JPH0831734A - Semiconductor exposure device and optimizing method of multiple imagery exposure method - Google Patents

Abstract

PURPOSE:To obtain the optimum value of an FLEX step automatically by optically observing a pattern latent image formed to a photochromic material acquired by irradiating a reticle with light from a light source for a projection optical system. CONSTITUTION:Light sources 10, 20 for a projection optical system, the projection optical system 62 for forming an optical image obtained by irradiating a reticle 61 with light from the light sources 10, 20 for the projection optical system onto a substrate, and a substrate stage 66, on which the substrate is placed, are provided. A photochromic material 70 arranged on the substrate stage 66 and a latent-image detection system for optically observing a pattern latent image acquired by irradiating the reticle 61 with light from the light sources for the projection optical system and formed on the photochromic material 70 are provided. The pattern latent image is observed optically, thus automatically obtaining the optimum value of an FLEX step DELTAZ.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a so-called multiple imaging exposure method (FLEX method, Focus Lath) for giving a large depth of focus.
A mechanism for optimizing the
More specifically, a semiconductor exposure apparatus having a mechanism for automatically optimizing a FLEX step in a multiple imaging exposure method,
Also, the present invention relates to an optimization method for a multiple imaging exposure method using such a semiconductor exposure apparatus.

[0002]

2. Description of the Related Art In a semiconductor exposure apparatus, when the wavelength of light used for exposing a resist is λ and the numerical aperture of a lens is NA, resolution and depth of focus DOF are expressed by the following equations. Resolution = k ₁ λ / NA Equation (1) DOF = k ₂ λ / NA ² Equation (2) where k ₁ and k ₂ are constants depending on the semiconductor exposure apparatus.

In the conventional semiconductor exposure apparatus, it is necessary to use an exposure light source having a short wavelength in order to increase the integration degree of the semiconductor integrated circuit, as is apparent from the equation (1). As the exposure light source, for example, an ultrahigh pressure mercury arc lamp or an excimer laser can be used.

In the conventional exposure light source, since the wavelength of the emitted exposure light is constant and has a single focal point, as the wavelength λ becomes shorter, the depth of focus obtained by the equation (2) becomes insufficient. There is. Therefore, it is becoming difficult to obtain sufficient resolution when a pattern is printed on a resist on a region having a step (for example, a contact hole portion).

One of the means for solving the problem of the depth of focus is a multiplex imaging exposure method. The outline of this multiple imaging exposure method will be briefly described below. Usually, the pattern formed on the reticle is transferred to a resist formed on a wafer mounted on a wafer stage by using a reduction projection optical system (reduction projection lens or the like). At this time, in the multiple imaging exposure method, exposure is performed at least twice by changing the distance between the resist formed on the wafer and the reduction projection optical system.
The distance (z) between the resist formed on the wafer and the reduction projection optical system can be changed by moving the wafer stage in the Z-axis direction.

14 and 15 schematically show the light intensity distribution in the thickness direction of the resist obtained by such a multiple imaging exposure method. When the distance between the resist formed on the wafer and the reduction projection optical system is z, the light intensity distribution as shown in (a) on the left side of FIG. 14 is obtained by exposure. In addition, by changing the distance between the resist formed on the wafer and the reduction projection optical system to z−ΔZ (or z + ΔZ) and performing exposure, the light intensity distribution as shown in the center (b) of FIG. 14 is obtained. Is obtained. In the multiple imaging exposure method, the distance between the resist formed on the wafer and the reduction projection optical system is changed for each exposure as described above, and the amount of change in such distance is referred to as the FLEX step. This is referred to below as ΔZ.

As a result of such double exposure, a light intensity distribution in the thickness direction of the resist can be obtained as shown in (c) on the right side of FIG. The light intensity distribution obtained by connecting the peaks of the light intensity distribution shown in FIG. 14C is shown in FIG. By performing such exposure at least twice,
Although the peak contrast of the light intensity is lower than that of the normal single exposure, the depth of focus DOF can be increased. The number of exposures is not limited to 2 times,
It can be modified appropriately to obtain the optimum depth of focus and peak contrast. In the case of k times of exposure,
The distance between the resist formed on the wafer and the reduction projection optical system is, for example, z, z + ΔZ, z + 2ΔZ, ...
・, Change to z + kΔZ.

[0008] Such a multiple imaging exposure method is an effective method for increasing the value of the depth of focus. Then, it is important to optimize the value of the FLEX step ΔZ. If the value of the FLEX step ΔZ is too large, the contrast in the vicinity of the best focus is lowered as shown in FIG. On the other hand, if the value of the FLEX step ΔZ is too small, the depth of focus DOF becomes narrow as shown in FIG.

[0009]

The problem in adopting the multiple imaging exposure method is that the FLEX step Δ is dependent on the pattern size and pattern density to be formed on the resist.
There is a point where the optimum value of Z changes. Pattern size 0.4
FIG. 16 shows the relationship between the deviation of the distance from the best focus and the pattern size of the exposed resist when the FLEX step ΔZ is variously changed and the pattern density is 1: 2. 17 and 18 show the same relationship when the pattern size is 0.32 μm, the pattern density is 1: 2, and the pattern size is 0.32 μm and the pattern density is 1: 1.7. Further, FIG.
The optimum values of the FLEX step ΔZ under each condition obtained from FIG. 18 are shown in the table below. The pattern size means the size (width) of the pattern formed on the resist formed on the wafer, and the pattern density means the size (width) of the pattern formed on the resist and the pattern It means the ratio of the size (width) of the space between the patterns.

Pattern size Pattern density Optimum value of FLEX step ΔZ 0.4 μm 1: 2 2.4 μm 0.32 μm 1: 2 1.8 μm 0.32 μm 1: 1.7 2.4 μm

As is apparent from the above table, when the pattern size and the pattern density change, the optimum value of the FLEX step ΔZ also changes. Therefore, in the conventional technique, the value of the FLEX step ΔZ is changed every time the pattern size and the pattern density are changed, and the pattern is actually printed on the resist formed on the wafer, and the measurement of the baked pattern is performed. For example, length measurement SEM
It has been necessary to perform a complicated work such as obtaining the optimum value of the FLEX step ΔZ by using.

Therefore, an object of the present invention is to provide a kind of self-timer which can automatically obtain an optimum value of FLEX step ΔZ, which is suitable for optimizing a so-called multiple imaging exposure method for giving a large depth of focus. It is an object to provide a semiconductor exposure apparatus having a diagnostic mechanism and an optimization method of a multiple imaging exposure method using such a semiconductor exposure apparatus.

[0013]

In order to achieve the above object, a semiconductor exposure apparatus of the present invention comprises: (a) a projection optical system light source and (b) a projection optical system light source for irradiating a reticle. A projection optical system for forming an optical image obtained on the substrate, (c) a substrate stage on which the substrate is mounted, (d) a photochromic material arranged on the substrate stage, and (e) projection optics. A latent image detection system for optically observing a pattern latent image formed on a photochromic material obtained by irradiating a reticle with light from a system light source.

