JPH11143048A - Phase shift mask - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a phase shift mask capable of equalizing the sizes of the patterns formed by respective transparent regions when a light shielding film formed as a thin film is used. SOLUTION: The mask patterns by the light shielding film 2 are formed on a transparent substrate 1 consisting of synthetic quartz. The mask patterns are L/S patterns. The transparent regions 4 and 5 having the equal width are alternately formed across the light shielding regions formed with the light shielding film 2 having the width equal to their width. Further, the thickness of the transparent substrate 1 in the transparent regions 5 is thinner by 241 nm than the thickness in the transparent regions 4. The phase of the transmitted light transmitted through the transparent regions 4 and 5 varies by 175 deg.. The light shielding film 2 comprises a chromium film having a film thickness of 30 μm and a chromium oxide film of 30 μm in the film thickness formed on this chromium film. The transmission intensity of the KrF excimer laser beam transmitted through the light shielding film 2 is 0.2%.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a phase shift mask used in an optical lithography technique during a manufacturing process of a semiconductor device, and in particular, the dimensions of adjacent patterns are equal even when the focal position is shifted. The present invention relates to a phase shift mask.

[0002]

2. Description of the Related Art Conventionally, in a manufacturing process of a semiconductor device,
Optical lithography is mainly used to form a pattern on a semiconductor substrate. According to the optical lithography technology, a predetermined pattern of a photosensitive resin is transferred by transferring a pattern of a semiconductor element formed on a photomask onto a semiconductor substrate coated with a photosensitive resin by a reduction projection exposure apparatus and developing the pattern. Can be obtained.

A photomask is an exposure master in which a pattern of a transparent region and a light-shielding region is formed on a transparent substrate with a light-shielding material. This exposure master is called a reticle when a pattern on a photomask is transferred onto a semiconductor substrate by an exposure apparatus. In addition, the material of the transparent substrate is required to have characteristics such as a high transmittance to exposure light, a low coefficient of thermal expansion, a pattern position that is not shifted by a temperature rise due to exposure light irradiation, and a small deformation due to its own weight. ing. Synthetic quartz is most suitable for a transparent substrate requiring such characteristics, and synthetic quartz is used for all high-precision products for the most advanced semiconductor device manufacturing. In addition, when the positional accuracy of the mask pattern is not required to be very high and the pattern size is not fine, a less expensive material such as soda glass is used for manufacturing a semiconductor element.

Further, a chrome light-shielding film which is easily etched by a wet etching solution (aqueous cerium ammonium nitrate solution) is used as a light-shielding film in a light-shielding region of a photomask. This is because the mask pattern is drawn by an electron beam lithography apparatus, and the photosensitive resin (resist) for electron beam lithography has low dry etching property and wet etching is applied. The chrome light-shielding film is generally composed of a chromium film having a thickness of 70 nm and a chromium oxide film having a thickness of 30 nm formed on the chromium film.
It has a layer structure, and the upper chromium oxide film functions as an antireflection film. This anti-reflection film reflects the light reflected by the semiconductor substrate back to the photomask through the projection lens system when the semiconductor substrate is exposed to light, and the reflected light is reflected on the photomask surface and irradiated again on the semiconductor substrate. It is to prevent that.

As the miniaturization of semiconductor elements further progresses and the pattern size of the photomask becomes 1 μm or less, it is necessary to improve the dimensional accuracy of the mask pattern. In the conventional optical lithography technology, the development of an exposure apparatus has been adapted to miniaturization of a semiconductor element pattern, particularly by increasing the numerical aperture (NA) of a projection lens system.
The numerical aperture (NA) is a value corresponding to how much the lens can collect the spread light, and the larger the value, the more the spread light can be collected and the higher the performance of the lens. In addition, there is a relation between the critical resolution R and the NA, which are the dimensions of the critical fine pattern that can be separated, by the well-known Rayleigh equation represented by the following equation (1).

