JPH0541342A - Fine pattern projection aligner - Google Patents

Abstract

(57) [Abstract] [Purpose] To prevent the reduction of the light intensity due to the aperture stop and to realize the oblique incidence in the limited direction with easy alignment. [Structure] For example, an optical fiber 4 or a cone lens is used as a means for inclining a light beam illuminating a mask by an angle corresponding to the numerical aperture of a projection lens. At the same time, the special aperture stop 6 is used as a means for illuminating from a specific one direction or a limited number of directions. As a result, the effective radius of the light source is expanded by using the optical fiber 4 and the cone lens, and the light intensity is concentrated on the peripheral portion or some areas within the peripheral portion, and a special aperture stop having a desired shape is inserted behind the concentrated light intensity. As a result, the light intensity due to the aperture stop can be prevented. Further, the oblique incidence in the limited direction can improve the resolution suitable for an LSI pattern without inserting an optical element.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a so-called projection exposure apparatus for forming a fine pattern such as an LSI on a wafer (substrate) using a projection lens.

[0002]

2. Description of the Related Art Conventionally, a projection exposure apparatus for forming a fine pattern such as an LSI is required to have a high resolution. Therefore, the projection lens of the recent projection exposure apparatus has a resolution close to the theoretical limit determined by the wavelength of light.
Nevertheless, in order to cope with the recent miniaturization of LSI patterns, higher resolution is required. In order to meet this demand, in recent years, it has been necessary to add 1 to the adjacent light transmitting parts on the reticle.
A phase shift method has been proposed in which a phase difference close to 80 degrees is provided to bring the light intensity in the light-shielding portion close to zero, and it has been shown that the resolution is improved. However, the phase shift method has drawbacks such as difficulty in manufacturing and inspecting a reticle, and difficulty in arranging an effective shifter.

On the other hand, the same applicant proposes a method of irradiating the fine pattern projection exposure apparatus by irradiating the light incident on the reticle with an angle corresponding to the numerical aperture of the projection optical system from the optical axis. (Japanese Patent Application No. 3-99822)
issue). It was theoretically and numerically shown that this exposure method can improve the resolution equivalent to that of the phase shift method. .. This method will be briefly described with reference to FIG. 8 in comparison with the conventional method of FIG.

FIG. 8 shows the above-mentioned oblique incidence projection method (Japanese Patent Application No. 3-99822) proposed by the same applicant, and FIG. 9 shows a conventional projection method. In the conventional projection method, FIG.
As shown in, the incident light I (wave number k ₀ ) is perpendicularly incident on the surface of the reticle 31 and diffracted light (wave number k ₁ ) on both sides of the optical axis z.
Cause Aperture stop under reticle 31
32, the maximum wave number that can pass through the projection system in the figure is k ₁ , which causes light with a large wave number, that is, 2
Light diffracted by a pattern with a period shorter than π / k ₁ is blocked. Here, if the diffraction angle is α '

K ₁ = k ₀ · sin α ′ (1)

Therefore, the minimum resolution dimension is 2π / (k ₀
・ It becomes 1/2 of sin α '). On the other hand, in the grazing incidence projection method shown in FIG. 8, when the 0th-order light obtained by going straight on the incident light I (wave number k ₀ ) has an inclination that passes through the outermost periphery of the projection system aperture stop, as shown in the figure. The diffracted light having the diffraction angle 2α ′ and the 0th-order light give the highest resolution. The wave number of this diffracted light is

K ₁ ′ = k ₀ · sin (2α ′) ≈2k ₀ · sin (α ′) (2)

Therefore, the minimum resolution dimension is 2π /
Half of (2k ₀ · sin α ′), that is, half the size of the conventional exposure method can be resolved. By tilting the light irradiating the reticle with respect to the optical axis in this way, it is possible to pass light having a larger diffraction angle and improve the resolution. When the 0th-order light passes through the outermost circumference, the light with the largest diffraction angle can be transmitted, so that the maximum resolution can be realized. However, the above-mentioned prior application does not mention a specific method of realizing the oblique incidence projection method.

