JP3085288B2 - Exposure method and device manufacturing method using the method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure method for transferring an original pattern formed on a mask onto a sensitive substrate in order to manufacture a semiconductor device or a liquid crystal device.

[0002]

2. Description of the Related Art In the manufacture of semiconductor devices, miniaturization and high integration are progressing year by year, and a lithography process of a 1-bit memory, a 4-bit memory and an increasingly narrower line width is required. In order to respond to this demand, a reduction projection type exposure apparatus (stepper) is mainly used as an exposure apparatus currently used in a lithography process. In particular, a method in which a reticle having an original image pattern is reduced to about 15 × 15 mm square by a 1/5 reduction projection lens and exposed to a resist layer on a wafer is often used.

The projection lens of this stepper has a high numerical aperture (NA) every year in order to increase the resolution. When the wavelength of the illumination light for exposure is 436 nm (g-line), the numerical aperture of the projection lens increases. A. =
The one with about 0.48 has been put to practical use. Increasing the numerical aperture of the projection lens in this way means that the effective focal depth decreases accordingly. A. =
The depth of focus of the projection lens set to 0.48 is, for example, ± 0.
8 μm or less. That is, assuming that one shot area on the wafer is 15 × 15 mm square, the entire surface (resist layer) of this area is within ± 0.8 μm (preferably ± 0.2 μm) with respect to the best imaging plane of the projection lens. Within).

In order to cope with the shortage of the depth of focus of the projection lens, a method has been proposed in which the pattern of the same reticle is multiple-exposed while the wafer is displaced in the optical axis direction with respect to the projection lens. This method increases the apparent depth of focus of the projection lens, and is one effective exposure method.

[0005]

This multi-focus exposure method is intended to guarantee the contrast over a wide focal range, although the best focus contrast is slightly reduced. This method is based on experimental results.
This is effective for a pattern for a so-called contact hole process in which the pattern surface of the reticle is almost a dark area (shielding area) and rectangular openings (transmission areas) are scattered therein. In particular, at present, it is not as effective as a contact hole for a reticle pattern such as a wiring layer in which bright and dark linear patterns are repeated. In such a reticle pattern such as a wiring layer, when the focal position is changed, light intensity due to a defocused image of a bright line portion is given to a portion that should be a dark line on the wafer, resulting in a sharp decrease in contrast and a decrease in resist. This is because the film is reduced. In the projection exposure method, the cycle of the transferable repetitive pattern is limited to a certain value or more due to the performance of the projection lens. This value is also referred to as the resolution limit of the projection lens. A. = 0.45, the line width of the bright line and the dark line of the repetitive pattern is 0. 0 on the wafer.
It is about 8 μm (4 μm on a reticle).

Therefore, even if the line width of the pattern on the reticle is reduced, a pattern having a line width smaller than that is not normally exposed, and the limitation of lithography by the projection exposure method is mainly due to the performance of the projection lens ( Resolution). Also in the proximity exposure method,
Due to the diffraction phenomenon that occurs according to the wavelength of the illumination light, it is difficult to make the repetition period of the bright line and the dark line on the mask smaller than a certain value, and the wavelength is shortened as much as possible. For this reason, special energy rays such as soft X-rays were required.

The present invention has been made in view of these problems, and enables a finer pattern to be transferred without extremely increasing the numerical aperture of a projection optical system and reducing the wavelength of illumination light to an extremely short wavelength. The first purpose is to do so. Further, a second object of the present invention is to obtain a method capable of transferring a finer pattern irrespective of a projection exposure method or a proximity exposure method.

It is a third object of the present invention to provide a method capable of sufficiently obtaining the effect of the multifocal exposure method even for most patterns other than contact holes.

[0009]

In order to achieve the above object, according to the first aspect of the present invention, a mask is irradiated with an energy beam, and the substrate is exposed to the energy beam through the mask and a projection optical system. In the method, a main pattern to be transferred onto a substrate and an auxiliary pattern for assisting imaging of the main pattern are formed on the mask, and the mask on which the main pattern and the auxiliary pattern are formed is used. When exposing the substrate, it is located on the pupil plane of the projection optical system.
Through the pupil plane of the projection optical system
The line is limited to an annular shape, and the pattern imaging surface of the projection optical system and the substrate are relatively displaced in the optical axis direction of the projection optical system.

According to a third aspect of the present invention, there is provided an exposure method for irradiating an energy beam from an illumination system to a mask and exposing the substrate via the mask and the projection optical system, wherein the mask is transferred onto the substrate. Including a main pattern to be and an auxiliary pattern provided to assist the imaging of the main pattern,
The energy beam irradiated on the substrate via the main pattern and the auxiliary pattern is limited to an annular shape by an optical member provided on a pupil plane of the projection optical system.

According to a fourth aspect of the present invention, there is provided an exposure method for irradiating a mask with an energy ray and exposing the substrate through the mask and a projection optical system, wherein the mask includes a line pattern to be formed on the substrate, An auxiliary pattern provided along the longitudinal direction of the line pattern to assist in the formation of the line pattern, and when a substrate is exposed using the mask, a pattern image forming surface by a projection optical system. And the substrate are relatively displaced in the direction of the optical axis of the projection optical system. According to a fifth aspect of the present invention, there is provided an exposure method for irradiating a mask with an energy ray and exposing the substrate through the mask and the projection optical system, wherein the mask includes a line pattern to be formed on the substrate, And an auxiliary pattern provided along the longitudinal direction of the line pattern to assist the formation of the substrate. When the substrate is exposed using the mask, the energy beam from the annular secondary energy beam source is The mask was irradiated. The invention according to claim 6 is an exposure method for irradiating a mask with an energy ray and exposing the substrate through the mask and the projection optical system, wherein the mask includes a line pattern to be formed on the substrate, And an auxiliary pattern provided along the longitudinal direction of the line pattern to assist in the formation of the substrate.When the substrate is exposed using the mask, an optical member provided on a pupil plane of the projection optical system is used. The energy ray passing through the pupil plane of the projection optical system is limited to an annular shape. Still further, the invention according to claim 8 is an exposure method for exposing a predetermined sensitive layer on a substrate through a projection optical system and forming a predetermined pattern on the predetermined sensitive layer, wherein a line pattern to be formed on the substrate, A first step of exposing the predetermined sensitive layer using a first pattern including an auxiliary pattern provided along a longitudinal direction of the line pattern to assist in forming the line pattern; A second step of exposing the predetermined sensitive layer to be exposed by using a second pattern, and forming the predetermined pattern on the predetermined sensitive layer through the first step and the second step. did.

