. Each case is described below.

(1) As shown in FIG. 9 (a), in the non-scanning direction (Y direction), exposing from the shooting region Si to the adjacent shooting region Sj, the scanning direction for the reticle is reversed. .

In this case, the operations (A) and (B) are performed, since stepping in the non-scanning direction of (R) can be started simultaneously when the scanning exposure is completed and can be completed within the acceleration time t _A. From the scanning exposure completion time for the shooting region Si, the time T ₁ required to start the exposure for the next shooting region Sj is

to be.

(2) As shown in FIG. 9 (b), the exposure step from the shooting area Sp to the adjacent shooting area Sq in the scanning direction (X direction), and the scanning direction for the reticle are reversed.

In this case, as shown in Figs. 7 (a) and 7 (b), the XY stage 20 of the wafer 15 operates completely in synchronization with the reticle stage 9, but the wafer 15 Two possible operations can be taken: The first operation is to move the wafer 15 to the acceleration start position (scanning start position) for the next shot region at the same time as the completion of the scanning exposure for the shot region, and the second operation, Since the moving speed of the wafer 15 slows the wafer 15 in synchronization with the reticle 7, the wafer 15 is stamped only after it becomes zero. The former is advantageous for productivity, but the latter is discussed here. In this case, the time T ₂ required from the completion of the scanning exposure to the shooting area Sp to the start of the next shooting area Sq is

to be.

In this case, even though an operation for stepping the wafer 15 in the non-scanning direction is included, T ₂ = t _A + t _C , and the time T ₂ is the same since (t _A + t _C )> t _B. Is maintained.

(3) As shown in ninth (c), when stepping from the shooting area Si to the adjacent shooting area Sj in the non-scanning direction, the scanning exposure idle-returns the reticle 7, thus scanning the same. Direction is carried out continuously.

In this case, the above-described operations (A), (C) and (D) are performed. That is, the reticle 7 is decelerated after the scanning exposure is completed, and the reticle 7 returns to the acceleration start position for the next shot area after the movement speed V _R reaches zero. Therefore, the time T ₃ required from the completion of the exposure to the previous shot region Si to the start of the exposure to the next shot region Sj is

to be.

When the relationship of equation (5) is replaced by equations (6)-(8), exposure starts to the next shoot area from completion of exposure to the previous shoot area in the operation of Figs. 9 (a) -9 (c). The time needed to time T ₁ -Time T ₃ has the following relationship.

As can be seen from equation (9), the exposure order (shooting map) of stepping in the non-scanning direction and inverting in the scanning direction between adjacent shot regions is most preferable for productivity, as shown in Fig. 9 (a). It can be said. On this principle, the most favorable exposure sequence for productivity is the conventional exposure sequence shown in FIG. However, the exposure order of FIG. 5 is not used in this embodiment because of poor control accuracy with respect to the focus position and the inclination of the outer periphery of the wafer 15.

Further, as shown in FIG. 9 (e), the exposure sequence for scanning the next shot region in the same direction as the previous shot region after the idle return of the mask or the reticle 7 in order to idle return the reticle 7 It takes a very long time, approximately equal to the time from the start of acceleration to the completion of deceleration. Therefore, it is desirable to avoid the idle return operation on the reticle 7. Next, the present embodiment provides an exposure sequence that can prevent sagging at the outer periphery of the wafer 15 and does not include an idle return to the reticle 7.

6A and 6B are flowcharts showing an example of a method for determining the exposure order in this embodiment. Exposure is performed to the shot regions S1-S32 on the wafer 15 of FIG. 10 by the scanning exposure system according to this exposure sequence.

Typically, the main control system 13 is optimized for the external shape of the wafer based on the external dimensions of the wafer, the size of the shoot area, the step pitch and the like prior to the first printing (exposure to the first layer). Form a shoot map. This shot map data represents a method of arranging a shot region in a wafer from an external shape of the wafer, which is automatically determined by calculation and stored in memory as a stepping coordinate of the XY stage 20. The main control system 13 is stored in memory and based on the determined shot map data to determine the scanning direction for each shot region. 10 shows an example of shoot map data determined corresponding to an external shape of a wafer.

First, in step 81 of FIG. 6A, whether the scanning direction for the shoot regions S1-S4, S5, S10, S23, S28, S29-S32 at the outer periphery of the wafer 15 is calculated from equation (2). It is determined by using the accelerating acceleration distance Lc. On the other hand, when the acceleration distance is taken into account in step 85, the shoot area in which the lead-ahead area 34 (see FIG. 2) is not included in the generally exposed area is the wafer 15 with respect to the slit exposure area 16. In the acceleration start position. If the acceleration distance Lc is long, such a shot region includes all the shot regions S1-S4, S5, S10, S23, S28, and S29-S32 of FIG. 10, so that the scanning direction for the wafer 15 is determined by the exposure region ( 16 is determined to move from the inside of the wafer 15 to the outside. The operation then enters step 84.

On the other hand, when the acceleration distance Lc is not taken into account in step 81, the operation goes to step 82 to determine the scanning direction for the wafer 15, whereby the exposure area 16 moves in the scanning direction. In the outer periphery of the wafer 15 with respect to the (X direction), the shot regions S1-S4, S5, S10, S23, S28, S29, and S32 are equally moved from the inside of the wafer 15 to the outside. Next, in step 83, the scanning direction with respect to the wafer 15 is determined so as to shoot the shot region from the shoot regions S1-S4, S5, S10, S23, S28, S29-S32 at the outer periphery with respect to the scanning direction. (S9-S6, S16, S11, S22, S17, S27-S24) each have a scanning direction opposite to the adjacent shoot area at the outer periphery. Thereafter, step 84 is entered.

In step 84, the total number of shoot regions on the wafer 15 (32 in the embodiment of FIG. 10), the number of shoot regions whose scan direction is defined (24 after execution of steps 82 and 83) and the scanning direction The number of this unlimited shoot area is calculated. In the embodiment of FIG. 10, after the execution of steps 82 and 83, the shoot regions of which the scanning direction is not limited are eight shoot regions scanned along the trajectories U13-U16 and U17-U20. Thereafter, in step 85, the scanning direction for the wafer 15 is determined for a shoot area in which the scanning direction is not limited so that the number of shoot areas having different scanning directions are equal to each other; Alternatively, the number of shot regions having two scanning directions (X direction or -X direction) is larger than that of one shot region, and the scanning direction for the shot regions arranged in the non-scanning direction (Y direction) is Alternately reversed.

After performing steps 82 and 83 in the embodiment of FIG. 10, the eight shot regions scanned along the trajectories U13-U16 and U17-U20 have four shot regions in the X direction and the remaining four shots. The area is sufficient to make the scanning direction in the -X direction. The scanning direction is alternately inverted with respect to the shoot area scanned along the trajectories U13-U16, and the scanning direction is sufficient to be inverted alternately with respect to the trace U17-20.

Operation then proceeds to step 87 where it calculates the total number of shot regions on the wafer 15 (32 in FIG. 10) and decides whether they are even or odd, and the minor step as in FIG. And any shot region (e.g., shot region S10) is first exposed using a scanning exposure system. On the other hand, the subsequent operation exposes the shooting area S11 in the opposite scanning direction with respect to the previous shooting area S10, and exposes the shooting areas S1, S9, S3 .... S22 and S23 in this order. In step 90, the scanning directions are alternately reversed. On the other hand, if the total number of shot regions in step 87 is odd, the operation goes to step 89 of first exposing any shot regions belonging to the scanning direction for the larger number of shot regions, and then to step 90. Enter This completes the exposure to all the shot regions S1-S32 on the wafer 15.

In this case, the movement trajectory of the exposure area 16 with respect to the wafer 15 in the i-th (i = 1-32) scanning exposure is represented by the trajectories U1-U32 of FIG. That is, as indicated by the trajectory U1, the wafer 15 is first scanned so that the exposure area 16 moves in the X direction with respect to the shoot area 10. That is, since the exposure area 16 is actually stopped, the wafer 15 is scanned in the direction opposite to the arrow of the trajectory U1 (-X direction). In this movement, the focus position and the inclination of the wafer 15 are corrected according to the focus position measured in the lead head region 35. Then, after exposure is performed to the shooting region S11 as indicated by the trajectory U2, the wafer 15 is scanned and the shot region at the outer periphery as indicated by the trace U3 to expose the shooting region S11. The exposure 16 is moved in the X direction with respect to S1. In this movement, the focus position and the inclination of the wafer 15 are corrected based on the focus position measured in the lead head area 35. Then, exposure is performed in the order of traces U4, U5 ... U31, U32.

