JP2010034331A - Exposure device and device manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce variation of overlay accuracy due to variation of thickness of a substrate. <P>SOLUTION: Reference marks include a plurality of reference marks arranged at different positions in a second direction orthogonal to an optical axis direction and a first direction, and in the optical axis direction. A control means moves a stage in the optical axis direction and the second direction while keeping a rotational angle constant in accordance with a measurement value of a first interferometer, causes a first measurement means to measure the respective positions of the plurality of reference marks, moves the stage in the optical axis direction and the second direction while keeping the rotational angle constant in accordance with a measurement value of a second interferometer to cause a second measurement means to measure the respective positions of the plurality of reference marks, and calculates an amount for correcting the position of the stage in a plane in accordance with the position of the stage in the optical direction based on the respective positions of the plurality of reference marks measured by the first measurement means and those measured by the second measurement means. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention includes a stage that holds a substrate, an exposure station that includes a projection optical system, and a measurement station that measures the position of a mark formed on the substrate held on the stage, via the projection optical system. The present invention relates to an exposure apparatus that exposes a substrate held on the stage.

  In recent years, along with miniaturization and high integration of semiconductor integrated circuits such as ICs and LSIs and liquid crystal panels, exposure apparatuses such as semiconductor exposure apparatuses have become highly accurate and highly functional. In particular, in aligning an original plate such as a mask or reticle and a substrate such as a semiconductor wafer, a technique for superposing the original plate and the substrate on the order of several nanometers is required. As an exposure apparatus used for manufacturing such a device such as a semiconductor integrated circuit, an exposure apparatus called a stepper or a scanner is often used. These exposure apparatuses sequentially transfer a pattern formed on an original plate (eg, a reticle) to a plurality of locations on the substrate while stepping the substrate (eg, a semiconductor wafer). An exposure apparatus that performs this transfer all at once is called a stepper, and an exposure apparatus that performs transfer while scanning the substrate stage is called a scanner.

Next, the alignment between the original plate and the substrate in the exposure apparatus will be described.
For aligning the original plate and the substrate in the exposure apparatus, there is a die-by-die alignment method in which an exposure position is measured for each exposure to perform alignment (alignment).
Furthermore, there is a global alignment method in which position measurement is performed at an appropriate number of measurement points in advance, and an exposure position correction formula is created from the measurement result to perform alignment.
The global alignment method is a method that provides high throughput and high accuracy, and performs alignment according to the same correction formula for the entire area of the substrate. It has advantages in terms of usability, such as being able to judge. TTL (Through-the-lens) projection that measures the position of the alignment mark via a projection optical system as a detection method of the alignment target itself or an alignment mark arranged in the vicinity thereof in order to perform alignment There is an optical system.
Furthermore, there is an OA (off-axis) system that directly measures the position of the alignment mark without using a projection optical system.
As described above, miniaturization of ICs and LSIs is progressing at an accelerating rate, and higher device performance is required year by year in semiconductor manufacturing apparatuses. In recent years, there has been a great demand for improvement in productivity accompanying an increase in demand for semiconductors typified by DRAMs, and semiconductor manufacturing apparatuses are required not only to improve accuracy but also to improve throughput. For this reason, an exposure apparatus having a so-called two-stage configuration that individually has a measurement station that measures a pattern position on a wafer and an exposure station that performs exposure on the wafer and performs measurement processing and exposure processing in parallel is disclosed in, for example, Patent Literature 2 is proposed.

With reference to FIG. 10, a conventional two-stage exposure apparatus will be described.
This conventional example includes a measurement station 2, an exposure station 3, and an apparatus control unit 1. The measurement station 2 is a station that measures the relative positional relationship between the stage 6a that is a substrate stage and the pattern on the wafer 4a that is a substrate to be exposed. The exposure station 3 is a station for projecting and exposing the pattern of the reticle 12 onto the wafer 6b after measuring the relative positional relationship between the reticle 12 as an original and the stage 6b as a substrate stage.
The apparatus control unit 1 is a control unit that controls the measurement station 2 and the exposure station 3. The measurement station 2 includes a wafer 4a and an OA alignment detection system 5. The stage 6a mounts the wafer 4a, and the position is measured by the interferometer 7a to position the wafer 4a. The height detector 8a and the reference mark 9a are placed on the stage 6a. Interferometer bar mirror 10a-1 is a mirror that reflects light emitted from interferometer 7a. The interferometer bar mirror 10a-2 is a mirror that reflects light emitted from the interferometer 7b when the stage 6a moves to the exposure station 3 side.