Here, an outline of decomposing an entire pattern to be formed on a sensitive substrate (substrate having a layer sensitive to energy rays) into a plurality of patterns, aligning the decomposed patterns with each other, and exposing the exposed patterns. A description will be given based on FIG. In FIG. 1, the whole patterns to be formed on the sensitive substrate are patterns PA and PB formed in a chip (or shot) region CP, and the pattern PA is 90 ° in a line-and-space (L / S) shape. The pattern PB is a simple L / S pattern.

The patterns PA and PB are each divided into three decomposition patterns, and each decomposition pattern is formed on three reticles R1, R2 and R3. Each reticle R1, R
2 and R3, around which a light-shielding band SB corresponding to the chip area CP is formed, and three patterns PTA1, PTA2, and PTA3 obtained by decomposing the pattern PA, and three patterns PTB1 and PTB2 obtained by decomposing the pattern PB.
, PTB3 are formed. Each reticle R1
, R2, R3 have alignment marks RM1, RM
2, RM3, RM4 are provided, and marks WM1, WM2, WM3, WM4 provided in association with the chip area CP are provided.
Used for alignment with

The patterns PTA1, PTA2, PTA3,
PTB1, PTB2, and PTB3 are indicated by dark lines in the figure.
Actually, it is a bright line by the light transmitting portion. Pattern PTA1
, PTB1 are positioned in the chip area CP and exposed, and then replaced with a reticle R2 to form patterns PTA2, PTB.
2 is positioned in the chip area CP and exposed, and then the reticle R3 is positioned to expose the patterns PTA3 and PTB3.

Each of the patterns PTB1, PTB2, and PTB3 is an L / S pattern of the pattern PB, which is obtained by taking out a line pattern corresponding to a bright line every two lines, and having a pitch of line and space. Is three times (duty is 1/3) that of the whole pattern. The same applies to each of the patterns PTA1, PTA2, and PTA3.
Are broken so that a continuous line is not formed by bending at 90 ° like each line of FIG. The 90 ° bend is set so that the ends of two lines (each line is formed on a different reticle) orthogonal to each other partially overlap. As described above, in the case of the line-and-space pattern, adjacent bright lines are formed on different reticles, and the pattern density of the bright lines is reduced in one reticle (see FIG. 1). Was 1/3) to isolate the bright line.

[0016]

FIG. 2A shows a case where the entire pattern Pa in a line-and-space pattern is directly formed on one reticle R, and FIG. 2B shows a bright line of the pattern Pa of FIG. 2A. In the case of the decomposition pattern Pb formed every other line. Here, the widths of the bright lines Pa and Pb are equal and are d.
When these reticles R are irradiated with illumination light, diffracted light is generated in a direction corresponding to each pattern pitch P.
The diffraction angle θ of the n-th order diffracted light is defined assuming that the wavelength of the illumination light is λ.
sin θ = nλ / P (where n = 0, ± 1, ± 2...). That is, in the case of the decomposed pattern Pb having a large pitch with the pattern, the diffraction angle of the same diffraction order is smaller, and as a result, the amount of diffracted light contributing to the imaging of the first or higher order increases, and the image contrast increases.
An example is shown below.

FIGS. 2C, 2D and 2E show the g-line and the N.D.
A. = 0.45, σ = 0.5, the best focus at the time of projecting and exposing 0.4 μmL / S (repeated pattern of 0.4 μm wide bright line and dark line) on the photosensitive substrate. The calculated value (simulation) of the aerial image is shown. Here, the σ value represents a ratio between the area of the entrance pupil of the projection lens and the area of the light source image. FIG. 2C shows the intensity distribution of the aerial image when exposed by one reticle. The horizontal axis indicates the position (μm) on the photosensitive substrate with the center of a certain bright line as the origin, and the vertical axis indicates the relative position. Strength. FIG. 2F shows the sum of the intensities of the aerial images separated into two reticles and exposed respectively, and FIGS. 2D and 2E show the intensity distribution of the aerial image of the separated pattern. As is clear from this simulation, the contrast of the aerial image is improved by dividing and exposing the pattern.

That is, in the case of an L / S pattern,
By using two or more decomposition patterns, even if a projection lens having the same numerical aperture is used, more high-order light can be used for imaging. This means that a finer linear pattern is formed as much as possible up to the resolution limit determined by the performance of the projection lens, and the image quality of the pattern (image quality of the resist pattern) is improved.

Further, by setting the pattern Pb in which the ratio of the bright portion to the entire pattern Pa is reduced, even if the best imaging surface of the projection lens and the surface of the photosensitive substrate are defocused, the dark portion of the pattern Pb is defocused. The focus image keeps the dark part to the last, does not become brighter, and only the contrast of the brighter image is reduced. Therefore, if the multifocal exposure method is performed for each of the separation patterns, the effect of increasing the apparent depth of focus can be obtained as in the case of the contact hole.

[0020]

FIG. 3 is a perspective view showing a configuration of a projection type exposure apparatus (stepper) suitable for an embodiment of the present invention. The basic configuration of this stepper is described, for example, in Japanese Patent Application Laid-Open No. 62-145730.
Since this is the same as that disclosed in Japanese Patent Application Laid-Open Publication No. H10-209, it will be briefly described below. Illumination light from the exposure light source 2 passes through an illumination optical system 4 having a reticle blind (illumination field stop) and the like, and illuminates one reticle on a reticle stage 6. On the reticle stage 6, four reticles R1, R2, R3, and R4 can be placed at the same time and move two-dimensionally in the x and y directions. On the reticle stage 6, movable mirrors 8x and 8y for reflecting a laser beam from a laser interferometer 10 for position measurement are fixed at right angles to each other. The reticle alignment system 12 is provided so as to detect the alignment marks RM1 to RM4 of the reticle and also detect the marks WM1 to WM4 on the wafer W.
For this reason, the alignment system 12 is used both when positioning one of the four reticles with respect to the apparatus and when detecting the marks RM1 to RM4 and the marks WM1 to WM4 simultaneously to perform die-by-die alignment. Available to Although only one alignment system 12 is provided in FIG. 3, each mark RM1 shown in FIG.
, RM2, RM3, and RM4. Mark RM1 to RM4 or mark WM1
.. WM4 are detected by the mark detection system 14.