[0006]

R = K ₁ × λ / NA Here, K ₁ is a constant depending on a process such as resist performance, and λ is a wavelength of light. As shown in Equation 1, the larger the NA, the finer the limit resolution R.

However, as the NA is increased, the depth of focus, which indicates a range in which the deviation of the focal position can be tolerated, decreases, and this phenomenon of the depth of focus makes it difficult to miniaturize the pattern. The depth of focus DOF and NA have the relationship of the Rayleigh equation shown in the following Expression 2.

[0008]

DOF = K ₂ × λ / (NA) ² where K ₂ is a process-dependent constant. As shown in the above formula 2, as the NA increases, the depth of focus DOF decreases, and even a slight shift of the focal position cannot be tolerated.

Accordingly, various super-resolution techniques have been studied to increase the depth of focus. The super-resolution technique is a technique for improving the light intensity distribution on the imaging plane by controlling the transmittance or the phase on the pupil plane of the illumination optical system, the photomask, or the projection lens system.

A phase shift mask which is a super-resolution technique by improving a photomask has been proposed (JP-A-62-5).
0811). The phase shift mask described in this publication is called a Shibuya-Levenson type phase shift mask, and is a type that alternately changes the phase of light transmitted through a transparent region by 180 ° in a periodic pattern. FIG. 9A is a plan view showing a first Shibuya-Levenson type phase shift mask, and FIG.
It is sectional drawing in the -A line, (c) is a graph which shows the position of a phase shift mask on a horizontal axis, and the amplitude of transmitted light on a vertical axis | shaft, and is a graph which shows the amplitude distribution of transmitted light. In the first method of manufacturing a Shibuya-Levenson type phase shift mask, first, an etching stopper 22 is formed on a transparent substrate 21. Next, the light shielding film 2 is formed on the etching stopper 22.
The transparent regions 24 and 25 are formed alternately by forming a film 3 and selectively removing the film. Then, a phase shifter 26 is formed on the transparent region 25. The thickness of the phase shifter 26 is λ / (2 × (n ₁ −1)), where λ is the wavelength of the exposure light and n ₁ is the refractive index of the phase shifter 26. Further, the phase shifter 26 is, for example, a silicon oxide (S
iO ₂ ).

Assuming that the wavelength of the exposure light in air is λ ₁ and the refractive index of a substance through which the exposure light passes is n, the wavelength of the exposure light in this substance is λ / n. In the phase shift mask configured as described above, the exposure light passing through the transparent region 24 and the exposure light passing through the transparent region 25 have a 180 ° phase shift. Therefore, as shown in FIG. 9C, the cycle of the amplitude distribution is twice as long as the original one. Therefore, the diffraction angle of the diffracted light of this phase shift mask is の of a normal angle, and even in a pattern having a resolution lower than the limit resolution up to that time, the diffracted light can be collected by the projection lens. Then, due to the interference between the light beams whose phases are shifted by 180 °, the light intensity is reduced between the adjacent openings, and the fine pattern can be separated.

Further, since the etching stopper 22 is provided on the transparent substrate 21, it is possible to prevent the transparent substrate 21 from being etched when performing the etching for forming the phase shifter 26. When the transparent substrate 21 is etched, a phase error occurs. Therefore, the etching stopper 22 does not affect the operation of the phase shift mask, but is required in the manufacturing process.

In the Shibuya-Levenson type phase shift mask, not only an example in which a phase shifter is provided on a light shielding film serving as a mask, but also an example in which one of transparent regions is etched. FIG. 10A is a sectional view showing a second Shibuya-Levenson type phase shift mask, and FIG.
FIG. 2 is a cross-sectional view showing the Shibuya-Levenson type phase shift mask of FIG. In the second Shibuya-Levenson type phase shift mask, as shown in FIG.
A phase shifter 36 is formed between the etching stopper 32 formed on the substrate a and the light shielding film 33a serving as a mask. On the other hand, in the third Shibuya-Levenson type phase shift mask, as shown in FIG.
1b is etched on the surface of the phase shifter to form irregularities, and the phase shifter and the etching stopper are not formed. Furthermore, in these Shibuya-Levenson type phase shift masks, the etched phase shifters or the side walls of the transparent substrate are recessed below the light shielding films 33a and 33b.

