JPH10125578A - Exposure method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the precision of alignment by previously measuring the shift quantity of a projected image owing to distortion in first and second optical systems and reducing the difference of shift quantity first and second graphic at the transfer size of the second optical system with a wide field until it becomes a value not more than a previous value which is previously decided. SOLUTION: The first graphic by a first aligner is previously transferred on a wafer 1a. The field size of the first aligner is narrower than that of the second aligner. A storage device 40 stores shift quantity by the distortion of the first optical system and that by the distortion of the second optical system with the wide field. In the first graphic on the wafer 1a, the shift quantity and the position of the projected image in the maximum transfer size of the second alighner with the wide field are called from the storage means 40 and the difference of shift quantity between the first optical system and the second optical system is operated in an operation means 42. Then, the transfer size of the second aligner is repetitively reduced until the difference of shift quantity becomes a previously decided specification so as to search the optimum size.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure method in a photolithography process in a manufacturing process of a semiconductor device such as an LSI, and more particularly, to a method of transferring figures by using reduced projection type exposure machines having different field sizes. The exposure method when performing the exposure.

[0002]

2. Description of the Related Art For example, in order to manufacture an LSI by processing a silicon single crystal substrate (wafer), a pattern on a photomask is exposed on a photoresist film applied on the wafer, and the resist film is drawn with the resist. A figure (hereinafter referred to as a resist figure) is formed. Thereafter, using the resist figure as a mask, the wafer is etched or impurities are introduced. Then, a resist film is applied again, and the figure on the second photomask is formed. Photolithography is performed almost at each of many manufacturing steps, such as exposing and forming a resist pattern. In this case, at the time of the second resist pattern formation, a figure that has already been transferred to the underlying wafer surface (this figure is a resist figure formed in the first photolithography step, The second resist pattern must be accurately aligned (aligned) with respect to the unevenness generated on the surface or the difference in optical properties.

An exposure apparatus used in the above-described photolithography includes an apparatus of a contact exposure type in which a wafer and a photomask are brought into close contact with each other, and a wafer and a photomask (reticle) are separated from each other to form a figure image on the reticle. There is a projection-type apparatus that forms an image on a wafer, and a projection-type exposure apparatus is usually used for manufacturing an LSI with a remarkably high integration in recent years. This is because there is no damage to the wafer and the photomask in the contact type exposure apparatus, and there is no decrease in the yield rate due to the damage. FIG. 10 schematically shows a positional relationship between a wafer, a reticle, and an optical system when a projection exposure apparatus is used.
Referring to FIG. 10, wafer 1 and reticle 2 are vertically arranged with a distance therebetween. On the entire surface of the wafer 1, a photoresist film (not shown) is formed. On the reticle 2, a figure (not shown) to be transferred to the photoresist film is drawn in advance. In the space between the reticle 2 and the wafer 1, a projection optical system 3 is arranged. In this arrangement, visible light or ultraviolet light is irradiated from above the reticle 2, an image of a figure on the reticle is formed on the wafer 1 by the projection optical system 3, and the photoresist film is exposed and transferred.

Next, the projection type exposure apparatus shown in FIG.
This is an equal-size projection exposure apparatus in which a figure on the reticle 2 is transferred to a resist film on the wafer at an equal magnification, but a reduced-projection exposure apparatus is generally used for manufacturing an LSI. . In a projection type exposure apparatus of this type, a figure on a reticle is reduced by a projection optical system to form an image on a wafer, and the wafer is sequentially exposed (shot) by a step-and-repeat method as described later. . Therefore, although the productivity is reduced, the dimensional error in the figure on the reticle is reduced on the wafer, the exposure area (transfer size) for each shot is small, and the focus can be adjusted for each shot. This is because the effect of warpage is small and the alignment accuracy (alignment accuracy) and the dimensional accuracy are excellent.

FIG. 11 shows a configuration diagram of a reduction projection type exposure apparatus. In the following description, the projection optical system 3 shown in FIG. 11 is used as the first projection optical system 3a in the first photolithography step for transferring the first figure on the wafer.
In the second photolithography step for aligning and transferring the second graphic with respect to the first graphic transferred on the wafer, a second projection different from the first projection optical system is performed. The optical system 3b is shown, but for simplicity of description, it is simply shown as "projection optical system 3" in FIG. The same applies to the wafer 1 and the reticle 2.

