JPH04225357A - Projection type exposure device - Google Patents

Abstract

PURPOSE:To enhance resolution and depth of focus at the time of projecting and exposing a circuit pattern, etc. CONSTITUTION:A reticle 16 is irradiated with an illuminating luminous flux from a light source 1 through plural fly-eye lens groups 11A and 11B separating from each other, and then, the image of a reticle pattern 17 is formed and projected on a wafer (photosensitive substrate) 20 by a projecting optical system 18. Each outgoing end side 11b of the lens groups 11A and 11B is conjugated with the pupil 19 of the projecting optical system 18, respective lens groups 11A and 11B are held in one body at a discretizing position decentering from an optical axis AX by the amount which is decided in accordance with the periodicity of the reticle pattern 17. High resolution and the great depth of focus can be attained, and also, the uniformity of illuminance distribution on the reticle can be maintained in a good state.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a projection exposure apparatus used for transferring circuit patterns of semiconductor integrated devices or the like or patterns of liquid crystal devices.

0002

[Prior Art] For forming circuit patterns of semiconductor devices, etc.,
A process generally called photolithography is required. This step typically involves transferring a reticle (mask) pattern onto a sample substrate such as a semiconductor wafer. A photosensitive photoresist is coated on the sample substrate, and a circuit pattern is transferred to the photoresist in accordance with the irradiation light image, that is, the pattern shape of the transparent portion of the reticle pattern. In a projection exposure apparatus (for example, a stepper), an image of a circuit pattern to be transferred drawn on a reticle is projected and formed onto a sample substrate (wafer) via a projection optical system.

Further, an optical integrator such as a fly's eye lens or a fiber is used in an illumination optical system for illuminating the reticle, and the intensity distribution of the illumination light irradiated onto the reticle is made uniform. When using a fly-eye lens to achieve optimal uniformity,
The reticle-side focal plane and the reticle surface (pattern surface) are connected almost in a Fourier transform relationship, and the reticle-side focal plane and the light source side focal plane are also connected in a Fourier transform relationship. Therefore, the pattern surface of the reticle and the light source side focal plane of the fly's eye lens (more precisely, the light source side focal plane of each lens of the fly's eye lens) are connected in an imaging relationship (conjugate relationship). Therefore, on the reticle, the illumination light from each element (secondary light source image) of the fly-eye lens is added (superimposed) and averaged, making it possible to improve the uniformity of illumination on the reticle. It becomes.

In a conventional projection exposure apparatus, the light intensity distribution of the illumination light beam incident on the entrance surface of an optical integrator such as the above-mentioned fly's eye lens is distributed within a substantially circular (or rectangular) center around the optical axis of the illumination optical system. I tried to make it almost uniform. FIG. 13 shows a schematic configuration of the conventional projection exposure apparatus (stepper) as described above, and the illumination light beam L130 is transmitted through the fly-eye lens 11 in the illumination optical system,
A pattern 17 on a reticle 16 is irradiated via a spatial filter 12 and a condenser lens 15. here,
The spatial filter 12 is arranged at or near the reticle-side focal plane 11b of the fly-eye lens 11, that is, the Fourier transform plane (hereinafter abbreviated as pupil plane) for the reticle 16, and is centered on the optical axis AX of the projection optical system 18. It has an aperture with an approximately circular area, and a secondary light source (
Surface light source) Limits the image to a circle. The illumination light that has passed through the pattern 17 of the reticle 16 is imaged on the resist layer of the wafer 20 via the projection optical system 18. Here, the solid line representing the luminous flux represents the chief ray of light emitted from one point. At this time, the ratio of the numerical aperture of the illumination optical system (11, 12, 15) to the reticle side numerical aperture of the projection optical system 18, so-called σ
The value is determined by the aperture stop (for example, the aperture diameter of the spatial filter 12), and the value is generally about 0.3 to 0.6.

Now, the illumination light L130 is diffracted by a pattern 17 patterned on the reticle 16, and from the pattern 17, 0th-order diffracted light D0, +1st-order diffracted light DP,
and -1st-order diffracted light Dm is generated. The respective diffracted lights (D0, Dm, DP) are focused by the projection optical system 18 to generate interference fringes on the wafer (sample substrate) 20. This interference fringe is an image of the pattern 17. At this time, the angle θ (on the reticle side) formed by the 0th-order diffracted light D0 and the ±1st-order diffracted lights DP and Dm is sinθ=λ/P (λ: exposure wavelength,
P: pattern pitch).

By the way, as the pattern pitch becomes finer, sin θ increases, and when sin θ becomes larger than the reticle-side numerical aperture (NAR) of the projection optical system 18, ±1
The second order diffracted lights DP and Dm can no longer pass through the projection optical system 18. At this time, only the 0th order diffracted light D0 reaches the wafer 20, and no interference fringes are generated. In other words, sinθ>NAR
In this case, an image of the pattern 17 cannot be obtained, and the pattern 17 cannot be transferred onto the wafer 20.

From the above, in conventional projection exposure apparatuses, the pitch P such that sin θ=λ/P≈NAR is given by the following equation. P≒λ/NAR (1) From this, the minimum pattern size is half the pitch P, so the minimum pattern size is about 0.5·λ/NAR, but in the actual photolithography process, wafer curvature, Effects of wafer steps etc. due to process,
Or some depth of focus is required due to the thickness of the photoresist itself. Therefore, the practical minimum resolution pattern size is expressed as k·λ/NA. Here, k is called a process coefficient and is approximately 0.6 to 0.8. Since the ratio of the reticle-side numerical aperture NAR to the wafer-side numerical aperture NAW is the same as the imaging magnification of the projection optical system, the minimum resolution pattern size on the reticle is k·λ/NA
R, the minimum pattern size on the wafer is k・λ/NA
W = k·λ/B·NAR (where B is the imaging magnification (reduction ratio)).

Therefore, in order to transfer a finer pattern, it is necessary to choose between using an exposure light source with a shorter wavelength or using a projection optical system with a larger numerical aperture. Of course, efforts can also be made to optimize both the exposure wavelength and numerical aperture. In addition, a so-called phase shift reticle, which shifts the phase of transmitted light from a specific part of the transmitted part of the circuit pattern of the reticle by π from the phase of transmitted light from other transmitted parts, is available, for example, in Japanese Patent Publication No. 62
This is proposed in Publication No.-50811 and the like. By using this phase shift reticle, it is possible to transfer finer patterns than before.

[0009]

[Problems to be Solved by the Invention] However, in the conventional projection exposure apparatus as described above, it is difficult to make the illumination light source have a shorter wavelength than the current one (for example, 200 nm or less) by using a suitable optical system that can be used as a transmissive optical member. This is currently difficult due to the lack of materials. Furthermore, the numerical aperture of the projection optical system is already close to its theoretical limit at present, and it is almost impossible to increase the aperture further.

Furthermore, even if it were possible to make the aperture larger than the current one, the depth of focus expressed by ±λ/2NA2 would rapidly decrease as the numerical aperture increases, and the depth of focus required for actual use would decrease. The problem of fewer and fewer people is becoming more and more obvious. On the other hand, many problems remain regarding phase shift reticles, such as the complicated manufacturing process and high cost, and the methods for inspection and repair that have not yet been established.

The present invention has been made in view of the above problems, and an object of the present invention is to realize a projection type exposure apparatus that can obtain high resolution and a large depth of focus even when using a normal reticle.

0012

[Means for Solving the Problems] A projection type exposure apparatus according to the present invention is basically constructed as shown in FIG. Figure 1
2, the same members as in the prior art are given the same reference numerals. In FIG. 12, fly-eye lenses (11A, 11B
) is arranged so that its reticle-side focal plane 11b is approximately at the position of the Fourier transform plane with respect to the circuit pattern (reticle pattern) 17 on the reticle 16 (a position conjugate with the pupil plane 19 of the projection lens 18), Further, the above fly-eye lenses (11A, 11B) are arranged in a dispersed manner into a plurality of fly-eye lens groups. Also, fly eye lens 1
The illumination light amount distribution at the reticle-side focal plane 11b of lenses 1A and 11B is determined by
In order to make the light almost zero except for the individual fly-eye lens positions B, a light shielding member 10 is provided on the light source side (or on the reticle side, or integrally with the fly-eye lens) of the fly-eye lenses 11A and 11B. For this reason, fly eye lens 1
The illumination light amount distribution on the reticle-side focal plane 11b of lenses 1A and 11B exists only at the positions of each fly-eye lens group 11A and 11B, and is almost zero at other locations.