In the semiconductor exposure apparatus of the present invention, the light source for the projection optical system emits the laser light source and the light having the wavelength based on the second harmonic of the light which is emitted from the laser light source. It is preferable that the laser light source is composed of a second harmonic wave generating device that emits light, and that the laser light source also serves as a light source for a latent image detection system and a color erasing light source for erasing the pattern latent image. In this case, the laser light source includes a laser diode, a solid-state laser medium made of Nd: YAG, and an LD-pumped solid-state laser made up of a nonlinear optical crystal element. And a light source for erasing. Alternatively, the laser light source is a laser diode, Nd: YA
A solid-state laser medium made of G and an LD-pumped solid-state laser made of a nonlinear optical crystal element are used, and light emitted from a laser diode can be used as a latent image detection system light source and a color erasing light source.

Alternatively, in the semiconductor exposure apparatus of the present invention, the light source for the projection optical system is a laser light source, and light emitted from the laser light source is incident on the light source and a wavelength based on the second harmonic of the light is emitted. The second harmonic generation device for emitting light, and further comprising a bandpass filter for splitting light having a long wavelength component from the light emitted from the second harmonic generation device. It is preferably used as a light source for a latent image detection system and a light source for erasing.

In the semiconductor exposure apparatus of the present invention, the latent image detection system and the off-axis alignment detection system can be combined. In this case, it is desirable to irradiate the photochromic material with light from the light source for the projection optical system through the off-axis alignment detection system.

In order to achieve the above object, the method of optimizing the multiplex imaging exposure method of the present invention is: (a) The initial distance from the projection optical system to the photochromic material placed on the substrate stage is Z _0. , The distance Z from the projection optical system to the photochromic material arranged on the substrate stage is Z ₀ + Δ
A pattern optical image obtained by irradiating a reticle on which a pattern is formed with light from a light source for a projection optical system by sequentially changing to Z × i (where i = 0, 1, 2, ..., K). A step of forming a latent image of a pattern on the photochromic material by a multiple imaging exposure method of forming (K + 1) times on the photochromic material via a projection optical system, with the value of ΔZ being constant and the value of the initial distance value Z ₀ being Is changed and repeated M times, Z
Forming M pattern latent images having different values of (b), (b) repeating the step (a) N times by changing the value of ΔZ, and finally forming M × (N + 1) pattern latent images. (C) The obtained pattern latent image is optically observed, and (d) the optimum ΔZ value is obtained based on the result of the pattern latent image observation.

In the method for optimizing the multiple imaging exposure method of the present invention, the step of erasing the pattern latent image by irradiating the pattern latent image formed on the photochromic material with light from the erasing light source is further included. Can be included. The optical observation of the latent pattern image consists of measuring the contrast of the latent pattern image and can be performed using an off-axis alignment detection system.

[0019]

In the present invention, the optimum value of the FLEX step ΔZ is determined by optically observing the pattern latent image formed on the photochromic material obtained by irradiating the reticle with the light from the light source for the projection optical system. It can be automatically requested. Therefore, even when the pattern size and the pattern density to be formed on the resist are changed due to the exchange of the reticle and the like, the multiple imaging exposure method can be performed easily and in a short time without spending time, labor and cost unlike the conventional technique. It is possible to optimize the FLEX step ΔZ in. Further, by decoloring the photochromic material on which the pattern latent image is formed, the optimization method of the multiple imaging exposure method can be repeatedly executed. In a preferred embodiment of the semiconductor exposure apparatus of the present invention, the light source for the projection optical system also serves as the light source for the latent image detection system and the light source for erasing, so that the structure of the semiconductor exposure apparatus can be simplified.

[0020]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be described based on embodiments with reference to the drawings.

(Embodiment 1) A semiconductor exposure apparatus of Embodiment 1 is as shown in the conceptual diagrams of FIGS. 1, 2, 3 and 9.
Projection optical system light sources 10 and 20, and projection optical system light source 1
A projection optical system 62 for forming an optical image obtained by irradiating the reticle 61 with light from 0 and 20 on a substrate 64, and a substrate stage 66 on which the substrate 64 is mounted. Further, the photochromic material 70 disposed on the substrate stage 66 and the pattern latent image formed on the photochromic material 70 obtained by irradiating the reticle 61 with light from the light source for the projection optical system are optically observed. It is equipped with a latent image detection system for doing so. still,
In the first embodiment, the photochromic material 70 is further added.
An erasing light source is provided for erasing the pattern latent image formed in 1., and the projection optical system light source also serves as the latent image detecting system light source and the erasing light source. Further, the latent image detection system for optically observing the pattern latent image (specifically, for measuring the contrast of the pattern latent image) also serves as the off-axis alignment detection system.

The pattern latent image formed on the photochromic material 70 is composed of, for example, a part of a circuit pattern. Then, as shown in FIG. 1, the light sources 10, 20 for the projection optical system are used.
A part of the reticle 61 on which the circuit pattern is formed is irradiated with the light from the circuit pattern optical image to form an image on the photochromic material 70 by the projection optical system 62. Thereby, a pattern latent image can be formed on the photochromic material 70. A test pattern may be formed on the reticle 61. The formation of the pattern latent image will be described later.

The photochromic material 70 is made of, for example, a spiropyran-based material, and can be formed on the surface of the substrate stage 66 near the substrate chuck 67 by vapor deposition or sputtering. It is desirable to cover the surface of the photochromic material 70 with PVA or a glass plate to protect the photochromic material 70.

The light source for the projection optical system is, for example, as shown in the schematic view in FIG. 11, based on the laser light source 10, the light emitted from the laser light source 10, and the second harmonic of the incident light. Second, which emits light with different wavelengths
It is composed of a harmonic generator 20. The laser light source 10 is an LD including a laser diode 11, a solid-state laser medium 12 made of Nd: YAG, and a nonlinear optical crystal element 13.
It consists of a pumped solid-state laser. In addition, the second harmonic generation device 2
Reference numeral 0 is composed of a nonlinear optical crystal element 21, an optical resonator 22, and a resonator length control device 30. These laser light sources 10,
Details of the second harmonic generation device 20 and the resonator length control device 30 will be described later.

In the first embodiment, the laser light source 10 constituting the light source for the projection optical system also serves as the light source for the latent image detection system and the light source for erasing, and the laser light source 10 (more specifically, nonlinear optical The light emitted from the crystal element 13) is used for the optical observation of the pattern latent image and the decolorization of the photochromic material 70. The laser light source 10 also serves as an off-axis alignment detection system light source.

Off-axis, which also serves as a latent image detection system
As shown in FIG. 10, the alignment detection system includes an alignment microscope 40, a first photodetector 41 for detecting 0th-order diffracted light, and a second photodetector 4 for detecting ± 1st-order diffracted light.
2. Processing a signal output from the half mirror 43 and the photodetectors 41 and 42 for causing the 0th-order diffracted light reflected by the alignment mark 65 formed on the substrate 64 to enter the first photodetector 41. Signal processing device 44. The alignment microscope 40 is an ordinary optical microscope having a small numerical aperture NA. The illumination light for the off-axis alignment detection system can be obtained by splitting the light emitted from the laser light source 10 (specifically, the nonlinear optical crystal element 13) by the optical path splitting means 50.

