JP2006073915A - Mark, conveying equipment, aligner, position detecting method, conveying method, and process for fabricating device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect positional information of a mark with high precision in a short time. <P>SOLUTION: Since an index pattern T1 consisting of a combination of any three of marks m0-m5 included in a mark 50M is arranged in the vicinity of the center of the mark 50M, at least one mark such as the index pattern T1 exists in the imaging field of view fv of a mark detection system 42 when it falls within the region of the mark 50M. Consequently, the part of the mark 50M being caught by the imaging field of view fv can be grasped by detecting the index pattern from the imaging results of the mark 50M. Based on that index pattern, position of the mark 50M can be detected with high precision from L/S patterns LSx and LSy. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a mark, a transport apparatus, an exposure apparatus, a position detection method, a transport method, and a device manufacturing method, and more particularly, a mark provided on an object and used when detecting position information of the object, Conveying device having a provided conveying member, an exposure device for transferring a device pattern onto a substrate, a position detecting method for detecting position information of an object, a conveying method using the position detecting method, and a device manufacturing method using the conveying method About.

  In a lithography process for manufacturing a semiconductor element, a liquid crystal display element, or the like, a pattern formed on a mask or a reticle (hereinafter, collectively referred to as “reticle”) is applied to a wafer coated with a resist or the like via a projection optical system. An exposure apparatus that transfers onto a substrate such as a glass plate (hereinafter collectively referred to as a “wafer”), such as a step-and-repeat reduction projection exposure apparatus (so-called stepper), or a step-and-step with an improved stepper. A sequential movement type projection exposure apparatus such as a scanning type scanning projection exposure apparatus (so-called scanning stepper) is mainly used.

  In such an exposure apparatus, wafer alignment is performed in order to optimize the relative position between the shot area already formed on the wafer and the shot area to be transferred and formed next. Prior to the wafer alignment, the wafer position and rotation are adjusted so that the wafer alignment marks already formed on the wafer are within the capture range (detection field of view) of the detection system. Alignment is performed.

  In this pre-alignment, the applicant forms a mark on a loading arm for loading a wafer, and detects the positional relationship between the mark and the wafer in a state where the wafer is held on the loading arm. Immediately before loading the wafer, only the mark position information is detected, the wafer position is estimated based on the detected mark position information and the positional relationship, and the wafer delivery position is determined based on the estimated position. A technique for adjustment was proposed (for example, Japanese Patent Application No. 2003-36395). By adopting this technology, the loading position and the transfer position can be brought close to each other, and the throughput can be dramatically improved, and the light source for pre-alignment and the like can be moved away from the projection optical system. Therefore, it is possible to reduce the influence of the heat generated from the light source or the like on the optical characteristics of the projection optical system, and the flow of the atmospheric gas in the vicinity of the transfer position in the exposure apparatus can be made smooth.

  However, when the above technique is adopted, for example, an imaging device for imaging the mark on the loading arm must be installed near the wafer transfer position so that the imaging field of view accurately captures the mark. There is. In order to achieve this installation, for example, the magnification of the imaging device is lowered to enable imaging of a wide range of areas so that the marks on the loading arm can be easily captured, and then the image obtained from the imaging device is confirmed. However, it is necessary to adjust the installation position of the imaging device so that the mark is located near the center of the imaging field of view, and gradually increase the magnification of the imaging device while repeatedly adjusting the installation position. Met.

  From the first viewpoint, the present invention made under the above circumstances is a mark (50M) provided on the object (50) and used for detecting the position information of the object. A plurality of index patterns (T1 to T11 of different types) having a known positional relationship between the first pattern (Ar1) where the measurement patterns (LSx, LSy) having a measurable shape are arranged and the measurement pattern. ) Include second areas (Ar2) arranged at different locations, and the mark as a whole has a mark area larger than the imaging field of view of the predetermined imaging device, The mark in which the plurality of index patterns are arranged so that at least one index pattern of the plurality of index patterns is included in the imaging result when captured. .

  According to this, the mark of the present invention is a mark that is generally larger than the imaging field of view of the predetermined imaging device, but the index pattern is necessarily included in the imaging result of the mark by the imaging device. Become. Since the index patterns are different from each other, if the type of the index pattern included in the imaging result is detected, it becomes clear which area of the mark the imaging field of view of the imaging device captures. As a result, even if the imaging result is an image of only a part of the mark, the position information of the mark can be detected. If such a mark is used, at least a part of the mark can be captured by the imaging device and the position information can be detected without strictly defining the mounting accuracy of the imaging device.

  According to a second aspect of the present invention, there is provided a conveying member (50) provided for conveying a predetermined object and provided with the mark (50M) of the present invention; and when the conveying member conveys the object And at least one imaging device (42, 45) capable of imaging the mark; detection for detecting position information of the mark when the conveying member conveys the object based on an imaging result of the imaging device And a correction device (20) that corrects position information of the object based on the detection result. In such a case, the object is conveyed by the conveying member provided with the mark of the present invention, and when the image is conveyed, the mark is imaged by the imaging device, and the position information of the mark is detected based on the imaging result, Since the position information of the object is corrected based on the detection result, the object can be conveyed with high accuracy and in a short time.

  According to a third aspect of the present invention, there is provided an exposure apparatus (100) for transferring a device pattern onto a substrate (W), the transport apparatus of the present invention; and a substrate (W) delivered from the transport apparatus. ) Holding a stage (WST); and a transfer device (20) for transferring the device pattern onto the substrate held on the stage. In such a case, since the object is transported using the transport device of the present invention, high-throughput and high-precision exposure can be realized.

  According to a fourth aspect of the present invention, there is provided a position detection method for detecting position information of an object (50), wherein the mark (50M) of the present invention provided on the object is placed on an imaging device (42, 45). The position detection method includes: an imaging step of picking up an image by; a specifying step of specifying the position of the mark based on the imaging result; and a detecting step of detecting position information of the object based on the specifying result. It is. In such a case, the position of the mark is specified based on the imaging result of the mark of the present invention, and the position information of the object on which the mark is formed can be detected with high accuracy and in a short time based on the identification result. it can.

  From a fifth aspect, the present invention is a transport method for transporting the substrate so that the substrate (W) is held at a predetermined position on the stage (WST) and delivering it to the stage. Detecting the position information of the conveying member (50) provided with the mark (50M) using the position detecting method when the substrate is conveyed by the conveying member; and based on the detection result, the substrate A step of adjusting at least one of a relative position between the stage and the transport member at the time of delivery and position information of the substrate held on the stage. In such a case, since the position information of the conveying member provided with the mark is detected using the position detection method of the present invention, the position of the substrate held on the stage is determined with high accuracy and shortness based on the detection result. Can be adjusted to time.

  According to a sixth aspect of the present invention, there is provided a device manufacturing method for manufacturing a device, wherein a transfer position of a substrate and a transfer position of a device pattern via the projection optical system are determined using the transport method of the present invention. A step of delivering the substrate to a stage movable between; and a step of transferring the device pattern onto the substrate held on the stage. In such a case, since the position information of the object is detected using the transport method of the present invention, it is possible to realize high-throughput and high-accuracy exposure on the substrate, and thus the productivity of a highly integrated device. Can be improved.

  Hereinafter, an embodiment of the present invention will be described with reference to FIGS.

  FIG. 1 schematically shows a longitudinal sectional view of a part of an exposure apparatus 100 (particularly an exposure apparatus main body) according to an embodiment of the present invention. The exposure apparatus 100 includes a main body chamber 14 installed in a clean room, and a transfer chamber 15 installed adjacent to the left side of the main body chamber 14 in FIG. The main body chamber 14 and the transfer chamber 15 are connected to each other through the openings 14A and 15A.

  The main body chamber 14 accommodates most of the exposure apparatus main body. The exposure apparatus main body includes at least a part of an illumination system (not shown), a reticle stage RST that holds a reticle R as a mask, a projection optical system PL, a wafer stage WST as a stage that can hold a wafer W as a substrate, and alignment detection. The system AS and the main controller 20 as these control systems are included. The main controller 20 is disposed outside the main body chamber 14 and the transfer chamber 15.

  The exposure apparatus main body is configured around the projection optical system PL. Therefore, in the following, the vertical direction in the drawing in FIG. 1, that is, the direction of the optical axis AX of the projection optical system PL is defined as the Z-axis direction (the lower side in the drawing is positive), and the horizontal direction in FIG. The description will be made assuming that the left side of the drawing is positive, and the direction orthogonal to the drawing in FIG. 1 is the X-axis direction (the front side of the drawing is positive).

  The reticle stage RST located on the −Z side (upper side) of the projection optical system PL holds the reticle R by suction, for example, by vacuum suction or electrostatic suction. Reticle stage RST has its position information in the XY plane detected by an unillustrated interferometer or the like, and based on the detected position information, under the instruction of main controller 20, for example, a linear motor (not shown) or the like. Thus, at least minute driving is possible in an XY plane (including rotation around the Z axis) perpendicular to the optical axis of the illumination system (matching the optical axis AX of the projection optical system PL described later). When the circuit pattern drawn on the reticle R held on the reticle stage RST is illuminated with the exposure light IL from an illumination system (not shown), an illumination area having a substantially uniform illuminance is formed on the circuit pattern. The

  Projection optical system PL is arranged below reticle stage RST in FIG. As the projection optical system PL, for example, a birefringent optical system that is telecentric on both sides and has a predetermined reduction magnification (for example, 1/4 or 1/5) is used. For this reason, when the illumination area of the reticle R is formed by the exposure light IL, the reduced image of the circuit pattern of the reticle R in the illumination area (through the projection optical system PL) by the exposure light IL that has passed through the reticle R ( An inverted image) is formed in a region conjugate with the illumination region on the wafer W held by suction on the wafer stage WST disposed below (+ Z side) of the projection optical system PL. This area is also referred to as an exposure area.

  Wafer stage WST is moved on the wafer base 17 in the XY plane (including the rotation direction θz direction around the Z axis) and in the Z axis direction by a wafer stage drive unit (not shown) including a linear motor, a voice coil motor (VCM) and the like. It is also possible to finely drive in an inclination direction with respect to the XY plane (rotation direction around the X axis (θx direction) and rotation direction around the Y axis (θy direction)). The position of wafer stage WST in the XY plane (including rotation around the Z axis (θz rotation)) is a wafer laser interferometer having a plurality of measurement axes (hereinafter abbreviated as “wafer interferometer”). 18 is always detected with a resolution of, for example, about 0.5 to 1 nm. Position information (or velocity information) of wafer stage WST detected by wafer interferometer 18 is supplied to main controller 20. Main controller 20 controls the position (or speed) of wafer stage WST via a wafer stage drive unit (not shown) based on the position information (or speed information) of wafer stage WST. By this control, wafer stage WST, as shown in FIG. 1, starts from the exposure position (transfer position of the pattern via projection optical system PL) immediately below projection optical system PL indicated by the solid line, and is a two-dot chain line (virtual line). The wafer W can be moved at least to the delivery position, that is, the loading position.

  As shown in FIG. 1, a center table CT indicated by a dotted line is disposed in the vicinity of the center of wafer stage WST. When loading or unloading the wafer W to or from the wafer stage WST, the center table CT is driven by a drive mechanism (not shown) and moves up and down with the central portion of the wafer W being sucked and held from below. It is assumed that the center table CT sucks and holds the wafer W by a disk-like suction portion formed at the tip of the center table CT by vacuum suction or electrostatic suction. The center table CT is also driven under the instruction of the main controller 20.

  An off-axis alignment detection system AS is provided in the vicinity of the + Y side surface of the projection optical system PL. As the alignment detection system AS, for example, a digital image processing type FIA (Field Image Alignment) type sensor is used. The imaging result of the alignment detection system AS is output to the main controller 20.

  The main controller 20 includes a so-called microcomputer (or workstation) including a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), etc. To control. For example, an input device including a pointing device such as a keyboard and a mouse, a display device such as a CRT display (or liquid crystal display), and a storage device including a hard disk are externally connected to the main controller 20. Has been. These input device, display device, and storage device are not shown.

  Exposure apparatus 100 having an exposure apparatus body including the above-described components further includes a transfer system for transferring wafer W to wafer stage WST, and a pre-alignment system for performing pre-alignment of wafer W transferred by the transfer system. Yes. In the exposure apparatus 100, a main body chamber 14 is provided with a load slider 50 that is a part of the wafer W transfer system and a mark detection system 42 that is a part of the pre-alignment system.

  The load slider 50 can hold the wafer W by vacuum suction or electrostatic suction. The load slider 50 can move in the Y-axis direction while passing the openings 14A and 15A between the main body chamber 14 and the transfer chamber 15 by a transfer mechanism described later while holding the wafer W. The load slider 50 that has received the wafer W in the transfer chamber 15 moves in the −Y direction and moves above the loading position of the wafer stage WST in the main body chamber 14 as shown in FIG. Then, the cooperative operation of the load slider 50 and the center table CT provided on the wafer stage WST located at the loading position realizes the loading of the wafer W onto the wafer stage WST.

  Further, the load slider 50 is provided with a light transmission portion (not shown in FIG. 1) that transmits light in the Z-axis direction. A mark 50M as a mark for detecting a two-dimensional position (including rotation) in the XY plane is formed on the surface on the −Z side of the light transmitting portion. Details of the load slider 50 and the mark 50M will be described later.

  The mark detection system 42 is disposed at a position facing the position of the mark 50M on the load slider 50 when the load slider 50 is positioned above the loading position (that is, above the mark 50M). The mark detection system 42 includes a light source that illuminates a region in the XY plane including the mark 50M, a two-dimensional CCD camera, and the like. In the mark detection system 42, reflected light with respect to illumination light from the light source is received by a two-dimensional CCD camera or the like, and the mark 50M is imaged by a so-called epi-illumination type. Since the mark detection system 42 is fixed to a structure (not shown) that supports the exposure apparatus main body such as the projection optical system PL, the positional relationship with the projection optical system PL is constant, and the origin of the imaging field of view is fixed. The position in the XY plane is always constant. Therefore, the XY coordinate system and the coordinate system defined by the imaging field of view of the camera, that is, the camera coordinate system are always constant. The imaging result (digital two-dimensional image data) of the mark 50M by the mark detection system 42 is sent to the main controller 20.

  FIG. 2 schematically shows a cross-sectional view of a part of the exposure apparatus 100 centering on the wafer W transfer system and the pre-alignment system. The transfer system of the wafer W includes a load robot 92 that takes out the wafer W from a front opening unified pod (hereinafter referred to as “FOUP”) 27, and a wafer from the load robot 92 to the load slider 50. A pre-alignment stage 52 that relays transfer of W and pre-aligns the wafer W during the relay, a turntable 51 mounted on the pre-alignment stage 52, the load slider 50, and the load A Y drive mechanism 60 for driving the slider 50 in the Y-axis direction, an unload slider 62 for unloading the exposed wafer W from the wafer stage WST, and an unload robot 93 for receiving the wafer W from the unload slider 62 It is comprised including.