Furthermore, fly-eye lens groups 11A and 11B
Since the reticle-side focal plane 11b of is almost equal to the Fourier transform plane for the reticle pattern 17, the light intensity distribution (position coordinates of the light flux) at the reticle-side focal plane 11b of the fly-eye lens groups 11A and 11B is the illumination light flux for the reticle pattern 17. This corresponds to the angle of incidence ψ. Therefore, the individual positions of the fly-eye lens groups 11A and 11B (
The angle of incidence of the illumination light beam incident on the reticle pattern 17 can be adjusted depending on the position (in a plane perpendicular to the optical axis). Here, it is desirable that the fly-eye lens groups 11A and 11B are arranged symmetrically with respect to the optical axis AX, and each fly-eye lens group is composed of at least one lens element.

Furthermore, in the present invention, the eccentricity of the fly-eye lens groups 11A and 11B with respect to the optical axis of the illumination optical system or the projection optical system is adjusted according to the difference in the periodicity (pitch, arrangement direction, etc.) of the reticle pattern 17. Each of the plurality of holding members, which are different from each other and held together, is arranged replaceably in the optical path of the illumination optical system. For this reason,
Each of the plurality of holding members is placed in the optical path of the illumination optical system. Specifically, one of the plurality of holding members is selected that is most suitable for the reticle pattern based on the periodicity of the reticle pattern to be transferred. By arranging this selected holding member in the optical path, it becomes possible to control the incident angle of each irradiation light beam (multiple beams) incident on the reticle 16 in accordance with the periodicity of the reticle pattern 17. ing.

[0015]

[Operation] The circuit pattern 17 drawn on the reticle (mask) generally includes many periodic patterns. Therefore, from the reticle pattern 17 irradiated with the illumination light from one fly-eye lens group 11A, the 0th-order diffracted light component D0, ±1st-order diffracted light components DP, Dm, and higher-order diffracted light components are Occurs in the direction according to the degree. At this time, since the illumination light flux (principal ray) enters the reticle 16 at an inclined angle, the generated diffracted light components of each order also have a certain inclination (angular shift) in the reticle pattern, compared to when the illumination is perpendicular. It occurs from 17. Illumination light L120 in FIG. 12 is incident on the reticle 16 at an angle of ψ with respect to the optical axis.

The illumination light L120 is diffracted by the reticle pattern 17, and 0th-order diffracted light D0 travels in a direction tilted by ψ with respect to the optical axis AX, +1st-order diffracted light DP travels tilted by θP with respect to the 0th-order diffracted light, and 0 A −1st-order diffracted light Dm is generated that travels at an angle of θm with respect to the next-order diffracted light D0. Here, since the illumination light L120 enters the reticle pattern at an angle ψ with respect to the optical axis AX of the projection optical system 18, which is telecentric on both sides, the 0th order diffracted light D0 also enters the reticle pattern at an angle with respect to the optical axis AX of the projection optical system 18. It moves in a direction tilted by ψ.

Therefore, the +1st-order diffracted light DP travels in the direction of (θP +ψ) with respect to the optical axis AX, and the -1st-order diffracted light D
m travels in the direction of (θm −ψ) with respect to the optical axis AX. At this time, the diffraction angles θP and θm are si
n(θP +ψ) − sinψ=λ/P (2)
sin(θm −ψ)+sinψ=λ/P (
3). Here, it is assumed that both the +1st-order diffracted light DP and the -1st-order diffracted light Dm pass through the pupil 19 of the projection optical system 18.

When the diffraction angle increases with the miniaturization of the reticle pattern 17, the +1st-order diffracted light DP traveling in the direction of the angle (θP +ψ) becomes unable to pass through the pupil 19 of the projection optical system 18. That is, sin(θP +ψ)>
It has to do with NAR. However, the illumination light L120
Since it is incident at an angle with respect to the optical axis AX, the −1st-order diffracted light Dm can enter the projection optical system 18 even at this diffraction angle. That is, sin(θm−ψ)<NA
The relationship is R.

Therefore, the 0th order diffracted light D0 is on the wafer 20.
Interference fringes are generated by two beams of light and -1st-order diffracted light Dm. These interference fringes are an image of the reticle pattern 17, and when the reticle pattern 17 has a 1:1 line-and-space ratio, the image of the reticle pattern 17 is approximately 90% contrast on the resist layer coated on the wafer 20. It becomes possible to pattern.

[0020] The resolution limit at this time is when sin (θm - ψ) = NAR (4), therefore, NAR + sin ψ = λ/P
P=λ/(NAR +sinψ) (5) is the pitch of the minimum transferable pattern on the reticle side.

As an example, sinψ is set to 0.5×NAR
The minimum pitch of the pattern on the reticle that can be transferred is P=λ/(NAR +0.5NAR)=2λ/3NA
R (6).

On the other hand, as shown in FIG. 13, in the case of a conventional projection exposure apparatus in which the light intensity distribution of illumination light on the pupil 19 is within a circular region centered on the optical axis AX of the projection optical system 18,
The resolution limit was P≒λ/NAR as shown in equation (1). Therefore, it can be seen that higher resolution than the conventional projection exposure apparatus can be achieved. Next, by irradiating the reticle pattern with exposure light at a specific angle of incidence, the depth of focus can be increased by forming an imaged pattern on the wafer using the 0th-order diffracted light component and the 1st-order diffracted light component. I will explain the reason why.

As shown in FIG. 12, when the wafer 20 is aligned with the focal position (best image forming plane) of the projection optical system 18, the beam exits from one point in the reticle pattern 17 and reaches one point on the wafer 20. Each diffracted light component that reaches the projection optical system 18 has the same optical path length no matter which part of the projection optical system 18 it passes through. Therefore, as in the conventional case, the 0th order diffracted light component is transmitted to the projection optical system 1.
Even when passing approximately the center (near the optical axis) of the pupil plane 19 of No. 8, the optical path lengths of the 0th-order diffracted light component and other diffracted light components are equal, and mutual wavefront aberration is also zero. However, when the wafer 20 is in a defocused state where it does not match the focal position of the projection optical system 18, the optical path length of the high-order diffracted light component that is incident obliquely is focused relative to the 0th-order diffracted light component that passes near the optical axis. It is shorter in the front (away from the projection optical system 18) and longer in the rear of the focal point (closer to the projection optical system 18), and the difference therebetween corresponds to the difference in incidence angle. Therefore, the 0th order, ±
The first-order, ... diffracted light components mutually form wavefront aberrations,
This results in blurring in front and behind the focal point.

The wavefront aberration due to the above-mentioned defocus is expressed as follows: ΔF is the amount of deviation from the focal position of the wafer 20, and r is the sine of the incident angle θw when each diffracted light component is incident on the - (negative) side.
If (r=sinθw), then the amount is given by ΔFr2/2. At this time, r is the pupil plane 1 of each diffracted light component.
9 represents the distance from the optical axis AX. In the conventional projection exposure apparatus (stepper) shown in FIG.
0 passes near the optical axis AX, so r (0th order)=0. On the other hand, the ±1st-order diffracted lights DP and Dm are r (first-order) =
M·λ/P (M is the imaging magnification of the projection optical system).

Therefore, the wavefront aberration due to the defocus between the 0th-order diffracted light D0 and the ±1st-order diffracted lights DP and Dm is Δ
F・M2(λ/P)2/2. On the other hand, in the projection exposure apparatus according to the present invention, as shown in FIG.
The distance from is r (0th order)=M·sinψ.

Furthermore, the distance of the −1st order diffracted light component Dm from the optical axis on the pupil plane is r (−1st order)=M·sin(θ
m − ψ). And at this time, sinψ=sin
(θm −ψ), then the 0th order diffracted light component D0 and −1
The relative wavefront aberration due to the defocus of the next diffracted light component Dm becomes zero, and even if the wafer 20 is slightly shifted from the focal position in the optical axis direction, the image blur of the pattern 17 will not be as large as in the past. That is, the depth of focus will increase. Also, as in equation (3), sin(θm −ψ)
Since +sinψ=λ/P, the illumination luminous flux L120
If the incident angle ψ on the reticle 16 is set in the relationship sin ψ=λ/2P for a pattern with a pitch P, it is possible to significantly increase the depth of focus.