The off-axis alignment detection system is provided with, for example, a multi-faceted rotating mirror (polygon mirror) (not shown) so that the split light from the optical path splitting means 50 can be scanned. There is. That is, the illumination light for the off-axis alignment detection system emitted from the laser light source 10 and split by the optical path splitting means 50 passes through the multi-face rotation mirror, the half mirror 43 and the alignment microscope 40, and the resist 63. The substrate 64 on which the alignment mark 65 is formed is irradiated and scanned at a constant speed. The light diffracted, scattered, and refracted and reflected by the alignment mark 65 is reflected by the first and second photodetectors 41,
42.

A pattern latent image is formed on the photochromic material 70 in a pattern latent image forming step (details will be described later) in the optimization method of the multiple imaging exposure method. In addition, in the semiconductor exposure apparatus of the first embodiment, as shown in FIGS. 2 and 3, the photochromic material 70 is formed by the light emitted from the laser light source 10 (more specifically, the nonlinear optical crystal element 13). The formed pattern latent image is optically observed or the pattern latent image is erased. Specifically, as shown in FIGS. 2 and 3, the light is emitted from the laser light source 10 and transmitted onto the beam splitter or the half mirror 50, the half mirror 43, and the alignment microscope 40 to be arranged on the substrate stage 66. The photochromic material 70 is irradiated, whereby the pattern latent image formed on the photochromic material 70 is optically observed or the pattern latent image is erased.

Hereinafter, with reference to FIGS. 1, 2 and 4, an optimization method of the multiple imaging exposure method of the present invention, more specifically, an optimization method of the FLEX step ΔZ in the multiple imaging exposure method will be described. explain. A flow chart of the optimization method of this FLEX step ΔZ is shown in FIG.

[Pattern Latent Image Forming Step-1] First, referring to FIG.
As shown in, the initial value of the distance from the projection optical system 62 to the photochromic material 70 arranged on the substrate stage 66 is Z ₀ . Then, the distance Z from the projection optical system 62 to the photochromic material 70 arranged on the substrate stage 66 is Z ₀ + ΔZ × i (where i = 0, 1, 2, ...
.., K), and the reticle 61 on which the pattern is formed is irradiated with light from the light sources 10 and 20 for the projection optical system. A latent image of the pattern is formed on the photochromic material 70 by a multiple imaging exposure method in which the pattern optical image thus obtained is imaged (K + 1) times on the photochromic material via the projection optical system 62. In such a pattern latent image forming process, the distance initial value Z
By changing the value of ₀ and repeating M times, M pattern latent images having different values of Z are formed. Here, ΔZ is the FLEX step. To change the distance Z, move the substrate stage 66 to Z
This can be done by moving in the axial direction.

In the first embodiment, K = 1 and M = 21. That is, the distance Z from the projection optical system 62 to the photochromic material 70 arranged on the substrate stage 66 is
By sequentially changing to Z ₀ and Z ₀ + ΔZ, a pattern optical image is formed on the photochromic material twice (K + 1) = 2, and a pattern latent image is formed on the photochromic material 70. Then, the value of ΔZ is kept constant (for example, ΔZ = 2.0 μm), the value of the initial distance value Z ₀ is changed in steps of 0.2 μm, and the above-described pattern latent image forming step is repeated M = 21 times, Z M = 21 pattern latent images having different values of were formed. At this time, the substrate stage 66 was appropriately moved in the (X, Y) directions to change the pattern latent image forming position on the photochromic material 70.

When the pattern latent image is formed by irradiating the photochromic material 70 with light from the light source for the projection optical system, the light of the photochromic material 70 in the portion where the pattern latent image is formed is schematically shown in FIG. The absorption characteristic is a characteristic having an absorption peak in the short wavelength region (shown by (a) in FIG. 4).
Changes to a characteristic having an absorption peak in the long wavelength region (shown by (b) in FIG. 4).

[Pattern latent image forming step-2] Next, [Pattern latent image forming step-1] is repeated N times (N = 2 in the first embodiment) by changing ΔZ in steps of 0.2 μm,
Finally, M × (N + 1) = 63 pattern latent images were formed. When the [pattern latent image forming step-1] is repeated, the distance initial value Z ₀ and the amount of change may not always be the same, or the values of M and K may be changed.

[Contrast measurement process of pattern latent image]
Next, the M × (N + 1) pattern latent images thus obtained are optically observed. Specifically, the contrast of the pattern latent image is measured using an off-axis alignment detection system that also serves as a latent image detection system (see FIG. 2). For this purpose, first, the substrate stage 66 is moved so that the pattern latent image formed on the photochromic material 70 is located immediately below the alignment microscope 40. Then, the contrast of the pattern latent image formed on the photochromic material 70 is measured using an off-axis alignment detection system. That is, the light is emitted from the laser light source 10 and is split by the optical path splitting means 50.
The pattern latent image of the photochromic material 70 is irradiated and scanned with the light that has passed through the half mirror 43 and the alignment microscope 40 (this light is equivalent to the illumination light of the off-axis alignment detection system). The light reflected by the region of the photochromic material 70 on which the pattern latent image is formed is the alignment microscope 40 and the half mirror 43.
Via the first photodetector 41 consisting of a CCD camera
Incident on.

The signal output from the first photodetector 41 is signal-processed by the signal processing device 44, and the light absorption curve of the photochromic material 70 shown in FIG. 4B, that is, the contrast can be obtained.

For example, by moving the substrate stage 66 in the Z direction to change the distance from the alignment microscope 40 to the photochromic material 70, the sharpest contrast (absorption curve having the highest peak and And / or the contrast value (peak value) when the latent image of the photochromic material 70 having the absorption curve with the narrowest half-width) is obtained. Such an operation is executed for each pattern latent image. Then, for each value of ΔZ, the contrast value (peak value) is connected using the distance Z from the projection optical system 62 to the photochromic material 70 arranged on the substrate stage 66 as a parameter to obtain a contrast curve. Such a contrast curve is schematically shown in FIG. In Example 1, since N = 2, a total of three contrast curves are obtained. Note that such a contrast curve can be easily created using a small computer.

[Calculation Step of Optimal FLEX Step ΔZ]
After that, based on the optical observation result of the pattern latent image (specifically, the contrast measurement result of the pattern latent image),
The optimum value of the FLEX step ΔZ in the multiple imaging exposure method is obtained.

For example, in the contrast curve shown in the curve (c) of FIG. 5, since the value of the FLEX step ΔZ is too large, a concave portion is formed and two peaks exist. Therefore,
When the contrast curve like this is obtained because the contrast near the best focus is lowered.
The value of LEX step ΔZ is not the optimum value. The contrast curve shown by the curve (a) in FIG. 5 has a narrow flat portion because the value of the FLEX step ΔZ is too small. That is, the value of the depth of focus DOF is small. Therefore, the value of the FLEX step ΔZ when obtaining such a contrast curve is also not the optimum value. On the other hand, in the contrast curve shown in the curve (b) of FIG. 5, the value of the FLEX step ΔZ is optimum, the flat portion is wide, and there is no recess. That is, the depth of focus DOF is large, and the contrast curve is flat over a wide depth of focus region. Therefore, the value of the FLEX step ΔZ when obtaining such a contrast curve is the optimum value. In this way, FLEX
The optimum value of step ΔZ is determined.