The image of the pattern area of the reticle is formed and projected on the chip area CP formed on the wafer W via the projection lens system 16. The wafer W is placed on a wafer stage 26 that moves in the x and y directions. This wafer stage is mounted on a Y stage 26y and Y that moves in the y direction.
X stage 26 moving on stage 26y in x direction
The X stage 26x is configured by a Z stage 26z that moves slightly on the X stage 26x in the projection optical axis direction (Z direction). Z stage 26
On z, moving mirrors 28x and 28y that reflect laser beams from the laser interferometers 30x and 30y are fixed at right angles to each other.

A reference mark FM is fixed to the Z stage 26z so that the reference mark FM has substantially the same height as the wafer W. Driving of the X stage 26x and the Y stage 26y in the respective axial directions is performed by motors 27x and 27y. Here, the projection lens system 16 incorporates an image forming correction mechanism 18, and the optical characteristics (magnification, focus, certain types of distortion, etc.) of the projection lens system 16 that fluctuate depending on the energy accumulation state due to the exposure light exposure, environmental conditions, and the like. ) Is automatically corrected every moment. Since the image forming correction mechanism 18 is disclosed in detail in, for example, Japanese Patent Application Laid-Open No. 60-78454,
Here, the description is omitted. Also, this stepper has
An alignment optical system 20 for detecting marks (WM1 to WM4, etc.) on the wafer W from below the reticle stage 6 via only the projection lens system 16, and photoelectrically detects mark light information detected by the alignment optical system 20. A TTL (through-the-lens) type alignment system composed of a mark detection system 22 and an off-axis type alignment system 24 separately provided immediately adjacent to the projection lens system 16 are provided.

Also, although not shown in FIG.
As described in JP-A-78454, an oblique incident light type focus sensor for detecting the height position of the surface of the wafer W with high resolution is provided, and the best imaging of the projection lens system is performed together with the Z stage 26z. It operates as an automatic focusing mechanism that always matches the plane with the wafer surface. Here, the optical relationship between the illumination optical system 4 and the projection lens system 16 in the configuration of FIG. 3 will be described with reference to FIG. Illumination optical system 4
Is configured to project a secondary light source image (surface light source) into the pupil EP of the projection lens system 16, and a so-called Koehler illumination method is adopted. The surface light source image is set to be slightly smaller than the size of the pupil EP. Now, focusing on one point of the reticle R having the entire pattern Pa, the illumination light IL reaching this point has a certain solid angle θr / 2. This solid angle θr / 2 is preserved even after passing through the entire pattern Pa, and is incident on the projection lens system 16 as a light beam Da0 of zero-order light. The solid angle θr / 2 of this illumination light IL
Is also called the numerical aperture of the illumination light. Assuming that the projection lens system 16 is a bilateral telecentric system,
On each of the reticle R side and the wafer W side, the principal ray 11 passing through the center of the pupil EP (the point where the optical axis AX passes) is parallel to the optical axis AX. The light beam that has passed through the pupil EP in this way becomes an image forming light beam ILm on the wafer W side and forms an image on one point on the wafer W. In this case, if the reduction magnification of the projection lens system 16 is 1/5,
The solid angle θw / 2 of the light beam ILm has a relation of θw = 5 · θr.

The solid angle θw / 2 is also called the numerical aperture of the image forming light beam on the wafer W. The numerical aperture on the wafer side of the projection lens system 16 alone is defined by the solid angle of the light beam ILm when the light beam passes through the entire pupil EP.

Now, if the overall pattern Pa is equivalent to that shown in FIG. 2A, the first-order or higher-order diffracted light Da1,
Da2,... Occur. These higher-order lights include a zero-order light flux D
Some are generated by spreading outside a0, and others are generated by being distributed inside the zero-order light beam Da0. In particular, the zero-order light beam Da0
A part of the high-order light distributed outside the pupil will be cut off by the pupil EP even if the light enters the projection lens system 16, for example.
It does not reach the wafer W. Therefore, if more high-order diffracted light is used for imaging, the diameter of the pupil EP must be as large as possible, that is, the numerical aperture (NA) of the projection lens system 16 must be further increased. Alternatively, by reducing the numerical aperture (solid angle θr / 2) of the illumination light IL (reducing the diameter of the surface light source image), it is possible to reduce the spread angle of the higher-order lights Da1, Da2, etc. from the pattern Pa. It is. However, in this case, if the numerical aperture (solid angle θw / 2) of the 0-order imaging light flux ILm on the wafer W side is extremely reduced, the original resolution performance will be impaired.

Furthermore, originally, the diffraction angle of higher-order light is uniquely determined by the pitch and duty of the pattern Pa. Therefore, even if it is possible to make the solid angle θr / 2 of the illumination light IL close to zero, a higher angle is required. More than a certain order of the next-order diffracted light is cut off by the pupil EP. However, when the entire pattern is divided into a plurality of decomposed patterns as in the present embodiment, as is clear from FIG. 2B, the diffraction angle of the high-order light spreading outside the 0-order light beam can be suppressed small.
The pupil EP can be easily passed.

In FIG. 3, four reticles R1 to R4 are mounted on the same reticle stage 6.
The center of any one of the reticles is the projection lens system 1
6 can be exchanged so as to be located on the optical axis AX. The positioning accuracy of each reticle at the time of this replacement depends on the laser interferometer 1
Since 0 is used, extremely high precision (for example, ± 0.02
μm). Therefore, the four reticles R1 to R4
If the mutual positional relationship is precisely measured in advance, each reticle can be positioned by moving the reticle stage 6 based only on the coordinate measurement values of the laser interferometer 10.
Further, even when the mutual positional relationship between the reticles R1 to R4 is not measured in advance, the alignment system 1 is provided for each reticle.
2. Accurate positioning can be performed using the mark detection system 14, the reference mark FM, and the like.

Further, in this embodiment, the multi-focus exposure method is used together with the exposure of each of the reticles R1 to R4 having the decomposition pattern. For this reason, when one chip area (shot area) CP on the wafer W is exposed using a certain reticle, the height position Z0 of the wafer surface detected as the best focus point by the oblique incident light type focus sensor.
The exposure is repeatedly performed at each of three focal positions, that is, a height position Z1 which is, for example, about 0.5 μm above this position Z0 and a height position Z2 which is, for example, about 0.5 μm below Z0. Therefore, while a certain chip area CP is exposed with one reticle, the height of the wafer W is changed to the Z stage 26z.
Is moved up and down in 0.5 μm steps.