In general, a large number of figures having the same pattern, each having a chip as a repetition unit, are arranged in rows and columns over substantially the entire surface of the wafer. On the other hand, on the reticle, figures in which the pattern in the chip is enlarged are arranged in rows and columns. However, the number of rows and columns on the reticle is naturally smaller than the number of rows and columns of chips on the wafer. This is because the projection type exposure apparatus reduces a figure on a reticle and projects it on a wafer. Therefore, in an actual photolithography process, an operation of aligning a pattern on a reticle with a chip included in a certain region on a wafer and performing a shot is repeated in a step-and-repeat manner. Here, in understanding the present invention,
Since it is important that the pattern on the wafer or reticle is composed of an array of chips and rows (in the case of a reticle, enlarged chips), in the following description,
The arrangement of the chips is referred to as a “chip arrangement pattern”.

Referring to FIG. 11, the exposure apparatus shown in this figure is a reduction projection type exposure apparatus disclosed in Japanese Patent Application Laid-Open No. 3-241728, and includes a figure already transferred onto wafer 1a. This is a projection type exposure apparatus of a type that aligns a figure on a reticle 2b (to be transferred to a resist film on a wafer 1a) with a through-the-lens (TTL) method. In such a reduced projection type exposure apparatus,
In the case where the projection optical system 3a used for transferring a pattern to the wafer 1a first and the projection optical system 3b used for forming a resist pattern overlapping the pattern on the wafer 1a are different from each other, the respective projection optical systems Due to the unique distortion of the system, the two figures transferred on the wafer (wafer 1
The alignment accuracy between the first figure previously transferred to the surface a and the resist figure aligned with the first figure in the next photolithography step may decrease. The invention described in the above publication is intended to correct the alignment deviation due to the distortion of the projection optical system and improve the alignment accuracy. The description is given below.

In the invention described in the above publication, first, the amount of error of the projected image due to the distortion of the first projection optical system 3a is stored in the storage means 30 in advance. An example of a method for storing the error amount will be described with reference to FIG. Referring to FIG. 12A, the ideal grating 20 is formed by the projection optical system 3a.
Is projected as a projected image 20a via At this time, when the positive directions of the X axis and the Y axis are indicated by arrows in the figure, the point A is distorted by the point A.
It may represent an error amount of when projected ₁ (ΔX _1, -ΔY _1). Similarly, the error amount when the point B is projected on the point B ₁ and the error amount when the point C is projected on the point C ₁ are (ΔX ₃ , 0) and (ΔX ₄ , ΔY ₄ ), respectively.
It is expressed as In this way, the error amount based on the distortion of the projection optical system 3a is measured by designating a plurality of points and coordinates in the image field, and a map is created and stored in the storage means 30.

Similarly, an error amount map of the projected image based on the distortion of the second projection optical system 3b is stored in the storage means 30 in advance. An example of the error amount in this case is:
This is shown in FIG. Referring to FIG. 12B, it is assumed that ideal grating 20 is projected via projection optical system 3b as projected image 20b. Each error amount when the point D is an error amount and point E when it is projected to a point D ₁ is projected to a point _{E 1 (- ΔX 5, 0} ), (- ΔX 6, -ΔY 5)
It is expressed as In this way, the error amount based on the distortion of the second projection optical system 3b is stored in advance in the storage unit 30.
To memorize.

Next, an image of the second figure on the reticle 2b is formed using the second projection optical system 3b on the first figure transferred onto the wafer 1a in the first photolithography step. Alignment is performed by the TTL method. At the time of the alignment, just performing the normal alignment operation
An alignment shift occurs due to a difference in distortion between the two projection optical systems 3a and 3b. For example, FIG.
(A) shows a first projection optical system 1 for a figure on a wafer 1a.
Due to the distortion a, a pincushion-type error as shown in FIG. 12A occurs, while the projection image of the figure on the reticle 2b is distorted by the second projection optical system 3b as shown in FIG. 2) shows a state in which the two figures 21a and 21b are aligned in a desirable state on the assumption that a barrel-shaped error as shown in FIG. This alignment state is obtained as a result of adding a correction (described later) for an error based on the distortion of the projection optical systems 3a and 3b in addition to a normal alignment operation.
The centers of the figures 21a and 21b coincide.
On the other hand, the alignment mark 22a provided in the figure 21a on the wafer and the alignment mark 22b provided in the figure 21b on the reticle are separated by a distance ΔX ₁ + ΔX
₂ (however, ΔX ₁ is the first projection optical system 3
(a) is the amount of error generated at the position of the alignment mark 22a due to the distortion, and .DELTA.X2 is the amount of error generated at the position of the alignment mark 22b due to the distortion in the second projection optical system 3b.