[Pattern latent image erasing process] With the above operation,
The multiple imaging exposure method is optimized. That is, specifically, the optimum value of the FLEX step ΔZ is obtained. afterwards,
It is desirable to erase the pattern latent image by irradiating the pattern latent image formed on the photochromic material 70 with light from the erasing light source. In the semiconductor exposure apparatus of Example 1, as shown in FIG. 3, the light emitted from the laser light source 10 (more specifically, the nonlinear optical crystal element 13) that also serves as the light source for erasing is used to perform photochromic. The pattern latent image formed on the material 70 is erased.

That is, the light emitted from the laser light source 10 and passing through the beam splitter or the half mirror 50, the half mirror 43, and the alignment microscope 40 is used to dispose the photochromic material 70 on the substrate stage 66.
Is irradiated. [Pattern latent image contrast measurement process]
When measuring the contrast of the pattern latent image formed on the photochromic material 70, the irradiation time of the light from the laser light source 10 to the photochromic material 70 is shortened. This can prevent the pattern latent image formed on the photochromic material 70 from being erased. When the pattern latent image formed on the photochromic material 70 is erased, the photochromic material 70 may be exposed to the light from the laser light source 10 for a long time.

For reference, the resist exposure process and the alignment operation by the off-axis alignment method will be briefly described below.

In the resist exposure process, as shown in FIG. 9, the light emitted from the laser light source 10 is incident on the second harmonic generator 20. Second harmonic generation device 20
Emits light having a wavelength based on the second harmonic of the incident light. The light emitted from the second harmonic generation device 20 irradiates the reticle 61, and the circuit pattern or the like formed on the reticle 61 is formed on the substrate 64 via the projection optical system (for example, reduction projection lens) 62. Resist 63
Transfer to. The pattern formed on the reticle 61 is a pattern to be formed on the resist 63 magnified five times, for example. The projection optical system 62 transmits the incident light and projects, for example, an optical image reduced to ⅕ onto the resist 63 formed on the substrate 64. As a result, a fine circuit pattern or the like is formed on the resist 63. The substrate 64 is placed on the substrate stage 66.

In the resist exposure process, the [optimal FL
The step of calculating EX step ΔZ] is performed to expose the resist by the multiple imaging exposure method using the optimum value of FLEX step ΔZ. That is, with the distance between the resist 63 formed on the substrate 64 and the reduction projection optical system 62 set as z, the circuit pattern or the like formed on the reticle 61 is printed on the resist 63. Next, the distance between the resist 63 formed on the substrate 64 and the reduction projection optical system 62 is set to z-Δ.
Z × i (or z + ΔZ × i) (where i = 1, 2,
The resist exposure is performed by sequentially changing to (k). This is repeated k times.

In the alignment operation by the off-axis alignment method, as shown in FIG. 10, the alignment mark 65 is illuminated by the illumination light for the off-axis alignment detection system emitted from the laser light source 10. Then, the light diffracted, scattered, refracted and reflected by the alignment mark 65 is reflected by the first and second photodetectors 41,
42.

In this case, a bright field image can be obtained by signal-processing the output signal from the first photodetector 41 which has received the 0th-order diffracted light by the signal processing device 44.
This bright field image is observed on a CRT. A bright field image can be obtained by performing signal superposition processing,
An image with good reproducibility can be obtained by the averaging effect of the signal output, but the alignment mark detection resolution is low. The signal processing device 44 performs signal processing on the output signal from the second photodetector 42 that receives the ± 1st-order diffracted light from the alignment mark 65 formed of a grating formed on the substrate 64, and thereby the dark field image is obtained. Can be obtained. Compared with the bright field image, this dark field image has a higher alignment mark detection resolution, but the image reproducibility is lower. This is because the diffraction angle greatly depends on the shape of the alignment mark 65 and the like. FIG. 10 shows the alignment mark 6 formed on the substrate.
First and second photodetectors 41, 4 when 5 is observed
The signal output from 2 is shown schematically. By combining the bright field image and the dark field image, the position of the alignment mark 65 can be detected with high accuracy.

In this way, the laser light source 10 also serves as the light source for the latent image detection system, the light source for the erasing color, and the light source for the off-axis alignment detection system, so that the He-Ne used in the conventional semiconductor exposure apparatus is used. A separate light source for off-axis / alignment detection system, such as a laser, is not required, or a separate light source for latent image detection system or light source for erasing is not required, and the manufacturing cost and maintenance cost of the semiconductor exposure apparatus are reduced. You can The laser light source 1
Since the light emitted from 0 has strong coherence, it is possible to improve the alignment mark detection resolution of the off-axis alignment detection system.

(Embodiment 2) Embodiment 2 is a modification of Embodiment 1. The laser light source 10 is replaced by a laser diode 11, N
The solid-state laser medium 12 made of d: YAG and the LD-pumped solid-state laser made of the nonlinear optical crystal element 13 are the same as in the first embodiment. In Example 1, the light emitted from the nonlinear optical crystal element 13 was used as the latent image detection system light source and the erasing light source. On the other hand, in the second embodiment, as shown in the conceptual diagram of FIG. 7, the light emitted from the laser diode 11 is used as the latent image detection system light source and the erasing light source. This point is different from the first embodiment. That is, the light emitted from the laser diode 11 constituting the laser light source 10 and passing through the half mirror 43 and the alignment microscope 40 via the beam splitter or the half mirror 50 causes the photochromic material 70 arranged on the substrate stage 66 to be removed. Irradiate and perform optical observation or erasing of the latent pattern image. Other structures of the semiconductor exposure apparatus and the optimization method of the multiple imaging exposure method can be the same as those in the first embodiment, and thus detailed description will be omitted. The light emitted from the laser diode 11 is
It can also be used as illumination light for an off-axis alignment detection system.

(Third Embodiment) The light source for the projection optical system in the third embodiment is similar to the first or second embodiment in that the laser light source 10 and the light emitted from the laser light source are incident on the incident light. The second harmonic generation device 20 emits light having a wavelength based on the second harmonic of.
Example 3 is different from Example 1 or Example 2 in that
As shown in the conceptual diagram of FIG. 8, (A) the semiconductor exposure apparatus includes a bandpass filter 80 that disperses light having a long wavelength component from the light emitted from the second harmonic generation device, and (B) The light having such a long wavelength component is used as a light source for a latent image detection system and a light source for erasing.

The light emitted from the second harmonic generation device 20 contains a mixture of light having the wavelength (long wavelength component) of the light emitted from the laser light source 10. In the third embodiment, the bandpass filter 80 disperses the light having the long wavelength component. Then, the light thus obtained is used for observing the pattern latent image formed on the photochromic material 70 (for contrast measurement) or for erasing. That is, the photochromic material 70 disposed on the substrate stage 66 is irradiated with the light emitted from the second harmonic generation device 20 and passing through the bandpass filter 80 and the half mirror 43 and the alignment microscope 40. Other structures of the semiconductor exposure apparatus and the optimization method of the multiple imaging exposure method can be the same as those in the first embodiment, and thus detailed description will be omitted. The light having the long wavelength component obtained by the bandpass filter 80 can be used as the illumination light for the off-axis alignment detection system.