Note that, instead of moving the Z stage 26z up and down during the exposure operation, the best imaging plane (reticle conjugate plane) of the projection lens system 16 itself is moved up and down by using the imaging correction mechanism 18. The effect of is obtained. In this case, as disclosed in Japanese Patent Application Laid-Open No. 60-78454, the image forming correction mechanism 18 is a system for adjusting the gas pressure in the sealed lens space in the projection lens system 16, so that the original An offset pressure value for moving the image plane up and down by about ± 0.5 μm may be added to the pressure adjustment value for correction during the exposure operation. At this time, it is necessary to select a combination of lens spaces that changes only the focal plane due to the pressure offset and does not change the magnification, distortion, and the like.

Further, by taking advantage of the fact that the projection lens system 16 is telecentric on both sides, by moving the reticle up and down, the height position of the best imaging plane can be similarly changed. In general, in the case of reduction projection, the amount of defocus on the image side (wafer side) is determined by the square of the reduction magnification when converted into the amount of defocus on the object side (reticle side). Therefore, when ± 0.5 μm defocus is required on the wafer side,
Assuming that the reduction ratio is 1/5, the reticle side is ± 0.5 /
(1/5) 2 = ± 12.5 μm.

Next, as briefly described with reference to FIG. 1, several examples of dividing the entire pattern into decomposed patterns will be described with reference to FIGS. 5, 6, 7, and 8. FIG. FIG. 5 shows that the entire pattern is composed of a bright line pattern PLc having a width D1 and a dark line pattern PLs having a width D2 (D2 ≒ D1) as shown in FIG.
This is an example in which a separation pattern as shown in FIGS. 5B and 5C is formed on each of two reticles in the case of a line-and-space in which are alternately repeated. In both the decomposition pattern of FIG. 5B and the decomposition pattern of FIG. 5C, every other bright line pattern PLc is formed compared to the entire pattern. The position of the bright line pattern PLc is complementary between the two decomposition patterns.
In this case, the pitch in the entire pattern is D1 + D2 (≒ 2
D1), the duty is D1 / (D1 + D2) ≒ 1/2, but the pitch in the decomposition pan is 2D1 + 2D2 (≒ 4D1),
The duty becomes D1 / (2D1 + 22) ≒ 1/4. Therefore, the bright line pattern PLc on each reticle is isolated.

FIG. 6 shows that, when the entire pattern is in the L / S shape as shown in FIG. 6A, each bright line pattern PLc is not divided into separate reticles, but all the bright line patterns are minute. FIG. 6 (B)
(C) shows a state where they are arranged complementarily to each other. In this method, the two decomposition patterns are determined such that the rectangular bright portions PLd, both of which are isolated, are shifted in directions perpendicular to each other in the L / S pitch direction. Therefore, focusing on any one rectangular bright portion PLd, L / S
On both sides in the pitch direction, a dark portion having a width (D1 + 2D2) exists, and the duty in the pitch direction is about 1
/ 4.

FIG. 7 shows a linear pattern which is bent at a right angle in the whole pattern as shown in FIG. 7 (A).
As shown in FIG. 7, two linear patterns PTe and PTf are divided by the bending portion in each direction. Here, the inside of the patterns PTe and PTf is a transparent portion, and the periphery thereof is a shielding portion. Here, if the two patterns PTe and PTf are bright portions, they may be partially overlapped at the bent portions. However, the overlapping portion is set to be approximately 45 ° with respect to the longitudinal direction of each of the two patterns PTe and PTf. For this reason, the pattern PTe,
The connection of PTf should not be at a right angle,
Keep the shape cut off with ゜. In this manner, the linear pattern bent at 90 ° is formed into two patterns PTe and PTf.
When the overlapping exposure is performed, the image reproduction, particularly on the bent portion of the resist, is improved, and the shape of the inner corner portion surrounded by 90 ° is clearly exposed. The same method can be applied to a linear pattern bent at another angle. Furthermore, even if the pattern is not a linear pattern but has an edge bent at an acute angle (90 ° or less), the pattern may be decomposed into two patterns according to the two directions of the edge.

FIG. 8 shows that the entire pattern intersecting in a T-shape as shown in FIG. 8A is decomposed into two linear patterns PTg and PTh according to directions as shown in FIGS. 8B and 8C. Show the case. Assuming that both the linear patterns PTg and PTh are bright portions, the leading end of the linear pattern PTg is 90
An isosceles triangle having an angle of ゜ or more is formed, and this triangle part partially overlaps the straight edge of the pattern PTh as shown in FIG. 8C. By doing so, the 90-degree corners of the T-shaped pattern are extremely sharp on the resist image, and are less likely to be rounded.

Although several examples of the pattern decomposition have been described above, the whole pattern PA shown in FIG.
The method of FIG. 7 and the method of FIG.
It is divided into TA1, PTA2 and PTA3. The number to be decomposed may be two or more, and there is no particular limitation. However, if the number of disassembled patterns (reticles) is large, errors during superposition exposure will accumulate accordingly,
It is also disadvantageous in terms of throughput.

Each of the decomposed patterns is formed on a separate reticle R1 to R4. However, as disclosed in Japanese Patent Application Laid-Open No. Sho 62-145730, a single large glass substrate is used. A plurality of pattern regions of the same size may be provided, and each decomposed pattern may be provided in each pattern region. Next, a typical sequence of this embodiment will be described with reference to FIG.

[Step 100] First, each reticle R1 to R4 having a decomposition pattern is placed on the reticle stage 6, and each reticle R1 to R4 is accurately positioned on the reticle stage 6 by using the alignment system 12. In particular, the rotation error of each of the reticles R1 to R4 is reduced with sufficient accuracy. For this reason, a minute rotation mechanism is provided on a portion of the reticle stage 6 for holding each of the reticles R1 to R4. However, a mechanism for minutely moving the reticles R1 to R4 in the x and y directions can be omitted. This is because the coordinate position of the reticle stage 6 itself is precisely controlled by the laser interferometer 10, and each reticle R1 to R
The coordinate values when the reticle stage 6 is positioned so as to detect the marks RM1 to RM4 of No. 4 by the alignment system 12 may be stored. Further, each reticle R1 to R
Since the rotation reference 2 is actually a coordinate system defined by the laser interferometers 30x and 30y on the wafer stage side, the reference mark FM and the marks RM1 to RM4 are detected by the alignment system 12, and each reticle R1 is detected. It is necessary to reduce the rotation error of R4 to zero in the coordinate system on the wafer stage side.