On the other hand, FIG. 13B shows an alignment state (without correction) obtained by a normal operation. The graphic 21a on the wafer 1a is the graphic 21 on the reticle.
b is relatively shifted in the negative direction of the X axis. On the other hand, the alignment mark 22a on the wafer and the alignment mark 22b on the reticle have the same position in the X-axis direction as shown in FIG. I have. This is for the following reason. Referring to FIG. 11 showing the configuration of the projection type exposure apparatus, in general, in the alignment operation by the TTL method, the positional relationship between the alignment mark 22a on the wafer and the alignment mark 22b on the reticle observed by the ITV camera 4 is shown. From the image processing means 31 to the moving means 32 so that L ₁ = L ₂ (see FIG. 13C).
And the motors 5 and 6 are operated, and FIG.
Is performed as shown in FIG. As a result of this alignment operation, although the two alignment marks seem to coincide,
An alignment error of (ΔX ₁ + ΔX ₂ ) will occur along the X axis.

Therefore, with respect to the above alignment result,
A correction amount is obtained so as to be in the alignment state shown in FIG. 13A, and the wafer 1a and the reticle 2b are relatively moved based on the correction amount. First, the image processing means 31
Detects coordinates in the image fields of the alignment marks 22a and 22b based on an image taken by the ITV camera 4. Then, the error amount corresponding to the detected coordinates is the alignment mark 22a, 2
It is read from the storage means 30 every 2b. The correction amount calculating means 32 converts the value into a correction amount for correcting the alignment error, and provides the correction amount to the moving means 33. Transportation means 33
Drives the motors 5 and 6 according to the given correction amount, and moves the wafer 1a and the reticle 2b.

Referring again to FIG. 13A, for example, the error amount of the alignment mark 22b on the reticle read from the storage means 30 is (-ΔX ₂ , 0), and the alignment on the wafer is The error amount of the mark 22a is (ΔX ₁ ,
0), the correction amount in the X-axis direction indicated by the correction amount calculating means 32 is {(−ΔX ₂ ) − (ΔX ₁ )} = − (Δ
X ₁ + ΔX ₂ ), and the correction amount in the Y-axis direction is (0-0). That is, after the correction, it becomes (-(ΔX ₁ + ΔX ₂ ), 0). This correction amount is given to the moving means 33 and the moving means 3
Reference numeral 3 denotes a mode in which the alignment mark 22b on the reticle in FIG. 13B is shifted relative to the alignment mark 22a on the wafer by (ΔX ₁ + ΔX ₂ ) in the negative X-axis direction. The terminals 5 and 6 are driven. Thereafter, projection exposure of a figure on the reticle 2b is performed on the wafer 1a. As a result, the alignment result is corrected to the state shown in FIG. 13A, and the first and second projection optical systems 3a and 3b
The projection error caused by each distortion can be reduced.

[0014]

The first problem is that, in the alignment technique described in the above-mentioned publication, the figure to be aligned in an originally preferable state is not aligned with the alignment mark due to the difference in the distortion of the projection optical system. The purpose of this is to prevent the position from being shifted from the desired state due to the occurrence of an error in the position of the first figure, and the first figure on the wafer transferred by the first projection type exposure apparatus. In addition, the shift amount due to the distortion between the reticle and the second figure projected on the reticle projected by the second projection type exposure apparatus is not directly compressed.

That is, for example, when an alignment error in the X-axis direction is considered, in the alignment state shown in FIG. 13A, the figure 21a on the wafer and the figure 21b on the reticle are often not located at the center of each figure. However, there is an error of △ X ₁ + △ X _{2 on} the left side of the figure.辺 X ₃ + △ X ₄ (however, △ X ₃
Is an error amount at the right side position of the figure 21a on the wafer 1a, and ΔX ₄ is an error amount at the right side position of the figure 21b on the reticle 2b). On the other hand, in the alignment state shown in FIG. 13B where no correction is applied, the shift amount on the left side of the figure is zero, but ΔX ₁ at the center.
+ ΔX ₂ occurs, and on the right side, ΔX ₁ + ΔX ₂ +
A shift of ΔX ₃ + ΔX ₄ has occurred. In other words, the maximum deviation {X ₁ + {X ₂ +} that occurs in the normal alignment operation.
X ₃ + △ X ₄ is simply assigned to each part of the figures 21a and 21b as appropriate, and the maximum deviation amount ΔX ₁ + {X ₂ +} caused by the distortion of the two projection optical systems.
X ₃ + △ X ₄ itself is not compressed.

The second problem is that a second projection type exposure apparatus having a field size larger than the field size of the first projection type exposure apparatus is used and one shot in the second photolithography step is performed. When the number of transfer chips is larger than the number of transfer chips for one shot in the first photolithography process, it is difficult to correct the projection error of the alignment mark due to the distortion. This is because the distortion of the first projection optical system 3a includes an alignment error between shots.

That is, in the alignment between the projection type exposure apparatuses having different field sizes, when the second projection type exposure apparatus having a wider field is used in the first projection type exposure apparatus, the wide field type is used. Since the distortion of the second projection optical system is inevitably larger and the first figure on the wafer includes an alignment error between shots, the first figure on the wafer and the reticle The alignment accuracy with the above-mentioned second figure is poor, and the process to which the wide field projection type exposure apparatus can be applied is limited. This poses a major problem in terms of production efficiency.