Next, the exposure station 3 includes a projection optical system 11, a TTL alignment detection system 13a, 13b, and an illumination optical system 14 that project an image of the reticle 12 onto the wafer 4b. The stage 6b mounts the wafer 4b, is position-measured by the interferometer 7b, positions the wafer 4b, and the reference mark 9b is placed on the stage 6b. Interferometer bar mirror 10b-2 is a mirror that reflects light emitted from interferometer 7b. The interferometer bar mirror 10b-1 is a mirror that reflects light emitted from the interferometer 7a when the stage 6b moves to the measurement station 2 side.
Here, for simplification of explanation, only the interferometers 7a and 7b for controlling the Y-direction interferometer and the Z-direction moving axis are shown, but in reality, the stage has six axes (X, Y, Z, ωX, ωY, ωZ) are controlled.

In this conventional example, the reticle and wafer are aligned in the following procedure.
Usually, in an exposure apparatus of a two-stage mechanism, when the stage swap is performed, the interferometer that has measured the position of the stage is switched. Therefore, when the stage moves from the measurement station 2 to the exposure station 3, the position between the exposure center and the stage Relationship is lost.
Similarly, when moving from the exposure station 3 to the measurement station 2, the positional relationship of the stage with respect to the apparatus is lost. In order to accurately expose the pattern of the reticle 12 on the wafer 4a, it is necessary to obtain the relative positional relationship between the exposure center and the alignment mark on the wafer 4a.
Here, in the exposure station 3, the reference marks 9a on the stage 6a and the alignment marks on the wafer 4a are measured by the TTL alignment detection systems 13a and 13b, and the relative positional relationship between the exposure center and the alignment marks on the wafer 4a is determined. I can't ask for it. This is because the exposure light is used for the measurement using the TTL alignment detection systems 13a and 13b, and the wafer 4a is exposed.
In addition, even when using non-exposure light, direct alignment of the wafer inevitably degrades the image quality of the alignment mark due to the process on the wafer surface, and in order to maintain accuracy, it is necessary to perform multipoint measurement. Throughput deteriorates and does not hold.

Therefore, in the conventional example, the relative positional relationship between the exposure center and the alignment mark on the wafer 4a is obtained as follows. First, the measuring station 2 uses the OA alignment detection system 5 to measure the reference mark 9a on the stage 6a, and aligns the measurement center with the reference mark 9a. Next, the relative positional relationship between the reference mark 9a and the alignment mark on the wafer 4a is calculated by measuring the alignment mark on the wafer 4a using the OA alignment detection system 5 with the position as the reference (origin).
Next, after stage swapping, the exposure station 3 uses the TTL alignment detection systems 13a and 13b to measure the reference mark 9a on the stage 6a via the reticle 12, and aligns the exposure center with the reference mark 9a. Do.
Further, the relative position relationship between the exposure center and the alignment mark on the wafer 4a is calculated from the relative position relationship between the reference mark 9a obtained by the measurement station 2 and the alignment mark on the wafer 4a using the position as a reference (origin).
In this conventional example, the processing of the measurement station 2 and the exposure station 3 can be performed in parallel, and the total processing time can be shortened by combining precise alignment and wafer exposure processing.
JP-A-9-218714 Japanese Patent Laid-Open No. 10-16309

In the process of manufacturing a semiconductor integrated circuit, the thickness of the wafer varies depending on various processes (etching, sputtering, CMP, etc.) as well as variations in thickness due to manufacturing errors of the wafer itself. In this case, the height of the stage when observing the alignment mark on the wafer during wafer alignment and the height of the stage when observing the reference mark are different for each wafer.
On the other hand, the Z drive control in the measurement station 2 and the exposure station 3 is performed by interferometers 7a and 7b using two laser beams separated in the Z direction, respectively, as in the interferometers 7a and 7b shown in FIG. This is achieved by controlling the inclinations of the mirrors 10a-1 and 10b-2 to be constant.
Therefore, Z travel, which is the Z drive orthogonality with respect to the XY drive plane of the stages 6a, 6b, that is, the movement deviation in the XY direction when the stages 6a, 6b are driven on the Z axis is the interferometer bar mirror 10a- 1, 10b-2 is defined by the degree of orthogonality to the XY plane.
For this reason, when the bar mirror 10a-1 and the bar mirror 10b-2 that control the Z-running in the measurement station 2 and the exposure station 3 are different, the Z-running in the measurement station 2 and the exposure station 3 due to bar mirror manufacturing, assembly error, etc. The difference between them is an amount that cannot be ignored.