An alignment method relating to such reticle rotation is disclosed in, for example, Japanese Patent Application Laid-Open No. 60-1868.
No. 45 discloses this in detail.

[Step 101] Next, the opening shape and size of the reticle blind as an illumination field stop provided in the illumination optical system 4 are set so as to match the light shielding band SB of the reticle.

[Step 102] Subsequently, the wafer W coated with the photoresist is loaded on the wafer stage, and the wafer W is placed on the wafer W using the off-axis type alignment system 24 or the TTL type alignment optical system 20. Marks attached to some chip areas CP are detected, alignment of the entire wafer (global alignment) is performed, and the arrangement coordinates of the chip areas CP on the wafer W and the optical axis AX of the projection lens system 16 (reticle pattern area) (Center point) in the xy plane. Here, when the exposure on the wafer W is the first print, since the marks WM1 to WM4 do not exist, the step 102 is omitted.

[Step 103] Next, the number of decomposition patterns, that is, the pattern number n corresponding to the number of reticles
And the chip number m corresponding to the number of chip areas CP to be exposed on the wafer W is registered in the main controller including the computer. Here, the pattern number n is the number A of the reticle.
Is set to one of the numbers, and the chip number m
Is set to 9 at the maximum, and is set to 1 in the initial state.

[Step 104] Next, the reticle stage 6 is precisely positioned so that the reticle corresponding to the pattern number n is directly above the projection lens system 16.

[Step 105] Then, the wafer stage is stepped based on the chip number m, and the m-th chip area CP to be exposed is positioned immediately below the projection lens system 16. At this time, the center of the n-th reticle and the center of the m-th chip area CP are usually aligned within a range of about ± 1 μm according to the result of global alignment.

[Step 106] Next, assuming that die-by-die alignment is to be executed, the reticle marks of the marks WM1 to WM4 attached to the chip area CP using the alignment optical system 12 or the alignment optical system 20. The positional deviation with respect to RM1 to RM4 is precisely measured, and either the wafer stage 26 or the reticle stage 6 is finely moved until the positional deviation is within an allowable range.

The TTL alignment optical system 2
0 or die-by-
Instead of performing die alignment, use Japanese Patent Application Laid-Open No. 61-44
As disclosed in Japanese Patent Application Laid-Open No. 429,
Enhanced global alignment (EGA) that measures the positions of the marks WM1 to WM4 of the チ ッ プ 9 chip areas CP and obtains the stepping positions of all the chip areas by a statistical calculation method based on the measured values. ) Method may be adopted.

[Step 107] Next, the m-th chip area CP is exposed with the n-th reticle. Here, since the multifocal exposure method is applied to each chip area, first, the chip area is exposed. The oblique incident light type defocus sensor is operated to accurately measure the height position of the chip area surface with respect to the best imaging plane. Then, after adjusting to the best focus position by the Z stage 26z,
The reticle pattern is exposed at about 1/3 of the normal exposure amount. Next, for example, when a position where a 0.5 μm L / S pattern is accurately formed on the wafer W is set as the best focus, +0.5 μm, −0.5
The Z stage 26z is offset at each of the two positions changed by about μm, and exposure is performed at each height position with an exposure amount of ３. That is, in this embodiment, triple exposure is performed at a total of three height positions, that is, the best focus point and points before and after the best focus point. The exposure amount at each exposure of the multiple exposure may be approximately 1/3 of the normal exposure amount, but may be finely adjusted.
When the best imaging plane itself is moved up and down using the imaging correction mechanism 18, instead of fixing the image plane position stepwise, the image plane is continuously moved within ± 0.5 μm. Exposure can also be carried out.

In this case, the shutter provided in the illumination optical system 4 needs to be opened only once for one chip area CP, which is extremely advantageous in terms of throughput.

[Step 108] When the exposure of the m-th chip area is completed, the set value of m is incremented by one.

[Step 109] Here, it is determined whether or not exposure of all chip areas on the wafer W has been completed.
Here, the maximum value of m is set to 9;
If it is 0 or more, the process proceeds to the next step 110, and if it is 9 or less, the process returns to step 105 and stepping to the next chip area is performed.

[Step 110] When the n-th reticle is exposed on the wafer W, the wafer stage is reset to the exposure position for the first chip area, and the chip number m
Is set to 1.

[Step 111] If all the reticles of the prepared separation pattern have been exposed,
This means that the exposure for one wafer has been completed. If there are any remaining reticles, the process proceeds to step 112.

[Step 112] Next, the pattern number n is changed to a value corresponding to another reticle.
4 and the same operation is repeated. In each of the above steps, it goes without saying that step 106 is omitted in addition to step 102 in the first print.

As described above, wafers W are processed one after another. For example, when processing a plurality of wafers through the same process, the first reticle for all wafers in the lot is processed. , The reticle is replaced, and all wafers in the lot are exposed by the next reticle. Step 10
In Die-by-Die alignment, one type of mark attached to the chip area CP is commonly used when aligning with each of the reticles R1 to R4. The relative displacement between the patterns of each reticle to be transferred can be minimized.

Further, E. G. FIG. When using Method A,
Drift of each alignment system, drive system, etc. during the exposure sequence may be a problem, but by checking the drift of each system each time the reticle is replaced using the fiducial mark FM or the wafer exposure is completed. Even if drift occurs, it can be corrected immediately. As described above, in this embodiment, since each of the isolated separation patterns is subjected to multiple exposure at a plurality of focal positions, both the resolution limit and the depth of focus can be obtained. The resolution limit here means that the entire pattern on the reticle is L / S
Because of the denseness of the projection lens system, it means the limit where the bright line and the dark line when the pattern is transferred onto the resist are not well separated and resolved by the diffraction phenomenon.
This has a different meaning from the theoretical resolution of a single substance. In the present embodiment, each linear pattern in the entire pattern is decomposed so as to be isolated, and the isolated pattern is projected. Therefore, the linear pattern is almost fully used up to the theoretical resolving power of the projection lens system 16 and finer lines are used. Pattern can be transferred.
This effect is obtained when the multi-focus exposure method is not used, that is, in step 107 in FIG. 9, while the Z stage 26z is fixed at the best focus, the reticle R
Even when overlapping exposure of 1 to R4 is performed, the same can be obtained.