As described above, the reduction projection type exposure apparatus has a very great advantage that the transfer accuracy of a figure from a reticle is high, and the high precision is more remarkable as the reduction ratio is larger. However, on the other hand, when the reduction ratio increases, 1
Since the number of shots required to expose the entire wafer increases, the productivity naturally decreases. Therefore, in manufacturing an LSI, a reduction ratio is determined in consideration of dimensional accuracy and productivity of a figure to be transferred onto a wafer. At that time,
It is very inconvenient to limit the application of the wide field projection exposure apparatus.

Accordingly, the present invention provides a first projection exposure apparatus used for transferring a first figure onto a wafer and a second projection apparatus used for forming a resist figure in alignment with the first figure. It is an object of the present invention to reduce the maximum alignment error itself that would be caused by a normal alignment operation due to a difference in distortion peculiar to each exposure apparatus when the projection type exposure apparatus is different.

Another object of the present invention is to expand the photolithography process applicable to the wide field projection exposure apparatus.

[0021]

According to the present invention, a plurality of second projection exposure apparatuses are used by using a second projection exposure apparatus whose field size is larger than the field size of the first projection exposure apparatus in which the first chip arrangement pattern is transferred to the underlying substrate. In an exposure method including an operation of aligning and shooting the two chip arrangement patterns, the distortion of the first projection type exposure apparatus and the distortion of the second projection type exposure apparatus are measured in advance, and based on data of the measurement results. The second projection expression of the wide field is set such that the difference between the distortion of the plurality of first chip arrangement patterns corresponding to the second chip arrangement pattern and the distortion of the second chip arrangement pattern is reduced. After selecting the transfer size from the field of the exposure equipment and setting the blinds according to the transfer size Alignment, and executes a shot.

[0022]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the drawings using examples. FIG. 1 is a configuration diagram of a wide field projection type exposure apparatus used in the second photolithography step in the first and second embodiments of the present invention. Comparing FIG. 1 with FIG. 11, a wide-field projection exposure apparatus used in the second photolithography step when practicing the present invention is provided with a blind 45 on the reticle 2b (closer to the exposure light source). Is different from the conventional projection type exposure apparatus. The blind 45 is, for example, as shown in FIG. 2 and FIG. 3 (a) in which four blades are arranged two by two so that the interval can be adjusted in the vertical and horizontal directions. 3
It consists of two L-shaped wings as shown in FIG. In the blind 45 having such a structure, the opening area can be adjusted by driving the motor 46 (see FIG. 1) to move each wing plate back and forth in the direction indicated by the thick arrow in FIG. The adjustment of the opening area is performed by the blind control system 44 as described later in detail.
This is performed by driving the motor 46 based on data from 0.

The configuration other than the blind 45 and its control system is substantially the same as the configuration of the conventional projection type exposure apparatus shown in FIG. That is, the wafer stage 7 is
The position is measured by the interferometer system 8 and moved by the motor 6. A first figure (not shown) transferred by using the first projection type exposure apparatus is transferred onto the wafer 1a in advance. The field size of the first projection exposure apparatus used for the first graphic transfer is smaller than the field size of the second projection exposure apparatus shown in FIG. Folded mirror
9 changes the course of irradiation light. The lens 10 converges the irradiation light. The half mirror 11 transmits the irradiation light and changes the optical path of the reflected light of the mark for measuring the arrangement error between shots (described later) on the wafer 1a. The lens 12 causes the reflected light to form an image on the CCD camera 4. The mark image for measuring the arrangement error between shots of the CCD camera 4 is given to the image processing means 41. The image processing means 41 issues a command to the stage control system 43 and drives the motor 6 to move the wafer stage 7. Then, the mark position is measured, and the deviation from the ideal lattice is sent to the storage means 40 (described later). Storage means 4
0 also indicates the amount of deviation from the ideal grating due to the distortion in the first projection optical system and the second field of the wide field in advance.
And the shift amount from the ideal grating based on the distortion in the projection optical system 3b.

In the present invention, a first projection exposure apparatus used for transferring a first figure to an underlying substrate (wafer 1a) in alignment between projection exposure apparatuses having different field sizes from each other. When the second figure is transferred by the second projection type exposure apparatus having a larger field size, the displacement of the projection image based on the distortion in the first projection optical system and the second field of the second field The shift amount of the projected image based on the distortion in the system is measured in advance, and a map of the shift amount from the ideal lattice when each of the exposure devices is used is stored in the storage means 40.