[0027]

Embodiment FIG. 1 shows the configuration of a projection exposure apparatus according to a first embodiment of the present invention.
A diffraction grating pattern 5 is provided as an optical member (part of the input optical system) for concentrating the light intensity distribution of illumination light on each of the light source side focal planes 11a of B.

In FIG. 1, the illumination light flux IL generated from the mercury lamp 1 is condensed at the second focal point f0 of the elliptical mirror 2, and then passes through a relay system such as a mirror 3 and a lens system 4 to a diffraction grating pattern 5. is irradiated. The lighting method at this time is
Although the Kohler illumination method or the critical illumination method may be used, the critical illumination method is more desirable in order to obtain a strong amount of light. Diffracted light I generated from the diffraction grating pattern 5
La and ILb are concentratedly incident on each of the fly-eye lens groups 11A and 11B by the relay lens 9. At this time, the light source side focal planes 11a of the fly-eye lens groups 11A and 11B and the diffraction grating pattern 5 have a nearly Fourier transform relationship via the relay lens 9. On the other hand, the reticle-side focal plane 1 of the fly-eye lens groups 11A and 11B
1b is arranged in the in-plane direction perpendicular to the optical axis AX so as to substantially coincide with the Fourier transform plane (pupil conjugate plane) of the reticle pattern 17. Although the illumination light to the diffraction grating pattern 5 is illustrated as a parallel light beam in FIG. 1, it is actually a diverging light beam, so the fly-eye lens group 11A,
The incident light flux to 11B has a certain size (thickness).

Further, the holding member 11 is arranged such that each center of the fly-eye lens groups 11A and 11B (in other words, the center of gravity of each light quantity distribution formed by the secondary light source images in each of the fly-eye lens groups 11A and 11B) is located on the reticle pattern. The fly-eye lens groups 11A, 11 are set at discrete positions eccentric to the optical axis AX by an amount determined according to the periodicity.
B is held together. Furthermore, the movable member 24 (switching member of the present invention) together with the holding member 11 holds a plurality of fly-eye lens groups in different eccentric states with respect to the optical axis AX according to the difference in periodicity of the reticle pattern 17. A plurality of holding members (not shown) are fixed together, and by driving this movable member 24, each of the plurality of holding members can be exchangeably placed in the optical path of the illumination optical system. However, the details will be explained later.

Here, it is desirable that each of the plurality of fly's eye lens groups (11A, 11B) fixed to the same holding member has the same shape and the same material (refractive index). Furthermore, each lens element of the individual fly-eye lens groups 11A and 11B shown in FIG. Although this was an example, the lens elements of the fly-eye lens group do not have to strictly satisfy this relationship, and the lens elements may be plano-convex lenses, convex-plano lenses, or plano-concave lenses.

Note that the light source side focal plane 1 of the fly-eye lens group
1a and the reticle side focal plane 11b are, of course, in a Fourier transform relationship. Therefore, in the case of the example in Figure 1,
The reticle-side focal plane 11b of the fly-eye lens group, that is, the exit surface of the fly-eye lens groups 11A and 11B, is in an imaging relationship (conjugate) with the diffraction grating pattern 5. Now, the light flux emitted from the reticle-side focal plane 11b of the fly-eye lens groups 11A and 11B passes through the condenser lenses 13 and 15 and the mirror 14, and illuminates the reticle 16 with a uniform illuminance distribution. In this embodiment, a light shielding member 12 is arranged on the exit side of the fly-eye lens groups 11A and 11B,
The 0th order diffracted light ILc etc. from the diffraction grating pattern 5 are cut. The light shielding member 12 is a metal plate with an opening cut out to match the fly's eye lens group, or a glass, quartz substrate, or the like, on which an opaque material such as metal is patterned. The openings of the light shielding member 12 correspond to the positions of the fly-eye lens groups 11A and 11B, respectively. Therefore, the illumination light amount distribution in the vicinity of the reticle-side focal plane 11b of the fly-eye lens groups 11A and 11B can be made zero at positions other than the positions of the respective fly-eye lens groups 11A and 11B. Therefore, the illumination light irradiated onto the reticle pattern 17 is only the light flux emitted from the fly-eye lens groups 11A and 11B (the light flux from the secondary light source image), and the angle of incidence on the reticle pattern 17 is also limited to a specific angle of incidence. is restricted only to luminous fluxes with .

Here, in this embodiment, since the holding members (fly's eye lens groups 11A, 11B) are replaceable, the opening of the light shielding member 12 is also variable accordingly, or the light shielding member 12 must also be interchangeable. For example, it is desirable to fix the light shielding member 12 to a holding member together with the fly's eye lens groups 11A and 11B so that they can be replaced together. Note that the size (thickness) of the light beam incident on each of the fly-eye lens groups 11A and 11B is
If the size of the light source side focal plane 11a is set to be approximately equal to or smaller than the size of the light source side focal plane 11a, the light shielding member 12 is particularly
Needless to say, it is not necessary to provide the lens near the fly-eye lens group.

As described above, by irradiating the illumination light onto the reticle pattern 17 at a specific angle of incidence, the diffracted light generated from the reticle pattern 17 on the reticle 16 is
As described with reference to FIG. 12, the telecentric projection optical system 18 focuses and images the light, and the image of the reticle pattern 17 is transferred onto the wafer 20. When diffracting the illumination light beam using the above-mentioned diffraction grating pattern 5 and concentrating the diffracted light on a specific position (fly's eye lens group) within the light source side focal plane of the fly's eye lens groups 11A and 11B, The position changes depending on the pitch and directionality of the diffraction grating pattern 5. Therefore, each fly eye lens 11A
, 11B, the pitch and directionality of the diffraction grating pattern 5 are determined.

Further, as described above, an image of the diffraction grating pattern 5 is formed on the reticle side focal plane 11b of the fly eye lens 11, and the reticle pattern surface 17 and the reticle side focal plane of the fly eye lens groups 11A and 11B are formed. Surface 11b
Since this is a Fourier transform relationship, the illumination intensity distribution on the reticle 16 will not be made non-uniform due to defects, dust, etc. in the diffraction grating pattern 5. Further, the diffraction grating pattern 5 itself does not form an image on the reticle 16 and deteriorate the illuminance uniformity.

Here, the diffraction grating pattern 5 may be formed by patterning a light shielding film such as Cr on the surface of a transparent substrate, for example a glass substrate, or may be formed by patterning a dielectric film such as SiO2. Alternatively, it may be a so-called phase grating. For phase gratings, 0
Generation of the next diffracted light can be suppressed. Further, the diffraction grating pattern 5 may be not only a transparent pattern but also a reflective pattern. For example, a high reflectance film such as a metal film such as Al or a dielectric multilayer film patterned into a diffraction grating may be used on the surface of a flat reflecting mirror such as glass, and the steps to give a phase difference to the reflected light cause diffraction. It may also be a high reflectance mirror patterned in a grid pattern.

In the case where the diffraction grating pattern 5 is reflective, as shown in FIG. The diffracted light may be concentrated near the fly-eye lens groups 11A and 11B via the relay lens 9. Note that even if the individual fly-eye lens groups 11A, 11B are moved (that is, the holding members are replaced), the diffraction grating pattern is used so that the illumination light can be concentrated near the respective fly-eye lens groups 11A, 11B. 5 or 5A
can be exchanged for one with a different pitch.

Furthermore, the diffraction grating pattern 5 or 5A may be rotatable in any direction within a plane perpendicular to the optical axis AX. In this way, when the pitch direction of the line-and-space pattern in the reticle pattern 17 is different from the X and Y directions (that is, the fly-eye lens groups 11A and 11B move (rotate around the optical axis AX) according to the pitch direction). ) can also be accommodated.