Next, a description will be given of a pattern decomposition method according to a second embodiment of the present invention and an exposure method associated therewith. FIG.
0 (A) is a cross section schematically showing an example of a circuit pattern configuration formed on the wafer W, and minute irregularities are formed on the wafer surface in the latter half of the manufacturing. Depending on the case, this minute unevenness may be caused by the depth of focus of the projection lens system 16 (for example, ±
0.8 μm). FIG.
(A) shows a case where a resist layer PR is formed on the wafer surface, and the projections on the wafer are exposed to patterns Pr1, Pr2, and Pr4, and the recesses are exposed to pattern Pr3. In this case, in the conventional exposure method, all of the patterns Pr1 to Pr4 as the transparent portions are formed on one reticle, but in this embodiment, the patterns Pr1, Pr2,
r4 is a transparent portion Ps on the reticle R1 as shown in FIG.
1, Ps2 and Ps4 are formed, and the pattern Pr3 exposed at the concave portion is formed as a transmission portion Ps3 on the reticle R2 as shown in FIG.

When the reticle R1 is used for the overlap exposure, the reticle R1 is exposed so that the best image forming surface of the projection lens system 16 is aligned with the convex side of the wafer W. In the case of the reticle R2, exposure is performed so that the best image forming surface is aligned with the concave side. In this way, it is possible to prevent all the patterns in the chip area CP from being exposed with extremely high resolving power and to cause partial defocusing due to the influence of the projections and depressions.

In this embodiment, each reticle R 1, R
In the second exposure, the multifocal exposure method described in the first embodiment may be used together. When the linear pattern is exposed from the concave portion to the convex portion on the wafer W, the linear pattern is decomposed in the longitudinal direction on the reticle and divided into a portion corresponding to the convex portion and a portion corresponding to the concave portion. Good. Further, the convex portion and the concave portion on the wafer W may be divided into three stages to form three decomposition patterns, and exposure may be performed at three focal positions. Of course, the decomposition rules described with reference to FIGS.

FIG. 11 is a diagram for explaining a pattern decomposition technique according to the third embodiment. In recent years, sub-patterns have been used to precisely reproduce the shape of minute isolated patterns (contact holes, etc.) and corner edges formed on the reticle.
It has been proposed to include a space mark. FIG.
1 (A) shows a small rectangular opening Pcm formed on the reticle as a contact hole, and this opening Pcm is about 1 to 2 μm square when exposed on a wafer. When this kind of opening Pcm is exposed by projection, the corner of 90 ° is often crushed and rounded on the resist. Therefore, a sub-space mark Msp of a size (for example, 0.2 μm square on the wafer) so small that it cannot be resolved by the projection optical system is provided near the four corners of the opening Pcm.

As described above, in addition to the original opening Pcm,
When the space mark Msp is formed, if the arrangement pitch of the openings Pcm becomes narrow, it becomes difficult to form the sub space mark Msp with a conventional reticle. However, as in the present invention, the opening Pcm in the entire pattern is set to 1
If every other reticle (or separate decomposition pattern) is formed together with the sub space mark Msp,
Sufficient space around one opening Pcm (shielding part)
Therefore, there is an advantage that the degree of freedom in obtaining the sub space mark Msp can be obtained.

FIG. 11B shows a case where linear sub-space marks Msp are provided on both sides near the end of the line pattern PLm. When the entire pattern is divided into decomposed patterns, the sub space mark Msp accompanying the rectangular or line pattern to be exposed must be formed on the reticle together with the decomposed pattern. When one entire pattern (for example, a bent linear pattern) is decomposed into a plurality of patterns, and a corner edge is generated in each of the decomposed patterns, a new sub space mark is provided near the corner edge or the like. You may keep it.

FIG. 12 is a view for explaining a pattern decomposition method according to the fourth embodiment. In the present embodiment, in addition to the effects described in the previous embodiments, an effect is obtained that lithography with a fine line width exceeding the resolution limit of the projection optical system is achieved. FIG. 12A shows an example of a cross section of the wafer W, and shows a case where thin line patterns Pr5, Pr6, and Pr7 extending in a direction perpendicular to the paper surface of the resist layer PR are left as a resist image.

On the resist layer PR, patterns Pr5, Pr6,
Assuming that the entire area around Pr7 is exposed, the decomposition pattern on the reticle is divided into two as shown in FIGS. 12 (B) and 12 (C). In FIGS. 12B and 12C, a light-shielding portion is formed on each of the two reticles so as to overlap each other at the patterns Pr5, Pr6, and Pr7. The width ΔD of the overlapping light-shielding portion is the pattern Pr5, Pr6, Pr7.
Is determined. As is apparent from the above description, in the conventional method, since one dark line pattern corresponding to each of the patterns Pr5, Pr6, and Pr7 is exposed, each of the patterns Pr5 to Pr7 is exposed.
Is limited by the performance of the projection lens and the like. However, in this embodiment, the width of the dark portion on the pattern separated into each of the two reticles becomes extremely large,
Hardly affected by diffraction. Therefore, the width ΔD can be made extremely small without being limited by the performance, diffraction, etc. of the projection lens. For example, a 0.4 μm line pattern can be formed using an exposure apparatus having a resolution limit of 0.8 μm.

In the case of the present embodiment, the dimensional accuracy of the pattern image transferred onto the wafer W is as follows: the alignment accuracy of the two reticles (each decomposition pattern), the alignment accuracy with each chip area CP on the wafer W, In addition, it is conceivable that the image quality may worsen depending on a pattern region creation error between the two reticles and the like. However, alignment accuracy is improving year by year, and it is thought that there will be few practical problems if a sequence in which errors in creating the pattern area of each reticle, errors in marking, etc. are measured in advance and the position is corrected during alignment is taken. Can be Further, FIG.
Although it is clear from the pattern decomposition method of FIG.
The light quantity at the time of exposure in each of the three decomposition patterns may be set to a substantially appropriate exposure amount for either of the decomposition patterns. The resist layer PR may be either a positive type or a negative type, and it is effective to use the resist layer PR in combination with a multifocal exposure method.