The same applicant uses a method of tilting a light beam perpendicular to the mask by inserting an optical element in the exposure apparatus (Japanese Patent Application No. 3-142872), an optical fiber, a cone lens and a special aperture stop. It proposes a method (Japanese Patent Application No. 3-148133) of inclining the mask illumination light by widening the apparent radius of the light source and using only the light of the outer edge portion. The difference in illumination method between these methods will be described with reference to FIGS. 6 and 7.

FIG. 6 illustrates the effect of an oblique incidence light source realized by inserting an optical element, in which the light beam incident on the reticle is replaced on the exit surface of the secondary light source (FIG. 6 (a)).
1) and (b1)) and conceptually the angles at which the light flux enters the reticle (FIGS. 6 (a2) and (b2)).
The origin in FIG. 6 coincides with the intersection of the xy plane and the optical axis. 6 (a1) and 6 (a2) show the case of the right-angled triangular prism 21 shown in FIG. 6 (a), and FIGS. 6 (b1) and 6 (b2) show the isosceles triangular prism 22 or 1 shown in FIG. 6 (b). The case of a three-dimensional diffraction grating is shown. The radius of the small circle corresponds to the radius of the aperture stop of the secondary light source in an ordinary exposure apparatus, and the distance r from the origin to the center of the small circle corresponds to the tilt degree in oblique incidence illumination.

On the other hand, FIG. 7 is a diagram for explaining the effect of the light source when the oblique incidence illumination is realized by the optical fiber or the circular aperture stop. FIG. 7 (a) is a diagram of the aperture stop of the secondary light source, and FIG. 7 (b) is a diagram conceptually showing the angle of the reticle irradiation light from the light source as in FIG. As is clear from a comparison of both FIGS. 6 and 7, even in the case of illumination in which the inclination corresponding to the numerical aperture of the projection lens is similarly given, the nature of the light that actually illuminates the mask differs depending on the implementation method. .. If the illumination method using a prism as shown in FIG. 6 (a) is assumed to be one-way oblique incidence, the illumination by a circular aperture stop can be expressed as oblique incidence from all directions at a fixed angle.

Here, the inclination degree of the oblique incidence illumination in the reduction projection exposure is defined as a standardized incident angle R, and the spread degree of the light flux corresponding to the coherence σ of the light in the normal exposure is defined as σ _D.

R = m · sin β / NA σ _D = m · sin β ′ / NA (3)

Here, m is the reduction magnification, β is the angle of inclination of the center of the incident light beam from the optical axis, β ′ is the divergence angle of the light beam, and NA is the numerical aperture of the projection system. As shown in FIGS. 6 and 7, the meaning of σ _D is slightly different between the case of omnidirectional incidence such as a ring and the case of limited direction incidence such as a prism. 6 and 7, r and s are approximately represented by the following equations.

R≈sin β = (R · NA) / m s≈sin β ′ = (σ _D · NA) / m (4)

[0016]

Among the specific methods for realizing the exposure method by the grazing incidence illumination, particularly in the case of an isosceles triangular prism, the area where the light in both directions is superposed is narrowed, so that the reticle is bilaterally obliquely incident. In order to do so, it is necessary for a large area of light to be incident on both side prisms having a very large area, which is difficult at present. In addition, in the optical element insertion method that also includes right-angled triangular prisms, it is necessary to accurately align the optical element installation direction with the x and y directions of the pattern, and this can be used by replacing the element even in a removable device. If you do, there is always a problem. The omnidirectional oblique incidence method in which the light source is annular is intended to achieve optimum angular oblique incidence for patterns of all angles, so it is not so effective when there are many line patterns in the x and y directions, such as LSI patterns. It is not suitable, and the one-way oblique incidence method is highly effective in improving the resolution.

In the above two types of methods for realizing the grazing incidence illumination method as described above, the resolution improving effect is maximized in the LSI pattern having many line patterns orthogonal to the alignment problem regarding the insertion of the optical element. There was a problem that it should not happen.