  The FOUP 27 is the same as the transfer container disclosed in, for example, Japanese Patent Application Laid-Open No. 8-279546. The FOUP 27 is provided with an opening that allows a wafer W to be taken in and out only on one surface, and a door that can be opened and closed at the opening ( An open / close container (sealed wafer cassette) to which a lid is attached. In the FOUP 27, a plurality of wafers W are stored in the vertical direction at a predetermined interval. The FOUP 27 is set at the position shown in FIG. 2 by a FOUP transport device (not shown). With this setting, the opening 15B for the FOUP 27 disposed in the transfer chamber 15 is connected to the opening of the FOUP 27. When the door of the opening is opened, the inside of the FOUP 27 is passed through the opening and the opening 15B. The wafer W can be loaded into the transfer chamber 15.

  The load robot 92 is a horizontal articulated robot capable of attracting and holding the wafer W at the tip of its arm, and mainly transporting the wafer W from the FOUP 27 to the pre-alignment stage 52, from the unload robot 93. The exposed wafer is collected. The posture control of the load robot 92 is performed by driving a rotary motor (not shown) incorporated in a joint or the like of the load robot 92 under the instruction of the main controller 20.

  The pre-alignment stage 52 is a stage that can move in the XY plane. The pre-alignment stage 52 is at least movable between a position where the wafer W can be delivered to the load slider 50 and a position where the load robot 92 can deliver the wafer W in the Y-axis direction. It is configured. The pre-alignment stage 52 is controlled by driving a driving mechanism such as a linear motor as shown in FIG. FIG. 2 shows the pre-alignment stage 52 in a position where the wafer W can be delivered to the load slider 50.

  The turntable 51 is disposed at a substantially central portion on the surface of the pre-alignment stage 52 on the −Z side, and can move up and down within a predetermined range, and holds a wafer W and is a rotation axis parallel to the Z axis. It is a table that can rotate around. The end surface on the −Z side of the turntable 51 is provided with a disk-shaped wafer adsorption holding surface for adsorbing and holding the wafer W by vacuum adsorption or electrostatic adsorption. The wafer W sucked and held on the suction holding surface can be rotated. This rotation and vertical movement are performed by driving a drive mechanism (not shown) under the instruction of the main controller 20.

  Information about the XY position of the pre-alignment stage 52, the rotation position of the turntable 51, the height of the wafer suction holding surface (wafer W), and the like are detected by a position detection sensor (not shown) and sent to the main controller 20. ing. Main controller 20 controls the XY position of pre-alignment stage 52 and the position (rotation position, Z position) of turntable 51 based on the information.

  As shown in FIG. 2, the load slider 50 is connected to a Y drive mechanism 60 that extends in the Y-axis direction from the transfer chamber 15 side to the main body chamber 14 side. The load slider 50 can be moved (slid) in the Y-axis direction between the transfer chamber 15 and the main body chamber 14 by the drive of the Y drive mechanism 60 under the instruction of the main controller 20. Then, the wafer W held on the turntable 51 is received, moved to the −Y side, and the wafer W is transferred above the loading position. FIG. 2 shows the load slider 50 positioned above the loading position, that is, at a position where the mark detection system 42 can detect the mark 50M.

  FIG. 3A shows a top view of the load slider 50. As shown in FIG. 3A, in the load slider 50, a light transmission part 50A is formed in the vicinity of the end of the arm part extending in the X-axis direction on the −X side, and −Z of the light transmission part 50A is formed. A mark 50M is formed at a substantially central portion of the side surface. Further, the load slider 50 is provided with a pair of finger portions each provided with a suction mechanism for sucking and holding an object to be placed. The pair of finger portions are connected to one end and the other end of the arm portion, and the center of the wafer W is sandwiched from each other when viewed from the −Z side so that the wafer W can be transported in a stable state. Thus, the wafer W can be sucked and held. In the load slider 50, since it is necessary to transfer the wafer W to and from the turntable 51 and the center table CT, the distance between the pair of fingers is the disk-like suction holding surface of the turntable 51 and the center table CT. It is set to be larger than the diameter.

  Further, as will be described later, in the pre-alignment system, the wafer W held on the load slider 50 is detected in order to detect at least three edge positions of the wafer W in order to calculate the center position and rotation amount of the wafer W. A part of the outer edge is imaged from the −Z side. In FIG. 3A, regions including five edges of the wafer W to be imaged by the pre-alignment apparatus 45 described later are shown as regions VA to VE, respectively. That is, when the + Y direction is 6 o'clock and the + X direction is 3 o'clock with respect to the center of the wafer W, 6 o'clock (region VA), 7:30 (region VB), 4:30 (region VC), The area including the edge of the wafer W in the direction of 3 o'clock (area VD) and 1:30 (area VE) is the area to be imaged. Since this imaging is performed by so-called transmitted illumination, the pair of finger portions of the load slider 50 are arranged so as to avoid the transmitted illumination areas (areas VA to VE), respectively.

  As described above, there are a total of five areas to be imaged as described above. Actually, however, when the notch of the wafer W is in the 6 o'clock direction, it is 6 o'clock, 7:30, 4 o'clock. Three half areas VA, VB, VC are imaged. When the notch of the wafer W is in the 3 o'clock direction, three areas VD, VC, VE at 3 o'clock, 4:30, 1:30 are imaged. Become so. That is, all the five areas are not imaged on the same wafer. Hereinafter, the areas VA to VE are also referred to as imaging areas VA to VE.

  As shown in the perspective view of FIG. 3 (B), the connecting portion between the arm portion and each finger portion in the load slider 50 has a certain width in the Z-axis direction. Designed to be different in height from the part. The distance between the arm part and the finger part in the Z-axis direction is defined to be sufficiently larger than the thickness of the wafer W, and the distance between the connecting parts that connect the arm part and each finger part in the X-axis direction is The diameter is defined to be sufficiently larger than the diameter of the wafer W. Therefore, when the load slider 50 is viewed from the Y-axis direction, it appears that a space surrounding the wafer W is formed by the arm portion, each finger portion, and their connecting portion. As a result, the load slider 50 can be moved in the Y-axis direction without interfering with the wafer W held on the turntable 51, for example, by driving by the Y drive mechanism 60.

  FIG. 4 shows an enlarged view around the mark 50M. As shown in FIG. 4, the mark 50 </ b> M formed at the center of the light transmitting portion 50 made of glass has a mark region including a region Ar <b> 1 as a first region and a region Ar <b> 2 as a second region. ing. The region Ar1 is further divided into three regions. Of these three regions, the -X side and + X side regions have a line-and-space (hereinafter referred to as a pattern) having a shape in which the Y-axis direction is an arrangement direction and position information about the Y-axis direction can be measured. (Abbreviated as “L / S”). A pattern LSy is arranged, and the L region as a pattern having a shape in which the X-axis direction is an arrangement direction and position information about the X-axis direction can be measured in the central region. / S pattern LSx is arranged.

  In the L / S patterns LSx and LSy, the line portion is formed as a light-shielding portion by chromium vapor deposition, and the space portion is a light transmission portion, but the opposite may be possible. In any case, the line portion that is the chrome portion and the space portion that transmits light are set so that the luminance in the imaging result (gray image) when the mark 50M is imaged is different. If the image is taken, the so-called mirror symmetry (reversal symmetry) related to the luminance distribution of the partial region is maximized from the image of at least a partial region of the L / S pattern LSx (or L / S pattern LSy). The position can be obtained as the center X (Y) position of the region. In other words, the L / S pattern LSx and the L / S pattern LSy are line and space patterns, and therefore, if at least a part of the area is extracted, the pattern has a shape capable of measuring the position information of the area. It can be assumed that

  As shown in FIG. 4, two of the L / S patterns LSy are arranged so as to sandwich the L / S pattern LSx, but the two L / S patterns LSx are different from the L / S pattern LSy. You may arrange | position so that it may pinch | interpose.

  Note that the duty ratio between the line portion and the space portion of the L / S patterns LSx and LSy is 1: 1, and the distance between them is such that the position measurement accuracy for one pixel of the mark detection system 42 is, for example, about 1/100 pixel. The position information can be measured with very high accuracy.

The area Ar2 is further divided into two areas S _L and S _R. Marks m0 to m5 are provided in the areas S _L and S _R , respectively. In each of the regions S _L and S _R , the marks m0 to m5 are light shielding portions (chrome portions) and the other portions are light transmitting portions, but the reverse may be possible. The marks m0 to m5 are a combination mark of a rectangular mark or a small rectangular mark, and are arranged at equal intervals along the Y-axis direction. The arrangement order of the marks m0 to m5 is the same in the areas S _L and S _R in the second area Ar2. That is, the marks m0 to m5 are arranged at equal intervals in the order of m1, m4, m2, m4, m3, m4, m0, m5, m3, m5, m2, m5, m1 from the −Y side in the Y-axis direction. ing.

  Among these, the mark m0 is a mark indicating the center of the mark 50M in the Y-axis direction, and is therefore arranged only at the center. Further, the mark m4 is arranged only above the center (−Y side) with respect to the Y-axis direction of the mark 50M, and the mark m5 is arranged only below the center (+ Y side). The remaining marks m1, m2, and m3 are arranged in the order of the marks m3, m2, and m1 from the center in the Y-axis direction of the mark 50M so as to be sandwiched between the marks m0, m4, and m5.

  Here, a combination of several marks m0 to m5 is regarded as one mark. For example, as shown in FIG. 4, a mark composed of a combination of marks m1, m4, and m2 can be regarded as one mark T1 shown within a broken line. As described above, if a combination of three marks among the marks m0 to m5 is regarded as one mark, the combination marks can be extracted from the second region Ar2. That is, 11 marks as shown in FIG. 5 are extracted. These marks are referred to as marks T1 to T11, respectively.

Since the marks T1 to T11 are of different types, each mark T1 to T11 acts as an index pattern indicating which part of the mark 50M the field of view of the mark detection system 42 points to. For example, if the mark T1 is within the field of view, it can be seen that the mark detection system 42 captures the vicinity of the −Y side end of the mark 50M. Hereinafter, the marks T1 to T11 are referred to as index patterns T1 to T11. Further, since the index patterns T1 to T11 are arranged in a line at equal intervals in the Y-axis direction with respect to the regions S _L and S _R , the index patterns T1 to T11 act as scale marks for specifying the positions in the Y-axis direction. An example of the scales indicated by the index patterns T1 to T11 is shown in Table 1 below. However, there is a possibility that the captured image does not become the mark 50M due to the drawing of the optical system of the camera, the insertion of a mirror, and the like, and the image may be rotated, vertically rotated, or horizontally reversed. Each of the marks T <b> 1 to T <b> 11 is a mark according to the mark 50 </ b> M, but this is not the case when the image is inverted upside down with respect to the original mark or rotated 180 degrees. In this case, since the mark 50M is visually imaged upside down, each of the marks T1 to T11 needs to be converted upside down.

表１に示されるように、例えば指標パターンＴ６の中心位置は、Ｙ軸方向に関するマーク５０Ｍの中心位置、すなわち原点に一致しており、指標パターンＴ６は、０（原点）を示す目盛マークであるとすることができる。また、指標パターンＴ１〜Ｔ５は、指標パターンＴ６の−Ｙ側にそれぞれ、１ｍｍ間隔で配置されているので、上記原点からの距離がそれぞれ−５．０〜−１．０ｍｍの位置を示す目盛マークであるとみなすことができる。さらに、指標パターンＴ７〜Ｔ１１については、指標パターンＴ６の＋Ｙ側にそれぞれ１ｍｍ間隔で配置されているので、上記原点からの距離がそれぞれ＋１．０〜＋５．０ｍｍの位置を示す目盛マークであるとみなすことができる。図６には、指標パターンＴ１〜Ｔ１１と、その指標パターンが示す目盛との対応関係が模式的に示されている。

As shown in Table 1, for example, the center position of the index pattern T6 coincides with the center position of the mark 50M in the Y-axis direction, that is, the origin, and the index pattern T6 is a scale mark indicating 0 (origin). It can be. In addition, since the index patterns T1 to T5 are arranged at 1 mm intervals on the −Y side of the index pattern T6, the scale marks indicating the positions where the distance from the origin is −5.0 to −1.0 mm, respectively. Can be considered. Further, since the index patterns T7 to T11 are arranged at intervals of 1 mm on the + Y side of the index pattern T6, they are scale marks indicating positions where the distance from the origin is +1.0 to +5.0 mm, respectively. Can be considered. FIG. 6 schematically shows the correspondence between the index patterns T1 to T11 and the scales indicated by the index patterns.

  FIG. 4 shows an imaging field fv of the mark detection system 42 and the like. Since the index patterns T1 to T11 are arranged near the center of the mark 50M in the X-axis direction, when the imaging field of the mark detection system 42 is within the mark area of the mark 50M, At least one of the index patterns T1 to T11 is included. Therefore, by detecting the index pattern from the imaging result, it is possible to grasp which part of the mark 50M the imaging visual field fv captures. Note that which part of the mark 50M the imaging field of view fv captures is determined by the positional relationship between the load slider 50 and the mark detection system 42. Therefore, the imaging visual field fv is not always at the position shown in FIG.

  Note that the index patterns T1 to T11 have a known positional relationship with the L / S patterns LSx and LSy, and the position information of any one of the index patterns T1 to T11 is detected. The approximate positions of the L / S patterns LSx and LSy can be grasped from the position information.

  Returning to FIG. 2, the unload slider 62 is configured to be movable (slidable) in the Y-axis direction below (+ Z side) the load slider 50. The unload slider 62 receives the wafer W from the center table CT lifted by holding the wafer W by vacuum suction or the like when unloading the exposed wafer W from the wafer stage WST. To the transfer position of the wafer W. The unload slider 62 is also driven by a drive mechanism (not shown) under the instruction of the main controller 20.

  The unload robot 93 is a horizontal articulated robot that receives the wafer W from the unload slider 62 at the delivery position, and delivers the wafer W to the load robot 92, for example. The posture control of the unload robot 93 is also performed by driving a rotation motor (not shown) incorporated in a joint or the like of the unload robot 93 under the instruction of the main controller 20.

  That is, in the present embodiment, the transfer system for the wafer W is constituted by the load robot 92, the load slider 50, the pre-alignment stage 52 (including the turntable 51), the Y drive mechanism 60, the unload slider 62, the unload robot 93, and the like. It is configured.

  FIG. 7 is a perspective view schematically showing the configuration of the pre-alignment system. In FIG. 7, the pre-alignment stage 52 at the transfer position of the wafer W with the load robot 92 (this is referred to as “first position”) is indicated by a two-dot chain line (virtual line), and the wafer W with the load slider 50 is shown. The pre-alignment stage 52 at the transfer position (this is referred to as a “second position”) is indicated by a solid line.

  This pre-alignment system includes illumination devices 81A to 81G1 and 81G2, two line sensors 83A and 83B, and a pre-alignment device 45. These interfere with the movement of the pre-alignment stage 52 and the load slider 50 on a top plate (not shown) disposed above the first position and the second position or suspended from the top plate. It should be supported so that nothing happens. However, it is assumed that the illumination device 81A is formed not on the top plate but on the pre-alignment stage 52.