The difference in Z-run is that when the alignment mark on the wafer is measured when the thickness of the wafer (surface height) and the height of the reference mark that corrects (detects) the origin of the stage during stage swapping are the same. The height of the stage is almost the same when measuring the reference mark. That is, since the Z drive positions substantially coincide, the difference in Z running between the measurement station 2 and the exposure station 3 can be ignored.
Even if the heights are different, if the thicknesses of the wafers 4a and 4b are always constant, they will appear as a constant alignment offset. For example, correction by feedback of alignment errors caused by exposure of the preceding wafer or the like is not possible. Easy.
However, as described above, in the process of manufacturing an actual IC, the thickness of the wafers 4a and 4b cannot be constant, and it is inevitable that the wafers 4a and 4b or the process varies. As a result, the difference in Z travel between the measurement station 2 and the exposure station 3 becomes manifest as an alignment shift error for each wafer due to the variation in wafer thickness, and the overlay accuracy is lowered.
Therefore, an object of the present invention is to reduce fluctuations in overlay accuracy due to fluctuations in the thickness of a substrate, for example.

In order to solve the above problems, an exposure apparatus of the present invention includes a stage that holds a substrate, an exposure station that includes a projection optical system, and a measurement station that measures the position of a mark formed on the substrate held on the stage. An exposure apparatus that exposes a substrate held on the stage via the projection optical system, the exposure apparatus including: a first mirror included in the stage disposed in the measurement station; The first interferometer for measuring the rotation angle of the stage around a first direction substantially orthogonal to the optical axis, and the second mirror included in the stage disposed in the exposure station, A second interferometer for measuring a rotation angle of the stage around a direction, and a position of a reference mark on the stage disposed at the measurement station substantially perpendicular to the optical axis. First measuring means for measuring within the surface to be measured, second measuring means for measuring the position of the reference mark on the stage disposed at the exposure station within the surface, and control means, and the reference The mark includes a plurality of reference marks arranged at different positions in a second direction orthogonal to the optical axis direction and the first direction and in the optical axis direction, and the control means includes the first interference The position of each of the plurality of reference marks is measured by the first measuring means by moving the stage in the direction of the optical axis and the second direction while keeping the rotation angle constant according to the measured value of the meter. And moving the stage in the direction of the optical axis and the second direction while keeping the rotation angle constant according to the measurement value of the second interferometer, Serial plurality of reference marks of each position is measured in the second measuring means,
The position of the stage in the direction of the optical axis based on the position of each of the plurality of reference marks measured by the first measuring means and the position of each of the plurality of reference marks measured by the second measuring means. An amount for correcting the position of the stage in the plane is calculated according to

  According to the present invention, it is possible to reduce, for example, variations in overlay accuracy due to variations in substrate thickness.

Embodiments of the present invention will be described below with reference to the accompanying drawings.
FIG. 1 is a block diagram of the exposure apparatus of the present embodiment.
The exposure apparatus of this embodiment is formed on a stage 6b that is a substrate stage that holds a wafer 4b that is a substrate, an exposure station 3 that includes a projection optical system 11, and a wafer 4a that is held on the stage 6a that is a substrate stage. And a measuring station 2 for measuring the position of the mark, and an apparatus for exposing the wafer 4b held on the stage 6b via the projection optical system 11. That is, the projection optical system 11 projects and exposes the pattern of the reticle 12 that is the original plate onto the wafer 4b that is the substrate placed on the stage 6b that is the substrate stage.
The stages 6a and 6b move alternately between the measurement station 2 and the exposure station 3, that is, perform stage swap.
The OA alignment detection system 5 as the first measuring means detects the pattern position on the wafer 4a on the substrate and the position of the stage 6a as the substrate stage without using the projection optical system 11. That is, the OA alignment detection system 5 as the first measuring means uses the positions of the reference marks 9a-1, 9a-2, 9a-3 on the stage 6a arranged in the measurement station 2 as the optical axis of the projection optical system 11. Measure in a plane that is substantially orthogonal.
TTL alignment detection systems 13a and 13b as second measuring means detect the position of the stage 6b via the projection optical system 11. That is, the TTL alignment detection systems 13a and 13b serving as the second measuring means determine the positions of the reference marks 9b-1, 9b-2, and 9b-3 on the stage 6b disposed in the exposure station 3 and the light of the projection optical system 11. Measure in a plane substantially perpendicular to the axis.
A plurality of fiducial marks 9a-1, 9a-2, 9a-3 and 9b-1, 9b-2, 9b-3 are arranged in stages 6a so that their heights are different from each other in the normal direction of the surfaces of the stages 6a, 6b. , 6b.
That is, the reference marks are arranged at different positions in the Y direction, which is the second direction orthogonal to the direction of the optical axis of the projection optical system 11 and the X direction, which is the first direction, and in the direction of the optical axis of the projection optical system 11. The plurality of reference marks 9a-1, 9a-2, 9a-3 and 9b-1, 9b-2, 9b-3.