Next, a fifth embodiment of the present invention will be described with reference to FIG.
Description will be made with reference to (A) and (B). In recent years, attention has been paid to using an excimer laser light source as the light source of the stepper shown in FIG. The excimer laser light source uses a rare gas halide (XeCl, Krf, ArF
Etc.), the one with a high laser gain is used.
Therefore, when a high-voltage discharge is caused between the electrodes in the laser tube, strong light in the ultraviolet region can be stimulated and emitted without a special resonance mirror. In this case, the spectrum of the emitted light is broad, and the coherency is low both temporally and spatially. Such broadband light, depending on the material of the projection lens, generates extremely large chromatic aberration. In order to transmit ultraviolet light efficiently, a projection lens for an excimer laser is often made of only quartz. Therefore, the spectral width of the excimer laser light needs to be extremely narrow, and its absolute wavelength needs to be constant.

Therefore, in this embodiment, as shown in FIG. 13A, a total reflection mirror (rear mirror 201) and a low reflectance mirror (front mirror) 205 acting as a resonator are provided outside the excimer laser tube 202. To increase the coherency slightly,
Between the mirror 201 and the mirror 205 outside the mirror, two variable tilt Fabry-Perot etalons 203, 204
Was arranged to narrow the band of the laser beam. Here, the etalons 203 and 204 are two quartz plates opposed in parallel with a predetermined gap, and function as a kind of bandpass filter. Of the etalons 203 and 204, the etalon 203 is for coarse adjustment, and the etalon 204 is for fine adjustment.
By adjusting the tilt angle of the etalon 204, feedback control is sequentially performed while monitoring the wavelength fluctuation so that the absolute value of the wavelength of the output laser light becomes a constant value.

Therefore, in the present embodiment, the best image plane is optically moved up and down by positively utilizing the configuration of the excimer laser light source and the longitudinal chromatic aberration of the projection lens.
The multifocal exposure method was performed. That is, when exposing a certain chip area CP, an excimer laser (pulse or the like) is irradiated while shifting one of the etalons 204 and 203 in the excimer laser light source by a predetermined amount from the tilt angle required for absolute wavelength stabilization. I do. If the tilt angle of the etalon is shifted, the absolute wavelength slightly shifts, and the position of the best image plane changes in the optical axis direction in accordance with the axial chromatic aberration of the projection lens. Therefore, if the tilt angle of the etalon is changed discretely or continuously during exposure with an excimer laser of 50 to 100 pulses, a similar multifocal exposure method can be performed without any mechanical movement between the reticle and the wafer. Can be implemented.

FIG. 13B shows another configuration of a similar excimer laser, and a reflection type diffraction grating (grating) 206 as a wavelength selection element is used instead of the rear mirror 201.
Are provided so as to be tiltable. In this case, the grating 206 is used for coarse adjustment when setting the wavelength, and the etalon 204 is used.
Is used finely. Etalon 2 for multifocal exposure
If either one of the grating 04 and the grating 206 is inclined, the oscillation wavelength changes, and the best image plane moves up and down.

As described above, when an excimer laser is used, the image plane (focal position) can be changed by using a physical phenomenon called chromatic aberration. The chromatic aberration includes longitudinal chromatic aberration (on-axis chromatic aberration) and lateral chromatic aberration (magnification chromatic aberration). ), Each of which may occur simultaneously due to a change in wavelength. Since the chromatic aberration of magnification means to change the projection magnification, it is necessary to correct it to a negligible degree. Therefore, as an example, in the case of a projection lens that is telecentric on both sides, a field lens group (correction optical system) for maintaining telecentric provided on the reticle side in the projection lens is moved up and down in the optical axis direction. If the field lens group is moved up and down in synchronization with the inclination, lateral chromatic aberration can be corrected.

Further, even in a method in which an offset is added to the control pressure in the projection lens system 16 by using the image forming correction mechanism 18 shown in FIG. 3 in conjunction with each other, the lateral chromatic aberration (magnification error) is similarly corrected. be able to. Next, another sequence of the multifocal exposure method described above will be described as a sixth embodiment.

For this sequence, the stepper shown in FIG. 3 is provided with a differential interferometer for measuring the yawing of the wafer stage 26, and the movable mirror 28x or 2
At 8y, two length measuring beams arranged in parallel at regular intervals are projected, and the change in the optical path difference between the two length measuring beams is measured. This measured value corresponds to a minute rotation error amount generated during movement of the wafer stage 26 or after stepping.

Therefore, first, all the chip regions on the wafer W are sequentially exposed at one focus position by the step-and-repeat method. At this time, the yawing amount of the wafer stage 26 is measured and stored during the exposure of each chip area. Then, the height of the Z stage 26z is changed, or the wavelength of the excimer laser beam is shifted, and the like, and the exposure is sequentially performed from the first chip area at the second focal position in the same manner by the step-and-repeat method. At this time, the yawing amount when stepping each chip area is compared with the previously stored yawing amount at the time of exposing the chip area, and if there is no deviation within an allowable value, the exposure is performed as it is. If the comparison result shows a large difference, the rotation is corrected by the θ table that holds the wafer W and rotates slightly, or the rotation is corrected by rotating the θ table that holds the reticle.

At this time, the positional deviation between the reticle and the chip region in the x and y directions may be monitored in real time (correction of the positional deviation) while monitoring the die-by-die method using the alignment system 12 or the like. That is, x, y
The alignment error in the direction is detected by detecting the marks WM1 to WM4 and the reticle marks RM1 to RM4 attached to the chip area and controlling the reticle stage 6 or the wafer stage 26 so that the alignment error becomes zero. At the same time, the rotation of the reticle or wafer is corrected based on the yawing measurement value from the differential interferometer.

With such a sequence, the alignment time for each chip area is shortened, and
A reduction in overlay accuracy due to errors in chip rotation and wafer rotation can be ignored. Also, since the yawing amount of the wafer stage is stored, even when the multifocal exposure method is used from the time of the first layer exposure (first print), the accuracy of the overlay exposure using the disassembled reticle does not decrease at all. As described above, in the present embodiment, the focus position is not changed each time exposure of each chip region is performed, but the focus position is changed only when the first exposure for one wafer is completed, so that an improvement in throughput can be expected.