  Each of the illumination devices 81G1 and 81G2 is arranged to illuminate a part of the outer edge of the wafer W held on the turntable 51 from the + Z side when the pre-alignment stage 52 is in the first position. The line sensors 83A and 83B receive the illumination lights from the illumination devices 81G1 and 81G2 above the wafer W, respectively. As a result, the edge of the wafer W held on the turntable 51 of the pre-alignment stage 52 at the first position can be detected by the line sensors 83A and 83B. The detection result is sent to the main controller 20.

  When the pre-alignment stage 52 is in the second position, the illumination devices 81A to 81E, for example, the imaging area VA shown in FIG. 3A on the wafer W held on the turntable 51 (or the load slider 50). Illuminate the outer edges corresponding to ~ VE from the + Z side. The position of the illumination device 81A is not limited to the position on the pre-alignment stage 52. For example, the illumination device 81A is suspended from a top plate (not shown) so as to be able to retract / enter the position shown in FIG. You may make it arrange | position at the edge part of the L-shaped member supported by lowering. In this way, when the pre-alignment stage 52 moves between the first position and the second position, the member is rotated to retract the illumination device 81A, and the pre-alignment stage 52 is moved to the first position. After moving to the second position, the wafer W can be illuminated by the illumination device 81A if the illumination device 81A enters the position where the area VA can be illuminated from the + Z side.

  The Y drive mechanism 60 extends to the + Y side from the second position. When the pre-alignment stage 52 is at the second position, the load slider 50 moves to the + Y side up to that position (position indicated by the dotted line). It is possible to receive the wafer W that is slid and held on the turntable 51. In FIG. 7, as described above, the load slider 50 that has entered above the second position is indicated by a two-dot chain line (virtual line). The load slider 50 receives the wafer W from the turntable 51 at this position.

  The illumination device 81F illuminates the vicinity of the mark 50M on the load slider 50 from the + Z side after the load slider 50 receives the wafer W. Since the vicinity of the mark 50M is the light transmitting portion 50A as described above, the illumination light from the illumination device 81F passes through the load slider 50 and reaches the pre-alignment device 45. The pre-alignment device 45 is provided with light transmitting portions 45A to 45F that transmit the illumination light from the illumination devices 81A to 81F so that each illumination light can be taken inside.

  The pre-alignment device 45 receives the illumination light from the illumination devices 81A to 81F, and images the edge portions (regions VA to VE) and the mark 50M of the wafer W. This imaging result is sent to the main controller 20. In the present embodiment, the size of the imaging region in the mark 50M is the same as the imaging field of view fv of the mark detection system 42 shown in FIG. 4, and this is the imaging region VF. As described above, if the illumination devices 81A to 81E illuminate the wafer W from the + Z side and the pre-alignment device 45 images the imaging regions VA to VE from the −Z side, the portion corresponding to the wafer W in the imaging result is As a dark part, the part which is not the wafer W is imaged as a bright part. In this way, it becomes possible to accurately recognize the outer shape of the wafer W in a state in which the contrast is conspicuous from the imaging result. The configuration of the optical system inside the pre-alignment device 45 is not particularly limited as long as the regions VA to VF can be imaged.

  FIG. 8 is a block diagram of a control system related to the wafer transfer system and the pre-alignment system in this embodiment. As shown in FIG. 8, the wafer W transfer system and the pre-alignment system control system are mainly configured by the main controller 20. In FIG. 8, components used for detection (imaging) are shown on the left side of the paper from the main controller 20, and components used for the wafer W adjustment operation based on the transfer operation and the pre-alignment result are arranged on the right side of the paper. ing. The function (configuration and individual operation) of each component is as described above. In FIG. 8, the line sensors 83A and 83B are grouped as the line sensor 83, and the lighting devices 81A to 81G1 and 81G2 are grouped as the lighting device 81.

  In the pre-alignment system configured as described above, it is necessary to detect the edge information of the wafer W and the position information of the mark 50M from the imaging results of the pre-alignment device 45 and the mark detection system 42. However, the coordinate system defined by the individual imaging field of the pre-alignment device 45 and the imaging field of the mark detection system 42 is not completely coincident with the XY coordinate system, and a slight deviation occurs depending on the respective attachment conditions. . FIG. 9 shows various coordinate systems related to pre-alignment in the present embodiment. In the pre-alignment, first, it is necessary to define a coordinate system as a reference for aligning the wafer W. Since the pre-alignment is performed based on the imaging result of the edge of the wafer W by aligning the wafer W on the wafer stage WST, the reference coordinate system is determined based on the imaging field of view of the pre-alignment device 45. That is, for example, when the notch of the wafer W is in the 6 o'clock direction, the coordinate system defined by the positional relationship of the imaging field of view corresponding to the regions VA, VB, and VC is used as the pre-alignment reference coordinate system. When the notch is in the 3 o'clock direction, a coordinate system defined by the positional relationship of the imaging field of view corresponding to the regions VC, VD, and VE is set as a pre-alignment reference coordinate system. Hereinafter, this reference coordinate system is referred to as a wafer coordinate system. Hereinafter, only the case where the notch of the wafer W is oriented in the 6 o'clock direction will be described.

FIG. 9 shows the X _W axis and the Y _W axis which are coordinate axes as the wafer coordinate system. However, so-called individual camera coordinate systems defined by individual imaging fields when imaging the regions VA, VB, and VC have an offset component, a rotation component, and a magnification component, respectively, with respect to the wafer coordinate system. . In this embodiment, by a calibration process using the calibration reference wafer (glass wafer) W _S to be described later that the calibration mark is provided, it is assumed that these components are already computed.

Similarly, the coordinate system defined by each imaging field of view when the pre-alignment device 45 images the region VF is called a “pre-2TA camera coordinate system”, and the rotation component of this coordinate system with respect to the wafer coordinate system is θ _T2. The magnifications in the X-axis direction and Y-axis direction are MX _CT2 and MY _CT2 , respectively. Similarly, a coordinate system defined by the imaging field of view of the mark detection system 42 is referred to as a “pre-3 camera coordinate system”. A rotation component of this coordinate system with respect to the wafer coordinate system is θ _T3 , and the X axis direction and the Y axis direction are related. The magnification is MX _CT3 and MY _CT3 , respectively. It is assumed that the values of these rotational components θ _T2 and θ _T3 are obtained in advance by the calibration process using the calibration reference wafer. Hereinafter, in the present embodiment, the variable representing the rotation amount (that is, the angle) is a positive rotation amount in the direction in which the right screw rotates with respect to the −Z direction, and a negative rotation amount in the opposite direction. It is said.

  It is assumed that the origin position of the pre-2LA camera coordinate system and the origin position of the pre-3 camera coordinate system with respect to the wafer coordinate system are not clear. Therefore, in the present embodiment, a mark coordinate system in which the rotation with respect to the wafer coordinate system is 0 and the magnification is 1 is set, and the position of the mark 50M is determined with reference to the mark coordinate system.

  FIG. 10 shows the processing procedure of the CPU in the main controller 20 regarding the operation at the time of performing exposure processing by the exposure apparatus 100 of this embodiment in which the exposure apparatus main body, the transport system, and the pre-alignment system are configured as described above. It demonstrates along the flowchart of FIG. 15, and FIGS. 16-26B.

  As a premise, before this exposure operation is performed, the imaging fields of view VA to VE in the pre-alignment apparatus 45 are the edges (6) of the wafer W when the wafer W is held on the turntable 51 or the load slider 50. , 7:30, 4:30, 3:30, and 1:30) are adjusted so as to fall within the imaging field of view simultaneously, and before the subroutine 101 is executed, the pre-alignment stage 52 The mechanical attachment positions of the wafer transfer system, the mark detection system 42, the line sensors 83A and 83B, and the illumination devices 81A to 81G1 and 81G2 are also adjusted appropriately, and the subroutine 103 of FIG. Before executing (after executing the subroutine 101), it is assumed that the above-described calibration processing has already been performed. In this calibration process, the rotation amount α of the wafer coordinate system with respect to the XY coordinate system is also obtained.

  In each step to be described later, the main controller 20 issues an instruction to the transport system, the pre-alignment system, etc., but since the instruction transmission path is as described above, detailed description will not be given. . Further, main controller 20 waits until it is confirmed that the instructed operation is completed by a response from the transport system, the pre-alignment system, or the like, and does not proceed to the next step. In this embodiment, it is assumed that the wafer W is always loaded with the notch direction set at 6 o'clock, and only the procedure necessary for processing the wafer W loaded in that direction will be described below.

Also, among the wafers stored in FOUP, wafers initially retrieved by load robot 92 is assumed to be a tool wafer W _S shown in FIG. 16. The tool wafer W _S is the calibration reference wafer is a glass wafer as described above is a silicon wafer of the same material as the wafer of different processes. As shown in FIG. 16, fiducial marks SMa, SMb, SMc, and SMd are formed on the surface. The reference mark SMa, SMb, SMc, SMd, at the surface of the tool wafer W _S, has a parallelogram to X _W axis or Y _W-axis, a rectangular vertex positions around the center of the tool wafer W _S The formed shape is a shape having a function as a two-dimensional position detection mark such as a two-dimensional L / S pattern that can detect the two-dimensional position with high accuracy. Therefore, if the positions of the reference marks SMa, SMb, SMc, SMd are detected, the position and rotation amount of the center OJ of the tool wafer W _s can be calculated from these positions.

  Note that the tool wafer is not limited to the one having the above configuration, and may be one in which a mark is provided at the center of the wafer.

First, in subroutine 101 of FIG. 10, the wafer W is transferred and pre-aligned. FIG. 11 shows a flowchart of the subroutine 101. In this subroutine, as shown in FIG. 11, first, in step 201, to load the robot 92, and instructs the loading of the tool wafer W _S. Loading robot 92, for example, carries the tool wafer W _S which have been stored in FOUP27, passes the turntable 51.

In the next step 203, to illuminate the tool wafer W _S by the illumination device 81G1,81G2, line sensors 83A, by 83B, for detecting the eccentricity and the rotation amount of the tool wafer W _S on the turntable 51, so-called pre-1 Measurement Perform the process. In the next step 204, the eccentric amount and the rotation amount of the calculated tool wafer W _S is to be canceled, to adjust the pre-alignment stage 52 and the turntable 51. In the next step 205, the pre-alignment stage 52 is moved to the second position.

In the next step 207, to measure the residual amount of rotation of the tool wafer W _S of the second position by the pre-alignment apparatus 45. This measurement will be referred to as “pre-2TT measurement”. This measurement, the offset component on the wafer coordinate system of the imaging visual field VA~VC the pre-alignment apparatus 45, the rotational component, taking into account the magnification component, imaging results of three edge position of the tool wafer W _S including a notch by transmission illumination the detected luminance distribution, and detects the amount of rotation of the tool wafer W _S from the three edge positions. Then, this rotation amount is canceled by rotating the turntable 51. As a result, fine rotation adjustment of the wafer on the turntable 51 is realized as described above.

In the next step 209, the wafer is transferred from the turntable 51 to the load slider 50. In step 211, after passing the completion of the loading slider 50 of the tool wafer W _S, the center position coordinates O (X _0, Y ₀₎ of the wafer held by the loading slider 50 and measures the amount of rotation. This measurement is called pre-2LA measurement. FIG 23 (A), the pre-2LA is detected in the measurement step, the center position coordinate O of the tool wafer W _S in the wafer coordinate system (X _0, Y ₀₎ is shown schematically. The center coordinates O (X ₀ , Y ₀ ) and the rotation amount are stored in a storage device (not shown). It is assumed that all the wafer center position coordinates detected in the subsequent pre-2LA measurement process are position coordinates with the position coordinates O as the origin. Incidentally, in the pre 2TT measurement tool wafer W _S, since the fine rotation adjustment is performed, the amount of rotation has a normal substantially zero.

  In the next subroutine 213, the position of the mark 50M on the load slider 50 is measured by the pre-alignment device 45. This measurement is referred to as pre-2TA measurement. FIG. 12 is a flowchart showing the processing of the subroutine 213. As shown in FIG. 12, first, in step 301, the pre-alignment device 45 captures an image of the mark 50M. Image data obtained by this imaging is referred to as image data P. Hereinafter, the image data P is also referred to as an image P as appropriate. Here, for example, it is assumed that the imaging field of view fv of the pre-alignment device 45 is at the position shown in FIG. 4 and the image P as shown in FIG.

In the next step 303, it is determined whether or not the position measurement of the mark 50M of the load slider 50 is the initial measurement. Here, if the determination is affirmed, the process proceeds to a subroutine 305 that performs initial measurement of the mark 50M, and if the determination is negative, the process proceeds to a subroutine 310 that performs normal measurement of the mark 50M. Here, the first sheet of FOUP27, that is, initial measurement tool wafer W _S is loaded, the decision is affirmative, advances the story as the process advances to a subroutine 305.

  FIG. 13 shows a flowchart of the subroutine 305, that is, a flowchart showing an initial measurement procedure. As shown in FIG. 13, first, in step 401, Y-axis direction projection is performed on the image P. Here, the projection is an integration process in which the luminance values of the pixels in the image are integrated. In the Y-axis direction projection, the luminance values are integrated in the Y-axis direction. Here, for example, the luminance value at each pixel in the image P shown in FIG. 17A is integrated in the Y-axis direction. The Y-axis direction here is the Y-axis direction in the pre-2TA camera coordinate system. In this way, it is possible to obtain an added value of the luminance of the pixel regarding each X position in the pre-2TA camera coordinate system. Therefore, if the addition values of the luminance values are arranged in the X-axis direction, a waveform indicating a change in the integrated luminance value in the X-axis direction as shown in FIG. 17B (this is referred to as a “projection waveform”). Wp is obtained. As shown in FIG. 17B, since the projection waveform Wp is obtained by transmitted illumination, the portion corresponding to the second region Ar2 having a relatively large area of the light transmitting portion is the first region Ar1. The value is larger than the part corresponding to. Therefore, the projection waveform Wp is a waveform that protrudes upward with reference to the vicinity of the center X position of the mark 50M. When the projection ends, the process proceeds to step 403.

  As shown in FIG. 4, the mark 50M is line symmetric with respect to a straight line parallel to the Y axis passing through the center position of the mark 50M in the X axis direction. Therefore, if the imaging field of view of the pre-alignment device 45 captures the center of the mark 50M, the projection waveform Wp, which is the integrated luminance value of the imaging result, should be line symmetric with respect to the center X position of the mark 50M. In order to obtain the center X position of the line-symmetric mark 50M, it is effective to obtain the inversion symmetry correlation of the waveform. The center X position is obtained by inverting the waveform on the right side of the position with respect to the waveform on the left side of the position with respect to the position as a reference, and superimposing the waveforms on the X axis direction. This is because it is considered to exist in the vicinity of the position where the degree is maximum. Therefore, in the next step 403, the inverse symmetry correlation function C (x) in the X-axis direction for the projection waveform Wp obtained in step 401 is calculated.