As described above, according to the present embodiment, the wafer 15 is scanned so that the exposure area 16 is in the respective shot regions S1-S4, S5, S10, S23, S28, S29 at the outer periphery of the wafer 15. Scanned to move from inside the wafer 15 to outside with respect to S32. Thus, even if sagging occurs in the outer periphery of the wafer 15, the focus position data measured in the lead-ahead area 34 or 35 does not exhibit an error value at the initial stage of scanning exposure, so that The adjustment of the focus position (auto focus) and the adjustment of the tilt angle (auto leveling) are performed correctly. In addition, when the operation of step 85 is selected in FIG. 6, the lossy shot area in the exposure area 16 and at least some (or almost exposed area) lead-ahead area 34 not completely included in the exposure area. (Or 35), the wafer is scanned so that the exposure area 16 moves from the inside of the wafer 15 to the outside, so that autofocusing and autoleveling of the wafer 15 are performed accurately.

In the exposure sequence of FIG. 10, since the shot region of the trace U13 is exposed after the shot region S16 scanned along the trace U12, the exposure region 116 covers the trace U13 of the shot region S15. Relative to the X direction. In practice, since the shooting regions S12-S15 and S18-S21 sufficiently satisfy the conditions for alternately inverting the scanning direction, the scanning direction for the shooting region S15 can be arbitrary, and follows the trajectory U13. Direction or the opposite direction. In addition, the scanning direction may be arbitrary with respect to the shot region 21, and may be a direction along the trace U17 or a direction opposite thereto.

Moreover, even if exposure is started from the shooting area S10 in which the exposure area 16 moves in the X direction (wafer 15 is scanned in the -X direction) in FIG. 10, the entire shooting area in FIG. Since the number of is even, the exposure can be started from the shooting area in which the exposure area 16 moves in the -X direction (e.g., S11), and the order of exposure is arbitrarily provided to satisfy the condition for alternately inverting the scanning direction. do.

However, as in the step 83 of FIG. 6, as shown in the embodiment of FIG. 10, one group of shot regions S10, S1-S4, S5 and another group of shot regions are formed at the outer periphery of the wafer 15. FIG. If there are two or more shot regions in the scanning direction between (S23, S32-S29, S28), the scanning can be reversed from the scanning direction of the outer peripheral portion to the inside of the shot region at the outer peripheral portion by applying step 83. . However, if there are three rows or ranks of the shoot regions arranged in the scanning direction on the wafer 15, step 83 cannot be applied.

11 shows a wafer in which three shoot regions are arranged in the scanning direction (X direction). In the figure, twelve shoot regions S1-S12 are arranged on the wafer 15A with three rows in the X direction. In this case, the scanning direction for one group of the shoot regions S1-S4 and the other of the shoot regions S9-S12 at the outer periphery is such that the exposure area 16 moves from the inside of the wafer 15A to the outside. Direction is set. Thereafter, assuming that the scanning direction for the internal shot regions S5-S8 is opposite to the scanning direction for the shooting regions S1-S4, the scanning direction for the internal shot regions S5-S8 is the shot region S9-. It becomes the same as the scanning direction with respect to S12). In addition, the number of shot regions in which the scanning direction for the exposure area is in the X direction and the number of shot regions in which the scanning direction for the exposure area is in the -X direction by two are different, which requires a reticle return and consequently the productivity Is lowered.

Next, in the case as shown in Fig. 11, since the number of shooting areas is 12 (even), the number of shooting areas having opposite scanning directions is the same, and the number of shooting areas having opposite scanning directions is the same. The exposure order is set to satisfy the conditions to be exposed alternately. As a result, the wafer 15A is first exposed so that the exposure area 16 is moved in the X direction with respect to the shoot area S1 as indicated by the trajectory V2. In this movement, the focus position and the inclination of the wafer 15A are shifted. Is corrected based on the focus position measured in the lead-ahead area 35. The exposure is then performed by scanning the exposure area 16 with respect to the shot area S9 as indicated by the trajectory V2. Is executed in the order of the trajectories V3, V4, .... V11, V12 by alternately inverting the scanning direction, therefore, the exposure is performed in the interior shot area S5-S8 in such a manner that the scanning directions are alternately inverted. ) Is executed.

If there is a shoot region in one row on the wafer, autofocusing and autoleveling are possible without being likely to be followed up, but can be sufficiently scanned in any direction without departing from a control error.

The present invention is not limited by the above embodiments, but various arrangements can be made without departing from the scope of the present invention.

12 shows a projection exposure apparatus used in the second embodiment of the present invention. In the figure, the pattern on the reticle 7 is emitted to the exposure light EL by a rectangular irradiation area (hereinafter referred to as "slit or slit-shaped irradiation area") from an irradiation optical system (not shown in the figure), The pattern image is projected and exposed on the wafer 15 via the optical projection system PL. In this case, if the Y axis is taken in the resin direction with respect to the sheet of FIG. 12, the + Y direction (or the -Y direction) at a constant speed V with respect to the slit or slit-shaped irradiation area of the exposure light EL. Is scanned at a constant speed V · β ( β is a magnification of the optical projection system 8) in the -Y direction (or + Y direction) in synchronization with the scanning of the reticle 7. The projection magnification is 1/4 or 1/5.

A drive system of the reticle 7 and the wafer 15 will be described. The reticle Y-axis drive end 109 is installed on the reticle support or base 110 that can move in the Y-axis direction. On the reticle Y-axis drive stage 109, a vacuum chuck or the like is provided with a fixed reticle drive stage 111. In the plane perpendicular to the optical axis of the optical projection system PL and parallel to the sheet shown in FIG. 12, the position of the reticle 7 is accurately and highly accurate in the X and Y directions and in the rotational direction (the direction of [theta]). To control. The movable mirror 112 is positioned on the reticle fine drive end 111, and the interferometer 114 located on the reticle support 110 is located in the X and Y directions in the directions of θ of the reticle fine drive end 111. Monitor position continuously. The positional information S1 obtained from the interferometer 114 is supplied to the main control system 113A.

In addition, a wafer Y-axis driving end 120A moving in the Y-axis direction is provided on the wafer support or base 121, and the wafer X-axis driving end 120B is in the X-axis direction in which the Z leveling end 119 is installed. It is installed to be movable. The wafer 15 is fixed on the Z leveling stage 119 by vacuum suction. The movable mirror 122 is fixed to the Z leveling end. The external interferometer 123 monitors the position of the Z leveling end in the X and Y directions and the θ direction. The position information obtained from the interferometer 123 is supplied to the main control system 113. The main control system 113 controls the position of the wafer Y-axis drive stage 120A, the wafer X-axis drive stage 120B, and the Z leveling stage 119 through the wafer drive apparatus 124, and controls the operation of the entire apparatus. To control.

In addition, the existing mark plate 118 is a wafer coordinate system defined by the coordinates measured by the interferometer 123 for the wafer and a reticle coordinate system defined by the coordinates measured by the interferometer 14 for the reticle. It is fixed near the wafer on the Z leveling stage 119 to have this correlation. Various reference marks are formed on the reference mark play 118. The reference mark is a backlight reference mark by irradiation light guided to the Z leveling end 119, which is a light emission reference mark.

The reticle alignment microscopes 101 and 102 are installed on the reticle 7 of this embodiment in order to simultaneously observe the reference mark on the reference mark plate 118 and the mark on the reticle 7. In this case the deflection mirrors 103, 104 are movably positioned to direct deflected light from the reticle 7 with respect to each of the reticle alignment microscopes 101, 102. Once the exposure sequence is started, the deflection mirrors 103 and 104 are discharged from the exposure light EL path by the mirror drive devices 105 and 106 under the command from the main control system 113.

The scanning exposure projection exposure apparatus of FIG. 12 is mounted to a tilted incident multipoint focus position detection system. The multipoint focus position detection apparatus of this embodiment is described in US Patent Application No. 172,098 filed December 23, 1994 by the assignee of the present invention, which is in front of the exposure area and the exposure area, respectively, with respect to the scanning direction. It is a lead-ahead type which detects the focus position (position in the optical axial direction of the optical projection system PL) of the wafer.

13 shows an optical system of the multipoint focus position detection system of this embodiment. In the figure, the pattern forming plane (reticle plane) on the reticle 7 is combined with the exposure plane of the wafer 15 to the optical projection system PL, but the reticle plane is not significantly changed. Thus, whether the exposure plane of the wafer 15 coincides with the imaging plate of the optical projection system PL within the depth of focus is detected only by the tilted incident multipoint position detection system, and the wafer 15 The focus position and the inclination of the exposure plane of are controlled in accordance with the detection result.

In the multipoint position detection system, irradiated light other than the exposure light EL which is not exposed to the photoresist on the wafer 15 is led from the irradiated light source through the optical fiber bundle 160. The irradiation light emitted from the optical fiber bundle 160 irradiates the pattern forming plate 162A through the condenser lens 161. Irradiation light passing through the pattern forming plate 162A is projected onto the exposure plane of the wafer 15 through the lens 163, the mirror 164, and the irradiation objective lens 165. The image pattern on the pattern forming plate 162A is projected onto the exposure plane of the inclined wafer 15 in relation to the optical axis AX1 of the optical projection system PL to be imaged. The irradiated light reflected on the wafer 15 is imaged with the condenser objective 166 and the rotational vibration plate 167 while the pattern image on the pattern forming plate 162A is reimaged on the light receiving plane of the light receiving device 169A. Through the lens 168 is projected again on the light receiving plane of the light receiving device 169A. In this case, the main control system 113 provides the oscillator 170 with the vibration described later about the rotational direction vibrating plate 167, and the detection signals from the plurality of light receiving members of the light receiving device 169A are transmitted to the signal processor. 171A). The signal processor 171A supplies the main control system 113 with a plurality of focus signals obtained by the synchronous detection of each detection signal having the drive signal from the oscillator 170.