Furthermore, the relay lens 9 can be a zoom lens system (such as an afocal zoom expander) made up of a plurality of lenses, and the focal length can be changed to change the focusing position. However, in this case, the diffraction grating pattern 5 or 5A and the fly eye lens groups 11A and 11B
The relationship between the focal plane 11a on the light source side and the focal plane 11a on the light source side should not be changed.

Incidentally, FIG. 1 shows a main control system 50 that centrally controls the entire apparatus, and a bar code representing a name formed on the side of the reticle pattern 17 while the reticle 16 is being conveyed directly above the projection optical system 18. A barcode reader 52 for reading BC, a keyboard 54 for inputting commands and data from the operator, and a movable member to which a plurality of holding members (fly's eye lens groups 11A, 11B and light shielding member 12) are fixed are driven. A drive system (motor, gear train, etc.) 56 is provided. In the main control system 50,
Names of a plurality of reticles to be handled by this projection exposure apparatus (for example, a stepper) and operation parameters of the stepper corresponding to each name are registered in advance. When the barcode reader 52 reads the reticle barcode BC, the main control system 50 determines the pre-registered positions of the fly-eye lens groups 11A and 11B (pupil conjugate plane) as one of the operating parameters corresponding to the name. The holding member that best matches the information (corresponding to the periodicity of the reticle pattern) regarding the position within the reticle pattern is selected from among the plurality of holding members, and a predetermined drive command is output to the drive system 56. . As a result, the previously selected holding members (fly's eye lens groups 11A and 11B) are set in the position as explained in FIG. 12. The above operations can also be executed by the operator directly inputting commands and data to the main control system 50 from the keyboard 54.

The first embodiment has been described above, but
The optical member that concentrates the light intensity distribution on the light source side focal plane of the fly-eye lens group near the position of each fly-eye lens is not limited to the diffraction grating pattern 5 or 5A. The reflective diffraction grating pattern 5 shown in FIG. 2 above
Instead of A, the movable plane mirror 6 is arranged as shown in FIG. 3, and a drive member 6a such as a motor for rotating the movable plane mirror 6 is provided. If the plane mirror 6 is rotated or vibrated by the driving member 6a, the light amount distribution within the light source side focal plane (incidence plane) 11a of the fly-eye lens groups 11A and 11B can be changed over time. By rotating the plane mirror 6 to a plurality of appropriate angular positions during the exposure operation, the light source side focal plane 1 of the fly-eye lens groups 11A and 11B can be adjusted.
The light amount distribution within 1a can be concentrated only in the vicinity of any one of the plurality of fly-eye lens groups. In addition, such a movable reflecting mirror 6
When using the relay lens system 9, the relay lens system 9 may be omitted.

Furthermore, each fly-eye lens group 11A
, 11B moves (replaces the holding member 11),
Changing the angular coordinates of the plurality of angular positions of the above-mentioned plane mirror 6,
It is sufficient to concentrate the reflected light flux near the fly-eye lens group at the new position. By the way, the light shielding member 1 shown in FIG.
2 is provided on the entrance surface side of the fly-eye lens groups 11A and 11B, but it may be provided on the exit surface side as in FIG.

FIG. 4 shows an optical fiber bundle 7 as an optical member for converging the illumination light flux on each of the fly's eye lens groups.
It is a schematic diagram when using. It is assumed that the structure closer to the light source than the relay lens system 4 and closer to the reticle than the fly-eye lens groups 11A and 11B is the same as that in FIG. Generated from a light source,
The illumination light transmitted through the relay lens system 4 is adjusted to have a predetermined numerical aperture (NA) and enters the entrance portion 7a of the optical fiber bundle 7. The optical fiber bundle 7 is divided into a plurality of bundles corresponding to the number of fly-eye lens groups while reaching the injection part 7b.
Each of the injection parts 7b includes a fly-eye lens group 11A,
11B near the light source side focal plane 11a. At this time, a lens (for example, a field lens) may be provided between each emission part 7b of the optical fiber bundle 7 and the fly's eye lens group 11, and the lens may be used to control the light source side focal plane of the fly's eye lens group 11. 11a and the light exit surface of the optical fiber exit section 7b may be in a Fourier transform relationship. Furthermore, if each injection part 7b (or the lens between the injection part 7b and the fly's eye lens group 11b) is movable one-dimensionally or two-dimensionally within a plane perpendicular to the optical axis by a driving member such as a motor, Even when the fly's eye lens group is moved due to replacement of the holding member, the illumination light beam can be concentrated near the position of each fly's eye lens group after the movement.

FIG. 5 shows an example in which a prism 8 having a plurality of refractive surfaces is used as an optical member for concentrating the illumination light flux on each fly's eye lens group. The prism 8 in FIG. 5 has an optical axis A
It is divided into two refractive surfaces with X as the boundary, and the optical axis A
Illumination light incident above X is refracted upward and optical axis AX
Illumination light incident further downward is refracted downward. Therefore, the light source side focal plane 1 of the fly-eye lens groups 11A and 11B
Illumination light can be concentrated near the individual fly-eye lens groups 11A and 11B on 1a according to the refraction angle of the prism 8.

The number of divisions of the refractive surface of the prism 8 is not limited to two, but may be divided into any number of surfaces depending on the number of fly-eye lens groups. Furthermore, the divisional positions do not have to be symmetrical with respect to the optical axis AX. Furthermore, by replacing the prism 8, even if the fly-eye lens groups 11A and 11B move due to replacement of the holding member, the illumination light flux can be accurately focused on the positions of the respective fly-eye lens groups 11A and 11B. be able to.

Further, the prism 8 at this time may be a polarizing light splitter such as a Wollaston prism. However, in this case, since the polarization directions of the divided light beams are different, it is better to take the polarization characteristics of the resist of the wafer 20 into consideration and align the polarization characteristics in one direction. Further, if a reflecting mirror having a plurality of reflecting surfaces at different angles is arranged as shown in FIG. 3 instead of the prism 8, the driving member 6a becomes unnecessary. Needless to say, it is good to have a function for replacing the prism etc. in the apparatus. Also, when such a prism or the like is used, the relay lens system 9 can be omitted.

FIG. 6 shows a plurality of mirrors 8a, 8b,
This is an example using 8c and 8d. In FIG. 6, the illumination light transmitted through the relay lens system 4 is reflected by primary mirrors 8b and 8c so as to be separated into two directions, guided to secondary mirrors 8a and 8d, and reflected again to the fly's eye. The light reaches the light source side focal plane 11a of the lens groups 11A and 11B. If each of the mirrors 8a, 8b, 8c, and 8d is provided with a position adjustment mechanism and a rotation angle adjustment mechanism around the optical axis AX, fly-eye lens groups 11A and 11B can be easily adjusted when the holding member is replaced.
Even after moving, the illumination light flux can be concentrated in the vicinity of each of the fly-eye lens groups 11A and 11B. Furthermore, each of the mirrors 8a, 8b, 8c, and 8d may be a plane mirror, a convex mirror, or a concave mirror.

Furthermore, lenses may be provided between the secondary mirrors 8a, 8d and the fly-eye lens groups 11A, 11B, respectively. In FIG. 6, primary mirrors 8b, 8c, secondary mirror 8
Although two mirrors are used for both a and 8d, the number is not limited to this, and the mirrors may be appropriately arranged depending on the number of fly-eye lens groups. In each of the above embodiments,
Although the number of fly-eye lens groups is two in all cases, the number of fly-eye lens groups may of course be three or more. Regarding the optical components that concentrate the illumination light on each fly-eye lens group, we have mentioned that the light is mainly concentrated in two places, but the illumination light can be concentrated in multiple positions depending on the number of fly-eye lens groups. Needless to say, it's a shame. In all of the above embodiments, it is possible to concentrate the illumination light on any arbitrary position (corresponding to the position of the fly's eye lens group). Further, the optical member for concentrating illumination light on each fly's eye lens group is not limited to the type mentioned in the embodiment, and may be of any other type.

Further, the light shielding member 12 may be replaced with the light shielding member 10 provided near the light source side focal plane 11a of the fly-eye lens group as shown in FIG. 12, or as shown in FIGS.
Each of the embodiments shown above and the light shielding member 1 shown in FIG.
0 may be used in combination. In addition, the light shielding member 10
, 12 is the reticle-side focal plane 11b of the fly-eye lens group.
For example, the two focal planes 11a, 11
A suitable place is between ``b'' and ``b''.