  When calculating the inverse symmetry correlation function C (x), a correlation calculation window that defines a range in which the correlation is calculated is used. FIG. 18A shows a correlation calculation window Win as an example of the correlation calculation window. As shown in FIG. 18A, the width Rw of the correlation calculation window Win is such that the region where the L / S pattern LSx is formed and the second region Ar2 in the projection waveform, that is, the index patterns T1 to T11 are formed. It is set to a size that sufficiently includes a portion corresponding to the area.

  The reference position of the correlation calculation window Win is assumed to be the center position of the window. In the inverse symmetry correlation calculation, the correlation between the waveform on the left side of the reference position and the waveform obtained by inverting the waveform on the right side of the reference position in the X-axis direction is calculated.

  Here, the reference position of the correlation calculation window Win is set to a position P1 shown in FIG. 18B, and moved from the position P1 to the direction (+ X direction) indicated by the arrow in FIG. Meanwhile, the degree of inversion symmetry correlation in the correlation calculation window Win at each reference position is obtained. FIG. 18C shows an example of the inverted symmetry correlation function C (x) obtained in this way.

In the next step 405, it determines the outline center X position X _m of the mark 50M. As described above, since the mark 50M is line symmetric with respect to the X-axis direction, the position where the value of the inversion symmetry correlation function C (x) is maximum is estimated as the approximate center X position of the mark 50M. Therefore, as shown in FIG. 18C, the X position X _m where the value of the inverse symmetry correlation function C (x) is maximum (this value is C (x _m )) is set to the mark 50M. It can be determined as the approximate center position.

Subsequent processing is carried out the outline central position X _m as a reference. First, the basis of this outline central position X _m, the target pattern T1～T11, is searched by template matching. The main controller 20 stores template image data (hereinafter abbreviated as “template”) of the index patterns T1 to T11 of the mark 50M shown in Table 1 in a storage device (not shown). . These template image data are referred to as templates T1 ′ to T11 ′, respectively. The templates T1 ′ to T11 ′ may be obtained by calculation from design information of the index patterns T1 to T11 or the like, and the index patterns T1 to T11 may be obtained in advance by a predetermined imaging device (for example, the mark detection system 42). It is good also as what was extracted from the imaging result obtained by imaging.

In this embodiment, first, in the following Step 407 and Step 409, the range in which the template matching is performed is limited. First, in step 407, based on the outline central position X _m determined in step 405, to determine the X-axis direction of the range for matching template T1'~T11 '.

The range in which the index patterns T1 to T11 exist can be predicted to some extent from each dimension (design value) of the mark 50M. FIG. 19A shows various dimensions related to the mark 50M. Here, a straight line passing through the center of the mark 50M in the X-axis direction is defined as a straight line Lx0. A boundary line between the L / S pattern LSx and the second region Ar2 is a straight line Lx1. A design value of the distance from the straight line Lx0 to both ends of the L / S pattern LSx in the X-axis direction is L1. The boundary line between the L / S pattern LSy and the second region Ar2 is a straight line Lx2. The design value of the distance from the straight line Lx0 to the straight line Lx2 is L2. The index patterns T1 to T11 are considered to exist between the line Lx1 and the line Lx2. Therefore, in the present embodiment, using the design values L1 and L2 determined in advance based on the mark 50M, for example, from the position X _m -L2 to the position X _m -L1 in the X-axis direction, template matching is performed. Is determined as the range in the X-axis direction. The index patterns T1 to T11 are arranged symmetrically on both sides with respect to the X direction L / S pattern LSx. However, here, it is sufficient to search only one of the index patterns T1 to T11. Only the side) will be searched.

  In the next step 409, a range in the Y-axis direction for matching with the templates T1 'to T11' is determined. In the present embodiment, pre-alignment is performed using position information of the mark 50M that is larger than the imaging field of view of the pre-alignment device 45 and the mark detection system 42. However, detection is performed in consideration of reproducibility of position control of the load slider 50. The mark position is preferably near the center of the imaging field of view of the pre-alignment device 45. Therefore, it is desirable to detect an index pattern in the vicinity of the center of the image P among the index patterns T1 to T11. Therefore, in the present embodiment, the range in the Y-axis direction in which template matching is performed is a range near the center of the image P. Therefore, as shown in FIG. 19A, a straight line Ly0 that bisects the image P is set. Then, a straight line Ly1 and a straight line Ly2 are set at positions separated from the straight line Ly0 by a distance L3 in the ± Y direction. Then, a range between the straight line Ly1 and the straight line Ly2 is determined as a range for performing template matching in the Y-axis direction.

  In the present embodiment, as also shown in Table 1 and FIG. 6, in the mark 50M, the index patterns T1 to T11 are arranged at intervals of 1.0 mm in the Y-axis direction. Therefore, in order to detect the center of the index pattern, the size of the template matching range in the Y-axis direction (here, 2 × L3) needs to be at least 1.0 mm.

  With the above processing, the region designated by the range in the X-axis direction and the range in the Y-axis direction shown in FIG. 19A is determined as a region for matching the templates T1 'to T11'. When the intersections of the straight line Lx1 and the straight lines Ly1 and Ly2 are a and b, respectively, and the intersections of the straight line Lx2 and the straight lines Ly1 and Ly2 are c and d, respectively, an inner region of the rectangle having these four intersections a to d as vertices The template matching is performed while moving the centers of the templates T1 ′ to T11 ′. This area is called a matching area MA.

  In the following steps 411 to 423, template matching is performed to detect the type and position of the index pattern included in the matching area MA. In the matching area MA obtained in steps 407 and 409, it is unknown which index template center corresponds to which template among the 11 types of scale mark templates T1 ′ to T11 ′. Then, matching is performed for all templates, and the types and positions of the index patterns T1 to T11 existing near the center of the image P are detected.

  Therefore, in the next step 411, the value of the counter k indicating the template type (hereinafter abbreviated as “counter value k”) is initialized to 1.

In the next step 413, template matching between the image P and the template Tk ′ is performed. In this template matching, the center of the template Tk ′ is moved by, for example, one pixel at a time in the matching area MA, and is expressed by the following equation between the template Tk ′ and the image of the portion overlapping the template Tk ′ in the image P. The position P _kmax where the value of the function Cr _k (x, y) is maximized is searched using the expression of the normalized cross-correlation function Cr _k (x, y). In the following expression, x and y are the center positions of the template, the sizes of the template Tk ′ in the X-axis direction and the Y-axis direction are t _x and t _y , respectively, and the luminance of a predetermined pixel of the template is T (i , J), and the luminance of the pixel corresponding to T (i, j) of the image P is I (x + i, y + j).

ここで、Ｔ_avgは、テンプレートＴｋ’における輝度の平均値であり、Ｉ_avgは、画像Ｐにおける、そのテンプレートＴｋ’と重ね合わされる領域の輝度の平均値であり、それぞれ次式で表される。

Here, T _avg is an average value of the luminance in the template Tk ′, and I _avg is an average value of the luminance of the region that is overlapped with the template Tk ′ in the image P, and is expressed by the following equations, respectively. .

ステップ４１５では、カウンタ値ｋを参照し、カウンタ値ｋが１１以上であるか否かを判断する。この判断が肯定されれば、ステップ４１９に進み、否定されればステップ４１７に進む。カウンタ値ｋが１１以上であるということは、すべてのテンプレートについてのマッチングが終了したことを意味する。ここでは、カウンタ値ｋが１であり、最初のテンプレートＴ１’についてのマッチングが行われたのみであるので、ここでの判断は否定され、ステップ４１７に進む。ステップ４１７では、カウンタ値ｋの値を１だけインクリメントして（ｋ←ｋ＋１）して、ステップ４１３に戻る。ステップ４１３では、次のテンプレート（ここではテンプレートＴ２’）のマッチングを行う。このように、ステップ４１５での判断が肯定されるまで、ステップ４１３→ステップ４１５→ステップ４１７の処理、判断が繰り返され、テンプレートＴ２’〜Ｔ１１’における正規化相互相関度関数Ｃｒ_k（ｘ，ｙ）が最大値となる位置Ｐ_kmaxが求められる。

In step 415, the counter value k is referred to and it is determined whether or not the counter value k is 11 or more. If this determination is affirmed, the process proceeds to step 419, and if not, the process proceeds to step 417. That the counter value k is 11 or more means that matching for all templates has been completed. Here, since the counter value k is 1 and only the matching for the first template T1 ′ has been performed, the determination here is denied and the routine proceeds to step 417. In step 417, the counter value k is incremented by 1 (k ← k + 1), and the process returns to step 413. In step 413, matching of the next template (here, template T2 ′) is performed. In this way, until the determination in step 415 is affirmed, the processing and determination of step 413 → step 415 → step 417 are repeated, and the normalized cross-correlation function Cr _k (x, y) in the templates T2 ′ to T11 ′. ) Is the maximum value P _kmax .

  When matching of all the templates T1 'to T11' is completed, the determination in step 415 is affirmed, and the process proceeds to step 419.

In step 419, the normalized cross-correlation value obtained in step 413 Cr _k (x, y) maximum value Cr _kmax (x, y) of, selects the template Tk 'was maximal. This maximum value is Cr _kmax '(x, y), and the selected template Tk' is Tk _S '. For example, when the matching area MA is selected as shown in FIG. 19A, the template T7 ′ shown in FIG. 5 is selected.

In step 421, it is determined whether the maximum value Cr _kmax '(x, y) of the normalized cross-correlation of the selected template is equal to or greater than a predetermined threshold value. If this determination is affirmed, the process proceeds to step 423, and if the determination is negative, the process proceeds to step 425. In step 425, a display indicating that the index pattern has not been detected is displayed on a display device (not shown), and various operations of the exposure apparatus 100 are forcibly terminated. After forcibly ending, the operator checks the display on the display device, visually checks the cause, and makes adjustments.

On the other hand, in step 423, based on the template Tk _S 'selected in step 419, the type and position of the index pattern existing near the center of the matching area MA are determined. For example, when the matching area MA as shown in FIG. 19A is selected, the index pattern T7 corresponding to the template T7 ′ is determined, and the center position of the index pattern T7 is determined as P _kmax. Will be.

When the process of step 423 ends, the process proceeds to step 501 of FIG. In step 501, a temporary center _PT is determined based on the determined index pattern position _Pkmax . As shown in FIG. 19B, the design value L4 of the distance between the index pattern in the X axis direction and the center position in the mark 50M is used to determine the temporary center _PT . That is, a point separated by L4 in the + X direction from the index pattern position P _kmax is determined as the temporary center _PT .

In step 503, a rectangular measurement range of the L / S pattern LSx is determined. This X-axis direction measurement range is determined based on the temporary center _PT . A plurality of measurement ranges are designated. The number can be determined in advance, for example, six.

In FIG. 19C, as an example, six X-axis direction measurement ranges determined based on the temporary center P _T are indicated by two-dot chain lines. Here, a range of a predetermined width centered on the temporary center _PT with respect to the X-axis direction is designated as the size of the measurement range of the L / S pattern LSx in the X-axis direction. The size of the X-axis direction measurement range in the X-axis direction is set to a value sufficiently larger than the design value of the width in the X-axis direction of the L / S pattern LSx, and the most + X side and the most −X of the L / S pattern LSx. The line part on the side is set so as to be always included in the measurement range. The size of the measurement range in the X-axis direction in the Y-axis direction and the interval between them can also be set to an arbitrary size. In short, each X-axis direction measurement range may be arranged at the same position in the X-axis direction so as to completely include the L / S pattern LSx in the X-axis direction with the temporary center _PT as the center. The six X-axis direction measurement ranges are represented by measurement ranges m (m = 1 to 6), respectively.

In step 505, a plurality of rectangular measurement ranges of the L / S pattern LSy, that is, measurement ranges in the Y-axis direction are provisionally determined. The measurement range in the Y-axis direction is also a range centered on the Y position of the temporary center _PT , and the size and number in the X-axis direction may be arbitrary. The measurement range in the Y-axis direction is set in the Y-axis direction so that a predetermined number of lines constituting the Y-direction L / S pattern LSy are included. For example, the five measurement ranges in the Y-axis direction indicated by broken lines in FIG. 20A are set to include 12 lines constituting the L / S pattern LSy. These five measurement ranges in the Y-axis direction are defined as a measurement range n (n = 1 to 6).

In step 507, as shown in FIG. 20B, based on the L / S pattern LSx of the measurement range m in each X-axis direction determined in step 503, that as a representative point of the X-direction measurement range m The center point (X _cm , Y _cm ) of the range is determined, and a least square approximate line Lx based on the representative point of the L / S pattern LSx in the X-axis direction is obtained from the coordinates of each center point.

  Determination of the center position of the L / S pattern LSx within the measurement range m in the X-axis direction is based on the fact that the L / S pattern LSx is line-symmetric with respect to the X-axis direction, and the entire L / S pattern LSx is inverted symmetrically. This is performed based on the degree of reciprocal symmetry and the degree of reversal symmetry of each line part constituting the L / S pattern LSx. Hereinafter, a method for determining the center position of each measurement range m in the L / S pattern LSx will be described.

  First, the luminance of each pixel (pixel) included in each measurement range m is added in the Y-axis direction, and a one-dimensional waveform indicating each addition result is obtained. Further, SINC interpolation, which is a kind of digital filtering processing, is performed on the one-dimensional waveform, for example, to improve the position detection accuracy in the one-dimensional waveform, which is a discrete data waveform. Then, a correlation calculation using an observation window of a predetermined width that scans the one-dimensional waveform with improved detection accuracy shows the overall reflection (inversion) symmetry correlation degree with respect to the axial position of the one-dimensional waveform. A correlation function indicating a correlation function or a correlation function indicating the degree of inversion symmetry of each line portion of the L / S pattern LSx is obtained, and a waveform indicating the degree of correlation is sharpened by multiplying the correlation functions of the respective inversion symmetry correlation degrees. Or by applying a quadratic function fitting to increase the detection accuracy and determine the position where the inversion symmetry (reflection symmetry) is maximized as the center position of the L / S pattern LSx within the measurement range (ie, L / S position).

  Then, the least square approximation straight line Lx regarding the center point of the plurality of measurement ranges m in the X-axis direction obtained in this way is obtained. FIG. 20C shows an example of the straight line Lx.

  In the next step 509, the slope A of the straight line Lx obtained in step 507 is calculated. In the next step 511, the position of the measurement range n of the L / S pattern LSy is corrected based on the slope A of the straight line Lx obtained in step 509. This position correction is performed so that the same line group is always included from among the line groups constituting the L / S pattern LSy between the measurement ranges n in the Y-axis direction.

This position correction is performed based on the inclination A obtained in step 509. Specifically, in step 505, as shown in FIG. 21A, the Y direction measurement temporarily determined so that the center thereof is arranged on a straight line passing through the temporary center Pt and parallel to the X axis. The measurement range n in the Y-axis direction is moved in the + Y direction or the −Y direction so that the position of the center point of the range n is arranged on the straight line Ly ′ that passes through the temporary center _{PT and} has an inclination of 1 / A. To do. FIG. 21B shows the Y-axis direction measurement range arranged on the straight line Ly ′ by this position correction. Since the load slider 50 is controlled with high accuracy, the actual rotation amount of the mark 50M is very small. However, in FIGS. 20A to 22B, the rotation of the mark 50M is performed for convenience of explanation. It is shown with emphasis.