[0018]

The present invention has been made in view of the above points, and has the following main features. That is, in order to solve the above-mentioned problems, it is sufficient to realize oblique incidence irradiation from a specific one direction or a limited number of plural directions without depending on the insertion of the optical element. The tilt corresponding to the numerical aperture of the projection system is realized by devising the secondary light source or the magnification of the condenser lens as in the case of circular irradiation, and the shape of the emitted light from the secondary light source is set to 1
This is achieved by inserting a special aperture stop into the secondary light source emission part, which has a shape suitable for a desired incidence method such as point-and-point oblique incidence or two-point oblique incidence.

[0019]

According to the present invention, the effective radius of the light source is expanded by using an optical fiber or a cone lens, and the light intensity is concentrated in the peripheral portion or some areas in the peripheral portion, and a special shape having a desired shape is formed behind the peripheral portion. By inserting the diaphragm, it is possible to realize the oblique incidence in the limited direction easily while preventing the reduction of the light intensity due to the aperture diaphragm. Due to the oblique incidence in the limited direction, a resolution improving effect suitable for an LSI pattern can be obtained without inserting an optical element.

[0020]

EXAMPLES Example 1: The method of the present invention is basically adapted to widen the effective radius of the light source to obtain a desired tilt angle and to realize a limited method oblique incidence such as one direction. It includes two elements of providing a light source emission hole. In addition, in order to prevent a decrease in illuminance due to a reduction in the area of the emission hole, a function of concentrating the light beam in a portion to be emitted in advance can be added.
Specific examples of the method of the present invention will be described below for each of the above functions with reference to the drawings.

In the first embodiment, the effective radius of the light source is widened and the luminous flux is concentrated on the ring by the method described in Japanese Patent Application No. 3-148133 filed by the same applicant, and in the prior application thereof. In this method, various special aperture stops as shown in FIG. 1 are used instead of the annular aperture stop. Here, the method of expanding the effective radius of the light source is the three methods shown in the previous application, namely, the method using an optical fiber, the method using a cone lens, and the method for increasing the magnification of the condenser lens while keeping the actual radius. Any of the methods for realizing the angle of oblique incidence illumination by may be applied. A desired grazing incidence light source is realized by combining these methods with the following special aperture stop.

The special aperture stop shown in FIG. 1 is inserted in front of the output lens 7 behind the filter 3 in the exposure apparatus shown in FIG. This special aperture stop is removable and replaceable. FIG. 1 shows the aperture A of the aperture stop in the secondary light source emission part, where the x-axis and y-axis correspond to the x-axis and y-axis on the reticle on the emission surface, and the origin is the point where the optical axis intersects the emission surface. Is. The diaphragm of FIG. 1 (a) realizes oblique incidence from one direction, and corresponds to the case where a right-angled triangle (single roof) prism is inserted in the optical element. This method improves the resolution only in the corresponding one direction. For example, there are many thin line patterns only in the y direction, and there are only patterns having a line width that is about three times that of the x direction and patterns that are larger than that. In such a case, the highest resolution and the improvement in the depth of focus are obtained. When the characteristics of the pattern shapes in the x direction and the y direction are opposite and there are many fine line patterns in the x direction, the aperture stop shown in FIG. 1 (a) may be rotated by 90 degrees and attached.

The diaphragm of FIG. 1 (b) is an aperture diaphragm that realizes oblique incidence from two directions symmetrical with respect to the optical axis, and corresponds to the case where an isosceles triangle (double roof) prism is inserted in the optical element.
In the case of the prism insertion, it is difficult to widen the overlapping area of the incident light in both directions, but in comparison with this method using the aperture stop, there is an advantage that almost the entire surface can be irradiated with incident light in both directions. Also when the diaphragm of FIG. 1 (b) is used, FIG.
As in the case of (1), when there are patterns in both the x and y directions, the resolution of the pattern in either one of the directions can be significantly improved. Compared to the case of Fig. 1 (a), the improvement of the depth of focus is relatively small, but almost the same effect can be obtained for the improvement of resolution, and the area of the secondary light source emission hole is doubled, so the exposure intensity is large. There are advantages.