Furthermore, the individual fly-eye lens groups 11A,
The optical member (input optical system) that concentrates the illumination light only in the vicinity of the reticle 11B is used to prevent loss of the amount of illumination light that illuminates the reticle 16, and is used to achieve high resolution, which is a feature of the projection exposure apparatus of the present invention. This is not directly related to the configuration for obtaining the effect of large depth of focus. Therefore, the optical member may be a lens system having a large diameter enough to allow the illumination light to enter the flood into each of the plurality of fly's eye lens groups.

FIG. 7 is a diagram showing the configuration of a projection type exposure apparatus (stepper) according to another embodiment of the present invention, in which a mirror 14, a condenser lens 15, a reticle 16, and a projection optical system 18 are the same as those in FIG. be. Note that the holding member 11 and movable member 24 that hold the fly-eye lens groups 11A and 11B are omitted here. Further, the portion closer to the light source than the fly-eye lens groups 11A and 11B has one of the configurations shown in FIGS. 1 to 6 or FIG. 12 described above. moreover,
A light shielding member 12a having an arbitrary opening (transmissive part) is provided near the reticle-side focal plane 11b of the fly-eye lens groups 11A, 11B.
Limit the illumination light flux emitted from 1B.

Since the Fourier transform of the reticle-side focal plane 11b of the fly-eye lens groups 11A and 11B with respect to the relay lens 13a becomes a conjugate plane with the reticle pattern 17,
A variable field stop (reticle blind) 13d is provided here. Then, it is Fourier transformed again by the relay lens 13b and reaches the conjugate plane (Fourier plane) 12b of the reticle-side focal plane 11b of the fly-eye lens groups 11A and 11B. The aforementioned light shielding member 12a may be provided on this Fourier plane 12b.

The illumination light flux from each fly-eye lens group 11A, 11B is further transmitted to condenser lenses 13C, 15.
, and guided to the reticle 16 by the mirror 14. Incidentally, as long as the system is such that the illumination light beam incident on each of the fly's eye lens groups 11A and 11B can be effectively concentrated only there, there is no problem even if the light shielding member is not provided at the position 12a or 12b in the figure. Even in this case, the field aperture (
Reticle blind) 13d can be used.

In any of the above embodiments, the diameter of each opening of the light shielding members 10, 12, 12a (or the area of each exit end of the fly-eye lens group) is the diameter of each opening of the light shielding members 10, 12, 12a. It is desirable that the ratio between the numerical aperture for the reticle 16 and the reticle-side numerical aperture (NAR) of the projection optical system 18, the so-called σ value, be set to about 0.1 to 0.3. When the σ value is smaller than 0.1, the pattern fidelity of the transferred image deteriorates, and when it is larger than 0.3, the effect of improving resolution and increasing the depth of focus becomes weak. In addition, in order to satisfy the σ value condition (approximately 0.1<σ<0.3) determined by one of the fly-eye lens groups, the area of the exit end of each fly-eye lens group 11A, 11B (light The size in the in-plane direction perpendicular to the axis) may be determined according to the illumination light flux (emission light flux).

Furthermore, each fly-eye lens group 11A, 11
A variable aperture stop (equivalent to the light shielding member 12) may be provided in the vicinity of the reticle-side focal plane 11b of B, and the numerical aperture of the light beam from each fly-eye lens group may be made variable to change the σ value. At the same time, a variable aperture stop (N
By providing an A-limited aperture), the NA and σ value of the projection system can be further optimized.

[0055] Furthermore, the light beam incident on each fly-eye lens group is illuminated widely to a certain extent outside the entrance end face of each fly-eye lens group, and the distribution of the amount of light incident on each fly-eye lens group is uniform. It is preferable if there is one, since the uniformity of illuminance on the reticle pattern surface can be further improved. Next, the structure of the movable member 24 for replacing the holding member (switching member of the present invention) suitable for each of the above embodiments will be explained using FIGS. 8 and 9.

FIG. 8 is a diagram showing a specific configuration of the movable members, and here four holding members 11, 21, 22, 2 are shown.
3 are arranged at approximately 90° intervals on a movable member (turret plate) 24 that is rotatable about a rotating shaft 24a. FIG. 8 shows that the illumination light beams ILa and ILb (dotted lines) are incident on the fly-eye lens groups 11A and 11B, respectively, and the holding member 11 is arranged in the illumination optical system. At this time, the holding member 11 is placed in the illumination optical system so that its center substantially coincides with the optical axis AX. The plurality of fly-eye lens groups 11A and 11B are integrated so that their respective centers are set at discrete positions offset from the optical axis AX of the illumination optical system by an amount determined according to the periodicity of the reticle pattern. are held by the holding member 11, and here they are arranged substantially symmetrically with respect to the center of the holding member 11 (optical axis AX).

Now, the four holding members 11, 21, 22,
Each of 23 positions the plurality of fly-eye lens groups in an eccentric state with respect to the optical axis AX (the center of the holding member) (i.e., a position in a plane substantially perpendicular to the optical axis AX) according to the difference in periodicity of the reticle pattern 17. ) are kept different from each other. The holding members 11 and 21 both have two fly-eye lens groups (11A, 11B) and (21A, 21B), and when these fly-eye lens groups are arranged in the illumination optical system, the arrangement direction are fixed so that they are almost orthogonal to each other. The holding member 22 arranges and fixes the four fly-eye lens groups 22A to 22D at substantially equal distances from the center 22c (optical axis AX). The holding member 23 has one fly-eye lens group 23A fixed approximately at the center, and is used when performing conventional exposure.

As is clear from FIG. 8, the four holding members 11, 21, 22 are rotated by rotating the turret plate 24 by the driving element 25 consisting of a motor, gears, etc. according to the information of the reticle barcode BC as described above. ,
23 can be replaced, and it becomes possible to arrange a desired holding member in the illumination optical system according to the periodicity (pitch, arrangement direction, etc.) of the reticle pattern.

Here, since a plurality of fly's eye lens groups are fixed in a predetermined positional relationship in each of the four holding members, there is no need to adjust the position between the plurality of fly's eye lens groups when replacing the holding member. do not have. Therefore, since it is sufficient to align the entire holding member with respect to the optical axis AX of the illumination optical system, there is an advantage that a precise positioning mechanism is not required. At this time, since the drive element 25 is also used for positioning, it is desirable to provide a rotation angle measuring member such as a rotary encoder. Note that each of the plurality of fly-eye lens groups forming the holding member has 16 lenses (only 3 fly-eye lens groups 23A) as shown in FIG.
Although it is composed of 6 lens elements, it is not limited to this, and in extreme cases, a fly-eye lens group may be composed of 1 lens element.

Further, in FIG. 1, the light shielding member 12 is arranged behind the holding member 11 (on the reticle side), but if each of the holding members other than the fly-eye lens group is used as a light shielding part, the light shielding member 12 can be There is no need to provide it. At this time, the turret plate 24 may be a transmitting portion or a light blocking portion. Furthermore, the number of holding members to be fixed to the turret plate 24,
The eccentric states (positions) of the plurality of fly-eye lens groups are not limited to those shown in FIG. 8, but may be set arbitrarily according to the periodicity of the reticle pattern to be transferred. In addition, when it is necessary to strictly set the angle of incidence of the illumination light beam onto the reticle pattern, each of the multiple fly-eye lens groups is moved in the radial direction (radial direction) around the optical axis AX in the holding member. The holding member (fly-eye lens group 1) can be moved slightly around the optical axis AX.
1A, 11B) may be configured to be rotatable. On this occasion,
In particular, when using the optical fiber bundle 7 (FIG. 4) as an optical member (input optical system) for concentrating the illumination light flux near each of a plurality of fly-eye lens groups, as the fly-eye lens groups move, The exit end 7b may also be configured to move, for example, the exit end 7b and the fly's eye lens group may be fixed together. In addition, as the holding member rotates, the rectangular fly-eye lens group also tilts relatively, but when the holding member is rotated, only the position of the fly-eye lens group moves without causing the above-mentioned tilt. It is desirable to configure the

Furthermore, when replacing the holding member, it is also necessary to replace the input optical system, such as the diffraction grating pattern 5, the relay lens 9 (FIG. 1), and the optical fiber bundle 7 (FIG. 4). It is desirable to integrally configure an input optical system for each member according to the eccentricity of the plurality of fly-eye lens groups and to fix it to the movable member 24.