In the next step 513, as shown in FIG. 22A, the center point (X _cn , Y _cn ) of each image range n is determined based on the image of each measurement range n determined in step 511. The determination of the center point in each measurement range n in the Y-axis direction is performed by the same method as in step 507 above. After step 513 is completed, the subroutine 305 is ended, and the process proceeds to step 307 in FIG.

In step 307, the L / S patterns LSx and LSy within the measurement range m in the X-axis direction and the measurement range n in the Y-axis direction obtained in steps 507 and 513 in FIG. The position coordinates (X _cm , Y _cm ) and (X _cn , Y _cn ) of the center point are converted into position coordinates in the mark coordinate system. Since the expression representing the conversion to the mark coordinate system is obtained in advance by the calibration process using the calibration reference wafer, the rotation amount θ _CT2 , the magnification MX _CT2 , MY _CT2 , the mark coordinate system origin position in the pre-2TA camera coordinate system ( X ₀ , Y ₀ ) and is expressed as follows:

ここで、マーク座標系の原点位置（Ｘ₀，Ｙ₀）は、カメラ原点（０，０）とすることができる。なお、上記式（４）では、Ｘ軸方向の計測範囲の中心点（Ｘ_cm，Ｙ_cm）に関する式だけを示しているが、Ｙ軸方向の計測範囲の中心点（Ｘ_cn，Ｙ_cn）を求める場合には、その中心点（Ｘ_cn，Ｙ_cn）を、（Ｘ_cm，Ｙ_cm）と入れ替え、その変換座標値（Ｘ_n，Ｙ_n）を、（Ｘ_m，Ｙ_m）と入れ替えれば良い。

Here, the origin position (X ₀ , Y ₀ ) of the mark coordinate system can be the camera origin (0, 0). In the above formula (4), only the formula related to the center point (X _cm , Y _cm ) of the measurement range in the X axis direction is shown, but the center point (X _cn , Y _cn ) of the measurement range in the Y axis direction is shown. Is obtained, the center point (X _cn , Y _cn ) is replaced with (X _cm , Y _cm ), and the converted coordinate value (X _n , Y _n ) is replaced with (X _m , Y _m ). It ’s fine.

In the next step 309, by using the least square method, the formula (4) by the obtained center point (X _m, Y _m) from, determines the least square approximation line Lxm, the center point (X _n, Y _n ), the least square approximation straight line Lym is determined. FIG. 22B shows these two straight lines Lxm and Lym. Then, the intersection of the two straight lines Lxm and Lym is calculated, and the intersection is determined as the position P _T ′ of the mark 50M. Further, the inclination of the straight line Lxm or the straight line Lym or the average of the inclinations of the straight lines Lxm and Lym is defined as the rotation amount θ _CT2 of the mark 50M.

The position information (position P _T ′ and rotation amount θ _CT2 ) of the mark 50M is _assumed to be p (S _CAx , S _CAy , θ _CA ). S _CAx is the X coordinate, S _CAy is the Y coordinate, and θ _CA is the rotation amount. FIG. 23A schematically shows position information p (S _CAx , S _CAy , θ _CA ) in the mark coordinate system. The position information p (S _CAx , S _CAy , θ _CA ) of the mark 50M in the mark coordinate system is stored in a storage device (not shown).

In the next step 215, the load slider 50 is moved above the loading position by driving the Y drive mechanism 60, and the wafer stage WST is moved to the loading position. The loading position of wafer stage WST at this time is a designed position. Then, in the subroutine 217, the pre-3 measurement process is performed. Here, based on the imaging result obtained by imaging by the mark detection system 42, the position information of the mark 50M in the mark coordinate system is detected. The processing of this subroutine 217 is the same as that of the subroutine 213 (pre-2TA measurement) of FIG. 12 described above except that the image captured by the mark detection system 42 is not the image captured by the pre-alignment device 45 but is processed. Since it is the same as the process, detailed description is omitted. Also in this case, in the subroutine 213, the subroutine 305 is executed instead of the subroutine 310. In the mark detection system 42, since an image is taken in the epi-illumination system as described above, an image in which the bright part and the dark part are reversed from the image shown in FIG. 17A is obtained. But there is no impact on processing. Further, in the conversion formula to the mark coordinate system in the pre-3 measurement process, MX _CT3 and MY _CT3 are used as magnifications, θ _T3 is used as rotation, and the camera origin is set to (0, 0).

The position information (the position P _T ′ of the mark 50M and the rotation amount θ _CT3 ) of the mark 50M in the mark coordinate system at the time of the pre-3 measurement obtained in the subroutine 213 is defined as q (S _CBx , S _CBy , θ _CB ). . S _CBx is the X coordinate, S _CBy is the Y coordinate, and θ _CB is the rotation amount. FIG. _23B schematically shows position information q (S _CBx , S _CBy , θ _CB ) of the mark 50M in the mark coordinate system. The position information q (S _CBx , S _CBy , θ _CB ) of the mark 50M is stored in a storage device (not shown).

After the subroutine 217 end, back to FIG. 10, in step 103, to load the tool wafer W _S to the wafer stage WST. After the tool wafer W _S is completely delivered to the center table CT, center table CT is lowered, eventually tool wafer W _S is placed on the wafer stage WST, is attracted to and held by vacuum suction.

In the next step 105, a wafer loaded to determine whether the tool wafer W _S. If this determination is affirmed, the process proceeds to step 125; Here, the determination is affirmed and the routine proceeds to step 125.

In step 125, wafer stage WST is moved in the -Y direction and positioned below alignment detection system AS. The reference marks SMa on the tool wafer W _S, SMb, SMc, detects position information of SMd. Specifically, drives and controls the wafer stage WST based on the position information obtained from the wafer interferometer 18, a reference mark SMa on the tool wafer W _S, SMb, SMc, SMd are sequentially below the alignment detection system AS move the tool wafer W _S to be positioned, the reference mark SMa by alignment detection system aS, imaging SMb, SMc, the SMd. This imaging result is supplied to the main controller 20.

The main controller 20, from the position information of the reference mark SMa～SMd, calculates the position coordinates of the center of the tool wafer W _S in the XY coordinate system. The position coordinates, a load position coordinates of the tool wafer W _S in the XY coordinate system. Here, the calculated load position coordinates are LP (LP _X , LP _Y ), and this point is a reference point of the estimated load position when the wafer W is loaded. FIG. 23B shows the load position coordinates LP (LP _X , LP _Y ) of the tool wafer W _S in this XY coordinate system. The calculated information and the load position coordinates LP (LP _X , LP _Y ) are stored in a storage device (not shown).

In the next step 119, to unload the tool wafer W _S. Here, the tool wafer W _S is unloaded by the unload slider 62. Then, unload the slider 62, retreated to the delivery position of the unload robot 93, and passes the tool wafer W _S to unload robot 93, further the tool wafer W _S, load robot from the unload robot 93 92 To the FOUP 27 by the load robot 92.

In the next step 121, it is determined whether or not the exposure for all wafers has been completed. If this determination is affirmed, the process ends. If the determination is negative, the process returns to the subroutine 101. Here, since the process for the tool wafer W _S has only been completed, the determination is denied and the process returns to the subroutine 101.

  In the next subroutine 101, the processing of steps 201 to 217 in FIG. 11 is performed again. However, here, it is the normal wafer W that is transferred and pre-aligned. That is, in step 201, the load robot 92 takes out the wafer W from the FOUP 27 and delivers it to the turntable 51. In the next step 203, the pre-1 measurement process is performed. In step 204, the position of the wafer W is substantially adjusted by XY movement of the prealignment stage 52 and rotation of the turntable 51 based on the measurement result of the pre1 measurement process. To do.

  In the next step 205, the pre-alignment stage 52 is moved to the second position, and in step 207, a pre-2TT measurement process is performed. Here, the rotation amount of the wafer W is calculated by detecting the edge of the wafer W including the notch of the wafer W by transmitted illumination from the luminance distribution of the imaging result, and the turntable 51 is rotated so as to cancel the rotation amount. Adjust the fine rotation.

In the next step 209, the wafer W is transferred from the turntable 51 to the load slider 50. In step 211, the pre-2LA measurement process is performed. Note that the position information of the wafer W detected here, that is, the center coordinates and the rotation amount, is C (X _C , Y _C , θ _C ). X _C is the X coordinate, Y _C is the Y coordinate, and θ _C is the rotation amount. The X coordinate and the Y coordinate are coordinate values with O (X ₀ , Y ₀ ) as the origin. It shall be. FIG. 24A schematically shows position information (center coordinates and rotation amount) C (X _C , Y _C , θ _C ) of the wafer W in the wafer coordinate system. The position information C (X _C , Y _C , θ _C ) is stored in a storage device (not shown).

  In the next subroutine 213, a pre-2TA measurement process is performed. In this subroutine 213, as shown in FIG. 12, first, in step 301, the pre-alignment device 45 images the mark 50M.

  In the next step 303, it is determined whether or not the position measurement of the load slider 50 is an initial measurement. Here, since the initial measurement has already been performed, the determination is denied and the process proceeds to the normal measurement subroutine 310.

  FIG. 15 shows a flowchart of the subroutine 310, that is, a flowchart showing a procedure for normal measurement.

  As shown in FIG. 15, in step 601, information on the template selected in the initial measurement is acquired from the templates T1 'to T11'. Let this template be Ta '. This is because the reproducibility of the stop position of the load slider 50 at the time of pre-2TA measurement is high and the error is minute, and the same scale mark as the scale mark selected in step 423 in FIG. 13 is positioned at the center of the imaging range. This is because it is efficient to perform the matching from the template Ta ′ corresponding to the scale mark.

In step 603, the template Ta ′ is matched with the image P. This matching is performed in the matching area MA determined in the initial measurement subroutine 305, and is performed in the same manner as the matching in the subroutine 413. As a result of this matching, the maximum value Cr _kmax (x, y) of the normalized cross-correlation value Cr _k (x, y) and its X position and Y position are detected.

In step 605, the maximum value Cr _kmax (x, y) of the normalized cross-correlation function Cr _k (x, y) calculated as a result of matching with the template Ta ′ calculated in step 603 is larger than a predetermined threshold value. Determine whether or not. If this determination is affirmed, the process proceeds to step 607, and if not, the process proceeds to step 621. In step 621, an alarm display indicating that there is no scale mark is displayed on a display device (not shown), and the process is forcibly terminated.

On the other hand, in steps 607 to 615, in order to verify whether or not the information acquired in step 601 is correct, the maximum value Cr _kmax (x, y) of the normalized cross-correlation value Cr _k (x, y). Matching of the 11 types of templates T1 ′ to T11 ′ with the image P at the position where) is maximized. Hereinafter, the procedure will be described.

In step 607, 1 is set to the counter value k indicating the template type, and the first matching template is set to T1 ′. In step 609, the template Tk ′ (here, template T1 ′) is matched. This matching is also performed by calculating a normalized cross-correlation function Cr _k (x, y).

  In step 611, the counter value k is referred to, and it is determined whether the value is 11 or more, that is, whether or not matching has been performed for all templates. If the determination is negative, the process proceeds to step 613. If the determination is positive, the process proceeds to step 615. Here, since the counter value k is 1, the determination is negative and the routine proceeds to step 613.

  In step 613, the counter value k is incremented (k ← k + 1), and in step 609, the next template (here, template T2 ') is matched. Thereafter, the processing and determination from step 609 to step 611 to step 613 are repeated until the determination at step 611 is affirmed. When matching of all templates is completed, the determination in step 611 is affirmed, and the process proceeds to step 615. In the loop processing here, matching of the template Ta 'is performed again, but this is not necessary. It is only necessary to perform matching with 10 types of templates excluding the template Ta '.

In step 615, the maximum value Cr _kmax (x) of the inverted symmetry correlation function Cr _k (x, y) of each template when the templates are matched is respectively compared, and the inverted symmetry correlation function Cr _{k is} compared. The template having the largest (x, y) maximum value Cr _kmax (x) is determined as the template Tb ′ from the templates T1 ′ to T11 ′.

  In step 617, it is determined whether or not the template Tb 'determined in step 615 is the same as the template Ta'. If the determination is negative, the process proceeds to step 621, and if the determination is positive, the process proceeds to step 619. The processing after step 621 is as described above.

On the other hand, in step 619, it is assumed that the scale mark corresponding to the template Ta ′ based on the information acquired in step 601 is present at the position detected in step 603, and the type and position information of the scale mark are stored in a storage device (not shown). To remember. After step 619 is over,
Proceed to step 501 in FIG. Since the subsequent processing of step 501 to step 513 is as described above, description thereof will be omitted. After step 513 is completed, the process of subroutine 310 is terminated, and the process proceeds to step 307 in FIG.

Since the processing of step 307 and step 309 is as described above, description thereof is omitted. At the end of these processes, position information (position P ′ _T and rotation amount θ _CT2 ) of the mark 50M in the mark coordinate system is detected. Here, this position information is assumed to be p ′ (S _CAx ′, S _CAy ′, θ _CA ′). S _CAx ′ is the X coordinate, S _CAy ′ is the Y coordinate, and θ _CA ′ is the rotation amount. FIG. 24A schematically shows position information p ′ (S _CAx ′, S _CAy ′, θ _CA ′). The position information p ′ (S _CAx ′, S _CAy ′, θ _CA ′) is stored in a storage device (not shown). After step 309 is completed, the process proceeds to step 215 in FIG.

  In the next step 215, the load slider 50 is moved above the loading position by driving the Y drive mechanism 60, and the wafer stage WST is moved to the loading position. Here, the position of wafer stage WST when wafer W is delivered from load slider 50 is estimated, and the load position of wafer stage WST in the XY coordinate system is determined based on the estimation result. Below, the estimation method is demonstrated in detail.

As described above, at this time, a pre-2LA measurement process in step 211 has been performed with respect to (11) the tool wafer W _S, have been conducted and the pre 2TA measurement step in subroutine 213 (FIG. 11), pre 2LA measuring positional information of the tool wafer W _S in the wafer coordinate system in step O (X _0, Y ₀₎ and the position information p (S _CAx mark 50M in the mark coordinate system at the pre-2TA measurement step, S _CAy, θ _CA ) (see FIG. 23A). The load position of the tool wafer _{W S LP (LP X, LP} Y) is determined in the XY coordinate system (see FIG. 23 (B)).

Further, at this time, the pre-2LA measurement process for the wafer W in step 211 (FIG. 11) and the pre-2TA measurement process in the subroutine 213 (FIG. 11) are performed, and the wafer coordinates in the pre-2LA measurement process are performed. Position information C (X _C , Y _C , θ _C ) of the wafer W in the system and position information p ′ (S _CAx ′, S _CAy ′, θ _CA ′) of the mark 50M in the mark coordinate system in the pre-2TA measurement process (See FIG. 24A). These pieces of position information p, O, LP, C, and p ′ are used for estimating the delivery position of wafer W on wafer stage WST.