The diaphragm of FIG. 1 (c) is an aperture diaphragm which realizes oblique incidence from two directions orthogonal to each other when viewed in a plane orthogonal to the optical axis. With this method, the pattern resolution in both the x and y directions can be improved at the same time. The diaphragm shown in FIG. 1 (d) has four directions including two directions which are symmetrical with respect to the optical axis and two directions which are symmetrical with respect to the optical axis which is orthogonal to the optical axis when viewed in a plane orthogonal to the optical axis. It is an aperture stop that realizes oblique incidence from. This method is similar to the method in Fig. 1 (c), x,
The resolution of the patterns in both y directions can be improved, and the exposure intensity can be further increased and the exposure time can be shortened as compared with the case of FIG. 1 (c).

The diaphragm of FIG. 1 (e) is an aperture diaphragm for the oblique incidence illumination method by the incident light from six directions having the same angle (60 °) with respect to the optical axis of each incident light and having the same angle with each other. The diaphragm of FIG. 1 (f) is an aperture diaphragm with similar properties for oblique incidence illumination from 8 directions. Both have properties similar to a circular aperture stop. When any of the aperture diaphragms shown above is used, the resolution of the pattern in one direction or in multiple directions can be improved by combining with the light source radius expansion method using an optical fiber or cone lens, etc. Can be improved. Also, compared to the example shown below, when the number of limited incident directions is small, the exposure intensity is greatly reduced, but the direction of the reticle incident light can be easily changed by simply changing the aperture stop according to the pattern to be formed. It has the advantage that it can be changed.

Embodiment 2 Next, a secondary light source similar to some aperture diaphragms shown in FIG. 1 is realized by another method, without using the aperture diaphragm or only when it is used supplementarily. An example of the method for FIG. 3 shows a method of concentrating a light flux at one point using an optical fiber and performing oblique incidence in one direction. Reference numeral 1 in FIG. 3 denotes a second focal point of the elliptical reflecting mirror, which is a point where the light emitted from the light source lamp 11 in the projection exposure apparatus as shown in FIG. The light emitted from the focus 1 passes through the input lens 2 to become parallel light, and up to the point where the light passes through the filter 3, the exposure is the same as in normal exposure. After this, a bundle 4 of optical fibers whose radius (2r ₁ ) at one end is smaller than the radius (2r ₀ ) at the other end is installed in the form as shown in FIG. 3 so that the light flux is concentrated in a small area and a desired inclination is given. Move to a position like. For example, when the numerical aperture of the projection system is 0.5, the reduction ratio is ⅕, and the set σ value is 0.
Consider the case in which the radius of the entrance pupil corresponding to this is 1.5 cm.

As an example of the grazing incidence illumination method at this time, if the degree of spread of the light flux σ _{D is} 0.2 and the degree of inclination R is 0.8, then the optical fiber 4 concentrates the light flux on a circle having a radius of 0.6 cm. And the center of the circle is 2.4 cm from the optical axis in the x direction.
A desired grazing incidence light source can be realized by moving it to the point. In this way, the light emitted from the optical fiber 4 passes through the optical integrator 5 and enters the output lens 7 of the illumination system as in the case of a normal exposure apparatus. A system in which the optical integrator 5 is not included is also possible, and a method of attaching a small lens to each of the exit ends of the optical fibers 4 having the same role is also effective.

When an optical integrator is inserted, an aperture stop 6 having a circular hole with a radius of 0.6 cm centered on a point at 2.4 cm from the optical axis in the x direction is provided between the optical integrator 5 and the output lens 7. It is desirable to insert it. Here, it is desirable to insert the aperture stop 6 to obtain a desired oblique incidence light source, but since the light beam having a desired diameter is already formed at a desired position by the optical fiber 4, the light blocked by this aperture is stopped. Is few. Although the case where the luminous flux is concentrated at one point has been described here, for example, in the case of incidence from two directions as shown in FIG. 1 (c), the bundle 4 of optical fibers is divided into two equal parts and one of them is projected. Exit x
A desired light source can be obtained by guiding the other side in the y direction to a predetermined position. Similarly, the grazing incidence light source corresponding to all the aperture stops shown in FIG. 1 can be made by an optical fiber.