FIG. 14B shows the pattern forming plate 162A of the FIG. 13 embodiment. As shown in the figure, the first row of the pattern forming plate 162A has nine slit-shaped opening patterns 172-11-172-19 and the second to fifth rows have nine slit-shaped opening patterns ( 172-21-172-59). That is, the pattern forming plate 162A is formed with a total of 45 slit-shaped opening patterns, and these slit-shaped opening pattern images are projected onto the exposure plane of the wafer 15 shown in FIG. do.

Fig. 14A shows the exposure plane of the wafer 15 under the optical projection system PL of this embodiment. In the figure, the reticle 7 pattern shown in FIG. 13 is exposed to a longer rectangular exposure field or region 116 in the X direction, and registers a circular irradiation field of the optical projection system PL. The wafer 15 is scanned in the Y direction with respect to the exposure field 116. The multi-point focus position detection system of the present embodiment includes the lead-ahead area 135A including nine measured (AF11-AF19) on the first row extending in the X direction from the upper area of the exposure field 116 in the Y direction. And a measurement including the lead head region 135B including the measurement points AF21-AF29 on the second row of the exposure field 116, and the measurement points AF31-AF9 on the third row of the exposure field 116. The lead-ahead region 135D including the region 135C, nine measuring points AF41-AF49 on the fourth row in the lower region of the exposure field 116 in the Y direction, and the measuring point AF51 on the fifth row; Project an image of a slit-shaped opening pattern on the lead-ahead area 135E including -AF59.

14 (c) shows the light receiving device 169A of the multipoint focus position detection system of this embodiment. Nine light receiving members 175-11-175-19 are arranged on the first row of the light receiving apparatus 169A, and the light receiving members 175-21-175-59 have nine members for each row. Arranged in the second to fifth rows. That is, a total of 45 light receiving members are arranged on the light receiving device 169A, and a slit-shaped aperture (not shown) is arranged in each light receiving member. The slit-shaped opening pattern images projected on the measurement points AF11-AF59 shown in FIG. 14 (a) are reimaged on the light receiving members 175-11-175-59, respectively. The detection signal from each light receiving member 175-11-175-59 is supplied to the signal processor 171A. The position of each reimaged image vibrates on the light receiving device 169A in the aperture width RD by vibrating the light reflected on the exposure plane of the wafer 15 by the rotating vibration plate 167 in FIG.

In addition, the image of the slit device of each of the measuring points AF11-AF59 shown in FIG. 14 (a) is projected at an angle with respect to the optical axis of the optical projection system PL. Therefore, when the focus position of the exposure plane of the wafer 15 changes, the reimage position of the image projected on the light receiving device 169A changes in the RD direction. Therefore, in the signal processor 171A, each of the 45 focus signals corresponding to the focus positions of the measuring points AF11-AF59 is a light receiving member 175-11-175-59 having a vibration signal from the rotational vibration plate 167 direction. Can be obtained by synchronously detecting the detection signals from the respective devices. The inclination (leveling angle) and the average focus position of the wafer are calculated from the predetermined focus signals of these 45 focus signals in the manner described below. The measured leveling angle and focus position are supplied to the main control system 113 shown in FIG. The main control system 113 sets the leveling angle and the focus position of the wafer 15 according to the leveling angle and the focus position supplied through the drive unit 124 and the Z leveling stage 119.

Therefore, in this embodiment, it is possible to measure the focus positions for all of the 45 measurement points AF11-AF59 shown in FIG. 14 (a). However, in the present embodiment, as shown in Figs. 15 (a) and (b), the position of the point where the focus position is actually measured among the 45 measurement points (hereinafter referred to as "sampling point") is in the scanning direction of the wafer. Change accordingly. In addition, the present embodiment further includes a lead head mode in which only the focus positions are read in advance in the lead head regions 135A or 135B (or 135E or 135D), and the exposure field 116 is further added in these lead head regions. It includes an additional lead-ahead mode for measuring the focus position in the measurement area 135C at. First, in the simple read head mode, the measurement is performed by using all the measurement points AF41-AF49 in the lead head area 135D as the sample position when the wafer is scanned in the -Y direction. In contrast, when the wafer is scanned in the + Y direction, the measurement point in the lead-ahead area 135B or 135A becomes the sample point. Since the exposure field 116 is wide in the Y direction, the lead head area 135D or 135B is too close to the exposure field 116 or the lead head area in relation to the scanning direction if the scanning speed of the wafer is high. The measuring point in lead ahead area 135E or 135A before 135D or 135B is sometimes used as a sample point. This is true for the following additional read ahead modes.

Next, as shown in FIG. 15 (a), when the wafer is scanned in the + Y direction with respect to the exposure field 116 in the additional lead-ahead mode, an odd number of odd portions in the lead-ahead area 135B is observed. The even measuring points AF32, AF34,... AF38 in the measuring points AF21, AF23, ... AF29 and the measuring area 135C in the exposure field 116 become sample points. In contrast, as shown in FIG. 15 (b), when the wafer is scanned in the -Y direction with respect to the exposure area 116, the odd number of measuring points AF41, AF43, ... in the lead-ahead area 135D ... AF49) and even measuring points AF32, AF34,... AF38 in the measuring area 135C in the exposure area 116 become sample points.

Further, in both the lead head area or the additional lead head area, the measurement result of the focus position during the scanning exposure continues to change in accordance with the conversion coordinate of the stage in the wafer. The measurement result of the focus position is stored in the memory of the main control system 113 of FIG. 12 as a two-dimensional map composed of the coordinates of the scanning direction (Y direction) of the stage and the coordinates of the measurement point in the non-scanning direction (X direction). The focus position and leveling angle of the wafer during exposure are calculated using the stored measurement results. The focus position and the leveling angle of the wafer are then correctly installed by driving the Z leveling stage 119 shown in FIG. 12, and the operation of the Z leveling stage 119 is controlled according to the open circuit control in accordance with the measurement result.

In this case, in the read head mode, the exposure of the exposure field 116 is executed according to the previously measured result. That is, on the one hand, when the wafer is scanned in the + Y direction as shown in FIG. 16 (a), the focus position for the region 126 on the wafer is determined by the predetermined distance of the lead head region 135B in the second row. It is measured at the sample point. Thereafter, when the area 126 on the wafer enters the exposure field 126 as shown in FIG. 16 (b), the leveling and focusing for the area 116 on the wafer is determined according to the measurement result in FIG. 16 (a). Controlled. On the other hand, samples of the measurement area 135C in the exposure field 116 when the focusing and leveling in the additional lead-ahead mode are controlled with respect to the area 126 as shown in Figs. 15 (a) and (b). The data of the point, that is, the focus position measured in the area 126 during exposure, are used together. In this additional read-ahead mode, the measurement data in the measurement area 135C in the exposure field 116 is used to correct the follow up error (difference between the exposure plane of the wafer and the imaging plane of the optical projection system).

17 shows the Z leveling stage 119 of this embodiment and its control system. In this figure, the upper member of the Z leveling end 119 is supported on the lower member through three support points 141A-141C which may extend in the focusing direction. The adjustment of each support point 141A-141C is performed by adjusting the focus position of the exposure plane of the wafer 15 on the Z leveling end 119, the inclination in the scanning direction ( θ _v ) and the inclination in the nonscanning direction ( θ _x ). Allows you to set the target value. Near each support point 141A-141C, there is provided a respective height sensor 142A-142C capable of measuring the displacement of each support point with a resolution of, for example, 0.01 μm in the focusing direction. It is possible to provide a highly accurate mechanism with a longer stroke as the positioning mechanism in the focusing direction (Z direction).

In order to control the leveling of the Z leveling stage 119, the main control system 113 supplies the inclination θ _x to the filters 143A and 143B in the non-scanning direction in which it is installed, and the scanning direction in which each moment is changed. The slope θ _y is supplied. The filters 142A, 142B supply the gradient obtained by filtering with different filter characteristics to the arithmetic unit 144, and the main control system 113 calculates the coordinates W (X, Y) of the area exposed on the wafer 15. Supply to unit (144). In addition, the main control system 113 supplies the arithmetic unit 144 with information on the focus position of the exposure plane of the wafer to be installed. The arithmetic unit 144 supplies the driving units 145A-145C with information on the displacement to be set based on the coordinates W (X, Y), the focus position, and the two inclinations. The height sensors 142A-142C are the flow heights of the support points 141A-141C on each of the drive units 145A-145C which respectively set the height of the support points 141A-141C at the height to be set by the arithmetic unit 144. Supply information.