The measurement station 2 is a position for measuring the relative positional relationship between the stage 6a and the pattern on the wafer 4a. The measurement station 2 includes an OA alignment detection system 5 that measures the reference mark 9a on the stage 6a and the alignment mark of the wafer 4a on the stage 6a.
The measurement station 2 is further a height detector 8a, 8b that measures the height of the reference mark 9a on the stage 6a and the alignment mark on the wafer 4a, and a first interferometer that measures the position of the stage 6a. It has an interferometer 7a.
The interferometer 7a which is the first interferometer is connected to the optical axis of the projection optical system 11 via the interferometer bar mirrors 10a-1 and 10a-2 which are the first mirrors included in the stage 6a disposed in the measurement station 2. The rotation angle of the stage 6a around the X direction, which is a first direction that is substantially orthogonal, is measured.
The exposure station 3 is a position for projecting and exposing the pattern of the reticle 12 onto the wafer 4b after measuring the relative positional relationship between the reticle 12 and the stage. The exposure station 3 includes a projection optical system 11 that projects an image of the reticle 12 onto the wafer 4b, and TTL alignment detection systems 13a and 13b that align the pattern of the reticle 12 and the reference mark on the stage 6b.
The exposure station 3 further includes an illumination optical system 14 that oscillates exposure light, and an interferometer 7b that is a second interferometer that measures the position of the stage 6b.
The interferometer 7b which is the second interferometer is in the X direction which is the first direction via the interferometer bar mirrors 10b-1 and 10b-2 which are the second mirrors included in the stage 6b disposed in the exposure station 3. The rotation angle of the surrounding stage 6b is measured.

The apparatus control unit 1 that is a control unit is a control unit that controls the measurement station 2 and the exposure station 3.
The apparatus control unit 1 first maintains a constant rotation angle of the stage 6a around the X direction, which is the first direction substantially orthogonal to the optical axis of the projection optical system 11, according to the measurement value of the interferometer 7a. By moving the stage 6a in the Y direction, which is the second direction orthogonal to the direction of the optical axis of the projection optical system 11 and the direction of the optical axis of the projection optical system 11, and the X direction, which is the first direction, a plurality of references The interferometer 7a is caused to measure the positions of the marks 9a-1, 9a-2, 9a-3.
Next, the apparatus control unit 1 keeps the rotation angle of the stage 6b around the X direction, which is the first direction substantially orthogonal to the optical axis of the projection optical system 11, constant according to the measurement value of the interferometer 7b. However, by moving the stage 6b in the direction of the optical axis of the projection optical system 11 and the Y direction, which is the second direction, the positions of the plurality of reference marks 9b-1, 9b-2, 9b-3 are interferometer 7b. Let me measure.
Further, the apparatus control unit 1 determines the positions of the plurality of reference marks 9a-1, 9a-2, 9a-3 measured by the interferometer 7a and the plurality of reference marks 9b-1, 9b measured by the interferometer 7b. -2 and 9b-3, the amounts of correction of the positions of the stages 6a and 6b in the plane substantially perpendicular to the optical axis according to the positions of the stages 6a and 6b in the direction of the optical axis calculate.
Similarly to the conventional example, only the interferometers 7a and 7b for controlling the Y-direction interferometer and the Z-direction moving axis are shown here for the sake of simplicity. However, in this embodiment, the stage is controlled by six axes (X, Y, Z, ωX, ωY, ωZ) by the measurement station 2 and the exposure station 3 by a plurality of interferometers (not shown).
The present embodiment can be applied if there are two or more reference marks, and can also be applied by selecting a mark that performs a function corresponding to the reference mark in addition to the reference marks.
Further, the distance between the reference marks and the height interval are not necessarily constant. However, the distance and height of the reference mark are determined when the apparatus is manufactured and assembled.

Next, referring to FIG. 2, FIG. 3, and FIG. 4, in the measurement station 2 and the exposure station 3, the Z run of the stage is different, so that the alignment mark position on the wafer is changed according to the change in wafer thickness. A case where an error occurs in the relative amount of the reference mark position will be described.
The focus depth is shallow when measuring alignment marks with an optical microscope. For this reason, it is necessary to perform positioning so that the reference mark 9a comes to the best focus position by Z driving the stage 6a with reference to a change in the contrast value of the alignment mark by an auto focus mechanism (not shown).
Here, as shown in FIGS. 2A, 2B, 2C, and 2D, when the thickness of the wafer 4a is the same as the reference mark 9a on the stage 6a, the reference mark 9a is measured. The Z position coincides with the Z position when measuring the alignment mark on the wafer. For this reason, it is not necessary to Z-drive the stage 6a for focusing.
For this reason, even if the Z running is different, there is no error in the relative amount of the measured values of the alignment mark position on the wafer 6a and the position of the reference mark 9a. Therefore, the wafer 4a may be exposed by comparing the relationship between the alignment mark position on the wafer 4a calculated by the measurement station 2 and the position of the reference mark 9a.