The laser light source 10 and the second harmonic generation device 20 suitable for use as the light source for the projection optical system described in each of the above embodiments will be described below with reference to FIGS. 11, 12 and 13. ,explain.

As shown in FIG. 11, the laser light source 10 is
A laser diode 11, a solid-state laser medium 12 made of Nd: YAG, and a non-linear optical crystal element 13,
It consists of an LD-pumped solid-state laser capable of emitting the second harmonic.
The second harmonic generation device 20 includes a nonlinear optical crystal element 21 and an optical resonator 22. Second harmonic generator 2
0 further includes a resonator length control device 30 for controlling the resonator length of the optical resonator 22.

In the first and second embodiments, the light emitted from the laser light source 10 is split by the optical path splitting means 50, and is off-axis during the execution of the optimization method of the multiple imaging exposure method and during the alignment operation. It is sent to the alignment detection system, and on the other hand, at the time of resist exposure, it is incident on the optical resonator 22 which constitutes the second harmonic generation device 20. The second harmonic generation device 20 has light having a wavelength based on the second harmonic of the light incident on the optical resonator 22 (fourth harmonic when the laser light generated by the solid-state laser medium is used as a reference). Inject. In the third embodiment, the second harmonic generation device 20
The light emitted from the reticle 61 is dispersed by the bandpass filter 80 and is sent to the off-axis alignment detection system when the optimization method of the multiple imaging exposure method is performed and when the alignment operation is performed.
Sent to. Note that FIG. 11 illustrates a light source for a projection optical system suitable for use in the first embodiment.

As shown in FIG. 11, the laser light source 10 is
For example, a plurality of laser diodes 11 (wavelength of emitted light:
808 nm), Nd: YAG solid-state laser medium 1
2 (wavelength of emitted light: 1064 nm), and KTP (KT
It is composed of a nonlinear optical crystal element 13 made of iOPO ₄ ). The solid-state laser medium 12 is an end face excitation type. With this configuration, the laser light source 10 emits N
Light of 532 nm, which is the second harmonic of the solid-state laser medium made of d: YAG, is emitted. The laser light source 10 has N
1/4 in front of the solid-state laser medium 12 made of d: YAG
The wave plate 14 is arranged. Thereby, in the laser light source, multimode oscillation due to the so-called hole burning effect can be suppressed.

The non-linear optical crystal element 13 is arranged in the optical path of the optical resonator composed of the plane mirror 15 and the concave mirror 16, and is a so-called external SHG method (a method of arranging in the optical resonator formed outside the laser oscillator). Make up. Plane mirror 15
Reflects most of the light. The concave mirror 16 is Nd: YA
It transmits most of the second harmonics of the solid-state laser medium made of G and reflects most of the light having other wavelengths. Concave mirror 1
6 can be constituted by a dichroic mirror, for example.

As shown in FIG. 11, the second harmonic generation device 20 is composed of a nonlinear optical crystal element 21 made of, for example, BBO (β-BaB ₂ O ₄ ) and an optical resonator 22. The nonlinear optical crystal element 21 that constitutes the second harmonic generation device 20 is arranged in the optical path of the optical resonator 22.
That is, the second harmonic generation device 20 is a so-called external SHG system. In this optical resonator 22, the so-called finesse value (corresponding to the Q value of resonance) is increased to, for example, about 100 to 1000, and the optical density inside the optical resonator 22 is changed to the light incident on the optical resonator 22. By setting the light density to several hundred times, it is possible to effectively utilize the nonlinear effect of the nonlinear optical crystal element 21 arranged in the optical resonator 22.

The optical resonator 22 comprises a pair of concave mirrors 23, 24.
And a pair of plane mirrors 25 and 26. Second
Light incident on the harmonic generator 20 (for example, 532 nm
Light having a wavelength of 2) passes through the first concave mirror 23, the nonlinear optical crystal element 21, and at least a part of the second
After being made into a harmonic wave (for example, light having a wavelength of 266 nm), it is reflected by the second concave mirror 24, and then the plane mirror 25,
It is reflected by 26 and further by the first concave mirror 23. In such a state, at least a part of the light (for example, light having a wavelength of 266 nm) incident on the second concave mirror 24 is transmitted through the second concave mirror 24 and is directed from the second harmonic generation device 20 toward the reticle 61. Is ejected. In addition, a part (for example, light having a wavelength of 532 nm) of the light that has entered the first concave mirror 23 from the plane mirror 26 is the first light.
Is transmitted through the concave mirror 23 and the resonator length control device 30 to be described later.
Incident on. Incidentally, the first and second concave mirrors 23, 24,
The plane mirrors 25 and 26 are designed to reflect / transmit light as described above. The second concave mirror 24 can be composed of, for example, a dichroic mirror. The optical resonator 22 suitable for use in the third embodiment has a structure in which a part of the light incident on the optical resonator 22 can be emitted from the second concave mirror 24.

The wavelength of the light emitted from the second harmonic generation device 20 is the second harmonic of the incident light with reference to the light incident on the second harmonic generation device 20. That is, the wavelength of the incident light incident on the second harmonic generation device 20 is 532
and the light emitted from the second harmonic generation device 20 is 266 nm. The wavelength of the laser light emitted from the solid-state laser medium 12 made of Nd: YAG (1064n
Based on m), the light emitted from the second harmonic generation device 20 corresponds to the fourth harmonic. Laser light having a narrow band with a wavelength of 266 nm is continuously emitted from the second harmonic generation device 20, and the mode uniformity of such light is high.
The light emitted from the second harmonic generation device 20 has a
Light having a wavelength of 32 nm is also mixed.

The second harmonic generator 20 is further provided with a resonator length controller 30. The resonator length (L) of the optical resonator 22 is precisely controlled by the resonator length control device 30 and is maintained at a constant length. By precisely maintaining the resonator length (L) of the optical resonator 22 at a constant length, the second
The intensity of the emitted light emitted from the harmonic generator 20 can be kept constant. The resonator length (L) is the first
Of the concave mirror 23, the second concave mirror 24, the plane mirror 25, the plane mirror 26, and the first concave mirror 23.

Emitted light emitted from the second harmonic generation device 20 (second incident light incident on the second harmonic generation device 20
When the wavelength of the higher harmonics is λ, and the resonator length L ₀ of the optical resonator 22 satisfies λ = L ₀ / N (where N is a positive number) (also referred to as a lock state), the optical resonance The container 22 resonates, and the second harmonic generation device 20 stably emits high-intensity light.
In other words, the optical path phase difference Δ in the optical resonator 22 is 2
When it is an integral multiple of π, the optical resonator 22 forming the second harmonic generation device 20 is in a resonance state. That is, the lock state is set. Here, when the refractive index of the nonlinear optical crystal element 21 is n and the thickness is 1, the optical path phase difference Δ can be expressed by (4πnl / λ).