Therefore, the inclination of the exposure plane of the wafer 15 in the scanning direction and the nonscanning direction is set to the target value, respectively.

In addition, when the positions where the support points 141A, 141B and 141C are located are called drive points TL1 and TL2, the drive points TL1 and TL2 are then located on a straight line parallel to the Y axis, and the drive point ( TL3) is located on the vertical bisector between the driving points TL1 and TL2. Furthermore, if the slit-shaped exposure area 116 exposed by the optical projection system is located in the shooting area SAij of the wafer 15, the focus position of the shooting area SAij in this embodiment is the leveling of the wafer 15. It does not change when controlled through these support points 141A-141C. Thus, leveling and focusing are controlled separately. In addition, the focus position of the exposure plane of the wafer 15 is set by placing three support points 141A-141C by the same amount.

Leveling and focusing in this embodiment are described in more detail below. First, the method of calculating the inclination of the focus and the inclination for the leveling will be described.

(A) Calculation of the slope

As shown in Figs. 16 (a) and (b), in the row of each measurement point, the X coordinate of the m th sample point in the nonscanning direction is represented by X _m , and the Y coordinate of the n th sample point in the scanning direction. Is denoted by Y _n , and the value of the focus position measured at the sample point having X _m of X coordinate and Y _n of Y coordinate is expressed as AF (X _m , Y _n ). Moreover, the following calculation is performed assuming that the number of samples in the non-scanning direction is M and the number of samples in the scanning direction is N. In the calculation, the sum Σ m indicates the sum of 1-M for the script m:

At the same time, the following calculation is executed assuming that the sum Σ _n indicates the sum of 1-N for the script n.

The following calculation is then performed using equations (10, 11) above.

Afterwards, the slope (AL (Y _n )) of the nonscanning direction (X direction) at the _nth sampling point appears from each An by the least-squares approximation, and the slope of the scanning direction (Y direction) at the mth sampling point (AL). (X _m )) from each Am by the least-squares approximation. Thereafter, the inclination θ _x in the nonscanning direction and the inclination θ _y in the scanning direction can be known by averaging as follows.

(B) Calculation of the focus position

The calculation of the focus position includes an averaging method and a maximum and minimum detection method. In this embodiment, the focus position is detected by the maximum and minimum detection methods. Basically, the averaging method uses the above-described focus position AF (X _m , Y _n ) to calculate the focus position with respect to the exposure plane of the wafer 15 as a whole from the following equation.

Then, in the maximum and minimum value detection method, assuming that the functions representing the maximum and minimum values are Max () and Min (), the focus position AF 'with respect to the exposure plane of the wafer 15 as a whole is as follows.

Next, as shown in FIG. 16 (b), when the measurement area 126 reaches the exposure field 116, the three support points 141A-141C in FIG. 17 are formulas (14, 15, 17). Is driven in an open circuit using the measurement results of the height sensors 142A-142C as respective references based on the detection results θ _x , θ _y , AF ′. In particular, autofocusing control is executed by simultaneously driving three support points 141A-141C. Auto-leveling control is executed so that the focus position of the exposure field 116 shown in FIG. 17 does not change.

In other words, 17 also the distance between the exposed areas the distance between the center and the X direction as a supporting point (141A, 141B) of 116 as a center and the X direction of the X _1, the exposure area 116, the supporting point (141C) X in ₂ , assuming that the distance between the center of the exposure area 116 and the support point 141A in the Y direction is Y ₁ , and the distance between the center of the exposure area 116 and the support point 141B in the Y direction is Y ₂ , the support point 141A, 141B, and 141C represent displacements of the ratio X ₁ : X ₂ in opposite directions, respectively, as a result of the inclination θ _x in the non-scanning direction, and the support points 141A, 141B represent the inclination θ in the scanning direction. _y ) shows the displacement of the ratio Y ₁ : Y _{2 in} the opposite direction, respectively.

An example of the exposure operation of this embodiment is described with reference to FIGS. 18-20 as follows. First, FIG. 18 shows the shot array on the wafer 15 in this embodiment. In this figure, the shoot regions SA1, SA2, ..., SA20 are arranged on the wafer at a predetermined pitch in the X and Y directions. When the first layer is exposed, the shoot regions SA1-SA20 are imaged, and when the second layer is exposed, the shoot regions SA1-SA20 are already formed with the same or different circuit patterns.

First, when the first shot area SA1 is exposed, the exposure area or the exposure field 116 is scanned in the Y direction, and the reticle 7 is scanned in the -Y direction in synchronization with it. Thus, the exposure field 116 moves along the trajectory 146A for the wafer 15. In this case, in the additional lead-ahead mode, the focus position is measured at the predetermined measurement position of the lead head region 135D and the predetermined measurement position at the measurement region 135C in the exposure field 116 and the leveling of the wafer 15. And focusing are performed according to the measurement result. However, since the shoot area SA1 is at the periphery of the wafer, if data of the actually measured focus position is used, the focus position and the leveling angle are too many to be corrected, and the following accuracy is lowered. Thus, if the accurately measured focus position data exceeds a predetermined allowable value, the data is arranged to be ignored as follows.

That is, FIG. 19 (a) shows the following state after scanning of the wafer in the Y direction so that exposure starts in the shot region SA1. In the drawing, the exposure field 116 and the lead-ahead area 135D are located in a thin area of the outer periphery of the wafer (the outer periphery area in which the thickness of the photoresist layer applied to the wafer 15 changes significantly). Moreover, when the image reference plane is previously coincident with the image plane of the optical projection system PL at the measurement point in the multipoint focus position detection system, that is, the exposure plane (eg, surface) of the wafer 15 is the optical projection system ( Assume that the calibration is performed to zero the detected focus position when coinciding with the image plane of PL). In this case, the focus position ΔZ measured in the lead head region 135D represents a difference of the image plane 139 from the focus position, and the focus position ΔZ is a large value to be considered. Thus, this embodiment already limits the allowable value [Delta] max and ignores the focus position [Delta] Z when the absolute value of the measured focus position [Delta] z exceeds the allowable value [Delta] max. In particular, the focus position is measured at the center of the shoot area SA1 before the exposure is performed in the shoot area SA1. If the absolute value of the measured focus position ΔZ exceeds the allowable value Δmax in the state shown in FIG. 19 (a), the focus position and leveling angle of the Z leveling stage 119 are in FIG. 19 (a). It is controlled according to the data of the already measured focus position of the state shown.

Thereafter, as shown in FIG. 19 (b), the absolute value of the focus position measured when the lead-ahead area 135D reaches the shoot area SA1 by further scanning the wafer 15 in the + Y direction. Since is equal to or less than the allowable value Δmax, the focus position and the leveling angle of the Z leveling stage 119 are controlled according to the actually measured focus position. Therefore, the focus position and the leveling angle (tilt) of the wafer 15 do not change particularly in the process of changing from the state of FIG. 19 (a) to the state of FIG. 19 (b), so that the exposure plane is in the state of FIG. 19 (b). Is correctly assigned to the image plane when starting exposure in the shoot area SA1. Therefore, the reticle pattern is exposed in the whole shot area SA1 with good following accuracy of autofocusing and autoleveling without slowing down the scanning speed of the wafer 15.

Then, when the shot region SA2 adjacent to the shot region SA1 in FIG. 18 is exposed after completion of the exposure of the shot region SA1, the wafer 15 passes through the wafer X-axis driving stage 120B of FIG. Stepped in the X direction. This causes the exposure field 116 to move in the + X direction with respect to the wafer 15 and to reach the scanning start point for the shot area SA2. From this state, the wafer 15 is scanned in the -Y direction with respect to the exposure field 116, and the reticle 7 (Fig. 12) is scanned in the + Y direction which is synchronous with it. As a result, the exposure field 116 moves along the trajectory 146B for the wafer 15. In this case, in the additional read-ahead mode, the focus position is measured at the predetermined measuring point of the lead-ahead area 135B and the measuring area 135C in the exposure field 116, and the focusing and leveling of the wafer 15 is measured. Is executed according to. However, since there is a shoot area SA2 on the outer peripheral area of the wafer 15, if the actual measured focus position is used, the correction for the focus position and the leveling angle is exceeded, and the following accuracy may be degraded. Therefore, the measurement is performed as follows.

FIG. 20 (a) shows a state in which the shot area SA2 is exposed. In this figure, autofocusing and autoleveling are performed in accordance with the measurement result because the measurement result of the focus position in the lead head region 135B flows out of the image plane within the allowable value.

However, if the wafer 15 is scanned further in the -Y direction, as shown in FIG. 20 (b), the lead head region 135B moves from the outer periphery of the wafer 15 to the thin region, and the lead The absolute value of the focus position [Delta] Z (i.e. outflow from the image plane) measured in the head region 135B exceeds the allowable value [Delta] max as described above. Thereafter, if the absolute value of the focus position ΔZ measured in the lead head region 135B exceeds the allowable value Δmax, the exposure is at the Z leveling stage at the set value until the data of the focus position ΔZ can be ignored. It is executed by maintaining the leveling angle and the focus position of 119.