Here, the positional deviation of the wafer W is estimated by dividing it into a parallel movement component due to the positional deviation of the mark 50M and a rotational component that is a positional deviation component caused by the difference in the rotation amount of the mark 50M. First, estimation of the parallel component of the positional deviation of the wafer W will be described. FIG. 25A shows position information O and C of the wafer W in the wafer coordinate system detected in the pre-2LA measurement step (step 211) (hereinafter simply referred to as “position O” and “position C”). A vector diagram schematically showing the positional relationship between the positional information p and p ′ (hereinafter simply referred to as “position p” and “position p ′”) of the mark 50M detected in the pre-2TA measurement step (step 213) is shown. Has been. Figure 25, (A), the position p of the mark 50M in the pre-2LA measuring step when loading a tool wafer W _S, the vector indicating the relative positional relationship between the position O of the tool center the wafer W _S is a vector P shown A vector indicating the relative positional relationship between the position p ′ of the mark 50M and the center position C of the wafer W in the pre-2LA measurement process when the wafer W is loaded is shown as a vector P ′. In the present embodiment, the translational component of the positional deviation of the position of wafer stage WST when loading wafer W is estimated using vectors P and P ′ as a reference.

That is, the center of the mark 50M, located in a positional relationship in which the center of the tool wafer W _S is represented by a vector P, since the tool wafer W _S that were in the positional relationship is loaded position is at a position LP The position where the wafer W is to be loaded is estimated to be a position represented by a vector P′−P, which is the difference between the vector P ′ and the vector P, with the position LP as a base point. _A vector indicating this positional deviation is referred to as a vector WA.

A specific method for calculating the translation component will be described. First, as shown in FIG. _25A , the position p (S _CAx , S _CAy ) of the mark 50M detected in the pre-2LA measurement process at the time of loading the tool wafer W _S , and at the time of loading the wafer W position p of the pre-2LA measurement process in detected mark _{50M '(S CAx', S} CAy ') X and Y components of the vector a _p of the difference between the is represented by the following formula.
Swp _x = S _CAx '−S _CAx (5)
_{Swp y = S CAy '-S CAy} ... (6)

Next, determine the vector W _A. Assuming that the X and Y components of the vector W _A are (W _Ax , W _Ay ), (W _Ax , W _Ay ) are respectively expressed by the following equations.

したがって、ここでは、工具ウエハＷ_Sのロード位置ＬＰを基準とするベクトルＷ_Aを上記式（５）〜（７）を計算して求め、それをウエハステージＷＳＴのローディング位置の位置ずれ量の平行移動成分とし、不図示の記憶装置に格納する。

Thus, here, obtains a vector W _A relative to the loading position LP of the tool wafer W _S by calculating the above equation (5) to (7), parallel positional displacement amount of the loading position of the wafer stage WST it The moving component is stored in a storage device (not shown).

Next, the rotational component of the positional deviation is estimated. The position p of the mark 50M when holding the tool wafer W _S, the position p 'and the mark 50M when holding the wafer W has provisionally perfectly matched, even if the distance from the mark 50M, to the center of the wafer even were the same, different amount of rotation of the mark 50M, so the center position of the tool wafer W _S and the wafer W is deviated. The rotation component means a positional deviation of the wafer W due to the rotation of the mark 50M.

In the present embodiment, the mark of the pre-alignment apparatus 45 and the mark-detecting system 42, and the rotation amount theta _CA mark 50M of the pre 2TA measurement step during tool wafer W _S conveyed by pre 2TA measurement step during wafer W conveyed A rotation amount θ _CA ′ of 50M has already been detected. Therefore, the rotational component of the positional deviation of the wafer W is estimated from the rotational amounts θ _CA and θ _CA ′ of these marks 50M.

Since the center of the mark 50M and the center of the tool wafer W _S are in a positional relationship represented by the vector P, the position where the tool wafer W _S in the positional relationship is loaded is the position LP. The position where W is to be loaded is represented by a vector of the difference between the rotation amount θ _CA of the mark 50M and the rotation amount θ _CA ′ of the mark 50M with the position C as the base point, as shown in FIG. It is estimated that it will be a position to be. Accordingly, it is assumed that the difference between the rotation amount θ _CA of the mark 50M and the rotation amount θ _CA ′ of the mark 50M is, for example, θ _p (= θ _CA ′ −θ _CA ) shown in FIG.

Rotation of the mark 50M in the pre-2LA measurement process, tool wafer W _S or wafer W or the like is determined by the attitude of the load slider 50 before it is held in the loading slider 50. This is because the wafer is transferred to the load slider 50 by driving the turntable 51, so that the posture of the load slider 50 does not change during that time, and the rotation amount of the mark 50M can be regarded as constant. Therefore, if the rotation amount is corrected after transporting to the loading position, the load position at the rotation amount θ _CA ′ of the mark 50M is changed from the load position LP at the rotation amount θ _CA of the mark 50M to the mark 50M. It is considered that the rotation direction is the direction opposite to the direction of θ _p (the rotation direction) with the center of the rotation as the rotation center. That is, the rotation amount of the center position of the wafer W to be corrected is estimated to be −θ _p . In FIG. 25B, θ _p is assumed to be positive (counterclockwise), but θ _p may be negative (clockwise). In this case, −θ _p Is positive (counterclockwise).

In the present embodiment, based on this corrected rotation amount −θ _p , the positional deviation vector Dt ₁ of the wafer W on the wafer stage WST accompanying the rotation component of the mark 50M obtained as described above is obtained using the following equation. calculate.

Ｄｔ_1x，Ｄｔ_1yは、ベクトルＤｔ₁のＸ軸成分及びＹ軸成分である。ここでは、マーク５０ＭとウエハＷとの距離の設計値Ｌと、ウエハ座標系のＸ_W軸に対するマーク５０Ｍ中心と工具ウエハＷ_Sの中心とを結ぶ線分の回転量の設計値θ_Dを用いている。

Dt _1x and Dt _1y are the X-axis component and the Y-axis component of the vector Dt ₁ . Here, using the design value L of the distance between the mark 50M and the wafer W, the design value theta _D of rotation of a line segment connecting the center of the mark 50M center and tool wafer W _S for X _W axis of the wafer coordinate system ing.

Calculated as described above, the vector W _A as a translation component of the wafer W, if adding the rotation component Dt ₁ as follows, to estimate the displacements loading position of the wafer stage WST Can do.

ここで、Ｄ_1x，Ｄ_1yは、ベクトルＤ₁のＸ成分及びＹ成分である。なお、ここで、ウエハ座標系と、ＸＹ座標系との回転量αが無視できない場合には、この回転量αにより、このベクトルＤ₁を回転させたベクトルをロード位置の推定に用いるようにしても良い。

Here, D _1x and D _1y are the X component and Y component of the vector D ₁ . Here, when the rotation amount α between the wafer coordinate system and the XY coordinate system cannot be ignored, the vector obtained by rotating the vector D ₁ based on the rotation amount α is used for estimating the load position. Also good.

The main control unit 20 stores the positional displacement vector D ₁ in a storage device (not shown). In the present embodiment, the position corresponding to the vector D ₁ is estimated as the load position of wafer stage WST, wafer stage WST is moved to the estimated load position, and wafer W is loaded.

Returning to FIG. 11, in the subroutine 217, the pre-3 measurement process is performed. The imaging result of the mark detection system 42 is sent to the main controller 20. The main controller 20, based on the imaging result, the positional information of the mark 50M in the mark coordinate system (position and rotation) are detected in the same manner as the tool wafer W _S subroutine 213 during loading. The position P _T ′ and the rotation amount θ _CT3 of the mark 50M detected in this subroutine 213 are set as position information q ′ (S _CBx ′, S _CBy ′, θ _CB ′). S _CBx ′ is the X coordinate, S _CBy ′ is the Y coordinate, and θ _CB ′ is the rotation amount. The position information q ′ (S _CBx ′, S _CBy ′, θ _CB ′) is stored in a storage device (not shown). FIG. 24B shows an example of position information q ′ of the mark 50M of the load slider 50 in the pre-3 measurement process in the subroutine 217. After the subroutine 217 is completed, the process of the subroutine 101 is terminated and the process proceeds to step 103 in FIG.

In the next step 103, wafer W is loaded on wafer stage WST. The load position of wafer W on wafer stage WST at this time is LP ′ (LP _x ′, LP _y ′). FIG. 24B shows a load position LP ′ of wafer W on wafer stage WST. When an exposed wafer is held on wafer stage WST, it is necessary to unload the wafer from wafer stage WST before performing this loading operation.

  In the next step 106, the amount of positional deviation of wafer W on wafer stage WST is calculated. Here again, the amount of misalignment is estimated separately for the translation component due to the misalignment of the mark 50M and the rotation component due to the difference in the amount of rotation of the mark 50M.

FIG. 26A shows a vector diagram schematically showing a method for estimating the amount of positional deviation of wafer W on wafer stage WST. In FIG. 26 (A), the position of the mark 50M in the mark coordinate system when loading the tool wafer W _S information q (hereinafter, shortly referred to as "location q") and the loading position LP of the tool wafer W _S A vector indicating the relative positional relationship is shown as a vector Q. In the present embodiment, the translation component of the positional deviation of the load position of the wafer W on the wafer stage WST is estimated using the vector Q as a reference.

First, a method for estimating the translation component will be described. Loading position of the wafer stage WST in step 103, the loading position LP (i.e. vector Q), has a position obtained by adding the vector W _A (LP '). However, the position q (S _CBx, S _CBy) pre 3 detected by the measuring process marks 50M when loading tool wafer W _S and, mark 50M detected by pre 3 Measurement when loading the wafer W The loading position of the wafer W on the wafer stage WST is the initially planned position due to deviation from the position information q ′ (S _CBx ′, S _CBy ′) (hereinafter abbreviated as “position q ′”). Therefore, the position shifts by the amount. Therefore, the parallel movement component of the positional deviation of the wafer W on the wafer stage WST includes the position q (S _CBx , S _CBy ) of the mark 50M detected by the mark detection system 42 when the tool wafer W _S is loaded, and the wafer W This is the vector A _q of the difference from the position q ′ (S _CBx ′, S _CBy ′) of the mark 50M detected by the mark detection system 42 during loading. The coordinate axis components Swq _x and Swq _y of the vector A _q are expressed by the following equations (10) and (11).
Swq _x = S _CBx '−S _CBx (10)
Swq _y = S _CBy '−S _CBy (11)
Therefore, main controller 20 stores this vector A _q in a storage device (not shown) as the amount of displacement of the translation component.

Next, the rotational component of the positional deviation is estimated. In the present embodiment, the mark detection system 42, the rotation amount of the mark 50M in the pre 3 Measurement process when loading the rotation amount theta _CB, the wafer W of the mark 50M in the pre 3 measuring step when loading a tool wafer W _S θ _CB 'has been detected. Therefore, the rotational component of the positional shift of the center position of the wafer W is estimated from the rotation amount of these marks 50M.

FIG. 26B shows an example of the rotation amount θ _q indicating the difference between the rotation θ _CB of the mark 50M at the position q and the rotation θ _CB ′ of the mark 50M at the position q ′. In the present embodiment, the difference in the rotation amount of the mark 50M detected by the position q and the position q ′ is the rotational deviation of the mark 50M after being held by the load slider 50. Rotation of the mark 50M in the pre 3 measurement process, tool wafer W _S or wafer W or the like is determined by the attitude of the load slider 50 after being held in the loading slider 50. In this case, if the difference between the rotation of the mark 50M at the position q ′ and the rotation of the mark 50M at the position q is θ _q (= θ _CB '−θ _CB ), the rotation in the direction of the wafer to be corrected as a result. The deviation is + θ _q as it is. That is, the correction rotation amount of the estimated position of the center of the wafer is + θ _q .

In the present embodiment, the distance between the mark 50M and the wafer W is set to L (design value) so that the rotational component of the positional deviation of the wafer W on the wafer stage WST is calculated based on the corrected rotation amount + θ _q . If the design values of the rotational component of the tool center the wafer W _S for the mark 50M for X _W axis of the wafer coordinate system and theta _D, vector Dt ₂ showing the rotational component is as follows.

ここで、Ｄｔ_2x，Ｄｔ_2yは、ベクトルＤｔ₂のＸ成分及びＹ成分である。上記式（１０）〜式（１２）より、最終的なウエハステージＷＳＴ上のウエハＷの位置ずれベクトルＤ₂は、次式のようになる。

Here, Dt _2x and Dt _2y are the X component and the Y component of the vector Dt ₂ . The above equation (10) to (12), positional displacement vector D ₂ of the wafer W on the final wafer stage WST is given by the following equation.

ここで、Ｄ_2x，Ｄ_2yは、ベクトルＤ₂のＸ成分及びＹ成分である。主制御装置２０は、この位置ずれベクトルＤ₂を不図示の記憶装置に格納する。なお、ここで、ウエハ座標系と、ＸＹ座標系との回転量αが無視できない場合には、この回転量αにより、このベクトルＤ₂を回転させたベクトルを求め、これを不図示の記憶装置に格納するようにしても良い。

Here, D _2x and D _2y are the X component and Y component of the vector D ₂ . The main control unit 20 stores the positional displacement vector D ₂ in a storage device (not shown). Here, when the rotation amount α between the wafer coordinate system and the XY coordinate system cannot be ignored, a vector obtained by rotating the vector D ₂ is obtained from the rotation amount α, and this is stored in a storage device (not shown). You may make it store in.

In addition, the rotation amount θ of the wafer W after loading can be estimated by a calculation formula θ = (θ _C + θ _q −θ _p ). Here, this rotation amount θ is also calculated and stored together in a storage device (not shown). Note that θ _C is a rotational component of the wafer W measured by the pre-2LA measurement.

In the next step 107, search alignment is performed. Here, wafer stage WST is moved in the XY plane so that the search alignment mark formed on wafer W is positioned below alignment detection system AS. At this time, wafer stage WST is moved to A position obtained by correcting the position alignment vector D ₂ of the wafer W (or a vector obtained by rotating the vector by the rotation amount α) and the wafer rotation amount θ to the design position coordinates of the search alignment mark. . Then, here, the position information (positional deviation amount and position information) of wafer W on wafer stage WST is calculated from the difference between the measured position information of the search alignment mark calculated from the imaging result of alignment detection system AS and the design position information. Rotation amount). Note that the rotation amount of the wafer W calculated by the search alignment is canceled by the rotation of the reticle stage WST.

  In the next step 109, wafer alignment is performed. That is, in consideration of the result of search alignment, the position of alignment marks attached to a plurality of sample shot areas on the wafer W is measured by an alignment detection system AS (not shown), and a statistical processing method is performed based on the measurement result. The so-called EGA calculation is performed to calculate the array coordinates of all shot areas. Thereby, arrangement coordinates on the XY coordinate system of all shot areas on the wafer W are calculated. Since this process is disclosed in, for example, Japanese Patent Laid-Open No. 61-44429, detailed description thereof is omitted.

  In the next step 111, 1 is set to the counter p indicating the array number of the shot area, and the first shot area is set as the exposure target area.