Also, as shown in FIG. 4, when the Z-running coincides at the measurement station 2 and the exposure station 3, the relative position between the alignment mark position on the wafer 4b and the position of the reference mark 9a at the measurement station 2 and the exposure station 3 is relative. The quantities 20 and 21 will be the same.
Therefore, even when the thickness of the wafer 4b is different from the height of the reference mark 9a, the wafer 4b is exposed by comparing the relationship between the alignment mark position on the wafer 4b calculated by the measurement station 2 and the reference mark 9a position. As a result, exposure is performed at a position shifted by the same amount as at the time of measurement, and there is no need to consider shift errors.
On the other hand, as shown in FIG. 3, a case will be described in which the Z runs are different from each other in the measurement station 2 and the exposure station 3, and the thickness of the wafer 4b is different from the height of the reference mark 9a on the stage 6a.
For each measurement station 2 and exposure station 3, an error occurs in the relative amounts 20 and 21 between the alignment mark position on the wafer 4b and the position of the reference mark 9a, and the difference between the relative amounts 20 and 21 is a shift during exposure of the wafer 4b. Since this is an error, correction is required.

Next, with reference to FIGS. 5 and 6, a flow for obtaining a difference between the amount of deviation in the X direction and the Y direction with respect to the Z drive amount of the stage in the measurement station and the exposure station will be described.
In the embodiment of FIG. 5, attention is paid to one stage (in this embodiment, stage 6a), and the state when the stage 6a is on the measurement station 2 side and the state when the stage 6a is on the exposure station 3 side are the same for convenience. Expressed on the drawing.
As shown in FIG. 6, a plurality of reference marks 9 a-1 and 9 a-2 are respectively measured by the first measurement process by the OA alignment detection system 5 which is the first detection system, and the first plurality of measurements is performed. The values Xm1, Ym1, Zm1, Xm2, Ym2, and Zm2 are measured.
Furthermore, a plurality of reference marks 9a-1 and 9a-2 are respectively measured by the second measurement step by the TTL alignment detection systems 13a and 13b as the second detection system, and the second plurality of measurement values Xe1, Ye1, Ze1, Xe2, Ye2, and Ze2 are measured.
Further, the first plurality of measurement values Xm1, Ym1, Zm1, Xm2, Ym2, and Zm2 are compared with the second plurality of measurement values Xe1, Ye1, Ze1, Xe2, Ye2, and Ze2.
Deviation amounts ΔXm and ΔYm in the X direction and Y direction of the plurality of reference marks 9a-1 and 9a-2 from positions determined at the time of manufacturing or assembling the apparatus calculated by this comparison, and a plurality of reference marks 9a- The stage 6a is controlled so that the relationship between the heights 1 and 9a-2 (Z) is the same in the first measurement step and the second measurement step.

As shown in FIG. 6A, the apparatus control unit 1 measures the reference mark 9a-1 included in the stage 6a by using the OA alignment detection system 5 as the first measuring means on the measurement station 2 side. Further, Xm1 and Ym1, which are measurement results at this time, and the value of the stage height Zm1 at the best focus position are stored. (Step 101)
Next, similarly, the reference mark 9a-2 of the stage 6a is measured using the OA alignment detection system 5 on the measurement station 2 side. Furthermore, the values of Xm2, Ym2, and the stage height Zm2 at the best focus position, which are measurement results at this time, are stored. (Step 102)
The fiducial marks 9a-1 and 9a-2 have X, Y, and Z components as positions determined at the time of device manufacture or assembly, respectively, as X1org, Y1org, Z1org, and X2org, Y2org, Z2org.
The apparatus control unit 1 compares this value with the values obtained in Steps 101 and 102, so that the displacements ΔXm and ΔYm in the X and Y directions with respect to the drive amount ΔZm of the stage 6a in the measurement station 2 in the Z direction are compared. Is calculated. (Step 103)
Specifically, the amount of deviation in the ΔXm and ΔYm directions relative to ΔZm can be obtained by the following calculation formula. Of course, the following calculation formula can also be established by using one of the measurement results of the reference marks 9a-1 and 9a-2 as the origin and the other measurement result as the distance from the origin.
Zm1 and Zm2 are values including manufacturing errors in the height direction of the reference mark 9a at the time of manufacturing, and are substantially the same values as Z1org and Z2org, respectively.
In the following correction formula, if the focus measurement accuracy is sufficiently high, it may be preferable to use Zm1 and Zm2 instead of the positions Z1org and Z2org determined at the time of device manufacture or assembly.