However, the absolute value of the focus position ΔZ measured in the lead-ahead region 135B in the state of FIG. 20 (b) is the allowable value Δmax in the outer peripheral region of the wafer 15 in which the circuit pattern does not exceed. There is a delay of time until the data which is read in advance and actually used is actually used, and even if the data of the absolute value exceeding the allowable value Δmax is ignored, the exposure of the shot area SA2 is hardly affected. However, ignoring the absolute data exceeding the allowable value Δmax eliminates significant changes in the leveling angle and focus position of the Z leveling stage 119 in the state shown in FIG. 20 (b). That is, it eliminates wasteful operation of the Z leveling stage and improves the stability of the control system.

In consideration of FIG. 18, scanning exposure is similarly performed for the shot area SA3-SA20 on the wafer 15. As shown in FIG. The relative scanning direction of the exposure field 116 for each of the shot areas SA3-SA20 is shown by an arrow. In this case, while autofocusing and autoleveling are performed with respect to the shoot area SA3-SA18 at the center by using the measurement data in the lead-ahead area, the measurement data in the pre-read shoot area is recorded in the shoot area in the outer peripheral area. SA19, SA20) can be partially ignored. This can effectively perform the exposure by the scanning exposure system with each shot region on the wafer 15 assigned to each image region plane and without unnecessary movement of the Z leveling stage 119.

18 illustrates a case where exposure is performed in the outer peripheral region on the shoot region with respect to the scanning direction on the wafer. FIG. 21 illustrates a case where exposure is performed in the outer peripheral region on the shoot region with respect to the non-scanning region (X direction) on the wafer.

21 shows the shoot area SA located adjacent to the edge of the wafer in the X direction. In FIG. 21, the exposure to the shooting area SA is performed by scanning the shooting area SA in the + Y direction (right side of the sheet) with respect to the exposure field 116. Thus, the additional read head mode includes the exposure field and the data about the focus position measured at the measurement points AF21, AF23, ... AF29 of the lead head area 135B before the exposure field 116 in the scanning direction. Data on the focus position measured at the measuring point of the measuring area in (116) are used.

In addition, assume that in FIG. 21, the end of each of the shot regions SA protrudes from the wafer 15 in the scanning direction. In this case, for example, if two identical circuit patterns 136A and 136B are formed in the shoot area SA, at least the circuit pattern 147A area is normally exposed so that the entire shoot area SA is not wasted. However, when the shot area SA is scanned in the + Y direction with respect to the exposure field 116, the absolute value of the focus position at the measuring point AF21 at the end of the lead-ahead area 135B is the initial position 137A. Exceeds the allowable value? Max, in the intermediate position is within the allowable value? Max, and again at the position 137C near the scanning end, again exceeds the allowable value? Max. That is, the absolute value of the focus position at the measuring point AF21 exceeds the allowable value Δmax at the portions ΔY1 and ΔY3 of the respective ends of the scanning direction Y, and within the allowable value Δmax at the central portion ΔY2. have.

In this case thereafter, the measurement data of the measuring point AF21 is ignored when the measurement data of the portions ΔY1 and ΔY3 are used, and the measurement data of the measuring point AF21 is used when the measuring data of the portion ΔY2 is used. This prevents the inclination and the focus position of the shoot area SA from being set incorrectly by the measurement data for the area at the outer periphery of the wafer 15 which is not suitable for exposure, and the pattern of the reticle has a high resolution and at least It is exposed in the circuit pattern 136A.

In this embodiment, a multipoint focus position detection system for projecting a slit-shaped opening pattern arranged in two dimensions on the wafer is used to measure the focus position at the multipoint of the exposure plane of the wafer. Instead, a focus position detection system may be used that projects a pattern image that forms a slit extending in the scanning direction on a lead-ahead area in one row on the wafer and measures the entire focus position in the non-scanning direction. In addition, a two-dimensional distribution of the focus position on the exposure plane of the wafer can be measured by using an image processing focus position detection system that projects a shading pattern and detects lateral drift of the pattern image. Even when focusing and leveling are performed accurately by using an additional lead-ahead system similar to the above embodiment.

A fourth embodiment of the present invention will be described with reference to FIGS. 22 to 26. In the present embodiment, the present invention uses the exposure method of the step-and-scanning projection exposure apparatus.

22 shows a projection exposure apparatus using this embodiment. In FIG. 22 an optically illuminated light-shaping system for the reticle 7 with respect to the relay lens 206 using the rectangular slit-shaped irradiation area 221 and the uniform intensity illumination by the light source 201. Irradiated from an optical irradiation system consisting of 202. In the slit-shaped irradiation area 221, the circuit pattern image of the reticle 7 is transferred onto the wafer 15 through the optical projection system PL. An excimer laser light source such as an ArF excimer laser light source or a KrF excimer laser or metal vapor laser light source or a pulse laser light source for a harmonic YAG laser generator or a structure combining a mercury lamp and an oval reflector is used as the light source 201.

In the case of a pulsed light source, the exposure of light is switched on and off by controlling the power supply from the power supply unit for the pulsed light source. In the case of a continuous light source, the exposure from the light is switched on and off by the shutter of the optically irradiated light forming system 202. However, since the movable blind (variable field stop) 252 is provided in the embodiments described below, the exposure of light can be switched on and off by switching of the movable blind 252.

In FIG. 22, the irradiation light from the light source 201 reaches a fly-eye lens 203 having a diameter of the light beam set by the optical irradiation light forming system 202 to a predetermined size. The number of two-dimensional light sources is formed on the exit surface of the fly's eye lens 203. The irradiation light of this two-dimensional light source is collected by the condenser lens 204 and reaches the movable blind (variable field aperture) 252. The light beam passing through the opening of the movable blind 252 becomes a light beam having a square slit cross section and enters the relay lens system 206. The relay lens system 206 is a lens system adjacent to the patterned surface of the movable blind 252 and the reticle 7. The movable blind 252 has two blades (light blocking panels) 252a and 252b for adjusting the width of the scanning direction (X direction) described later and two blades for adjusting the width in the non-scanning direction perpendicular to the scanning direction. (Not shown). The blades 252a and 252b, which limit the width in the scanning direction, are respectively supported by the driving portions 251a and 251b to be movable in the scanning direction individually. The two blades, which limit the width in the nonscanning direction, are supported to be movable individually. The relay lens system 206 is a dual telecentric optical system and the telecentric characteristic is maintained in the slit irradiation area 221 on the reticle 7.

The reticle 7 according to this embodiment is integrated and fixed to the reticle end 209. In the slit-shaped irradiation area 221 on the reticle 7, the circuit pattern image is exposed and projected onto the wafer 15 via the optical projection system PL. Assume that the region on the wafer 15 that is coupled to the slit type irradiation area 221 is the slit type exposure area (exposure field) 216. Further, the scanning direction of the reticle 7 with respect to the irradiation area 221 in the two-dimensional plane perpendicular to the optical axis AX of the optical projection system PL (direction parallel to the ground in Fig. 22) is + X direction ( Or -X direction), and a direction parallel to the optical axis AX is a Z direction. Further, in FIG. 1, the nonscanning direction perpendicular to the ground and perpendicular to the X axis is the Y direction.

In this case, the reticle stage 209 is driven by the reticle stage driver 214 and the reticle 7 is scanned in the scanning direction (X direction or -X direction). The operation of the driving units 251a and 251b of the movable blind 252 and the operation of the driving unit in the non-scanning direction are controlled by the movable blind control unit 253. The operation of the reticle stage driver and the movable blind control unit 253 is controlled by the main control system 213 which controls the operation of the entire apparatus. On the one hand, the wafer 15 is integrated and fixed to the Z leveling end 217. Z leveling stage 217 is located on XY stage 220 through three support points 219A-219C that are flexible in the Z direction. The position of the Z leveling end 217 is adjusted by adjusting the amount of flexibility at three parallel points 219A-219C. Adjustment of the inclination (leveling) of the Z leveling stage 217 may be made by adjusting the amount of flexibility at the three support points 219A-219C, respectively. The XY stage 220 positions the wafer 15 in the X and Y directions on a surface perpendicular to the optical axis AX, and scans the wafer 15 at a predetermined scanning speed in the + X direction (or -X direction). do.

Furthermore, a multipoint focus position detection system (hereinafter referred to as a multipoint AF sensor) consisting of the light emitting system 225 and the light receiving system 231 is arranged on the side of the optical projection system PL. In a multi AF sensor, for example, the slit image is projected diagonally onto the optical axis AX from the light emission system 225 on a number of scheduled measurement points on the wafer. Light reflected from the plurality of slit images enters the light receiving system 231, which in turn is imaged in the light receiving system 231. The detection light emitted from the light emission system 225 is light of a wavelength band having weak photosensitivity to the photoresist on the wafer 15. When the position of the surface of the wafer 15 changes in the Z direction, the position of the slit image again imaged in the light receiving system 231 is laterally disposed. Accordingly, the light receiving system 231 creates a plurality of focus signals corresponding to the lateral arrangement amount, and supplies them to the signal processing system 233. The signal processing system 233 calculates the position (focus position) of each measurement point in the Z direction corresponding to the supplied focus signal, and describes the focus position calculated by the main control system 213. In this case, when the surface 15 of the wafer coincides with the image position in advance by the optical projection system PL, a plurality of focus signals output by the light receiving system 231 are corrected to become "0".