In the thus constructed Shibuya-Levenson type phase shift mask, the scattering of light by the side walls is suppressed, so that even if the phase shift mask is formed by dry etching, the light whose phase is disturbed is transparent. Substrate 3
Passing through 1a or 31b is prevented. Thereby, there is no influence by the side wall.

However, even the use of the Shibuya-Levenson type phase shift mask cannot sufficiently cope with recent miniaturization of semiconductor devices. Therefore, thinning of the light-shielding film is being studied to improve the mask dimensions (Proceedi
ngs of SPIE Vol. 2793, Photomask and X-ray Mask Te
chnology III "Resolution Enhancement with Thin Cr
for Chrome Mask Making ", pp.53-61, 1996).
It is described that Å improves the dimensional accuracy. In other words, in the case of processing by wet etching, the amount of side etching is reduced by thinning, so that dimensional control is facilitated. In addition, dry etching can be easily performed even with a resist for electron beams having low dry etching resistance, and further improvement in dimensional accuracy is expected by applying dry etching.

[0016]

However, the transmission intensity of the exposure light in the chrome light-shielding film having a thickness of about 1000 ° is not more than 10 ^{−4 of} the transmission intensity of the transparent region (hereinafter, the transmission intensity is the exposure light transmitted through the transparent region). This is a value obtained by converting the light transmission intensity to 1), so that the light leaked from the light-shielding film did not affect the exposure result, but the transmission intensity in the light-shielding film was 10 ⁻ due to the thinning of the light-shielding film. ^{As a} result, the transfer characteristic has been affected. In particular, in the case of a phase shift mask, the pattern size is significantly changed due to a shift in the focal position, so that the influence is large.

The present invention has been made in view of such a problem, and a phase shift mask capable of equalizing the size of a pattern formed by each transparent region even when a thinned light shielding film is used. The purpose is to provide.

[0018]

A phase shift mask according to the present invention is provided between a first transparent area, a second transparent area, and the first transparent area and the second transparent area. And a mask pattern having a light-shielded region,
In a phase shift mask having a phase difference between transmitted light transmitted through the transparent region and transmitted light transmitted through the second transparent region, the phase difference is less than 180 °.

In the present invention, since the phase difference between the transmitted light transmitted through the first transparent region and the transmitted light transmitted through the second transparent region is intentionally made smaller than 180 °, the focal position is changed. Even if there is a deviation, the intensity of the transmitted light passing through each transparent region can be made equal. Thereby, the dimensions of the pattern formed by each transparent region can be made equal.

In the present invention, when the area ratio between the light-shielding region and the first transparent region is α, and the intensity of the light transmitted through the transparent region is 1, the intensity of the light transmitted through the light-shielding region is equal to 1. When the intensity is T and the difference between the second phase and the third phase of the transmitted light transmitted through the light-shielding region is θ, the phase difference is (180−sin ⁻¹ (α × √ (T) × sin
θ)) °. The phase difference between the second phase and the third phase is (180−sin ⁻¹ (α × √ (T) × s
In θ) °, the intensity of the transmitted light that passes through each transparent region can be reliably equalized.

Another phase shift mask according to the present invention comprises a first transparent region, a second transparent region, and a light-shielding region provided between the first transparent region and the second transparent region. A phase shift mask having a phase difference between transmitted light transmitted through the first transparent region and transmitted light transmitted through the second transparent region. The phase of the transmitted light passing through the region is 0 °, and the phase of the transmitted light passing through the second transmission region is 18 °.
0 °, and the phase of the transmitted light transmitted through the light-shielding region is 9 °.
0 °, and the width of the first transparent region is the second transparent region.
Is larger than the width of the transparent region. Further, the phase of the transmitted light passing through the light-shielding region may be less than 90 °, and the width of the first transparent region may be smaller than the width of the second transparent region.