Then, in the first figure on the wafer 1a, a plurality of areas included in an area corresponding to the maximum area (maximum transfer size) that can be exposed by one shot of the second wide field projection type exposure apparatus. The shift amount and the position of the projected image are called from the storage means 40 for the shot of. The shift amount and the position of the projected image at the maximum transfer size of the second wide field projection type exposure apparatus are called from the storage means 40. Then, the difference between the amount of shift of the projected image in the first projection optical system and the amount of shift of the projected image in the second projection optical system is calculated by the arithmetic processing means 42. If the difference between the deviation amounts exceeds a predetermined standard, the transfer size of the second wide field projection type exposure apparatus (the area where the wafer 1a is irradiated with ultraviolet light in one shot) is changed from the maximum transfer size to the chip size. The unit is reduced in units, and the difference in the amount of displacement is calculated again. This operation is repeated until the difference between the deviation amounts falls within the standard, and an optimum transfer size is found.

After the optimum transfer size is determined, the remaining first figures formed on the wafer 1a are subjected to the second operation for a plurality of chips adjacent to the area where the optimum transfer size has been obtained by the above operation. Similarly, for a plurality of shots included in an area corresponding to the maximum transfer size of the wide field projection type exposure apparatus, the projected image between the first projection type exposure apparatus and the second projection type exposure apparatus Then, the optimum transfer size is found such that the difference between the deviation amounts is within the standard.

With the above method, the optimum transfer size is found for all the areas on the wafer and stored in the storage means 40.

Then, the second step-and-repeat exposure in a wide field is started. At this time, the optimum transfer size stored in the storage means 40 is called,
Set the opening shape of the blind 45 to the optimal transfer size,
Moving the reticle 2b and the wafer 1a to the transfer position,
Execute the shot.

In the present invention, as described above, the first figure on the wafer transferred by the first projection type exposure apparatus and the reticle to be transferred by the wide field second projection type exposure apparatus are used. An actual shot is executed after an optimum transfer size is determined so that the difference between the amounts of displacement due to distortion in the respective exposure apparatuses and the second figure is equal to or smaller than a predetermined value. Therefore, the maximum value of the alignment error caused by the distortion is reduced. Moreover, the effect can be easily obtained only by distortion measurement and calculation processing.

Next, the above alignment operation is an operation in the case where there is no positional deviation between shots (inter-shot arrangement error) when the first figure is transferred onto the wafer 1a in advance. Above, it is inevitable that an inter-shot arrangement error occurs in the first figure. Hereinafter, an alignment operation performed when an inter-shot arrangement error occurs in the first figure will be described. As can be seen from the above description, when aligning the second figure with the first figure using the second exposure apparatus having a wider field than the first exposure apparatus, the transfer at the time of the second projection exposure is performed. The size usually includes a plurality of shots when the first figure is projected and exposed. Therefore, when aligning the second graphic on the reticle with the first graphic on the wafer 1a, the alignment is performed in consideration of the arrangement error between shots in the first graphic (see FIG. 9A, which will be described later). There must be.

In the present invention, a mark 46 for measuring an inter-shot arrangement error is provided in a chip included in the first figure 23a to be transferred onto the wafer 1a (see FIG. 9B). ) Is inserted. Then, based on the mark position, the shift amount of the projected image due to the distortion of the first projection optical system including the arrangement error between shots is measured, and the shift amount map is stored in the storage means 40. Further, a map of a shift amount of a projected image due to distortion in the second wide-field projection optical system is stored in the storage unit 40.

Then, of the first figures on the wafer,
Of a plurality of shots included in an area corresponding to the maximum transfer size of the wide field projection type optical system, including the arrangement error between shots, and the amount of deviation in the second projection optical system. . Until the difference between the shift amount including the alignment error between shots in the plurality of shots of the first projection optical system and the shift amount of the second wide field projection type exposure apparatus falls within a predetermined standard value. Then, the transfer size of the second wide field projection type optical system is reduced in units of a chip, and the optimum transfer size is obtained. Of the remaining first figures,
The optimum transfer size is similarly obtained for the area adjacent to the area for which the optimum transfer size has been obtained by the above operation. By repeating this operation, the optimum transfer size is determined for the entire area on the wafer, and is stored in the storage means 40.

Then, the second projection type optical system performs step
I will make a shot with and repeat,
The operation of setting the opening shape of the blind 45 in accordance with the optimal transfer size selected in advance, moving the reticle 2b and the wafer 1a to predetermined positions, and then performing a shot is repeated.

Hereinafter, a specific description will be given using examples.

(Embodiment 1) In this embodiment, first, FIG.
(A) A 4-imposition reticle (four chips C _a1 to C _a1)
a4 is, for example, a reticle arranged in an array pattern of 2 rows and 2 columns = a first chip array pattern =) and a first projection exposure apparatus, and a wafer 1a including four chips is used. One chip array pattern is transferred by 70 shots in 10 rows and 7 columns. FIG. 5 shows the resulting chip arrangement state. Referring to FIG. 5, 1 indicated by a broken line
The area is one chip. One area divided by a solid line is
This is the transfer size at the time of the first projection exposure. On the wafer, 280 chips are arranged in 20 rows and 14 columns.