The main control system 213 performs stepping and scanning on the wafer 15 by controlling the operation of the XY stage 220 through the wafer stage driver 214. The main control system 213 also displays the amount of flexibility at the three support points 219A-219C through the wafer stage drive unit 214 simultaneously and in parallel along based on the information at the focus position supplied by the signal processing system 233. By controlling a, the degree of inclination with respect to the image surface coincides with the Z coordinate in the exposure area 216 of the wafer 15 fixed to the Z leveling end 217. Thus, autofocusing and autoleveling are executed in this manner.

As shown in FIG. 23, the reticle 7 exposes the pattern image on the reticle 7 to the shot region 234 on the wafer 15 through the optical exposure system PL by the scanning exposure method. Scanned at the speed of V ₁ in the -X direction (or + X direction) to the irradiation area 221 at the width D set by the movable blind 252 of FIG. Further, the projection magnification of the optical projection system PL is β (eg β is 1/4, or 1/5). Wafer 15 are scanned at the speed of the reticle to the exposure area 216, in synchronization with the scanning of the (7) + X direction _{V 2 (= β · V1)} . In this way, the circuit pattern image of the reticle 7 is continuously switched to the shoot area 234. In this embodiment, the positions of the edges 221a and 221b in the scanning direction of the irradiation area 221 can be moved in the scanning direction by operating the blades 252a and 252b of FIG. Exposure of an unnecessary pattern near the light blocking area ST on the reticle 7 can be prevented. In this case, in this embodiment, the slit image is projected at a plurality of measuring points in the two rows of linear lead-ahead areas 235B and 235C arranged so that the linear measuring area (at the center of the exposure area on the wafer 15) 235A and the exposure area 216 overlap in the X direction from the light emission system 225 of the multi-AF sensor of FIG. 22.

FIG. 24 shows the relationship between the slit-shaped exposure area 216 on the wafer and the measurement point by the multi AF sensor. In FIG. 24, the slit image is projected at each of the seven measurement points P _A1 to P _A7 in the measurement area 235A that crosses the center position of the exposure area 216 in the Y direction. Further, the slit image is projected at seven measurement points P _B1 to P _B7 of the lead head region 235B in the -X direction in the exposure area 216. Similarly, the slit image is projected at seven measurement points P _c1 to P _c7 of the lead head region 235C in the + X direction in the exposure area 216. In this embodiment, when the wafer is scanned in the + X direction indicated by the arrow A, autofocusing and autoleveling are performed in the -A direction in the measurement area 235A and the exposure area 216 in the -ahead area 235B. The whole is performed by using information on the focus position at 14 measurement points.

In particular, if the focus position obtained when the center point Q of the wafer is in the lead-ahead area 235B at the center point t _Q is as low as δ Z as compared to the focus position of the imaging plane, Z leveling in the Z direction in FIG. 22. The position of the stage 217 is raised by δ Z from the position at the point tQ when the point Q reaches the center of the exposure area 216. In addition, since the information on the focus position around the point Q is obtained, leveling can be performed according to the information. Moreover, when the point Q reaches the center of the exposure area 216, if the focus position obtained in the measurement area 235A is lower than, for example, the value of the imaging plane, the Z leveling end 217 is in the Z direction. It should only move about the part. That is, on the one hand, it is used to correct the follow-up error remaining when the information on the focus position obtained in the measurement area 235A in the exposure area 216 is the Z coordinate, and the inclination is used in the lead head area 235B. It is controlled according to the obtained focus position.

On the other hand, when the wafer is scanned in the -X direction indicated by the arrow B, autofocusing and autoleveling are entirely in the lead-ahead area 235C in the + X direction with respect to the measurement area 235A and the exposure area 216. This is done by using the information of the focus position at 14 measurement points.

Here, as shown in FIG. 24, the width of the scanning direction (X direction) of the exposure area 216 is Ls, from the edge of the exposure area 216 in the -X direction in the lead-ahead area 235B. Assume that the spacing between the edges of the exposure area 216 in the + A direction in the lead-ahead area 235C is Lp.

Moreover, FIG. 25 shows a state in which the wafer is set at the scanning start position for the exposure area 216 to perform exposure of the wafer on the shoot area 234. Assume that the scanning direction of the wafer in FIG. 25 is the + X direction. The shoot area 234 is separated from the exposure area 216 in the -X direction by the width Lac of the accelerator 239. The width Lac of the acceleration unit 239 in the scanning direction is determined by the set time for accelerating the reticle stage 209 and setting the speed of FIG. This is because the magnification β of the optical projection system PL is generally a shrinkage magnification, and the reticle end 209 is scanned at a higher speed than the XY end 220 on the wafer.

Next, as shown in FIG. 23, as an example, see FIG. 26 for an example of the procedure in the case where the exposure is continuously performed on the shoot regions 234 and SB arranged adjacent to each other on the wafer in the non-scanning direction. Explain. Fig. 26 (a) shows the moving speed V _R of the reticle end 209 in the scanning direction (-X direction) in this case. FIG. 26B shows the moving speed Vwx of the XY stage 220 on the wafer in the scanning direction (+ X direction). FIG. 26C shows the movement speed Vwy of the XY stage 220 in the non-scanning direction (-Y direction).

First, as shown in FIG. 25, assume that the shoot area 234 is set at the scanning starting point at the point Po. The reticle end 209 and the XY end 220 on the wafer start synchronously in the scanning direction from the point Po. The acceleration ratio between the reticle stage 209 and the XY stage 220 is equal to the projection magnification β of the optical projection system PL. The scanning exposure starts at the point P ₁ at which the inner end reaches the predetermined scanning speed and passes the set time TSE for a stable speed. Assume that the time required for this is T ₁ , that is, the distance the XY stage 220 moves during the time T ₁ corresponding to the portion R ₁ saddled in FIG. 26 (b) is Lac. This distance Lac is the same as for the acceleration part 239 of FIG.

At this time, the reticle stage 209 and the XY stage 220 need not always be synchronized in the acceleration time in time, since only the time of the exposure start time (point P ₁ ) should be synchronized. That is, the acceleration ratio of the reticle stage 209 and the XY stage 220 in the X direction does not always need to coincide with the projection magnification β of the optical projection system.

The scanning exposure ends at the point P _{2 where} only the scanning exposure time for T ₂ . As shown in FIG. 26 (c), the XY stage 220 starts stepping in the non-scanning direction at the same time when the scanning exposure is completed. The speed of the reticle stage 209 is reduced from time T ₄ required for stepping above to time T ₃ . The XY stage 220 on the wafer in the scanning direction reduces the speed and moves to the next scanning start position of the shoot area SB in parallel with the above-described operation.

As shown in FIG. 26 (c), the stepping operation in the non-scanning direction ends at the point P ₅ when the time T ₄ passes. The shot region SB and the exposure region 216 of FIG. 23 overlap in the Y direction. However, the 26 (a), 26 (b) as shown in Fig., The acceleration of the reticle stage 209 and the XY stage 220 on the wafer in the scanning direction of points (P ₅₎ before the point (P ₄₎ in Can begin with Subsequently, the scanning exposure on the shot region SB is performed at the point where the time T ₆ elapses while autofocusing (including auto-leveling) is performed using the focus position of the lead-ahead region 235C on the right side of FIG. P ₇ ). At this time, assume that the distance of the XY end 220 of the wafer going in the scanning direction, that is, the distance of the XY end 220 going in the scanning direction to the shadow portion R ₂ in FIG. 26 (b) is Lb.

In this case, as shown in FIG. 26 (c), positioning of the XY end on the wafer in the non-scanning direction is performed at the positioning time T ₅ from the point P ₅ to the point P ₆ . However, the time T ₅ need not be taken into account in particular. Since this non small compared to the width of the detection light beam to form a slit image projected from a scanning direction at each measuring point as the variation in the positioning time of the XY-stage 220 and the second 24 degree P _A1, respectively, such as P _A1 This is because the position of the measuring point does not change. However, it can be assumed that the distance traveled in the X direction by the XY stage 220 after the point P ₅ at which the positioning operation is completed in the non-scanning direction is also Lb.