本実施例においてウェハ上のアライメントマークを計測する時は、まず、任意の基準マークを選び、その高さとウェハの厚さとの差から、（１）と（２）の式を利用して補正値を求める。
次に、装置制御部１は、ステージ６ａをスワップさせ、露光ステーション３でも同様に、露光ステーション３におけるステージ６ｂのＺ方向の駆動量△Ｚｅに対するＸ及びＹ方向のずれ量△Ｘｅ及び△Ｙｅを算出する。
なお、図６（ｂ）に示されるように露光ステーション３側の△Ｘｅ、及び△Ｙｅの算出方法は、計測ステーション２の△Ｘｍ、及び△Ｙｍと同様なのでここでは省略する。(ステップ１０４、１０５、１０６)。
但し、装置制御部１は、ＯＡアライメント検出系５ではなく、露光ステーション３側でＴＴＬアライメント検出系１３ａ、１３ｂを用いてステージ６ａが有する基準マーク９ａ−１、９ａ−２を計測する。

In this embodiment, when measuring the alignment mark on the wafer, first, an arbitrary reference mark is selected, and the correction value is calculated from the difference between the height and the thickness of the wafer using the equations (1) and (2). Ask for.
Next, the apparatus control unit 1 swaps the stage 6a, and similarly in the exposure station 3, the deviation amounts ΔXe and ΔY in the X and Y directions with respect to the drive amount ΔZe of the stage 6b in the exposure station 3 in the Z direction. calculate.
As shown in FIG. 6B, the calculation method of ΔXe and ΔYe on the exposure station 3 side is the same as ΔXm and ΔYm of the measurement station 2, and therefore will be omitted here. (Steps 104, 105, 106).
However, the apparatus control unit 1 measures the reference marks 9a-1 and 9a-2 of the stage 6a using the TTL alignment detection systems 13a and 13b on the exposure station 3 side instead of the OA alignment detection system 5.

FIG. 8 is a graph showing the amount of deviation ΔXm in the X direction on the measurement station 2 side and the amount of deviation ΔXe in the X direction on the exposure station 3 side with respect to the Z drive amount obtained in the flow of steps 101 to 106. Is.
FIG. 9 is also a graph showing the amount of deviation ΔYm in the Y direction on the measurement station side and the amount of deviation ΔYe in the Y direction on the exposure station 3 side with respect to the Z drive amount obtained in the flow of steps 101 to 106. Is.
8 and 9, the reference mark 9a-1 is set as the origin. When the thickness of the wafer does not coincide with the height of the reference mark 9a-1, the alignment mark on the wafer is determined from the difference in height and the correction coefficient obtained from the graph. A correction value for the distance to the reference mark on the stage can be derived.
By reflecting this in the control, the position of the alignment mark on the wafer in the measurement station 2 and the exposure station 3 can be accurately calculated.

By the way, it has been described above that the shift error at the time of exposure of the wafer does not occur by matching the Z-runs of the measurement station 2 and the exposure station 3 (with the same deviation).
It has also been described above that the correction coefficients in the X and Y directions with respect to the Z drive amount of the stage can be obtained by measuring a plurality of reference marks at the measurement station 2 and the exposure station 3.
8 and 9, in the X direction and Y direction obtained by the first measurement process of the plurality of reference marks, and the X direction obtained by the second measurement process of the plurality of reference marks. The difference from the deviation amount in the Y direction is controlled to be always constant in the X direction and the Y direction.
That is, the apparatus control unit 1 sets the stage at the same target position in the direction of the optical axis at the same target position in the plane substantially orthogonal to the optical axis of the projection optical system 11 in each of the measurement station 2 and the exposure station 3. A correction amount is calculated so that the difference between the positions of the stages 6a and 6b in the plane when the positions 6a and 6b are positioned is constant regardless of the target position in the plane.
That is, in the embodiment of FIGS. 8 and 9, the inclination of the shift amount with respect to the Z drive amount in the exposure station 3 is matched with the slope of the shift amount with respect to the Z drive amount of the stage on the measurement station 2 side.
That is, the Z running can be adjusted by controlling the stage drive on the exposure station 3 side so as to match the correction coefficients in the X direction and the Y direction. That is, the apparatus control unit 1 calculates the correction amount for each of the X direction that is the first direction and the Y direction that is the second direction.
In this case, the Z-running of the stage differs between the measurement station 2 and the exposure station 3, and the thickness of the wafer is thicker or thinner than the reference mark on an arbitrary stage.
In this case, a correction amount that is a correction coefficient in the X direction and the Y direction in the measurement station 2 and the exposure station 3 is obtained, and the stage drive is controlled so that they match, thereby generating a shift error during wafer exposure. Can be suppressed. Thereby, high-precision alignment between the exposure center and the alignment mark on the wafer becomes possible.