Next, a second projection type exposure apparatus having a wider field than the first projection type exposure apparatus, and a second projection type exposure apparatus shown in FIG.
5 Imposition of a second reticle (25 chips C _b1 to CB
₂₅ is used as an example, using an array pattern of 5 rows and 5 columns = a reticle arranged in a second chip array pattern =) and a chip array already formed on the wafer 1a (by a 4-sided reticle). The second chip arrangement pattern on the reticle 2b is aligned and shot. FIG. 6 shows a state in which the first exposure apparatus and the second exposure apparatus have no distortion and alignment is ideally performed. In FIG. 6, one area divided by a thin solid line is the transfer size of the first figure formed on the wafer 1a. The area 1 divided by the thick solid line is the maximum transfer size of the second chip array pattern on the reticle 2a. Therefore, originally,
In the second photolithographic process, one wafer is exposed with 12 shots in 4 rows and 3 columns using a 25-imposition reticle.

In this embodiment, since the first exposure apparatus and the second exposure apparatus have distortion, the shot is executed while correcting the alignment error caused by the distortion as follows. I will do it. In other words, when the second chip arrangement pattern on the reticle is shot by overlapping the chips (FIG. 5) arranged on the wafer as shown in FIG. The positions indicated by the circles in FIGS. 7A and 7B respectively indicate the amount of displacement of the projected image due to distortion and the amount of displacement of the projected image due to distortion in the second wide field projection optical system 2b. Measured with
The measurement position and the amount of deviation of the projected image in each case are stored in the storage means 40.

Next, of the chips arranged on the wafer,
An image projected by the first projection optical system for several first chip arrangement patterns included in an area corresponding to a maximum area (maximum transfer size) that can be transferred by one shot of the second projection optical system. The shift amount and the measurement position are called from the storage means 40. In addition, the shift amount and the measurement position of the projected image in the area (maximum transfer size) for one second arrangement pattern in the wide field second projection type exposure apparatus are called from the storage means 40. Then, the difference between the amount of shift of the projected image in the first projection optical system and the amount of shift of the projected image in the second projection optical system is calculated by the arithmetic processing means 42. The difference between the deviation amounts is a predetermined standard value (for example, ± 100 n
If it exceeds m), the size transferred from the second chip array pattern on the second reticle on the second wide field projection type exposure apparatus is reduced in chip units. That is, the transfer size is reduced in chip units from the maximum transfer size. Then, the shift amount difference is calculated again for the reduced transfer size. When this operation is performed, the difference in deviation
The process is repeated until the size becomes 0 nm or less, and the optimum transfer size (see FIG. 8) is searched.

After the determination of the optimum transfer size, of the remaining first figures, the chip exposure pattern adjacent to the area for which the optimum transfer size has been determined by the above operation is also subjected to two exposure apparatuses in the same manner as above. A difference in the amount of displacement of the projected image is calculated between the two, and an optimum transfer size is searched so that the difference in the amount of displacement becomes a standard value ± 100 nm or less. By repeating the above operation, the optimum transfer size is found for all the areas on the wafer-1a, and stored in the storage means 40.

Then, alignment and shot are performed by a wide field projection type exposure apparatus. At this time, as shown in FIG. 2, the optimum transfer size stored in the storage means 40 is called for each shot, the transfer area is set by the blind 45, and the reticle 2b and the wafer stage 1a are moved to the transfer position. Perform alignment, shots. This operation is repeated in a step-and-repeat manner.

(Embodiment 2) Next, an alignment operation in the case where a shift between shots has occurred in the first figure on the wafer 1a will be described with reference to Embodiment 2. Usually a wide field
When transferring a second chip arrangement pattern having a larger number of chips than the first chip arrangement pattern transferred by the first projection exposure apparatus using the second projection type exposure apparatus of
FIG. 9A shows an example of an arrangement error between chip arrangement patterns between the first chip arrangement patterns formed on the wafer, that is, a shot at the time of the first projection exposure. An alignment error due to the deviation between the two is included.

Therefore, in this embodiment, as shown in FIG. 9B, a mark 46 for measuring an inter-shot arrangement error is inserted in advance into a chip included in the first chip arrangement pattern. Keep it. Then, based on the mark position, a shift amount and a position including an error between shots in the first projection optical system and a shift amount (X, Y) and a position in the wide field projection optical system are determined. A map of the amount of deviation obtained from both measurement results (distortion map)
Is stored in the storage means 40.