In the above sequence, the width Lac of the accelerator 29 shown in FIG. 25 is shorter than the distance Lp between the exposure area 216 and the lead-ahead area 235B (Lac <Lp). In this case, the information on the focus position at the edge of the shoot area 234 cannot be obtained because the lead head area 235B is in the shoot area 234 at the scanning start time. Therefore, there is a possibility that an auto focusing follow up error occurs when the edge of the shoot area 234 enters the exposure area 216. In contrast, the lead head region 216 is present in the exposure region 216 instead of the shot region 234 at the scanning start position, so that the auto focusing is performed very accurately in the entire shot region 234. This can be expressed as:

The distance Lb at which the XY stage 220 advances from the stepping end point P ₅ to the exposure start point P ₇ in the non-scanning direction is the distance Lp between the exposure area 216 and the lead-ahead area 235C. (Lb <Lp). Each measurement point in the lead ahead area 235C runs obliquely at the edge of the shoot area SB because the wafer also moves in the nonscanning direction. In this case, since the focus position measured in the lead ahead region 235C is moved laterally in the Y direction. Accurate autofocusing is difficult to achieve by the measurement result. Thus, the condition for performing accurate autofocusing without being affected by the stepping operation in the nonscanning direction is that the distance Lb is longer than the distance Lp, as shown in the following equation:

The state of equation (19) satisfies this case.

In addition, in the embodiment, the moving speed of the XY stage 220 may be increased during the stepping operation of the XY stage 220 on the wafer, and the response speed of the autofocusing system may not keep up with the speed. Preferably, the operation of the autofocusing mechanism is turned off during the stepping operation. In particular, the scanning exposure in the sequence in FIG. 26 fixes the state of the Z leveling end 217 in FIG. 22, turns off the autofocusing control at the end point P ₂ of the scanning exposure, and the stepping end point P ₅ ) (or in some cases point P ₆ ) is performed in the opposite direction by turning on the autofocusing control. In this way, unstabilization of the autofocusing mechanism is avoided.

Further, the above embodiment shows a sequence in which the shot regions 234 and SB are aligned in series in the non-scanning direction on the wafer, but in the case where the shot regions are aligned in the scanning direction in series on the wafer, For example, all operations in FIG. 26 are the same except for the stepping operation of the XY stage 220 in the scanning direction on the wafer performed before the acceleration start point P ₄ .

Further, the above embodiment shows a sequence in which the stage for scanning starts acceleration in the non-scanning direction from the point P ₄ before the stepping end point P ₅ (or P ₆ ), but of course, non-scanning A sequence of starting acceleration for scanning exposure after the stepping ends in the direction can be used.

The present invention is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present invention.

Of course, the present invention is not limited to the above embodiments, and various changes can be made within the scope of the present invention without departing from the spirit of the present invention.

According to the first embodiment of the present invention, when the pattern of the mask (reticle) is projected onto the respective shoot area on the substrate (wafer) by the scanning exposure system, the shoot area on the outer periphery is scanned to expose the exposure area. Relatively move from inside to outside of the substrate. Thus, an excellent control accuracy can be obtained with respect to the focus position and tilt even for the shot region on the outer periphery of the wafer, which means that scanning can be applied to sagging on the periphery of the substrate even if the height is calibrated by the lead-ahead mode. This is because the influence is small.

When the total number of shot regions on the wafer is even, the number of shot regions in the opposite direction is made equal, and the scanning direction is reversed for each shot region while performing exposure, or when the total number of shot regions is odd. The difference between the number of shot regions in the opposite direction is 1, scanning is started from a direction having a plurality of shot regions, and the scanning direction for each shot region is reversed to no longer perform an idle return on the wafer. This eliminates idle returns in the mask and improves yield.

Further, when the scanning direction with respect to the inner shot region in contact with the shot region on the outer periphery of the substrate is inverted with respect to the shot region on the outer periphery, and the shot region and the inner shot region on the outer peripheral portion with respect to other shot regions, The scanning direction for the shoot regions arranged perpendicular to the scanning direction with respect to the scanning direction is selectively reversed, unnecessary movement (idle return) in the mask is eliminated, and stepping on the substrate is reduced, thereby improving yield.

Further, according to the second embodiment of the present invention, if the pattern of the mask is exposed to each shot area on the substrate by the scanning exposure system, the shot lost when the height detection area or the exposure area is not completely included in the exposure area. The exposure area is arranged to scan the area so that the exposure area moves from inside to outside with respect to the substrate. Therefore, an accurate accuracy of autofocusing and tilting can be obtained even for the shot region lost on the outer periphery of the wafer, which means that scanning is performed on the region lost at the outer periphery of the substrate even if the height is calibrated by the lead-ahead mode. Because it is less affected by.

According to the third embodiment of the present invention, since the height is pre-read toward the height of the substrate and controlled by using the pre-read result, when the exposure is performed by the scanning exposure system, the optical speed is not reduced. It is possible to precisely align the height of the actual exposure area on the photosensitive substrate, such as a wafer, with the height of the image plane of the projection system. In addition, since the height of the substrate is fixed to the height set by then, if the previously read height is outside the predetermined allowable range, the follow-up accuracy of autofocusing is not totally distorted even if the height of the substrate surface changes considerably.

Furthermore, by the above method, by fixing the height of the substrate to a set height until the previously read height deviates from the predetermined allowable range, it is also excellent when setting the scanning direction with respect to the exposure area in the peripheral area of the substrate in any scanning direction. Follow up accuracy can be obtained.

In addition, the follow-up accuracy of the leveling has the advantage that it does not degrade overall, which uses the pre-read height and the slope set when the pre-read height deviates beyond the allowable range predetermined for the image plane of the optical projection system. By controlling the inclination of the substrate by fixing the inclination of the substrate.

Further, before the exposure area is joined to the radiation area for the scanning direction, the substrate height is arranged in a direction crossing the scanning direction in the exposure area in parallel with the substrate height measurement in an array of measuring points arranged in the direction crossing the scanning direction. When measured in an array of measuring points, autofocusing and autoleveling can be performed accurately by using the height read in advance and the correct height. In this case, follow up of autofocusing and autoleveling as a whole by fixing the height and inclination of the substrate to a set value until the height measured in the array of measuring points arranged before the exposure area with respect to the scanning direction exceeds a predetermined acceptable range. Accuracy does not worsen overall.

In addition, when the substrate height is measured at a plurality of measuring points arranged in a direction intersecting the scanning direction before the exposure area for the scanning direction, and at a plurality of measuring points, a portion of the measuring height is allowed for the image plane of the optical projection system. Even if it exceeds the possible range, the height of the substrate is controlled by excluding height data outside the allowable range, for example when the exposure is performed to the exposure area of the peripheral area on the substrate in the non-scanning direction, and the exposure is controlled with the image plane. The exposure can be performed by aligning the flat areas in the exposure area.

According to the fourth embodiment, the position (focus position) of the optical projection system in the optical axis is detected in the lead-ahead area lying before the exposure area with respect to the scanning direction on the wafer, and autofocusing is performed as a result of the detection. Thus, autofocusing can be performed with a small follow up error even when the substrate is scanned. Further, the focus position at the edge of the shoot area is detected in the lead head area when the substrate moves to the accelerator, since the lead head area is between the predetermined shoot area and the exposure area when the substrate moves to the scanning start position. Thus, the advantage arises that autofocusing can be performed to have a small follow up error over the entire surface of the shoot area.

And the positioning (stepping) of the substrate in the direction perpendicular to the second direction (non-scanning direction) is completed after scanning of the substrate starts exposing the mask pattern to the substrate before the lead-ahead region reaches the predetermined shoot region. In this case, the lead head area is moved parallel to the scanning direction in the shot area. Thus, accurate autofocusing can be performed over the entire surface of the shoot area.

Also, if the positioning adjustment of the substrate in the optical axis direction is started after the positioning adjustment to the substrate in the optical axis direction is stopped immediately after the exposure is terminated on the predetermined shooting area, the second adjustment is performed for the exposure of the substrate on the next shooting area. Positioning in the direction to the next scanning start position of the substrate is completed, positioning of the substrate in the direction perpendicular to the second direction is terminated, and autofocusing is turned off while the substrate is moved at high speed by a stepping operation. Thus, the substrate does not perform unstable operation even if the focus position detected by the lead head region is moved at high speed.

1 is an apparatus diagram showing a scanning projection exposure apparatus according to a conventional application of the present invention.

2 is a plan view of the slit exposure and the lead-ahead area in the projection exposure apparatus shown in FIG.

3 is a conceptual diagram showing a lead head region used in correspondence with the scanning direction of the wafer 15.

4 is a cross-sectional view illustrating a "sagging" state of the outer periphery of the wafer.

FIG. 5 is a plan diagram showing a conventional exposure procedure for each shot region on a wafer by the projection exposure apparatus shown in FIG.

6 shows a relationship between FIGS. 6A and 6B.

6 (A) and 6 (B) are flowcharts showing the exposure sequence in the embodiment of the scanning exposure method according to the present invention.

7 (a) to 7 (c) are timing charts showing the moving speeds of the reticle stage (9) and the XY stage (20) in a wafer performing exposure in a scanning exposure type.

8 illustrates a calculation method for an acceleration distance LC.

9 (a) to 9 (c) are diagrams illustrating the combination of the shot array and the scanning direction in successive exposure of adjacent shot areas in the scanning exposure type.