While the embodiments of the present invention have been described above, each of the decomposed patterns necessarily has a different image intensity because the pattern shape is different. Therefore, the proper exposure amount may differ for each decomposition pattern. Therefore, for each of the decomposed patterns, the transmittance or the like of the pattern area of the reticle may be measured to determine an appropriate exposure amount for each decomposed pattern. Also, reducing the numerical aperture of the imaging light beam during projection exposure also helps to increase the depth of focus. The numerical aperture of the imaging light beam is determined by the pupil EP of the projection lens.
Can be adjusted by providing a variable aperture stop plate, changing the size of the secondary light source image in the illumination optical system using a stop, a variable power optical system, or the like. Further, the luminous flux passing through the pupil EP may be limited to a ring shape (ring zone shape) by a diaphragm as shown in FIG. Alternatively, the secondary light source image may be formed in a ring shape whose diameter and width are variable or switchable.

[0075]

As described above, according to the present invention, a finer pattern can be accurately transferred without measuring an increase in the numerical aperture of the projection optical system.

Further, by performing multiple exposure while changing the wavelength by excimer exposure or the like, the choice of a method of expanding the depth of focus is expanded. These are powerful methods for solving the physical limit of how to increase the depth of focus in lithography of 0.5 μm or less using light. Furthermore, in recent years, a technique for dividing the reticle has been developed in which a sub-space mark is added to each pattern, and it is difficult to put a sub-space mark together with the present pattern on the same reticle. Also.

[Brief description of the drawings]

FIG. 1 is a diagram schematically showing a method of the present invention.

FIGS. 2A and 2B are diagrams showing a state of generation of diffracted light between a line-and-space pattern and a thinning pattern thereof. (C) is a graph showing a simulation result of an image intensity distribution in a line and space pattern. (D), (E) is a graph showing the simulation of the image intensity distribution at the time of the thinning pattern. (F) is FIG.
7 is a graph showing a simulation result obtained by superimposing the image intensities of (D) and (E).

FIG. 3 is a perspective view showing a configuration of a stepper suitable for carrying out the present invention.

FIG. 4 is a diagram showing a state of image formation in a projection optical system of a stepper.

FIG. 5 is a view for explaining a pattern decomposition method of the method of the present invention.

FIG. 6 is a view for explaining a pattern decomposition method of the method of the present invention.

FIG. 7 is a view for explaining a pattern decomposition method of the method of the present invention.

FIG. 8 is a view for explaining a pattern decomposition method of the method of the present invention.

FIG. 9 is a flowchart illustrating one exposure procedure using the method of the present invention.

FIG. 10 is a diagram illustrating a pattern decomposition method according to a second embodiment.

FIG. 11 is a diagram illustrating a pattern forming method according to a third embodiment.

FIG. 12 is a diagram illustrating a pattern decomposition method according to a fourth embodiment.

FIG. 13 is a diagram showing a configuration of a laser light source suitable for performing an exposure method according to a fifth embodiment.

FIG. 14 is a plan view showing an annular filter for adjusting the numerical aperture of an image forming light beam.

[Description of Signs of Main Parts]

 R, R1, R2, R3, R4 Reticle W Wafer CP Shot Area PA, PB Whole Pattern PTA1, PTA2, PTA3 PA Decomposition Pattern PTB1, PTB2, PTB3 PB Decomposition Pattern 2 Light Source 4 Illumination Optical System 6 Reticle Stage 16 Projection Lens 18 Image formation correction mechanism

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H01L 21/027 G03F 1/08 G03F 7/20

Claims (10)

(57) [Claims] 1. A method of irradiating a mask with an energy beam and exposing a substrate with the energy beam through the mask and a projection optical system, wherein the mask includes a main pattern to be transferred onto the substrate, An auxiliary pattern for assisting the imaging of the main pattern is formed. When the substrate is exposed using a mask on which the main pattern and the auxiliary pattern are formed, the projection light
The projection optical system is provided by an optical member provided on a pupil plane of a scientific system.
To limit the energy rays passing through the pupil plane to a ring shape
The exposure method for causing relative displacement between the substrate and the pattern image plane by the projection optical system in the optical axis direction of the projection optical system. 2. The method according to claim 1, wherein the relative displacement between the pattern image plane and the substrate is performed by changing a wavelength of an energy beam applied to the mask. 3. An exposure method for irradiating a mask with an energy beam from an illumination system and exposing a substrate via the mask and a projection optical system, wherein the mask includes a main pattern to be transferred onto the substrate and a main pattern. An auxiliary pattern provided to assist the imaging of the main pattern, wherein an energy beam irradiated onto the substrate via the main pattern and the auxiliary pattern is provided on a pupil plane of the projection optical system. An exposure method, wherein the light is limited to a ring shape by an optical member provided. 4. An exposure method for irradiating a mask with an energy ray and exposing a substrate through the mask and a projection optical system, the mask comprising: a line pattern to be formed on the substrate; An auxiliary pattern provided along the longitudinal direction of the line pattern to assist the pattern, and when exposing the substrate using the mask, a pattern image forming surface by the projection optical system and the substrate The exposure optical system is relatively displaced in the optical axis direction of the projection optical system. 5. An exposure method for irradiating a mask with an energy ray and exposing a substrate through the mask and a projection optical system, wherein the mask includes a line pattern to be formed on the substrate, and a line pattern to be formed on the substrate. And an auxiliary pattern provided along the longitudinal direction of the line pattern to assist, when exposing the substrate using the mask, the energy beam from the annular secondary energy ray source is An exposure method comprising irradiating a mask. 6. An exposure method for irradiating a mask with an energy ray and exposing a substrate via the mask and a projection optical system, the mask comprising: a line pattern to be formed on the substrate; And an auxiliary pattern provided along the longitudinal direction of the line pattern to assist, when exposing the substrate using the mask, by an optical member provided on a pupil plane of the projection optical system. An exposure method, wherein an energy ray passing through a pupil plane of the projection optical system is limited in an annular shape. 7. The method according to claim 4, wherein the auxiliary pattern is not imaged by the projection optical system. 8. An exposure method for exposing a predetermined sensitive layer on a substrate via a projection optical system and forming a predetermined pattern on the predetermined sensitive layer, comprising: a line pattern to be formed on the substrate; A first step of exposing the predetermined sensitive layer using a first pattern including an auxiliary pattern provided along a longitudinal direction of the line pattern to assist the formation; and exposing the predetermined sensitive layer using the first pattern. A second step of exposing the predetermined sensitive layer using a second pattern, wherein the predetermined pattern is formed on the predetermined sensitive layer through the first step and the second step. Exposure method. 9. The method according to claim 4, wherein the auxiliary pattern is provided on both sides of the line pattern. 10. An element manufacturing method using the method according to claim 1.