Then, the maximum number of the first number included in the area corresponding to the maximum transfer size of the wide field projection type exposure apparatus
The difference in the amount of deviation in the wide field projection type exposure apparatus is determined in advance from the deviation amount map including an error between shots and the deviation amount map of the second projection optical system in the chip arrangement pattern of FIG. The transfer size of the wide field projection type optical system is selected and stored so as to be smaller than the value. And
At the stage of actually performing exposure, the opening shape of the blind 45 is set in accordance with the selected optimum transfer size, the reticle 2b and the wafer 1a are moved to predetermined transfer positions, and alignment and shot are executed.

[0044]

As described above, according to the present invention, the first pattern transferred onto the wafer by the first projection type exposure apparatus is
A second field having a larger field size than the first projection exposure apparatus
When aligning a second figure on a reticle with the projection type exposure apparatus, the amount of displacement of the projected image due to distortion in the first projection type optical system and the amount of displacement of the projected image due to distortion in the second projection type optical system are determined in advance. And the transfer size in the wide-field second projection optical system is reduced in chip units until the difference between the shift amounts of the first graphic and the second graphic is equal to or smaller than a predetermined fixed value. .

Thus, according to the present invention, it is possible to reduce the deviation of the alignment itself between the first figure on the wafer and the second figure on the reticle due to the difference in distortion between the two exposure apparatuses. Therefore, alignment accuracy can be improved.

According to the present invention, there is provided a mark for measuring a deviation amount of a projection image including an alignment error between shots in the first figure with respect to the first figure on the wafer.

Thus, according to the present invention, even when there is an alignment error between shots in the first figure on the wafer, it is possible to accurately align the second figure on the reticle.

[Brief description of the drawings]

FIG. 1 is a second view of a wide field used in the practice of the present invention.
FIG. 1 is a diagram showing a configuration of a projection type exposure apparatus.

FIG. 2 is a diagram schematically showing a relationship between a field size and an optimum transfer size at the time of alignment exposure in the wide field projection optical device shown in FIG.

FIG. 3 is a view schematically showing a structure of a blind used in the wide-field projection optical device shown in FIG.

FIG. 4 is a diagram showing a chip arrangement pattern on a first reticle used in a first projection optical system, and a chip arrangement pattern on a second reticle used in a wide field projection optical system. FIG.

FIG. 5 is a view showing a chip arrangement pattern transferred onto a wafer by a first projection type exposure apparatus.

6 is a view showing a state in which a second chip array pattern is aligned with the chip array pattern shown in FIG. 5 by a wide field second projection type exposure apparatus.

FIGS. 7A and 7B are diagrams showing a displacement measuring position from an ideal grating in the first projection optical system and a diagram showing a displacement measuring position from the ideal grating in a wide field second projection optical system; is there.

FIG. 8 is a schematic diagram for explaining an optimum transfer size.

FIG. 9 is a view schematically showing an inter-shot arrangement error generated in the first projection exposure apparatus at the time of pattern transfer, and a view showing a mark for measuring an inter-shot arrangement error.

FIG. 10 is a diagram showing a basic arrangement of a reticle, a projection optical system, and a wafer when a resist pattern is formed on a wafer.

FIG. 11 is a diagram showing an example of a configuration of a projection exposure apparatus conventionally used for manufacturing an LSI.

FIG. 12 is a diagram illustrating an example of an error of a projection image due to a distortion of the projection optical system, and a diagram illustrating another example.

FIG. 13 is a diagram showing a state in which a first graphic on a wafer and a second graphic on a second reticle are aligned in a preferable state, and the first graphic on the wafer and the second graphic on the second reticle. FIG. 3 is a view showing a state in which the figure is aligned by a normal alignment operation and a positional relationship between an alignment mark on a wafer and an alignment mark on a reticle when alignment is normally performed by the normal alignment operation. FIG.

[Explanation of symbols]

 Reference Signs List 1, 1a Wafer 2, 2b Reticle 3, 3b Projection optical system 4 ITV camera 5, 6 Motor 7 Wafer stage 8 Laser interferometer system 9 Folding mirror 10, 12 Lens 11 Half mirror 20a, 20b Projected image 21, 21a, 21b Figures 22a, 22b Alignment marks 23a, 23b Chip arrangement pattern 30, 40 Storage means 31, 41 Image processing means 32, 42 Correction amount calculation means 33 Moving means 43 Reticle and wafer stage control means 44 Blind control means 45 Blind 46 Array between shots Error measurement mark

Claims (6)