In the present invention, the width of the first transparent region or the second transparent region is made appropriate by the phase of the transmitted light transmitted through the light shielding region. The dimensions of the patterns formed by the regions can be made equal to each other.

The transparent region may be a reflection region, the light shielding region may be an absorption region, and the transmitted light may be reflected light. This phase shift mask is particularly suitable for an X-ray exposure apparatus or the like.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION The inventors of the present invention have conducted intensive experiments and researches to solve the above-mentioned problems. As a result, the phase shift of the transmitted light passing through the light-shielding film was set to about 90 °, so that the focal position was shifted. When there is no transmission light, the intensity of the transmitted light passing through the transparent region that causes a phase difference of 0 ° and the intensity of the transmitted light passing through the 180 ° transparent region can be equalized, and the phase difference between adjacent transparent regions can be reduced from 180 °. It has been found that the focus characteristic of transmitted light passing through adjacent transparent regions can be matched by intentionally causing a phase error by correcting an appropriate value. Note that the phase difference refers to a phase difference from the phase of transmitted light transmitted through the air, and the focus characteristic refers to a change in dimension of a line pattern due to a change in a focal position.

Hereinafter, a phase shift mask according to an embodiment of the present invention will be specifically described with reference to the accompanying drawings. FIG. 1 is a sectional view showing a phase shift mask according to a first embodiment of the present invention. In this embodiment, a mask pattern is formed by a light shielding film 2 on a transparent substrate 1 made of synthetic quartz. Note that the mask pattern in the present embodiment is a line and space pattern (hereinafter, referred to as an L / S pattern), and a repeated pattern of lines and spaces having the same dimensions is formed. That is, the transparent regions 4 and 5 having the same width are alternately formed with the light-shielding region in which the light-shielding film 2 having the same width is formed. Further, the transparent substrate 1 in the transparent area 5
Is 241 nm thinner than the thickness of the transparent region 4, and is 1 nm below the light shielding film 2 adjacent to the transparent region 5.
It has the same thickness with a width of 50 nm. On the other hand, the light shielding film 2 is composed of a chromium film having a thickness of 30 μm and a chromium oxide film having a thickness of 30 μm formed on the chromium film.

Next, the operation of the thus constructed phase shift mask will be described. Here, NA is 0.5
5, coherence factor σ is 0.3, reduction ratio is 5 times,
That is, using a KrF excimer laser device in which the ratio of the mask pattern dimension to the pattern on the image plane is 5: 1, exposure is performed on a 0.7 μm-thick silicon substrate coated with a negative resist. An L / S pattern is formed. In this embodiment, the transmission intensity of the KrF excimer laser light of the light shielding film 2 is 2 × 10 ⁻³ , the refractive index of the chromium film with respect to the KrF excimer laser light having a wavelength of 248 nm is 2.1, and chromium oxide is used. The refractive index of the film is 2.0.
Therefore, a phase difference of 91.4 ° occurs in the transmitted light transmitted through the light shielding film 2. Further, since the thicknesses of the transparent substrates in the transparent region 5 and the transparent region 4 are different by 241 nm, the light transmitted through the transparent region 4 and the light transmitted through the transparent region 5 have an angle of about 175 °. A phase difference occurs.

In this embodiment, since the phase difference of the transmitted light passing through the light shielding film 2 is set to about 90 °, the transmitted light passing through the transparent region 4 and the transmitting light passing through the transparent region 5 when the focal position is not shifted. The intensity with transmitted light can be made substantially equal. For this reason, it is possible to form line patterns of substantially the same size in regions on the silicon substrate corresponding to the respective transparent regions. Furthermore, since the phase difference of the transparent region 4 is corrected to an appropriate value from 180 ° to be 175 °, even when the focal position is shifted, the transmitted light passing through the transparent region 4 and the transmitted light passing through the transparent region 5 are transmitted. The intensity with light can be matched. That is, it is possible to make the focus characteristics of the transmitted light transmitted through the transparent regions 4 and 5 coincide.