  In step 113, the pattern area of the reticle R is illuminated with exposure light IL from an illumination system (not shown) based on the array coordinates of the exposure target area calculated by the EGA calculation, and exposure is performed. Thereby, the pattern of the reticle R is reduced and transferred to the exposure target area on the wafer W via the projection optical system PL.

  In step 115, the value of the counter p (hereinafter referred to as “counter value p”) is referred to, and it is determined whether or not exposure has been performed on all shot areas. Here, the counter value p is 1, and only the first shot area has been exposed, so the determination in step 115 is negative and the routine proceeds to step 117.

  In step 117, the counter value p is incremented (p ← p + 1), the next shot area is set as the exposure target area, and the process returns to step 113. Thereafter, the processing and determination of step 113 → step 115 → step 117 are repeated until the determination in step 115 is affirmed.

  When the transfer of the pattern to all the shot areas on the wafer W is completed, the determination in step 115 is affirmed and the process proceeds to step 119.

  In step 119, the wafer W is unloaded. The wafer W is unloaded by the unload slider 62 and returned to the FOUP 27 by the unload robot 93 or is transferred to a coater / developer (C / D) connected inline by a transfer system (not shown). In the next step 121, it is determined whether or not the exposure of all the wafers W in the lot has been completed. If this determination is negative, the process returns to the subroutine 101, and if the determination is positive, the process is terminated. Here, since the exposure of the first wafer W has only been completed, the determination is denied and the process returns to the subroutine 101. Thereafter, until the determination in step 121 is affirmed, the processing of the subroutines 101 to 121 is repeated, and the wafer W in the FOUP 27 is exposed.

In this embodiment, after the wafer stage WST is moved to the estimated load position estimated by the positional deviation amount D ₁ in step 215 in FIG. 11, the pre-3 measurement process is performed in the subroutine 217 in FIG. and the wafer position deviation amount D ₂ of W on wafer stage WST was to absorb the positional deviation by correcting the design position coordinates of the search alignment marks, but is not limited thereto. For example, before the wafer stage WST is moved to the estimated load position, the pre-3 measurement process is performed to obtain the positional deviation amount D ₁ + D ₂ , and the wafer stage WST is moved to the estimated loading position estimated by the positional deviation amount. Then, the wafer W may be loaded. Further, the position of wafer stage WST at the time of loading wafer W is always set to the same load position (that is, origin O), and the position coordinate of the search alignment mark is corrected in consideration of the positional deviation amount D ₁ + D _2. May be. An estimate of the loading position of the wafer stage WST determines only the estimated load position translation component W _A, the rotation component Dt _1, similar to the rotary component Dt _2, included in the correction of the position coordinates of the search alignment mark You may do it. At this time, the rotational component vector of the positional deviation amount is a vector Dt _LP shown in FIG. The equation for calculating the vector Dt _LP is as follows.

ここで、Ｄｔ_LPxはＸ成分であり、Ｄｔ_LPyはＹ成分である。このときのサーチアライメントマークの位置座標の補正ベクトルは、Ａ_q＋Ｄｔ_LPとすれば良い。

Here, Dt _LPx is the X component, and Dt _LPy is the Y component. The correction vector for the position coordinate of the search alignment mark at this time may be A _q + Dt _LP .

  In the present embodiment, since the position of the wafer W is estimated only by the difference in position information in the same coordinate system (pre-2LA camera coordinate system, pre-3 camera coordinate system and wafer coordinate system (XY coordinate system)), the wafer coordinates Regardless of whether or not the positional relationship between the system and the pre-2TA camera coordinate system and the pre-3 camera coordinate system is partially unknown, prealignment of the wafer W can be realized. In this way, it is not necessary to completely obtain the positional relationship between the pre-2TA camera coordinate system and the pre-3 camera coordinate system in the wafer coordinate system, so that the adjustment time of the exposure apparatus 100 can be shortened.

In the present embodiment, by using a tool wafer W _S, without strictly performing calibration for each camera coordinate system, although a method for performing pre-alignment of the wafer W with high accuracy, not limited to this, the wafer Of course, if the relationship between the coordinate system and each camera coordinate system is known, the relative position relationship between the mark 50M and the wafer W may be directly obtained to estimate the load position of the wafer W. Also in this case, it is desirable to estimate the estimated load position of the wafer W and the correction amount of the position coordinates of the search alignment mark separately for the parallel component and the rotation component.

  Note that the calculation formulas in the present embodiment have been described on the assumption that the imaging results of the pre-alignment device 45 are all normal images in order to simplify the description. However, as described above, the imaging results are actually mirror images. In this case, it is necessary to consider the rotation direction of the rotation matrix and the reversal of the position coordinates in the coordinate transformation equation such as the above equation (4).

  As described above in detail, according to the present embodiment, when the mark is imaged by the mark detection system 42 and the pre-alignment device 45, at least one of the index patterns T1 to T11 is included in the imaging result. A pattern is always included. Since the index patterns T1 to T11 are different from each other, it is clear based on this index pattern which part of the mark 50M the imaging field of view of the mark detection system 42 and the pre-alignment device 45 captures. Become. Thereby, even if the imaging result of the mark 50M captures only a part of the mark 50M, the position information of the mark 50M can be detected. If such a mark 50M is used, the mark 50M can be captured by the mark detection system 42 and the pre-alignment device 45 even if the mounting accuracy is not strictly defined. Therefore, it is possible to detect the position information of the mark 50M with high accuracy in a short time without performing a complicated operation of strictly attaching the mark detection system 42 and the pre-alignment device 45.

  In addition to the index pattern, the mark 50M of the present embodiment includes L / S patterns LSx and LSy as measurement patterns having a shape capable of measuring position information with high accuracy. In the present embodiment, the index patterns T1 to T11 are for roughly detecting the position of the mark 50M, and the final position of the mark 50M is detected by the L / S patterns LSx and LSy. The position information of the mark 50M can be detected with high accuracy. Therefore, it is possible to detect the position information with higher accuracy than a mark having only a pattern that acts as a scale indicating which part of the mark the imaging field of view of the imaging apparatus captures.

  Further, in the present embodiment, the index patterns T1 to T11 are detected by template matching, but the index patterns T1 to T11 are patterns composed of combinations of a plurality of types of patterns. In this way, a part of the index pattern can be shared, or a part of the template matching process can be shared, so that the storage capacity of the storage device for storing the template can be reduced and the process can be performed. Can be speeded up. Note that the patterns used for the combination of index patterns do not necessarily have to combine patterns of different types, and are not limited to the number of combinations. Moreover, the combination method may be multilayer. For example, the patterns m1 to m3 used for the combination of the index patterns can be a combination of smaller rectangular marks.

  In the present embodiment, since the index patterns T1 to T11 are arranged in the Y-axis direction, they function as scale marks. In this way, it is possible to grasp the distance to the mark center from the imaging results of the mark detection system 42 and the pre-alignment device 45. There is no limitation on the arrangement direction of the index pattern, and the indicator pattern may be arranged in a plurality of directions, such as the X-axis direction alone, the XY both-axis directions, or the directions intersecting with them.

  In the present embodiment, the mark 50M includes an L / S pattern LSx having a shape capable of detecting position information regarding the X-axis direction and an L / S pattern LSy having a shape capable of detecting position information regarding the Y-axis direction. Two are arranged. In this way, position information of the mark 50M in the XY two-dimensional plane can be detected. In the present embodiment, the area where the L / S pattern LSx is arranged and the area where the L / S pattern LSy are arranged are divided, and the index patterns T1 to T11 are arranged therebetween. . In this way, since the index patterns T1 to T11 can be arranged near the center of the mark 50M, the index patterns T1 to T11 can be easily captured in the imaging field of view of the imaging device.

  In the present embodiment, the mark 50M is arranged such that the L / S pattern LSx is sandwiched between the L / S patterns LSy. In this way, the mark 50M has a symmetrical shape with respect to the X-axis direction, and the projection waveform obtained by the luminance integration also has a symmetrical waveform with respect to the center of the mark 50M. Accordingly, the approximate center position of the mark 50M can be detected by a relatively simple process such as, for example, an inverse symmetry correlation calculation. Therefore, the position information of the mark 50M can be detected in a short time with high accuracy. it can. Of course, the mark 50M may be symmetrical in the Y-axis direction. Further, if the two measurement patterns for measuring the position in the same direction as the mark 50M are arranged away from each other, the rotation amount of the mark 50M can be detected with high accuracy.

  In this embodiment, since the measurement pattern is an L / S pattern in which line portions and space portions are regularly arranged, position information with high reliability can be detected.

  Further, in the present embodiment, at least three of the imaging result of the L / S pattern LSx and the imaging result of the L / S pattern LSy are extracted, and a representative point (center point) in the extraction result is detected, Least square approximate straight lines in the X-axis direction and Y-axis direction based on these representative points are obtained, and the intersection of these straight lines is specified as the position of the mark 50M. In this way, the position of the mark 50M can be specified with high accuracy using a statistical method. Note that it is not always necessary to extract three or more imaging results of the measurement pattern, and one may be used. In one case, the representative point of the extraction result may be the position of the mark 50M, and in two cases, the position of the mark 50M may be specified based on a straight line connecting the representative points of the extraction results.

  In the above embodiment, in the measurement of the mark 50M when the wafer W is actually loaded, the template selected at the time of initial measurement is given priority in principle based on the reproducibility of the position control of the load slider. Then, the index pattern is searched. In this way, the time required for the search can be shortened.

  As described above, in the exposure apparatus 100 of the present embodiment, the wafer W is transferred by the load slider 50 on which the mark 50M is formed, and after pre-alignment, the wafer W is loaded on the wafer stage WST. At the time of loading, the result of pre-alignment is reflected, and the relative position between wafer stage WST and load slider 50 at the time of wafer W delivery and the position information of wafer W held on wafer stage WST are calculated. Since at least one of them can be adjusted with high accuracy, high-accuracy exposure can be realized.

  In the above embodiment, the mark 50M includes two measurement patterns, that is, the L / S pattern LSx and the L / S pattern LSy. However, the measurement pattern includes position information in at least one direction. There should be a measurement pattern that can be measured. When it is necessary to detect the position of the mark 50M in the XY plane as in the above embodiment, a one-dimensional measurement pattern for each axial direction may be provided on the load slider 50.

  The index patterns T1 to T11 do not have to be arranged in one direction. In short, it is only necessary that any one of the index patterns be arranged within the imaging field of view of the imaging apparatus. For example, a plurality of index patterns may be arranged radially from the center of the mark.

  In the above embodiment, only the −X side index patterns T1 to T11 of the mark 50M are searched. However, only the + X side index patterns T1 to T11 may be searched, or both side indexes. Each pattern may be searched.

  In the above embodiment, the least-squares approximation line is obtained after converting the position coordinates of the representative points of the measurement ranges m and n on the L / S patterns LSx and LSy to the position coordinates of the mark coordinate system. Based on the position coordinates on each camera coordinate system, the least square approximation line of each of the XY axes is obtained, and the intersection point is obtained, and then the position coordinate of the intersection point is converted into the position coordinate of the mark coordinate system. May be.

  In the above embodiment, the mark 50M is disposed in the vicinity of the −X side end portion of the load slider 50. However, the mark 50M is not limited to this, and may be disposed in the substantially central portion of the arm portion, You may make it arrange | position at an edge part.

  In the above embodiment, the remaining rotation amount of the wafer W on the wafer stage WST, that is, the rotation amount of the wafer W detected by the search alignment is corrected by the rotation of the reticle stage RST. If the rotation range of θz of reticle stage RST is small and the amount of rotation cannot be canceled sufficiently, rotation of wafer table W can be performed by rotation of center table CT or rotation of reticle stage RST and center table CT. You may make it cancel rotation amount. Further, wafer stage WST itself may be rotated. Instead of rotating at least one of center table CT, wafer stage WST, and reticle stage RST, or in combination with this, load slider 50 may be rotated slightly before loading.

  In the above embodiment, the wafer is loaded and unloaded by moving the center table CT up and down. For example, the center table CT may be fixed, and a part of the wafer stage WST (such as a wafer holder) may be moved up and down. good. Furthermore, in the above embodiment, the center table CT may have three pins. In the above-described embodiment, the center table CT is provided on the wafer stage WST. However, the center table CT is not necessarily provided, and the exposure apparatus that loads and unloads the wafer without using the center table CT. The present invention can be applied. For example, as disclosed in Japanese Patent Application Laid-Open No. 11-284052 and the like, a gap is formed in which a load slider or an unload slider enters by cutting out two portions of the wafer holder, and the load slider or the unload slider is moved up and down in the gap. A method of loading and unloading the wafer W by moving it may be adopted. In this exposure apparatus, the rotational deviation amount of the wafer W may be canceled by rotating at least one of the load slider 50, the wafer stage WST, and the reticle stage RST.

  In the above embodiment, the load slider 50 is movable only in the Y-axis direction. However, the load slider 50 may be capable of adjusting the positions in the X-axis direction, the Z-axis direction, and the θz direction. In this case, even without adjusting the position of wafer stage WST and rotating the center table CT, the load slider 50 is moved from its reference position in the X-axis, Y-axis direction, and θz direction prior to wafer transfer. By fine movement, the above-described positional deviation amount and rotational deviation amount of the wafer W may be canceled. Whether to adjust with the load slider 50 or with the wafer stage WST or the like may be selected in consideration of the superiority or inferiority of the positioning accuracy.

  As described in the above embodiment, the mark 50M on the load slider 50 can be detected by either the epi-illumination system or the transmission illumination system. However, the outer shape of the wafer can be detected by applying illumination from the back surface of the wafer. It is desirable to detect. In the above embodiment, since the mark and the outer shape of the wafer are not imaged at the same time, there is an effect that one illumination system can prevent the other imaging result from being adversely affected.

  In addition, the load slider 50 in the above embodiment needs to have a shape that avoids the imaging regions of the imaging devices VA to VE. However, the image processing accuracy is high, and the outer shape of the wafer can be accurately detected by epi-illumination. In this case, the load slider may not have a shape that avoids the imaging region. Further, if the reflection characteristic of the load slider with respect to the light is made significantly different from the reflection characteristic of the wafer, the load slider fingers can be shaped to fit in the imaging areas VA to VE of the pre-alignment device 45. good. In short, the shape of the load slider 50 is not limited to the above embodiment, and may be arbitrary. The shape and structure of the transfer member is not limited to the load slider 50 of the above-described embodiment. For example, the transfer member may be configured to transfer a wafer while hanging it.

  Moreover, although the case where the wafer with a notch was processed was demonstrated in the said embodiment, it cannot be overemphasized that this invention is applicable also when processing a wafer with an orientation flat.

  In the above-described embodiment, when the wafer pre-alignment is performed, the outer shape of the wafer is detected at three locations. Only one location may be detected in a wide imaging range (a range that always includes a notch or the like), or all edges of the regions VA to VE may be detected.

  Further, in the above embodiment, the sensor used for detecting the mark 50M, such as the pre-alignment device 45 and the mark detection system 42, is not limited to the imaging device, and for example, a light amount sensor may be used. The same applies to the mark detection system 42. Furthermore, in the above-described embodiment, the pre-1 measurement step is performed. For example, when the wafer can be held on the turntable 51 with a relatively small amount of deviation between the center of the wafer and the rotation center of the turntable 51, The pre-1 measurement process may not be performed.