The resonator length of the optical resonator 22 L ₀ ± ΔL ₀
However, when λ ≠ (L ₀ ± ΔL ₀ ) / N ′ (where N ′ is a positive number) (also referred to as an unlocked state), the second harmonic generation device 20 emits light of low intensity. In other words, when the optical path phase difference Δ in the optical resonator 22 deviates from an integer multiple of 2π, the optical resonator 22 forming the second harmonic generation device 20 is in a non-resonant state. That is, the unlocked state is set.

Therefore, in order to stably emit the light of wavelength λ from the second harmonic generation device 20, the resonator length (L) of the optical resonator 22 changes with time (specifically, for example, It is necessary to minimize the fluctuation of the positions of the concave mirrors 23 and 24 and the plane mirrors 25 and 26. Therefore, under the control of the resonator length control device 30, the first concave mirror 23 is moved on the optical axis connecting the first concave mirror 23 and the second concave mirror 24, or the first concave mirror 23 with respect to the optical axis. By changing the arrangement angle of the optical resonator 22 to suppress the temporal change of the resonator length (L) of the optical resonator 22 and keep the resonator length (L) of the optical resonator 22 constant.

The resonator length control device 30 is a “laser light generator” for which a patent application was filed by the applicant of the present invention on March 2, 1992.
(Japanese Patent Application Laid-Open No. 5-243661).

A resonator length control device 30 of this type is shown in FIG.
1, a photodetector 31, such as a photodiode,
A voice coil motor (VCM) 32, a voice coil motor control circuit (VCM control circuit) 33, and a phase modulator 34 are included. The phase modulator 34 is arranged in the optical path between the laser light source 10 and the second harmonic generation device 20, and is a so-called EO (electro-optical) element or AO that phase-modulates the light emitted from the laser light source 10. (Acousto-optic) element. A condenser lens 35 is arranged between the phase modulator 34 and the second harmonic generation device 20. The voice coil motor 32 includes the first concave mirror 2 that constitutes the optical resonator 22.
3 is attached.

As shown in the schematic view of FIG. 12, a voice coil motor 32 includes a base 320 made of a magnetic material, one or more electromagnets (so-called voice coils) 322, a yoke 323 made of a magnetic material, and at least one coil spring. (Or a spiral leaf spring) 321 is an electromagnetic actuator. The coil spring 321 has one end attached to the base body 320 and the other end attached to the yoke 323.
Attached to. Further, the yoke 323 has a first
The concave mirror 23 and the electromagnet 322 are attached.
When a current is applied to the electromagnet 322, a magnetic field is formed and the distance between the yoke 323 and the base 320 changes. As a result, the position of the first concave mirror 23 can be moved. That is, the resonator length (L) of the optical resonator 22 can be changed by controlling the current flowing through the electromagnet 322. Servo control is performed on the voice coil motor 32.

The drive current of the voice coil motor 32 is about several tens to several hundreds mA. Therefore, the drive circuit configuration can be manufactured at low cost. Moreover, the frequency of multiple resonance of the servo loop can be set to several tens of kHz to 100 kHz or more, and since the frequency characteristics with few phase rotations are provided, the servo band can be widened to several tens of MHz and stable control can be performed. Obtainable.

When the optical resonator 22 is in the locked state, for example, the intensity of the light emitted from the first concave mirror 23 and reaching the photodetector 31 is minimized, and the phase of the light is greatly changed. Controlling an optical resonator using such changes is described in, for example, RWP Drever, et al. "Lase
r Phase and Frequency Stabilization Using an Optic
al Resonator ", Applied Physics B31. 97-105 (1983). Control of the locked state of the optical resonator 22 is described in
Basically, this technology is applied.

That is, for example, the light is transmitted through the first concave mirror 23,
If the VCM control circuit 33 drives the voice coil motor 32 to change the position of the first concave mirror 23 so that the intensity of the light reaching the photodetector 31 is always a minimum value (for example, 0), The locked state of the resonator 22 can be stably maintained. In other words, the light emitted from the laser light source 10 is phase-modulated based on the phase-modulated signal, is incident on the second harmonic generation device 20, and the return light from the second harmonic generation device 20 is detected by the photodetector 31. A detection signal is obtained by detecting with. Then, the detection signal is
Synchronous detection is performed using the phase modulation signal, and the error signal is extracted. The VCM control circuit 33 drives the voice coil motor 32 so that this error signal becomes 0, and the first concave mirror 2
Change the position of 3.

The VCM control circuit 33 has, for example, an oscillator 330, a phase modulator drive circuit 331, a synchronous detection circuit 332, a low pass filter 33, as shown in the configuration diagram of FIG.
3 and a voice coil motor drive circuit (VCM drive circuit) 334.

The frequency f output from the oscillator 330
_{The m} (for example, 10 MHz) modulation signal is sent to the phase modulator 34 via the phase modulator driving circuit 331. In the phase modulator 34, the light emitted from the laser light source 10 (frequency f _{O, on the} order of 10 ¹⁴ Hz) is subjected to phase modulation, and sidebands having a frequency f _O ± f _m are generated.

The first concave mirror 23 constituting the optical resonator 22.
The light (frequency: f _O and f _O ± f _m ) that has passed through and exited the system of the optical resonator 22 is detected by the photodetector 31. The FM sideband method for detecting the beat between the light having such a frequency (f _O and f _O ± f _m), it is possible to obtain an error signal having a polarity, such on the basis of the error signal optical resonator 22 Control the cavity length (L) of the.

That is, the signal output from the photodetector 31 is sent to the synchronous detection circuit 332. This signal is a signal in which a light intensity signal of frequency f _O and a signal corresponding to a modulation signal of frequency f _m are superimposed. Synchronous detection circuit 33
The modulated signal output from the oscillator 330 (with waveform shaping and phase delay applied as necessary) is also supplied to 2. The signal output from the photodetector 31 and the modulated signal are multiplied in the synchronous detection circuit 322 to perform synchronous detection. The detection output signal output from the synchronous detection circuit 332 is input to the low-pass filter 333. By removing the modulation signal component from the detection output signal in the low-pass filter 333, an error signal of the resonator length of the optical resonator 22 is generated. To be done. Here, the error signal means the measured resonator length (L ₀ ± ΔL ₀ ) with respect to the set resonator length (L ₀ ) of the optical resonator 22.
Is a signal representing the difference (± ΔL ₀ ).

This error signal is sent to the VCM drive circuit 334, and the voice coil motor 32 is based on this error signal.
Are driven (specifically, by controlling the current flowing through the electromagnet 322), the light is transmitted through the first concave mirror 23 and the photodetector 3
So that the light reaching 1 has a minimum value (in other words,
The resonator length (L) of the optical resonator 22 is adjusted so that the resonator length of the optical resonator 22 becomes L ₀ and the error signal becomes 0).