Reference numeral 13 in FIG. 2 is a cold mirror,
14 is a condenser lens, 15 is a reticle, 16 is a projection optical system, 17 is an aperture stop of the projection system, 18 is a wafer, 1
Reference numerals 9 respectively indicate lamphouses.

Embodiment 3 Next, a method of performing oblique incidence illumination by using a combination of a cone lens, a convex lens and a concave lens instead of an optical fiber will be described. FIG. 4 is a diagram for explaining this method by taking the case of oblique incidence from two directions symmetrical with respect to the optical axis as an example. The process up to the point where the light passing through the input lens 2 passes through the filter 3 is similar to the example of the optical fiber in FIG. The light that has passed through the filter 3 passes through a lens having a shape of overlapping cones, that is, a cone lens 8, and in this process, the radius of the light beam is expanded so as to obtain a desired degree of inclination. The light that has passed through the cone lens 8 is an annular parallel light flux with a wide cross section. After that, the convex lens 9 is placed at two points symmetrical with respect to the annular optical axis, and subsequently, the concave lens 10 is placed in front of the focal point of the convex lens 9 so as to be aligned with the central axis of the convex lens. When the convex lenses 9 are placed at two positions of the light flux having a wide annular shape, the light incident on the two convex lenses 9 is condensed respectively, and the concave lens 10 is provided before the focus.
Are placed, each becomes a circular parallel light flux with a small radius.

At this time, the position of this circular light beam and the radius of the circle correspond to the degree of inclination and the degree of spread of the light beam. For example, similarly to the above example, in an exposure apparatus having a setting σ value of 0.5 and a light source radius of 1.5 cm corresponding thereto, an oblique incidence illumination with a tilt degree R = 0.7 and a light beam spread degree σ _D = 0.2 is set. When trying to realize, the two light fluxes are each 2.1 cm from the optical axis.
Centered on two points symmetrical about the optical axis at a distance of 0.6
It should be a circle of cm. For that purpose, the maximum radius of the circular light flux in the cone lens 8 is expanded to 4.2 cm. In order to collect more light, it is preferable that the two convex lenses 9 have large radii, and the radius is set so as not to overlap with other convex lenses. In this case, two convex lenses having a radius of 2.1 cm are installed so that the center of each lens is located at a distance of 2.1 cm from the optical axis.

Assuming that the focal length of the convex lens 9 is f ₁ , it is 5/7 f ₁ away from the convex lens, and a concave lens having a radius of about 1 cm is arranged so that the center of the convex lens and the center of the concave lens are aligned with a 2.1 cm point from the optical axis like the convex lens. If 10 is placed, the light passing through the concave lens 10 becomes a parallel light flux with a radius of 0.6 cm. In such a lens system, since light that does not enter the convex lens after passing through the cone lens 8 and then entering the convex lens 9 does not contribute to the formation of a light source, an aperture stop that opens a light shielding plate or a convex lens portion around the convex lens. It is desirable to set up to block the surrounding light.

The shaded areas in some limited directions of oblique incidence are shown by the shaded areas in FIG. Here, reference numeral 601 in FIG. 5 denotes a light passing area from the cone lens 8, and 602 denotes a light incident area to the convex lens 9. Fig. 5 (a) shows the case of one-way incidence, and Fig. 5 (b) shows the case of two-way oblique incidence.
FIG. 5C shows the case of oblique incidence in four directions, and FIG. 5D shows the case of oblique incidence in one direction for suppressing the decrease in light intensity. At this time, if the light intensity after passing through the cone lens 8 is uniform over the entire surface, the decrease in light intensity due to light blocking is 50% in the above example. Similarly, when convex lenses are placed at four points symmetrical with respect to the optical axis, the light intensity is reduced by about 31%.