10 is a plan view showing an example of an exposure procedure for each shot region on the wafer 15 of one embodiment.

FIG. 11 is a plan view showing an example of an exposure procedure for a long shot region on a wafer 15A of the embodiment. FIG.

12 is an apparatus diagram of a projection exposure apparatus to which one embodiment of the scanning exposure method according to the present invention is applied.

FIG. 13 is an apparatus diagram showing an optical system of the multipoint focus position detection system of the projection exposure apparatus shown in FIG.

FIG. 14 (a) is a plan view showing a slit-shaped opening pattern projected in a two-dimensional array into an area including an exposure field by an optical projection system in the above embodiment, and FIG. 14 (b) is a multipoint focus position detection system Fig. 14 (C) is a diagram showing an array of light receiving elements of the light receiving element device.

FIG. 15 (a) is a diagram showing sampling points in performing additional read heads in the above embodiment, and FIG. 15 (b) shows sampling points in scanning in the opposite direction and performing division and read heads. The figure which shows.

16 (a) and 16 (b) are diagrams illustrating exposure using the lead head focus position.

FIG. 17 is an apparatus diagram showing a mechanism for autofocusing and autoleveling and control thereof.

18 is a plan view showing an example of an array of shot regions on a wafer;

19 (a) and 19 (b) are enlarged cross sectional views of essential parts for explaining the operation of exposing the shot area SA1 with an additional lead-ahead system.

20 (a) and 20 (b) are enlarged cross sectional views of essential parts for explaining the operation of exposing the shot area SA2 with an additional lead-ahead system.

21 is a partially enlarged plan view of a wafer for explaining the exposure of the shot region near the edge in the non-scanning direction of the wafer;

22 is an optical projection apparatus for performing one embodiment of the projection exposure method according to the fourth embodiment of the present invention.

FIG. 23 is a schematic perspective view provided for explaining a scanning exposure operation by the projection exposure apparatus shown in FIG. 22;

Fig. 24 is a plan view showing the positional relationship between the measuring point by the multi AF sensor and the slit-shaped exposure area 216 of the embodiment.

25 is a plan view showing the positional relationship between the slit-shaped exposure area 216 and the shoot area 234 on the wafer at the start of scanning.

26 (a) to 26 (c) show the moving speeds of the XY stage on the wafer and the reticle when the exposure is sequentially performed by the scanning exposure type to the adjacent shot region in the non-scanning direction.

Explanation of symbols on the main parts of the drawings

7: reticle 11: removable mirror

15: wafer 17: wafer holder

135: lead-ahead area 139: imaging plane

Claims (20)

Simultaneously projecting the pattern image of the disk illuminated by the exposure beam onto the substrate via a projection system, and in synchronization with scanning the mask in a first direction substantially perpendicular to the optical axis of the projection system. A projection exposure method in which a shot region on the substrate is exposed by scanning in a second direction corresponding to the direction by using a pattern of the mask, A measurement area for measuring position information of the substrate in an optical axis direction of the projection system, which is disposed at a position away from the second direction with respect to an exposure area to which the exposure beam is irradiated through the projection system; Starting to scan the substrate when positioned between the shoot area and the exposure area, And the position of the substrate in the optical axis direction of the projection system based on the positional information of the substrate measured in the measurement area during scanning of the substrate. 2. The method according to claim 1, wherein after the exposure of the shot region is finished, the substrate is stepped in a predetermined direction to expose the shot region next to the shot region, Scanning the substrate in a direction opposite to the second direction to expose the next shot area, Before the next shot area reaches the measurement area for measuring the positional information of the substrate in the optical axis direction arranged on the opposite side of the measurement area with respect to the exposure area, the direction of the substrate in the predetermined direction is reached. Stepping is completed, the projection exposure method characterized by the above-mentioned. The method according to claim 1, wherein after the exposure of the shot region is finished, the substrate is stepped in a predetermined direction to expose the shot region next to the shot region, After the stepping of the substrate in the predetermined direction is completed, based on the positional information of the substrate in the optical axis direction measured in the measurement region disposed on the opposite side of the measurement region with respect to the exposure region, the projection system And adjusting the position of the substrate in the optical axis direction of the projection exposure method. The method of claim 1, wherein the start of scanning comprises accelerating the substrate in the second direction. The substrate according to claim 1, wherein the substrate is exposed in a direction orthogonal to the second direction so as to expose a next shot region on the substrate during the deceleration movement of the substrate in the second direction after completion of the exposure of the shot region. The projection exposure method characterized by moving the. The method according to claim 1, wherein after the exposure of the shot region is finished, the substrate is moved in a direction orthogonal to the second direction to expose a next shot region on the substrate, And before the movement in the direction orthogonal to the second direction of the substrate is terminated, the substrate is accelerated to move in the direction opposite to the second direction to expose the next shot region. . 2. The method according to claim 1, wherein after the exposure of the shot region is finished, stepping of the substrate in a predetermined direction is performed to expose a next shot region on the substrate, And after the end of the exposure of the shot region, the control of the autofocusing to adjust the position of the substrate based on the measurement of the positional information of the substrate in the optical axis direction and the measurement result. A device manufacturing method using the projection exposure method according to claim 1. Simultaneously projecting the pattern image of the mask illuminated by the exposure beam onto the substrate through a projection system, and in synchronization with scanning the mask in a first direction substantially perpendicular to the optical axis of the projection system. A projection exposure method in which a shot region on the substrate is exposed using a pattern of the mask by scanning in a second direction corresponding to the direction. A measurement area for measuring positional information of the substrate in the optical axis direction of the projection system is defined at a position away from the second direction with respect to the exposure area to which the exposure beam is irradiated through the projection system; , The distance that the substrate travels in the second direction from the time point when the stepping in the predetermined direction of the substrate made after the end of the exposure of the previous shot region of the shot region to the starting point of the exposure of the shot region begins, The projection exposure method characterized by being more than the space | interval of the said exposure area | region and the said measurement area | region in a 2nd direction. 10. The substrate according to claim 9, wherein the substrate is exposed in a direction orthogonal to the second direction for exposing the shot region during the deceleration movement of the substrate in a direction opposite to the second direction after the end of the exposure of the previous shot region. The projection exposure method characterized by moving the. 10. The method of claim 9, wherein after the exposure of the previous shot region is finished, the substrate is moved in a direction orthogonal to a second direction to expose the shot region, And before the movement in the direction orthogonal to the second direction of the substrate ends, accelerated movement of the substrate in the second direction to expose the shot region. 10. The method according to claim 9, wherein after the exposure of the binary shot region is finished, the substrate is started to be stepped in a predetermined direction to expose the shot region, After the exposure of the previous shot region is finished, the measurement of the positional information of the substrate in the optical axis direction and the control of the auto focusing to adjust the position of the substrate in the optical axis direction based on the result of the measurement are stopped. The projection exposure method characterized by the above-mentioned. The element manufacturing method using the method of Claim 9. Simultaneously projecting the pattern image of the mask illuminated by the exposure beam onto the substrate through a projection system, and in synchronization with scanning the mask in a first direction substantially perpendicular to the optical axis of the projection system. A projection exposure method in which a shot region on the substrate is exposed using a pattern of the mask by scanning in a second direction corresponding to the direction. After the exposure of the shot region is finished, the measurement of the position information of the substrate in the optical axis direction of the projection system and the control of the auto focusing to adjust the position of the substrate in the optical axis direction based on the measurement result are stopped. , After the end of the exposure of the shot region, during the deceleration movement of the substrate with respect to the second direction, stepping of the substrate in a predetermined direction to expose the next shot region on the substrate, And after the stamping of the substrate is finished, exposing the next shot region. 15. The method of claim 14, wherein before the stepping of the substrate in the predetermined direction is finished, the substrate is accelerated to move in a direction opposite to the second direction to expose the next shot region. Projection exposure method. 15. The projection exposure method according to claim 14, wherein after the stepping of the substrate is finished, the control of the auto focusing is started. A device manufacturing method using the method according to claim 14. Simultaneously projecting the pattern image of the mask illuminated by the exposure beam onto the substrate through a projection system, and in synchronization with scanning the mask in a first direction substantially perpendicular to the optical axis of the projection system. A projection exposure method in which a shot region on the substrate is exposed using a pattern of the mask by scanning in a second direction corresponding to the direction. After the exposure of the shot region is terminated, the measurement of the positional information of the substrate in the optical axis direction of the projection system and the control of autofocusing for adjusting the position of the substrate in the optical axis direction based on the measurement result are stopped. , After the exposure of the shot region is finished, the substrate is started stepping in a predetermined direction to expose a next shot region on the substrate, During stepping of the substrate in the predetermined direction, to accelerate the substrate in a direction opposite to the second direction to expose the next shot region, And after the stepping of the substrate ends, exposing the next shot region. 19. The projection exposure method according to claim 18, wherein after the stepping of the substrate is finished, the control of the auto focusing is started. A device manufacturing method using the method according to claim 18.