When the resonator length (L) of the optical resonator 22 is set to L ₀ (that is, in the locked state), the resonator length control device 30 controls the resonator length of the optical resonator 22. The fluctuation of (L) over time is measured by the second harmonic generation device 2
1/1000 to 1/10000 of the wavelength of light incident on 0
Can be suppressed to

The present invention has been described above based on the preferred embodiments, but the present invention is not limited to these embodiments. [Pattern latent image forming step-
1], [pattern latent image forming step-2] and [pattern latent image contrast measuring step] may be changed as follows, which is also included in the optimization method of the multiple imaging exposure method of the present invention. . (1) [Pattern latent image forming step-1] (2) [Pattern latent image contrast measuring step] (3) [Pattern latent image forming step-2] (4) [Pattern latent image contrast measuring step] (5 ) Repeat (3) and (4) if necessary

Further, immediately after the execution of the pattern latent image forming step described in [Pattern latent image forming step-1], the [Pattern latent image contrast measuring step] is performed and these operations are repeated a necessary number of times. Is also included in the optimization method of the multiple imaging exposure method of the present invention.

Optical path splitting means 50 using an optical fiber
Or bandpass filter 80 to alignment microscope 4
Light can be directly transmitted to the photochromic material 70 without passing through 0. The optical path splitting means 50 may be any means capable of splitting the optical path other than the beam splitter and the half mirror.

The semiconductor exposure apparatus of the present invention is not limited to the projection exposure apparatus using the refraction-type optical system as in the above-mentioned embodiments, and for example, the semiconductor exposure apparatus using the reflection-type optical system. And a proximity exposure apparatus.

As the photochromic material, various organic photochromic materials or inorganic photochromic materials other than spiropyran can be used. As organic photochromic materials, photo-oxidation reduction (methylene blue + Fe ²⁺ ), generation of free radicals by photodissociation reaction (β-tetrachloro-1-ketodihydronaphthalene), tautomerization associated with intramolecular hydrogen transfer (salicylidene) Examples of the photochromic material include a reaction form such as aniline), cis-trans photoisomerization (azobenzene), and photopolymerization (anthracene). As an inorganic photochromic material, oxides such as SrTiO ₃ doped with transition metal elements, fluorides such as CaF ₂ doped with rare earth elements, and silver halides dispersed in glass such as silicate glass. What was made to use can be used. Examples of substances and dopants are shown below. Material Dopant SrTiO ₃ Fe / Mo, Ni / Mo CaTiO ₃ Fe, Zn, Sb, V, Ni / Mo TiO ₂ Fe, Cr, Cu, Na, Mn BaTiO ₃ Fe, Zn, Sb, V CaWO ₄ Bi Nb ₂ O ₅ Fe SnO ₂ Cu CaF ₂ Ce, Gd, Tb, La CaF ₂ Eu / Sm BaF ₂ Eu / Sm Silicate glass Eu, Ce, Zr Silicate glass AgBr, AgCl AgI · HgI ₂ ---

Laser light source 1 constituting a light source for projection optical system
0, the second harmonic generation device 20, and the resonator length control device 30
The above structure is an example, and the design can be changed as appropriate, and a light source for a projection optical system of another type can be used. The latent image detection system and the erasing light source may be provided independently of the projection optical system light source.

Laser light source 10 and second harmonic generator 20
In the case of configuring the projection optical system light source from the resonator length control device 30, the solid-state laser medium is not limited to Nd: YAG.
It can be composed of Nd: YVO ₄ , Nd: BEL, LNP and the like. The pumping method of the solid-state laser medium by the laser diode can be not only the edge pumping method but also the side pumping method and the surface pumping method, and further, a slab solid-state laser can be used. In addition to KTP and BBO, LN and QPM L are also used as nonlinear optical crystal elements.
N, LBO, KN or the like can be appropriately selected depending on the wavelength of light required for incident light or emitted light.

It is also possible to use a so-called internal SHG type laser light source in which a solid-state laser medium and a nonlinear optical crystal element are arranged in the optical path of an optical resonator composed of a pair of reflecting mirrors.
Further, it is also possible to adopt a structure in which light emitted from the solid-state laser medium 12 is passed through the nonlinear optical crystal element 13 (that is, a structure in which the optical resonator including the plane mirror 15 and the concave mirror 16 is omitted). Further, as the laser light source, for example, a blue semiconductor laser can be used instead of the LD pumped solid-state laser, and the emitted light of such a semiconductor laser can be directly incident on the second harmonic generation device, or it can be nonlinear with the semiconductor laser. It is also possible to have a combination structure of a laser light source of a so-called internal SHG system in which an optical crystal element is combined and a second harmonic generation device. Further, a resonator length control device 30 may be separately provided for controlling the resonator length of the optical resonator including the plane mirror 15 and the concave mirror 16.

The structure of the optical resonator 22 in the second harmonic generation device 20 may be, for example, a Fabry-Perot type resonator composed of a concave mirror and a plane mirror. In this case, a reflecting mirror that transmits the incident light incident on the second harmonic generation device 20 and reflects the returned light from the second harmonic generation device 20 is arranged in front of the second harmonic generation device 20. The light reflected by the reflecting mirror may be detected by the photodetector 31. In order to change the resonator length of the optical resonator 22, not only the first concave mirror 23 may be moved but other mirrors may be moved.

As another mode of the resonator length control device 30,
An example of the resonator length control device is PZT. That is, in order to move the first concave mirror 23 forming the optical resonator 22, a laminated piezoelectric element made of PZT or the like and a signal proportional to the length change of the resonator length (L) are supplied to this laminated piezoelectric element. A resonator length control device including a control device is used to feed back such a signal to form a servo loop. Thereby, the resonator length of the optical resonator 22 can be controlled, and the intensity of the emitted light emitted from the second harmonic generation device 20 can be controlled.

The light emitted from the second harmonic generator is
Although the light mainly has a wavelength based on the second harmonic of the incident light from the laser light source, the wavelength of the light emitted from the second harmonic generator is the solid state as described in the embodiment. Not only the fourth harmonic wave based on the light emitted from the laser medium but also the fifth harmonic wave can be used. In this case,
For example, the light (wavelength: 1064 nm) emitted from the solid-state laser medium made of Nd: YAG and the light (wavelength: 266 nm) emitted from the second harmonic generation device 20 are combined, and another second harmonic is generated again. Wave generator 20 (for example, urea CO (NH ₂ ) _{2 which} is an organic crystal as a nonlinear optical crystal element)
The fifth harmonic (wavelength: 213 nm) of the solid-state laser medium made of Nd: YAG can be generated.

As the substrate, a silicon semiconductor substrate or G
Examples thereof include compound semiconductor substrates such as aAs, glass substrates for forming TFTs, and the like.

[0086]

According to the optimization method of the multiple imaging exposure method of the present invention, even when the pattern size and the pattern density to be formed on the resist are changed due to the exchange of the reticle, the time, labor and cost as in the prior art are required. It is possible to easily optimize the FLEX step ΔZ in the multiplex imaging exposure method without spending time. Further, by decoloring the photochromic material on which the pattern latent image is formed, the optimization method of the multiple imaging exposure method can be repeatedly executed. Furthermore, in a preferred aspect of the semiconductor exposure apparatus of the present invention, the light source for the projection optical system also serves as the light source for the latent image detection system and the light source for erasing, so that the structure of the semiconductor exposure apparatus can be simplified. Therefore, it does not require much cost for increasing the cost and maintenance of the semiconductor exposure apparatus.