FIG. 9 is a diagram showing a modification of the movable member for replacing the holding member, in which the input optical system (optical fiber bundles 71, 72) and the holding member (32, 34) are integrated into the movable member (support rod). 36). Although a case will be described here in which an optical fiber bundle is used, the input optical system may be any other optical system shown in FIGS. 1, 5, or the like. The basic configuration (an example in which an optical fiber bundle is used as the input optical system) has been explained with reference to FIG. 4, so it will be briefly explained here.

In FIG. 9, two fly-eye lens groups 30A and 30B are held together by a holding member 32,
The optical fiber bundle 71 has an entrance part 71a and an exit part 71b.
are both held by a fixture 33, and the holding member 32 is integrally fixed to the fixture 33. Furthermore, the inside of the holding member 32 includes fly-eye lens groups 30A and 30B.
The remaining portions are light shielding portions (hatched portions in the figure, corresponding to the light shielding member 12 in FIG. 1, for example). On the other hand, the replacement fly-eye lens groups 31A and 31B are held together by a holding member 34, and the optical fiber bundle 72 has its entrance part 72a and exit part 72b both held by a fixture 35, and the holding member 34 It is integrally fixed to the fixture 35,
Similar to the above, the inside is a light shielding part. moreover,
The fixtures 33 and 35 are connected and fixed by a connecting member 37. Therefore, when replacing the holding member, it is sufficient to replace the entire fixture. In FIG. 9, the fixture 33 (holding member 32) is present in the illumination optical system, and the replacement fixture 35 is set at a position outside the illumination optical system. In addition, the light source side from the relay lens system 4, and the condenser lens 1
It is assumed that the reticle side from 3 has the same configuration as in FIG. 1, for example.

By the way, when replacing the holding member, the drive element 3
8 by pushing and pulling the support rod 36. Therefore, if the fly-eye lens group and the optical fiber bundle are constructed so that they can be replaced together when replacing the holding member as shown in FIG. This has the advantage that it is only necessary to align them, and there is no need to adjust the position between each component (fly's eye lens group, optical fiber bundle, etc.) each time it is replaced. At this time,
Since the driving element 38 is also used for positioning, it is desirable to provide a position measuring member such as a linear encoder or potentiometer.

Note that the number of fly-eye lens groups and lens elements constituting the fly-eye lens groups for each holding member shown in FIGS. 8 and 9 may be arbitrary. Also, the shape of the exit surface is not limited to a rectangle. Now, each position of the plurality of fly-eye lens groups shown in FIGS. 8 and 9 (in a plane perpendicular to the optical axis), in other words, the holding member to be selected depends on the reticle pattern to be transferred. It is better to decide (change). The determination (selection) method in this case is as described in the function section, so that the illumination light flux from each fly-eye lens group has the effect of improving the optimal resolution and depth of focus for the fineness (pitch) of the pattern to be transferred. The holding member may have a fly-eye lens group at or near the position where the light is incident on the reticle pattern (incidence angle ψ) so as to obtain the desired angle of incidence.

Next, a specific example of determining the position of each fly-eye lens group in order to select the optimum holding member will be explained using FIG. 10 and FIGS. 11(A) to 11(D). Figure 10 shows reticle pattern 1 from fly-eye lens groups 11A and 11B.
7, the reticle-side focal plane 11b of the fly-eye lens group 11 coincides with the Fourier transform plane 12c of the reticle pattern 17. Further, at this time, a lens or lens group that brings the two into a Fourier transform relationship is represented as a single lens 15. Further, from the fly-eye lens side principal point of the lens 15 to the reticle-side focal plane 11b of the fly-eye lens group 11,
It is assumed that the distance from the reticle pattern 17 to the reticle pattern 17 and the distance from the reticle side principal point of the lens 15 to the reticle pattern 17 are both f.

FIGS. 11(A) and 11(C) both show examples of a partial pattern formed in the reticle pattern 17, and FIG. 11(B) shows an example of a partial pattern formed in the reticle pattern 17. Indicates the position of the center of the optimal fly-eye lens group on the Fourier transform plane (or pupil plane of the projection optical system),
FIG. 11(D) is a diagram showing the optimum position of each fly-eye lens group in the case of the reticle pattern of FIG. 11(C).

FIG. 11A shows a so-called one-dimensional line-and-space pattern, in which transmitting parts and light-blocking parts are arranged in a band shape with equal width in the Y direction, and they are arranged at a pitch P in the X direction.
are arranged regularly. At this time, the optimal position of each fly-eye lens group is on the line segment Lα in the Y direction assumed in the Fourier transform plane and on the line segment L
Any position on β. FIG. 11(B) is a diagram of the Fourier transform surface 12c (11b) for the reticle pattern 17 viewed from the optical axis AX direction, and the coordinate system X, Y within the surface 12c is the same as that for the reticle pattern 17 viewed from the same direction. It is the same as FIG. 11(A). Now, in FIG. 11(B), the distances α and β from the center C through which the optical axis AX passes to each line segment Lα and Lβ are α=β, and when λ is the exposure wavelength, α=β=f・( 1/2)・(λ/P). If we express these distances α and β as f・sinψ, then sinψ=
λ/2P, which is consistent with the numerical value stated in the section of the effect. Therefore, if the center positions of each fly's eye lens are on the line segments Lα and Lβ, the 0 0 generated by the illumination light from each fly's eye lens group for the line-and-space pattern shown in FIG. The two diffracted lights, either the second-order diffracted light or the ±1st-order diffracted light, pass through positions on the projection optical system pupil plane 19 that are approximately equidistant from the optical axis AX. Therefore, as described above, the depth of focus for the line and space pattern (FIG. 11(A)) can be maximized, and high resolution can be obtained.

Next, FIG. 11C shows a case where the reticle pattern is a so-called isolated space pattern, and the pitch of the pattern in the X direction (horizontal direction) is Px and the pitch in the Y direction (vertical direction) is Py. It has become. FIG. 11(D) is a diagram showing the optimal position of each fly-eye lens group in this case, and the positional and rotational relationship with FIG. 11(C) is shown in FIG.
The relationship is the same as A) and (B). When illumination light is incident on a two-dimensional pattern as shown in FIG.
Diffracted light is generated in two-dimensional directions according to the periodicity in the dimensional directions (X: Px, Y: Py). Even in a two-dimensional pattern as shown in FIG. 11(C), either one of the 0th-order diffracted light and the ±1st-order diffracted light in the diffracted light is approximately equidistant from the optical axis AX on the projection optical system pupil plane 19. By doing so, the depth of focus can be maximized. In the pattern of FIG. 11(C), the pitch in the X direction is Px, so the pattern in FIG. 11(D)
As shown in , if the center of each fly-eye lens group is on the line segments Lα and Lβ where α=β=f・(1/2)・(λ/Px), then the depth of focus for the X direction component of the pattern can be calculated. can be maximum. Similarly, γ=ε=f・(1/2)
- If the center of each fly-eye lens group is located on the line segments Lγ and Lε that are (λ/Py), the depth of focus can be maximized for the Y-direction component of the pattern.

As described above, when the illumination light flux from the fly's eye lens group arranged at each position shown in FIG. Either the component DR or the -1st-order diffracted light component Dm passes through an optical path that is approximately equidistant from the optical axis AX in the pupil plane 19 in the projection optical system 18. Therefore, as described in the operation section, a projection exposure apparatus with high resolution and a large depth of focus can be realized.

As described above, the reticle pattern 17 shown in FIG.
Although we considered only the two examples shown in (A) or (C), we focused on the periodicity (fineness) of other patterns as well, and focused on the +1st-order diffracted light component or -1st-order diffracted light component from the pattern. The two luminous fluxes of one of them and the 0th order diffracted light component are
In the pupil plane 19 in the projection optical system, the center of each fly's eye lens group may be placed at a position where the optical path passes approximately equidistant from the optical axis AX. Furthermore, in the pattern examples shown in FIGS. 11A and 11C, the ratio of the line portion to the space portion (duty ratio) is 1:1, so the ±1st-order diffracted light is strong in the generated diffracted light. For this reason, we focused on the positional relationship between one of the ±1st-order diffracted lights and the 0th-order diffracted light,
In cases where the pattern differs from the duty ratio of 1:1, the positional relationship between the other diffracted lights, for example, one of the ±2nd-order diffracted lights and the 0th-order diffracted light, is set to the optical axis A in the projection optical system pupil plane 19.
It may be arranged to be approximately equidistant from X.