[Claims] 1. A plurality of second chip arrangements using a second projection exposure apparatus having a field size wider than a field size of a first projection exposure apparatus in which a first chip arrangement pattern is transferred to a base substrate. In an exposure method including an operation of aligning and shooting a pattern, the distortion of the first projection type exposure apparatus and the distortion of the second projection type exposure apparatus are measured in advance, and based on the measurement result data, The field of the wide field second projection type exposure apparatus is set such that the difference between the distortion of the plurality of first chip arrangement patterns corresponding to the two chip arrangement patterns and the distortion of the second chip arrangement pattern is reduced. Select the transfer size from, set the blinds according to the transfer size, and then align And performing a shot. 2. A plurality of second projection type exposure apparatuses using a second projection type exposure apparatus whose field size is larger than the field size of the first projection type exposure apparatus in which the first chip arrangement pattern is transferred to the underlying substrate. An exposure method including an operation of aligning and shot of the chip arrangement pattern of claim 1, wherein the first chip arrangement pattern includes a shot-to-shot arrangement error of a plurality of first chip arrangement patterns on a base substrate. A mark for measuring the position of the first chip arrangement pattern and the second chip arrangement pattern, the mark is measured, and the measured value is set in advance. A transfer size is selected from the wide field of the second projection type exposure apparatus so that a difference between the measured and the distortion of the second projection type exposure apparatus is reduced. An exposure method, wherein after setting a blind according to the conditions, an alignment and a shot are executed. 3. After transferring a first figure onto a substrate using a first projection exposure apparatus, a second projection exposure apparatus having a larger field size than the first projection exposure apparatus is used. In an exposure method including a step of aligning a second figure on a reticle with a first figure transferred onto the substrate and using a shot, a transfer size for transferring the second figure is set in advance. The above-described two values obtained from the measured shift amount and position based on the distortion of the projected image in the first projection type exposure apparatus and the measured shift amount and position based on the distortion of the projected image in the second projection type exposure apparatus. Reducing the maximum transfer size determined by the field size of the second projection type exposure apparatus so that the difference in the amount of displacement of the projected image between the two projection type exposure apparatuses is equal to or smaller than a predetermined value. Exposure method according to claim. 4. A first step of transferring a first graphic onto a substrate using a first projection type exposure apparatus, and after forming a photosensitive resin film on the substrate after the first step, By using a second projection type exposure apparatus having a larger field size than the first projection type exposure apparatus, the second figure on the reticle is aligned with the first figure transferred on the substrate, and the photosensitive pattern is exposed. A second step of exposing the conductive resin film, wherein in the second step, a shift amount and a position based on a distortion of a projection image in the first projection type exposure apparatus; The displacement amount and the position based on the distortion of the projection image in the type exposure apparatus are measured in advance, and based on the measured data, the displacement amount due to the distortion of the first figure and the distortion amount due to the distortion of the second figure are determined. The difference from the deviation amount is predetermined To be less than a certain value, Fi of the second projection type exposure apparatus - picks transcription size from field exposure method, which comprises carrying out a shot at the transfer size. 5. A step of repeating a first chip arrangement pattern on a first reticle on a substrate by step-and-repeat using a first projection exposure apparatus to transfer a first figure. After forming the photosensitive resin film on the substrate after the first step and the first step, the second projection type exposure apparatus having a larger field size than the first projection type exposure apparatus is used. A second step of aligning a second chip arrangement pattern on a reticle with a first pattern transferred onto a substrate in a step-and-repeat manner and shot. The second step includes: Measuring in advance the shift amount and position based on the distortion of the projected image in the first projection type exposure apparatus and the shift amount and position based on the distortion of the projected image in the second projection type exposure apparatus; Based on the data obtained, the amounts of shift in the plurality of first chip arrangement patterns included in the area corresponding to the maximum transfer size in the second projection type exposure apparatus on the substrate and the second projection type The difference between the maximum transfer size and the deviation amount in the exposure apparatus is obtained. When the deviation amount is larger than a predetermined value, the transfer size in the second projection type exposure apparatus is calculated in chip units from the maximum transfer size. The operation of obtaining the difference between the shift amount in the first chip array pattern and the shift amount in the second projection exposure process again with respect to the reduced transfer size is determined by the above-described method. Iteratively selecting the optimal transfer size until it becomes equal to or less than a certain value, and repeatedly performing the operation of selecting the optimal transfer size sequentially on adjacent regions on the substrate. After the area is divided into areas having the optimum transfer size, exposure is performed by step-and-repeat shots at the optimum transfer size for each area divided into the optimum transfer size. Method. 6. The exposure method according to claim 5, wherein a shift between shots when transferring a first figure on a substrate to the first chip arrangement pattern on the first reticle is measured. The position of the mark for each shot in the first figure transferred onto the substrate is used to determine a displacement due to a distortion including an alignment error due to a displacement between shots. An exposure method, wherein a shift due to a distortion including an alignment error between shots is used to calculate a difference in a shift amount between a first chip array pattern and the second chip array pattern.