That is, if the light is condensed using a large convex lens, the decrease in the exposure intensity due to the light shielding for realizing the oblique incidence light source is at most 3/4 even in the case of one-direction oblique incidence.
If you need a light intensity distribution that is biased in one direction, etc., you can further reduce the intensity reduction by using a prism instead of a cone lens to collect the light in the required part in advance. .. The method of directly producing the oblique incident light source in the limited direction by combining these optical fibers and lenses cannot easily change the limited direction, but has an advantage that the exposure intensity is not significantly reduced.

[0035]

As described above, according to the present invention, the resolution can be significantly improved by setting the range of the tilt of the reticle irradiation light used in the conventional apparatus according to the numerical aperture of the projection lens. Moreover, by realizing the limited-direction incidence corresponding to the line pattern often used in the LSI or the like by the light source, it is possible to obtain a higher resolution in the pattern of the LSI or the like without inserting an optical element whose alignment is difficult. it can. As described above, the use of the present invention has an effect that the degree of integration and the reliability can be improved in forming a fine pattern such as an LSI.

[Brief description of drawings]

FIG. 1 is a diagram showing a special aperture stop used in the present invention.

FIG. 2 is a schematic view of an ordinary exposure apparatus used in the present invention.

FIG. 3 is an explanatory diagram of a one-way oblique incidence light source using an optical fiber according to an embodiment of the present invention.

FIG. 4 is an explanatory diagram of a two-direction oblique incidence light source using a cone lens according to another embodiment of the present invention.

FIG. 5 is an explanatory diagram of a light blocking area around a convex lens in the present embodiment.

FIG. 6 is an explanatory view of an oblique incidence light source by inserting an optical element proposed by the same applicant.

FIG. 7 is an explanatory view of an oblique incidence light source using an optical fiber and an annular aperture stop proposed by the same applicant.

FIG. 8 is a principle explanatory diagram of reticle irradiation by an oblique incidence exposure method proposed by the same applicant.

9 is an explanatory view of reticle irradiation by a conventional method for comparison with FIG.

[Explanation of symbols]

 1 Second focus of elliptical reflecting mirror 2 Input lens 3 Filter 4 Optical fiber bundle 5 Optical integrator 6 Aperture stop 7 Output lens 8 Cone lens 9 Convex lens 10 Concave lens 11 Light source lamp 14 Condenser lens 15 Reticle 16 Projection optical system 17 Projection system aperture Aperture 18 Wafer 601 Light passing area from cone lens 602 Light incident area to convex lens

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Ryo Watanabe 1-6, Uchisaiwaicho, Chiyoda-ku, Tokyo Nihon Telegraph and Telephone Corporation (72) Inventor Katsuyuki Harada 1-6, Uchisaiwaicho, Chiyoda-ku, Tokyo No. Japan Telegraph and Telephone Corp. (72) Inventor Yoshiaki Mimura 1-1-6 Uchisaiwaicho, Chiyoda-ku, Tokyo Nihon Telegraph and Telephone Corp.

Claims (4)

[Claims] 1. A projection exposure apparatus for projecting and exposing a pattern on an object plane mask onto a wafer through a projection optical system, wherein a light beam irradiating a mask is tilted by an angle corresponding to a numerical aperture of a projection lens, and a specific value is set. A fine pattern projection exposure apparatus comprising means for illuminating from one direction or a limited number of directions. 2. The optical fiber according to claim 1, wherein the light beam irradiating the mask is inclined by an angle corresponding to the numerical aperture of the projection lens, and an optical fiber is used as a means for illuminating from a specific one direction or a limited number of plural directions. A fine pattern projection exposure apparatus characterized by the above. 3. A convex lens and a concave lens are combined as a means for inclining a light beam irradiating a mask by an angle corresponding to a numerical aperture of a projection lens and illuminating it from a specific one direction or a finite number of plural directions. A fine pattern projection exposure apparatus characterized by being used. 4. An optical fiber or a cone lens is used as a means for tilting a light beam illuminating a mask by an angle corresponding to a numerical aperture of a projection lens, or a condenser lens having a focal length shorter than usual is used. At the same time, a fine pattern projection exposure apparatus using a special aperture stop as means for illuminating from a specific one direction or a limited number of plural directions.