Furthermore, the reticle pattern 17 is shown in FIG.
), when focusing on a specific 0th-order diffracted light component, on the pupil plane 19 of the projection optical system, the X-direction (the 0th-order diffracted light component) There may be a first-order or higher order diffraction light component distributed in the Y direction (second direction) and a first-order or higher order diffraction light component distributed in the Y direction (second direction). Therefore, if a two-dimensional pattern is to be imaged well for one specific 0th-order diffracted light component, one of the higher-order diffracted light components distributed in the first direction is
One of the high-order diffracted light components distributed in the second direction and a specific 0th-order diffracted light component are transmitted along the optical axis AX on the pupil plane 19.
The position of a specific 0th-order diffracted light component (one fly's eye lens group) may be adjusted so that it is distributed approximately equidistant from . For example, in FIG. 11(D), it is preferable that the center position of each fly-eye lens group coincides with one of points Pζ, Pη, Pκ, and Pμ. The points Pζ, Pη, Pκ, and Pμ are all located on the line segment Lα or Lβ (the optimal position for the periodicity in the X direction, that is, one of the 0th-order diffracted light and the ±1st-order diffracted light in the Since it is the intersection of the line segments Lγ and Lε (the optimal position for periodicity in the Y direction), it is the optimal light source for both the X and Y pattern directions. It's the location.

[0073] In the above, it is assumed that the two-dimensional pattern is a pattern having two-dimensional directionality at the same location on the reticle, but if there are multiple patterns having different directionality at different positions on the same reticle pattern. The above method can also be applied. If the pattern on the reticle has multiple directions or degrees of fineness, the optimal position of the fly-eye lens group will correspond to each directionality and degree of fineness as described above, or The fly-eye lens group may be arranged at an average position of the optimum position. Further, this average position may be a weighted average with weights added depending on the degree of fineness and importance of the pattern.

An example of determining the position of a plurality of fly-eye lens groups has been shown above, but the illumination light flux is determined by the above-mentioned optical members (diffraction grating pattern, movable mirror, prism, fiber, etc.) as the holding member is replaced. Although the light is concentrated in accordance with the moving position of each fly-eye lens group, it is not necessary to provide an optical member for such concentration. Furthermore, the light beams emitted from each fly-eye lens group are incident on the reticle at an angle. At this time, if the direction of the light intensity center of gravity of these inclined incident light beams (plurality) is not perpendicular to the reticle, a problem will occur in which the position of the transferred image will shift in the in-plane direction of the wafer when the wafer 20 is slightly defocused. do. In order to prevent this, the direction of the center of gravity of the illumination light beams from each fly-eye lens group is perpendicular to the reticle pattern, that is, parallel to the optical axis AX. In other words, assuming that each fly-eye lens group has an optical axis (center line), the position vector of the optical axis (center line) in the Fourier transform plane with reference to the optical axis AX of the projection optical system 18, and It is sufficient that the vector sum of the products with the amount of light emitted from the fly-eye lens group becomes zero.

Furthermore, as a simpler method, the number of fly-eye lens groups is 2m (m is a natural number), and m
The position of each is determined by the optimization method described above (FIG. 12),
The remaining m pieces may be arranged at positions symmetrical to the above m pieces about the optical axis AX. Furthermore, if the device has, for example, n (n is a natural number) fly-eye lens groups, and the number of required fly-eye lens groups is m, which is less than n, then the remaining (n-m) It is not necessary to use several fly-eye lens groups. In order to avoid using the (n-m) fly-eye lens groups, it is sufficient to provide the light shielding member 10 or 12 at the position of the (n-m) fly-eye lens groups. Further, at this time, it is preferable that the optical member that concentrates illumination light on the position of each fly-eye lens group does not concentrate the illumination light on these (n−m) fly-eye lenses.

It is desirable that the position of the opening of the light shielding member 10 or 12 is variable in accordance with the movement of each fly's eye lens group as the holding member is replaced. Alternatively, a mechanism for replacing the light shielding members 10 and 12 depending on the position of each fly's eye lens may be provided, and several types of light shielding members may be included in the apparatus. In each of the above embodiments, it is assumed that the plurality of holding members (fly's eye lens group) are configured to be replaceable, but in the present invention, there is no particular need for the holding members to be configured to be replaceable. Needless to say, for example, simply placing only the holding member 11 shown in FIG. ) can be obtained. Note that when it is acceptable to have some loss in the amount of illumination light from the light source, there is no particular need to arrange an optical member (input optical system) for concentrating the illumination light flux on the fly's eye lens group.

In the above embodiments, the mercury lamp 1 was used as the light source, but other bright line lamps, lasers (such as excimer lasers), or continuous spectrum light sources may be used. Moreover, although most of the optical members in the illumination optical system are lenses, they may also be mirrors (concave mirror, convex mirror). The projection optical system may be a refractive system, a reflective system, or a catadioptric system. Further, in the above embodiments, a projection optical system that is telecentric on both sides is used, but a system that is telecentric on one side or a non-telecentric system may be used. Furthermore, monochromating means such as an interference filter may be provided in the illumination optical system in order to utilize only light of a specific wavelength among the illumination light generated from the light source.

Furthermore, fly-eye lens groups 11A and 11B
The illumination light may be made uniform by using a light scattering member such as a diffuser plate or an optical fiber bundle near the light source side focal plane 11a. Alternatively, in addition to the fly-eye lens group used in the examples of the present invention, an additional fly-eye lens (
Thereafter, the illumination light may be made uniform using an optical integrator such as a separate fly's eye lens. At this time, the separate fly-eye lens is the fly-eye lens group 1.
An optical member that can vary the illumination light amount distribution near the light source side focal plane 11a of 1A and 11B, for example, the diffraction grating pattern 5 shown in FIGS. 1 and 2, or a light source (lamp) than 5A.
It is desirable to be on the 1st side. Further, the cross-sectional shape of the lens element of the separate fly's eye lens is preferably polygonal, particularly regular hexagonal, rather than square (rectangular).

FIG. 14 shows the configuration around the wafer stage of a projection exposure apparatus applied to each embodiment of the present invention, in which a beam 100A is directed obliquely into the projection field of view on the wafer 20 of the projection optical system 18. and its reflected beam 100
An oblique incidence type autofocus sensor that receives B light is provided. This focus sensor detects the deviation in the direction of the optical axis AX between the surface of the wafer 20 and the best imaging plane of the projection optical system 18, and the focus sensor
The motor 112 of the Z stage 110 on which 0 is placed is servo-controlled. As a result, the Z stage 110 moves slightly in the vertical direction (optical axis direction) relative to the XY stage 114, and exposure is always performed in the best focus state. In an exposure apparatus capable of such focus control, the apparent depth of focus can be further expanded by moving the Z stage 110 in the optical axis direction with controlled speed characteristics during the exposure operation. This method uses the projection optical system 18
Any type of projection exposure apparatus (stepper) can be used as long as the image side (wafer side) of the image side (wafer side) is telecentric.

FIG. 15(A) shows the distribution or existence probability of the amount of light in the optical axis direction obtained in the resist layer as the Z stage 110 moves during exposure.
B) represents the speed characteristic of the Z stage 110 for obtaining the distribution as shown in FIG. 15(A). In both FIGS. 15(A) and 15(B), the vertical axis represents the wafer position in the Z (optical axis) direction.
) represents the existence probability, and the horizontal axis of FIG. 15(B) represents the speed V of the Z stage 110. Also in the same figure, position Z0
is the best focus position.