[Brief description of drawings]

FIG. 1 is a conceptual diagram of a semiconductor exposure apparatus according to a first embodiment and is a diagram showing a latent image forming step in an optimization method of a multiplex image formation exposure method.

FIG. 2 is a conceptual diagram of a semiconductor exposure apparatus according to a first embodiment, and is a diagram showing a latent image contrast measurement step in an optimization method of a multiple imaging exposure method.

FIG. 3 is a conceptual diagram of the semiconductor exposure apparatus according to the first embodiment, and is a diagram showing a latent image erasing step in the optimization method of the multiple imaging exposure method.

FIG. 4 is a diagram schematically showing changes in a light absorption curve of a photochromic material.

FIG. 5 is a diagram schematically showing contrast curves at various ΔZ.

FIG. 6 is a flow chart of steps of an optimization method of a multiplex imaging exposure method.

FIG. 7 is a conceptual diagram of a semiconductor exposure apparatus according to a second embodiment, and is a diagram showing a latent image contrast measurement step in an optimization method of a multiple imaging exposure method.

FIG. 8 is a conceptual diagram of a semiconductor exposure apparatus according to a third embodiment, and is a diagram showing a latent image contrast measurement step in an optimization method of a multiple imaging exposure method.

FIG. 9 is a conceptual diagram of a semiconductor exposure apparatus for explaining a resist exposure process.

FIG. 10 is a conceptual diagram of a semiconductor exposure apparatus for explaining an alignment operation by an off-axis alignment method.

FIG. 11 is a schematic diagram of a laser light source, a second harmonic generation device, and a resonator length control device.

FIG. 12 is a schematic view of a voice coil motor.

FIG. 13 is a configuration diagram of a VCM control circuit that constitutes a resonator length control device.

FIG. 14 is a diagram schematically showing a light intensity distribution in the thickness direction of a resist obtained by a multiple imaging exposure method.

FIG. 15 is a diagram schematically showing a light intensity distribution in the thickness direction of a resist obtained by a multiple imaging exposure method.

FIG. 16 is a diagram showing the relationship between the deviation of the distance from the best focus and the exposure pattern size when exposure is performed by changing the FLEX step ΔZ.

FIG. 17 is a diagram showing the relationship between the deviation of the distance from the best focus and the exposure pattern size when exposure is performed while changing the FLEX step ΔZ.

FIG. 18 is a diagram showing the relationship between the deviation of the distance from the best focus and the exposure pattern size when exposure is performed by changing the FLEX step ΔZ.

[Explanation of symbols]

 10 Laser Light Source 11 Laser Diode 12 Solid State Laser Medium 13 Nonlinear Optical Crystal Element 14 Quarter Wave Plate 15 Plane Mirror 16 Concave Mirror 20 Second Harmonic Generator 21 Nonlinear Optical Crystal Element 22 Optical Cavity 23 First Concave Mirror 24 Second Concave mirror 25, 26 Plane mirror 30 Resonator length control device 31 Photodetector 32 Voice coil motor 320 Base body 321 Coil spring 322 Electromagnet 323 Yoke 33 VCM control circuit 330 Oscillator 331 Phase modulator drive circuit 332 Synchronous detection circuit 333 Low-pass filter 334 VCM drive Circuit 34 Phase modulator 35 Condenser lens 40 Alignment microscope 41, 42 Photodetector 43, 71 Half mirror 44 Signal processing device 50 Optical path splitting means 61 Reticle 62 Projection optical system 63 Resist 64 Substrate 65 Alignment Over click 66 substrate stage 67 substrate chuck 70 photochromic material 80 band-pass filter

Claims (9)

[Claims] 1. A light source for a projection optical system, and a projection optical system for forming an optical image obtained by irradiating a reticle with light from the light source for a projection optical system on a substrate. (C) A substrate stage on which a substrate is placed, (d) a photochromic material placed on the substrate stage, and (e) a photochromic material obtained by irradiating a reticle with light from a light source for a projection optical system. Latent image detection system for optically observing the formed pattern latent image,
A semiconductor exposure apparatus comprising: 2. A light source for a projection optical system, a laser light source, and a second harmonic wave generating light which has a wavelength based on a second harmonic wave of the light when the light emitted from the laser light source is incident. 2. The semiconductor exposure apparatus according to claim 1, wherein the laser light source also serves as a latent image detection system light source and a color erasing light source for erasing the pattern latent image. 3. A laser light source is a laser diode, Nd:
A solid-state laser medium made of YAG and an LD-pumped solid-state laser made up of a non-linear optical crystal element, and light emitted from the non-linear optical crystal element is used as a light source for a latent image detection system and a light source for erasing. The semiconductor exposure apparatus according to claim 2. 4. The laser light source is a laser diode, Nd:
7. A solid-state laser medium made of YAG and an LD-pumped solid-state laser made up of a non-linear optical crystal element, wherein light emitted from a laser diode is used as a latent image detection system light source and a color erasing light source. 2. The semiconductor exposure apparatus according to item 2. 5. A light source for a projection optical system, a laser light source, and a second harmonic generation that emits light having a wavelength based on the second harmonic of the light upon receipt of the light emitted from the laser light source. The apparatus further comprises a bandpass filter for separating light having a long wavelength component from the light emitted from the second harmonic generation device, the light having such a long wavelength component being used as a light source for a latent image detection system and a pattern latent image. The semiconductor exposure apparatus according to claim 1, wherein the semiconductor exposure apparatus is used as a erasing light source for erasing an image. 6. The latent image detection system also serves as an off-axis alignment detection system.
The semiconductor exposure apparatus according to claim 5. 7. An initial value of the distance from the projection optical system to the photochromic material arranged on the substrate stage is Z _0.
Then, the pattern is formed by sequentially changing the distance Z from the projection optical system to the photochromic material arranged on the substrate stage to Z ₀ + ΔZ × i (where i = 0, 1, 2, ..., K). A pattern optical image obtained by irradiating the reticle with light from a light source for a projection optical system forms a pattern on the photochromic material by a multiple imaging exposure method that forms (K + 1) times on the photochromic material via the projection optical system. The step of forming the latent image is repeated M times while keeping the value of ΔZ constant and changing the value of the distance initial value Z ₀ to form M pattern latent images having different values of Z. (A) is repeated N times while changing the value of ΔZ, finally M × (N + 1) pattern latent images are formed, and (C) the obtained pattern latent image is optically observed. ) Based on the pattern latent image observation result, the optimum Δ Determination of the value, the optimization method of multiple-imaging exposure method, characterized in that it consists of the steps. 8. The multiplex according to claim 7, further comprising the step of erasing the pattern latent image by irradiating the pattern latent image formed on the photochromic material with light from a erasing light source. Optimizing the imaging exposure method. 9. The optical observation of the pattern latent image consists of measuring the contrast of the pattern latent image, and is performed using an off-axis alignment detection system.
Or the optimization method of the multiple imaging exposure method of Claim 8.