  In the above embodiment, the illumination device is disposed below the wafer W and the imaging device is disposed above the wafer W. However, this may be reversed.

  In the above-described embodiment, the exposure apparatus 100 includes one wafer stage. However, for example, the exposure includes two wafer stages as disclosed in International Publications WO98 / 24115 and WO98 / 40791. The present invention can also be applied to an apparatus. Since the wafer stage loaded with the wafer is often moved from the loading position to the alignment position prior to the transfer position, the loading position may be determined in consideration of the arrangement of the alignment detection system AS. preferable.

  The above embodiment relates to the pre-alignment of the wafer W, but it is of course applicable to the alignment of the reticle R. When the parts of the exposure apparatus such as the wafer holder are automatically replaced. Can also be applied.

  The exposure apparatus of the above embodiment can be a reduced projection exposure apparatus of any of the step-and-scan method, the step-and-repeat method, and the step-and-stitch method. Further, the present invention can be applied to an exposure apparatus such as a proximity system, a mirror projection aligner, and a photo repeater.

The present invention is not limited to the exposure light source. As a light source of an illumination system (not shown) that emits the exposure light IL, far ultraviolet light such as KrF excimer laser light source (oscillation wavelength 248 nm), ArF excimer laser light source (oscillation wavelength 193 nm), or F ₂ laser light source (oscillation wavelength 157 nm). Those emitting light such as vacuum ultraviolet light can be used. It is also possible to use an ultra-high pressure mercury lamp that generates ultraviolet emission lines (g-line, i-line, etc.). Further, it may be used other vacuum ultraviolet light source such as Ar ₂ laser light source (output wavelength 126 nm). In addition, for example, the laser light source output from each of the above light sources as vacuum ultraviolet light is not limited to the laser light source output from a DFB (Distributed Feedback) semiconductor laser or a fiber laser. A light source that irradiates, as illumination light, harmonics that are amplified with a fiber amplifier doped with, for example, erbium (Er) (or both erbium and ytterbium (Yb)) and converted into ultraviolet light using a nonlinear optical crystal. It may be used. In addition, the present invention may be applied to an exposure apparatus that uses EUV light, X-rays, or charged particle beams such as electron beams and ion beams as exposure beams. Furthermore, the present invention may be applied to an immersion type exposure apparatus disclosed in, for example, International Publication WO99 / 49504 and the like, in which a liquid is filled between the projection optical system PL and the wafer W.

  Moreover, although the said embodiment demonstrated the case where this invention was applied to the exposure apparatus for semiconductor manufacture, it is not restricted to this, For example, the exposure for liquid crystals which transfers a liquid crystal display element pattern to a square-shaped glass plate The present invention can be widely applied to an apparatus, an exposure apparatus for manufacturing a thin film magnetic head, an image sensor, a micromachine, an organic EL, a DNA chip, and the like.

  An illumination optical system and a projection optical system composed of a plurality of lenses are incorporated into the exposure apparatus main body, optical adjustment is performed, and a reticle stage or wafer stage consisting of a large number of mechanical parts is attached to the exposure apparatus main body to provide wiring and piping. , And further performing general adjustment (electrical adjustment, operation check, etc.), the exposure apparatus of the above embodiment can be manufactured. The exposure apparatus is preferably manufactured in a clean room where the temperature, cleanliness, etc. are controlled.

  Further, in order to manufacture reticles or masks used in not only microdevices such as semiconductor elements but also light exposure apparatuses, EUV exposure apparatuses, X-ray exposure apparatuses, electron beam exposure apparatuses, etc., glass substrates or silicon wafers, etc. The present invention can also be applied to an exposure apparatus that transfers a circuit pattern. Here, in an exposure apparatus using DUV (far ultraviolet) light, VUV (vacuum ultraviolet) light, or the like, a transmission type reticle is generally used. As a reticle substrate, quartz glass, fluorine-doped quartz glass, meteorite, Magnesium fluoride or quartz is used.

  Furthermore, the transport method of the present invention can be preferably applied to an apparatus other than the exposure apparatus such as an inspection apparatus or a processing apparatus as long as it is an apparatus that requires positioning accuracy of an object after transport.

<Device manufacturing method>
Next, an embodiment of a device manufacturing method using the exposure apparatus 100 described above in a lithography process will be described.

  FIG. 28 shows a flowchart of a manufacturing example of a device (a semiconductor chip such as an IC or LSI, a liquid crystal panel, a CCD, a thin film magnetic head, a micromachine, etc.). As shown in FIG. 28, first, in step 701 (design step), device function / performance design (for example, circuit design of a semiconductor device) is performed, and pattern design for realizing the function is performed. Subsequently, in step 702 (mask manufacturing step), a mask on which the designed circuit pattern is formed is manufactured. On the other hand, in step 703 (wafer manufacturing step), a wafer is manufactured using a material such as silicon.

  Next, in step 704 (wafer processing step), using the mask and wafer prepared in steps 701 to 703, an actual circuit or the like is formed on the wafer by lithography or the like, as will be described later. Next, in step 705 (device assembly step), device assembly is performed using the wafer processed in step 704. Step 705 includes processes such as a dicing process, a bonding process, and a packaging process (chip encapsulation) as necessary.

  Finally, in step 706 (inspection step), inspections such as an operation confirmation test and a durability test of the device created in step 705 are performed. After these steps, the device is completed and shipped.

  FIG. 29 shows a detailed flow example of step 704 in the semiconductor device. In FIG. 29, in step 711 (oxidation step), the surface of the wafer is oxidized. In step 712 (CVD step), an insulating film is formed on the wafer surface. In step 713 (electrode formation step), an electrode is formed on the wafer by vapor deposition. In step 714 (ion implantation step), ions are implanted into the wafer. Each of the above steps 711 to 714 constitutes a pretreatment process at each stage of the wafer processing, and is selected and executed according to a necessary process at each stage.

At each stage of the wafer process, when the above pre-process is completed, the post-process is executed as follows. In this post-processing step, first, in step 715 (resist formation step), a photosensitive agent is applied to the wafer. Subsequently, in step 716 (exposure step), the circuit pattern of the mask is transferred to the wafer using the exposure apparatus 100 of the above embodiment. Next, in step 717 (development step), the exposed wafer is developed,
In step 718 (etching step), the exposed member in a portion other than the portion where the resist remains is removed by etching. In step 719 (resist removal step), the resist that has become unnecessary after the etching is removed.

  By repeatedly performing these pre-processing steps and post-processing steps, multiple circuit patterns are formed on the wafer.

  If the device manufacturing method of this embodiment described above is used, the exposure apparatus 100 and the exposure method of the above embodiment are used in the exposure step (step 716), so that throughput can be improved and high-precision exposure is realized. can do. As a result, it becomes possible to improve the productivity (including yield) of highly integrated devices.

  As described above, the mark of the present invention is suitable for detecting the position information, and the position detection method according to the present invention is suitable for detecting the position information of the object. The transfer apparatus and method are suitable for transferring the substrate so that the substrate is held at a predetermined position on the stage and delivering it to the stage. The exposure apparatus of the present invention includes a semiconductor element, a liquid crystal display element, etc. The device manufacturing method of the present invention is suitable for the production of microdevices.

It is a longitudinal cross-sectional view of the exposure apparatus which concerns on one Embodiment of this invention. It is a cross-sectional view of the exposure apparatus which concerns on one Embodiment of this invention. 3A is a top view showing the structure of the load slider, and FIG. 3B is a perspective view showing the structure of the load slider. It is a figure which shows an example of a mark. It is a figure which shows an example of an index pattern. It is a figure which shows the correspondence of an index pattern and a scale. It is a perspective view which shows the structure of a pre-alignment system. It is a block diagram which shows the structure of a control system. It is a figure which shows the relationship of each coordinate system. It is a flowchart which shows the exposure operation | movement of one Embodiment of this invention. It is a flowchart which shows the subroutine 101 of FIG. 12 is a flowchart illustrating a subroutine 213 and a subroutine 217 in FIG. 11. 12 is a flowchart (part 1) illustrating a subroutine 305 in FIG. 12 is a flowchart (No. 2) showing subroutines 305 and 310 in FIG. 12 is a flowchart showing a subroutine 310 of FIG. It is a figure which shows an example of a tool wafer. FIG. 17A is a diagram illustrating an example of a mark imaging result, and FIG. 17B is a diagram illustrating a projection waveform. FIG. 18A is a diagram showing a correlation calculation window used for inversion symmetry correlation degree calculation, and FIG. 18B is a diagram showing a state in which there is a correlation calculation window at the calculation start position. 18 (C) is a diagram showing an inverted symmetry correlation function C (x). FIG. 19A is a diagram (part 1) illustrating a mark position center calculation method, and FIG. 19B is a diagram (part 2) illustrating a mark position center calculation method. ) Is a diagram (No. 3) illustrating the method of calculating the center of a mark position. 20A is a diagram (part 4) illustrating the mark position center calculation method, and FIG. 20B is a diagram (part 5) illustrating the mark position center calculation method. ) Is a diagram (No. 6) illustrating the method for calculating the center of a mark position. FIG. 21A is a diagram (part 7) illustrating the mark position center calculation method, and FIG. 21B is a diagram illustrating the mark position center calculation method (part 8). FIG. 22A is a diagram (No. 9) showing a mark position center calculation method, and FIG. 22B is a diagram (No. 10) showing a mark position center calculation method. FIG. 23A is a diagram showing the relationship between the center position of the tool wafer during pre-alignment and the position of the mark on the load slider, and FIG. 23B shows the position of the mark on the load slider during loading, It is a figure which shows the relationship with the center position of a tool wafer. FIG. 24A is a diagram showing the relationship between the center position of the wafer at the time of pre-alignment and the position of the mark on the load slider, and FIG. It is a figure which shows the relationship with the estimated position of the loading position of wafer stage WST. FIG. 25A is a vector diagram illustrating an example of a parallel component of wafer positional deviation, and FIG. 25B is a vector diagram illustrating an example of a rotational component of wafer positional deviation. FIG. 26A is a vector diagram illustrating an example of a parallel component of the residual positional deviation of the wafer, and FIG. 26B is a vector diagram illustrating an example of a rotational component of the residual positional deviation of the wafer. It is a figure which shows roughly the rotation component of the whole position shift of a wafer. It is a flowchart for demonstrating embodiment of the device manufacturing method which concerns on this invention. It is a flowchart which shows the detail of step 704 of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 20 ... Main control apparatus (transfer apparatus, detection apparatus, correction apparatus), 42 ... Mark detection system, 45 ... Pre-alignment apparatus, 50 ... Load slider (object, conveyance member), 50M ... Mark, 51 ... Turntable, 52 ... Pre-alignment stage, 60 ... Y drive mechanism, 81A to 81G1, 81G2 ... illumination device, 83A, 83B ... line sensor, 100 ... exposure device, Ar1 ... first region, Ar2 ... second region, LSx, LSy ... L / S Pattern (pattern for measurement), PL ... projection optical system, RST ... reticle stage, T1 to T11 ... index pattern, W ... wafer (substrate), WST ... wafer stage (stage).

Claims (16)

A mark provided on an object and used for detecting position information of the object,
The first area where the measurement pattern having a shape capable of measuring position information is arranged and the positional relationship between the measurement pattern and the plurality of index patterns of different types are arranged in different locations. And the mark as a whole has a mark area larger than the imaging field of view of the predetermined imaging device,
A mark in which the plurality of index patterns are arranged so that at least one index pattern among the plurality of index patterns is included in the imaging result when the imaging device captures an image of the mark area.   The mark according to claim 1, wherein each of the plurality of index patterns is formed from a combination of a plurality of types of marks. In the second region,
The mark according to claim 1, wherein the plurality of index patterns are arranged in at least one direction. In the first region,
The first pattern having a shape capable of measuring position information related to a predetermined direction and the second pattern having a shape capable of measuring position information related to a direction intersecting the predetermined direction are arranged. The mark as described in any one of 1-3. The first region is
Divided into at least one region in which the first pattern is formed and at least one region in which the second pattern is formed;
The mark according to claim 4, wherein the second region is arranged between a region where the first pattern is formed and a region where the second pattern is formed. The region where the second pattern is formed is arranged so as to be sandwiched between two regions where the first pattern is formed,
The mark according to claim 5, wherein the index patterns in the second region are arranged in the predetermined direction.   The mark according to claim 4, wherein at least one of the first pattern and the second pattern is a line and space pattern. A conveying member provided to convey a predetermined object and provided with the mark according to any one of claims 1 to 7;
At least one imaging device capable of imaging the mark when the conveying member conveys the object;
A detection device that detects position information of the mark when the conveying member conveys the object based on an imaging result of the imaging device;
And a correction device that corrects position information of the object based on the detection result. An exposure apparatus for transferring a device pattern onto a substrate,
A transport apparatus according to claim 8;
A stage for holding the substrate delivered from the transfer device;
An exposure apparatus comprising: a transfer device that transfers the device pattern onto a substrate held on the stage. A position detection method for detecting position information of an object,
An imaging step of imaging the mark according to claim 1 provided on the object by an imaging device;
A specifying step of specifying the position of the mark based on the imaging result;
A detection step of detecting position information of the object based on the identification result. The specific process includes
A first sub-step of detecting at least one index pattern of the plurality of index patterns based on the imaging result;
A second sub-process for extracting an imaging result corresponding to at least a part of the measurement pattern from the imaging result based on the detection result;
The position detection method according to claim 10, further comprising: a third sub-step of detecting position information of the mark based on the extraction result. The measurement pattern includes a first pattern for detecting position information related to a predetermined direction, and a second pattern for detecting position information related to a direction intersecting the predetermined direction,
In the second sub-step,
12. A plurality of imaging results respectively corresponding to a plurality of different parts on the first pattern and a plurality of imaging results respectively corresponding to a plurality of different parts on the second pattern are extracted. The position detection method as described in. In the second sub-step,
Extracting at least one of at least one of imaging results corresponding to a plurality of different parts on the first pattern and imaging results corresponding to a plurality of different parts on the second pattern;
In the third sub-process,
The position detection method according to claim 12, wherein position information of the mark is detected using a least square method. In the first sub-step,
The position detection method according to any one of claims 10 to 13, wherein search for index patterns detected in the past is prioritized. A transport method for transporting the substrate so that the substrate is held at a predetermined position on the stage and delivering it to the stage,
Using the position detection method according to any one of claims 10 to 14, detecting the position information of a conveyance member provided with a mark when the substrate is conveyed by the conveyance member;
A method of adjusting at least one of a relative position between the stage and the transport member when the substrate is transferred and position information of the substrate held on the stage based on the detection result. . A device manufacturing method for manufacturing a device, comprising:
Using the transport method according to claim 15, delivering the substrate to a stage movable between a substrate delivery position and a device pattern transfer position via the projection optical system;
Transferring the device pattern onto a substrate held on the stage.

Images