Here, the existence probabilities are set to approximately equal maximum values at two positions +Z1 and -Z1 which are vertically separated from the position Z0 by the theoretical depth of focus ±ΔD0 f of the projection optical system 18, and the positions between them are set to +Z3 to -Z3. In the range of , the probability of existence is kept to a small value. For that purpose, Z
The stage 110 moves up and down at a constant speed of low speed V1 at a position -Z2 when the shutter inside the illumination system starts opening, and immediately after the shutter is fully opened, it moves up and down at a high speed of V2.
accelerate to. While the Z stage 110 is moving up and down at a constant speed of V2, the existence probability is pushed to a low value, and when it reaches the position +Z3, the Z stage 110 starts decelerating toward the low speed V1, and the position +Z1 The probability of existence reaches its maximum value. At this time, a shutter closing command is output almost simultaneously, and the shutter is completely closed at position +Z2.

In this way, the light intensity distribution (existence probability) in the optical axis direction of the exposure amount given to the resist layer of the wafer 20 is determined by 2
If the speed of the Z stage 110 is controlled so that the maximum value is reached at a point, uniform resolution can be obtained over a wide range in the optical axis direction, although the contrast of the pattern formed on the resist layer is slightly reduced.

The progressive focus exposure method described above can be used in exactly the same way with a projection exposure apparatus that employs a special illumination method as shown in each embodiment of the present invention, and the apparent depth of focus is The image is enlarged by an amount corresponding to approximately the product of the enlargement obtained by the illumination method of the present invention and the enlargement obtained by the cumulative focus exposure method. Moreover, because it uses a special lighting method, the resolution itself is also high. For example, the minimum line width that can be exposed by combining a phase shift reticle with a conventional 1/5 reduction i-line stepper (projection lens NA 0.42) is about 0.3 to 0.35 μm, and the magnification rate of the depth of focus is The maximum is about 40%. On the other hand, when we incorporated a special illumination method like the present invention into the same i-line stepper and experimented with an ordinary reticle, we found that
The minimum line width was about 0.25 to 0.3 μm, and the depth of focus expansion rate was also about the same as when using a phase shift reticle.

[0084]

As described above, according to the present invention, it is possible to realize a projection type exposure apparatus having a higher resolution and a larger depth of focus than conventional ones, while using a normal mask. Moreover, according to the present invention, it is only necessary to change the illumination system of the projection exposure equipment already in operation at the semiconductor production site, and the projection optical system of the equipment in operation can be used as is to achieve higher resolution than before. becomes possible.

Further, according to the method of concentrating the illumination light on the fly-eye lens group shown in each embodiment of the present invention, the loss of the amount of illumination light from the light source can be minimized, so that it can be used as an exposure apparatus. There is also the effect that throughput does not drop significantly.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the configuration of a projection exposure apparatus according to a first embodiment of the present invention.

[Figure 2] First concentration of illumination light to the fly-eye lens group
It is a figure showing a modification of .

[Figure 3] Second concentration of illumination light on the fly-eye lens group
It is a figure showing a modification of .

[Figure 4] Third stage of concentration of illumination light on the fly-eye lens group
It is a figure showing a modification of .

[Figure 5] Fourth stage of concentration of illumination light on the fly-eye lens group
It is a figure showing a modification of .

[Figure 6] Fifth step of concentrating illumination light on the fly-eye lens group
It is a figure showing a modification of .

7 is a diagram showing an illumination system when a reticle blind is incorporated into the apparatus of FIG. 1. FIG.

FIG. 8 is a diagram showing a specific configuration of a movable member (switching member of the present invention) that exchanges four holding members comprising a plurality of fly-eye lens groups.

FIG. 9 is a diagram showing a modification of the movable member for exchanging a plurality of holding members.

FIG. 10 is a diagram schematically showing the optical path from the fly-eye lens group to the projection optical system.

FIGS. 11A and 11C are plan views showing an example of a reticle pattern formed on a mask. (B), (D
) is a diagram illustrating the arrangement of each fly-eye lens group on the pupil conjugate plane corresponding to each of (A) and (C).

FIG. 12 is a diagram illustrating the principle of the present invention.

FIG. 13 is a diagram showing the configuration of a conventional projection exposure apparatus.

FIG. 14 is a diagram showing a configuration around a wafer stage of a projection exposure apparatus.

FIG. 15 shows the existence probability of the exposure amount when executing the progressive focus exposure method using the Z stage of the wafer stages;
It is a graph which shows the speed characteristic of Z stage.

[Explanation of symbols]

5 Diffraction grating pattern 9 Relay lens system 11A, 11B Fly eye lens system 10, 12
Light blocking member (spatial filter) 15 Condenser lens 16 Reticle 17 Reticle pattern 18 Projection optical system 19 Pupil 20 Wafers 24, 36 Movable member

Claims (5)

[Claims] 1. An illumination optical system that shapes illumination light from a light source into a substantially uniform intensity distribution and irradiates the uniform illumination light onto a mask having a periodic pattern portion; A projection exposure apparatus comprising: a projection optical system that forms and projects an image onto a photosensitive substrate; and a stage that holds the photosensitive substrate so that the surface of the photosensitive substrate is placed near the image forming plane of the projection optical system; a plurality of fly-eye lens groups that form mutually separated secondary light source images in the optical path of the illumination optical system in the vicinity of a plane corresponding to the Fourier transform of the pattern on the mask or a conjugate plane thereof; The centers of each of the eye lens groups are set at discrete positions offset from the optical axis of the illumination optical system or the projection optical system by an amount determined according to the periodicity of the pattern on the mask. A projection exposure apparatus comprising: a holding member that holds the plurality of fly-eye lens groups together; 2. An illumination optical system that shapes illumination light from a light source into a substantially uniform intensity distribution and irradiates the uniform illumination light onto a mask having a periodic pattern portion; A projection exposure apparatus comprising: a projection optical system that forms and projects an image onto a photosensitive substrate; and a stage that holds the photosensitive substrate so that the surface of the photosensitive substrate is placed near the image forming plane of the projection optical system; a plurality of fly-eye lens groups that form mutually separated secondary light source images in the optical path of the illumination optical system in the vicinity of a plane corresponding to the Fourier transform of the pattern on the mask or a conjugate plane thereof; The centers of each of the eye lens groups are set at discrete positions offset from the optical axis of the illumination optical system or the projection optical system by an amount determined according to the periodicity of the pattern on the mask. a plurality of holding members that integrally hold the plurality of fly-eye lens groups; and a switching member that replaceably arranges each of the plurality of holding members in the optical path of the illumination optical system; A projection type exposure apparatus, wherein each of the holding members holds the plurality of fly-eye lens groups in different eccentric states according to a difference in periodicity of a pattern on the mask. 3. The holding member includes an input optical system that allows illumination light from the light source to enter each of the plurality of fly-eye lens groups set in the optical path of the illumination optical system. 3. The device according to claim 1, wherein the device is held integrally with a lens group. 4. The plurality of fly-eye lens groups have a length of 2 m.
(however, m≧1), and each center of the m fly-eye lens groups among the 2m fly-eye lens groups is connected to the 0th-order diffracted light component generated from the pattern of the mask, and the 0-order diffracted light component generated from the pattern of the mask. At least one of the ±1st-order diffracted light components spreading at an angle corresponding to the fineness of the pattern with respect to the second-order diffracted light component is distributed at approximately equal distance from the optical axis on the pupil plane of the projection optical system. , are eccentrically arranged within the Fourier transform equivalent plane or its conjugate plane, and each center of the remaining m fly-eye lens groups is
4. The apparatus according to claim 1, wherein the apparatus is arranged substantially symmetrically with respect to the optical axis and the center of each of the fly's eye lens groups. 5. When focusing on diffracted light generated from the mask by illumination light from any one of the plurality of fly-eye lens groups,
a 0th order diffracted light component distributed on the pupil plane of the projection optical system;
one of the first-order or higher-order diffracted light components distributed in a first direction on the pupil plane with the zero-order diffracted light component as the center depending on the two-dimensional periodic structure of the pattern of the mask; and the pupil plane. The three diffracted light components including one of the first-order or higher-order diffracted light components distributed in a second direction intersecting the first direction centering on the 0th-order diffracted light component are arranged on the optical axis on the pupil plane. 4. The apparatus according to claim 1, wherein the center of said arbitrary one fly's eye lens group is eccentrically arranged from said optical axis so that said fly's eye lens group is distributed approximately equidistant from said optical axis.