In some cases, control is performed such that the slope of the first straight line and the slope of the second straight line are the same in the X direction and the Y direction, respectively. The slope of the first straight line is a slope when the relationship between the height (Z) of the plurality of reference marks and the amount of deviation in the X direction and the Y direction obtained from the first measurement step is a linear regression model. . The slope of the second straight line is a slope when the relationship between the height (Z) of the plurality of reference marks and the amount of deviation in the X direction and Y direction obtained from the second measurement step is a linear regression model. .
That is, the drive of the stage on the measurement station 2 side may be controlled so that the correction coefficients in the X and Y directions of the stage on the measurement station 2 side are matched with the correction coefficients on the X and Y directions on the exposure station 3 side.
In some cases, the substrate stage is controlled by calculating a correction coefficient for XY driving accompanying Z driving of the substrate stage so that each inclination becomes 0, that is, zero. The inclination is the relationship between the height (Z) of a plurality of reference marks and the difference between the amount of deviation in the X direction obtained from the first measurement step and the amount of deviation in the X direction obtained from the second measurement step. Is the slope when is taken as a linear regression model.
Furthermore, the inclination is the difference between the height (Z) of the plurality of reference marks, the amount of deviation in the Y direction obtained from the first measurement step, and the amount of deviation in the Y direction obtained from the second measurement step. Is the slope when the relationship is a linear regression model.
That is, the stage drive may be controlled in the measurement station 2 and the exposure station 3 so that the correction amount, which is a correction coefficient in the X and Y directions of the stage, is zero, that is, zero. As described above, the apparatus control unit 1 may calculate an amount to be corrected so that the difference in stage position becomes zero.

Note that, by measuring three or more reference marks, the correction accuracy can be increased due to the averaging effect of the measurement values. Statistical processing using least square approximation, weighted least square approximation, or the like is desirable for processing three or more measured values.
The timing of correction may be at the time of equipment adjustment or regular maintenance, or at the beginning or end of a lot, the idle period between lots, or when two stages wait for synchronization to improve accuracy. May be.
In general, the change over time is a gradual trend, and the accuracy can be improved by increasing the number of measurement points. Therefore, in the case of periodic correction, a moving average that combines the most recent or most recent measured values is used. It is also possible to improve accuracy by performing processing.
In the present embodiment, the Z-run correction method has been described for only one stage, but it goes without saying that the other stage also has an inherent bar mirror manufacturing error, and thus needs to be corrected in the same manner. In this case, the correction time can be shortened by simultaneously measuring the reference marks of both stages at the measurement station and the exposure station, respectively.

Next, with reference to FIG. 7, an example of a flow when over-baking a wafer once processed in the present embodiment will be described.
First, correction coefficients in the X and Y directions with respect to the Z drive amount of the stage are obtained on the measurement station 2 side by the procedure shown in steps 101 to 103. (Step 111)
Next, the stage is swapped (step 112), and correction factors in the X and Y directions with respect to the Z drive amount of the stage are obtained on the exposure station 3 side by the procedure shown in steps 104 to 106. At this time, X and Y correction coefficients for the Z drive amount of the stage existing in the opposite measurement station 2 are also obtained. (Step 113)
Next, the stage drive correction parameter on the exposure station 3 side is changed so that the correction coefficients in the X and Y directions with respect to the Z drive amount of the stage at the measurement station 2 and the exposure station 3 are equivalent. Further, as described above, the other measurement station 2 also individually obtains correction parameters and changes the drive correction parameters on the exposure station 3 side as parameters for each stage. (Step 114)
In the above embodiment, the stage drive correction parameter on the exposure station 3 side is changed. However, even if the stage drive correction parameter on the measurement station 2 side is changed, the same effect can be obtained.
Even if the stage drive correction parameters in the measurement station 2 and the exposure station 3 are changed so that the correction amounts that are correction coefficients in the X and Y directions at the measurement station 2 and the exposure station 3 are 0, that is, zero. The same effect can be obtained.

Next, the wafer is loaded onto the stage on the measurement station 2 side (step 115), and the stage origin at the measurement station is set (alignment between the measurement center and the reference mark). (Step 116)
Subsequently, global alignment (relative position measurement between the origin and the alignment mark on the wafer) is performed to obtain a relative positional relationship between an arbitrary reference mark and the alignment mark on the wafer. (Step 117)
Thereafter, stage swap is performed (step 118), and the stage origin is set at the exposure station 3 (alignment between the exposure center and the reference mark). (Step 119)
Finally, the wafer is exposed by comparing the relative positional relationship between an arbitrary reference mark (selected on the measurement station 2 side) obtained on the measurement station 2 side and the alignment mark on the wafer. (Step 120)
As described above, the correction coefficients in the X and Y directions with respect to the Z drive amount of the stage are adjusted in advance at both stations so that no shift error occurs during wafer exposure. Can be aligned with high accuracy.

(Example of device manufacturing method)
A device (semiconductor integrated circuit element, liquid crystal display element, etc.) includes a step of exposing a substrate (wafer, glass plate, etc.) coated with a photosensitive agent using the exposure apparatus of any one of the embodiments described above, and the substrate It is manufactured by going through a step of developing and other known steps. Other well known processes include etching, resist stripping, dicing, bonding, packaging, and the like.

It is a block diagram of the exposure apparatus of the Example of this invention. In the exposure apparatus according to the embodiment of the present invention, the thickness of the wafer and the height of the reference mark on the stage are the same, and there is a difference in Z drive orthogonality (Z running) between the measurement station and the exposure station. It is explanatory drawing when a measurement error does not occur in the distance between the alignment mark and the reference mark on the stage. In the exposure apparatus according to the embodiment of the present invention, the thickness of the wafer and the height of the reference mark on the stage are different, and there is a difference in Z driving directivity (Z running) between the measurement station and the exposure station. It is explanatory drawing when a measurement error generate | occur | produces in the distance of an alignment mark and the reference mark on a stage. In the exposure apparatus according to the embodiment of the present invention, the thickness of the wafer and the height of the reference mark on the stage are different, and the deviation of the Z drive straightness (Z running) is equivalent between the measurement station and the exposure station. It is explanatory drawing of the state which does not generate | occur | produce a measurement error in the distance of the reference mark on a stage. It is explanatory drawing when a measurement error arises in the distance of the alignment mark on a wafer and the reference mark on a stage by the shift | offset | difference of the Z drive orthogonality of a stage in each station of the exposure apparatus of the Example of this invention. It is a flowchart which calculates | requires the measurement error of X and Y with respect to the Z drive amount of a stage in the measurement station and exposure station of the exposure apparatus of the Example of this invention. It is a flowchart at the time of carrying out the overprinting of the wafer once processed after matching the Z run of the stage in the measurement station of the exposure apparatus of an Example of this invention, and an exposure station. It is explanatory drawing which shows the deviation | shift amount to the X direction for every reference mark in the exposure apparatus of the Example of this invention. It is explanatory drawing which shows the deviation | shift amount to the Y direction for every reference mark in the exposure apparatus of the Example of this invention. It is a block diagram of the exposure apparatus of a prior art example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Apparatus control part 2 Measurement station 3 Exposure station 4 Wafer 5 OA alignment detection system 6 Stage 7 Interferometer 8 Height detector 9 Reference mark 10 Interferometer bar mirror 11 Projection optical system 12 Reticle 13 TTL alignment detection system 14 Illumination optical system

Claims (5)

A stage for holding a substrate, an exposure station including a projection optical system, and a measurement station for measuring the position of a mark formed on the substrate held on the stage, and held on the stage via the projection optical system An exposure apparatus for exposing a formed substrate,
A first interferometer that measures a rotation angle of the stage around a first direction substantially orthogonal to the optical axis of the projection optical system via a first mirror included in the stage disposed in the measurement station. When,
A second interferometer that measures a rotation angle of the stage around the first direction via a second mirror included in the stage disposed in the exposure station;
First measurement means for measuring a position of a reference mark on the stage arranged at the measurement station in a plane substantially perpendicular to the optical axis;
Second measuring means for measuring a position of a reference mark on the stage arranged in the exposure station in the plane;
Control means;
Have
The reference mark includes a plurality of reference marks arranged at different positions in a second direction orthogonal to the direction of the optical axis and the first direction and in the direction of the optical axis,
The control means includes
By moving the stage in the direction of the optical axis and the second direction while keeping the rotation angle constant according to the measurement value of the first interferometer, the position of each of the plurality of reference marks is changed to the first position. Let the measuring means measure,
By moving the stage in the direction of the optical axis and the second direction while keeping the rotation angle constant according to the measurement value of the second interferometer, the position of each of the plurality of reference marks is changed to the second position. Let the measuring means measure,
The position of the stage in the direction of the optical axis based on the position of each of the plurality of reference marks measured by the first measuring means and the position of each of the plurality of reference marks measured by the second measuring means. Calculating an amount for correcting the position of the stage in the plane according to
An exposure apparatus characterized by that.   The control means determines the position of the stage in the plane when the stage is positioned at the same target position in the optical axis direction at the same target position in the plane at each of the measurement station and the exposure station. The exposure apparatus according to claim 1, wherein the amount is calculated so that the difference is constant regardless of the target position in the plane.   The exposure apparatus according to claim 2, wherein the control unit calculates the amount so that the difference becomes zero.   The exposure apparatus according to claim 1, wherein the control unit calculates the amount for each of the first direction and the second direction. A step of exposing the substrate using the exposure apparatus according to claim 1;
Developing the substrate exposed in the step;
A device manufacturing method comprising:

Images