Next, a method of manufacturing the phase shift mask thus configured will be described. First, a chromium film and a chromium oxide film are sequentially formed on the transparent substrate 1 to a thickness of 30 nm. Next, the chromium film and the chromium oxide film are selectively removed so that the width of the light-shielding film 2 is equal to the width of the transparent region. After that, every other transparent region is 241 nm.
Etch. At this time, R using CHF ₃ gas
The dry etching by the IE method and the wet etching using the diluted hydrofluoric acid are also performed under the light shielding film 2 by 150 n.
Side-etch with a width of m. Thus, the unetched transparent region 4 and the etched transparent region 5
Is manufactured with a phase difference of 175 °.

Next, the effect of setting the phase difference of the transmitted light passing through the light shielding film to 90 ° will be described. First,
An L / S pattern having a pattern size of 0.2 μm was formed. At this time, the transmission intensity of the transmitted light leaking from the light shielding film is set to 2
× 10 ^-3 and the phase difference is 90 °, 0 ° and 1
80 °, and set the focus position to 0μ at each phase difference.
The relative light intensity was measured while changing to m, 0.5 μm, and −0.5 μm. When the focal position is 0.5 μm, the image plane is located at a position 0.5 μm away from the front surface toward the back surface.
There is an imaging plane at a position 5 μm away. These results are shown in FIG.
To 5 are shown. FIG. 3 is a graph showing the relative light intensity distribution when the horizontal axis indicates the position on the semiconductor substrate and the vertical axis indicates the relative light intensity, and the phase difference of the transmitted light leaking from the light shielding film is 90 °. FIG. 4 is a graph showing the relative light intensity distribution when the phase difference is 0 °, and FIG. 5 is a graph showing the relative light intensity distribution when the phase difference is 180 °. In the figure, the position on the semiconductor substrate is 0.2 μm at the position of 0 μm.
The center of the μm line pattern corresponds to -300 to -1
A transparent region having a phase difference of 180 ° corresponds to a region of 00 μm, and a transparent region having a phase difference of 0 ° corresponds to a region of 100 to 300 μm.

As shown in FIG. 3, when the focal position is 0 μm and the phase difference of the transmitted light leaking from the light shielding film is 90 °, it corresponds to a transparent region having a phase difference of 0 ° and a transparent region having a phase difference of 180 °. The distribution of the relative light intensity in the region where the light intensity changes is equal. Therefore, the width of the pattern formed in each region can be made equal.

On the other hand, as shown in FIG.
When the phase difference of the transmitted light leaking from the light-shielding film at m is 0 °, the relative light intensity in a region corresponding to the transparent region having a phase difference of 0 ° is stronger than that at a phase difference of 180 °. Therefore, when this mask is transferred to a negative resist, the line width of the line pattern becomes large in a region corresponding to a transparent region having a phase difference of 0 °. When the focus position is 0 μm and the phase difference of the transmitted light leaking from the light shielding film is 180 °, the relative light intensity distribution is opposite to the case where the phase difference is 0 °, as shown in FIG. For this reason, when this mask is transferred to a negative resist, the line width of the line pattern is reduced in a region corresponding to the transparent region having a phase difference of 0 °.

As a method of reducing the dimensional difference between adjacent patterns other than the above-described method, a method of correcting a mask dimension may be applied. That is, the mask size of the transparent region where the size of the formed line pattern becomes thinner is increased, or the mask size of the transparent region where the size of the line pattern becomes thicker is reduced. More specifically, when the phase of the transmitted light transmitted through the light-shielding region is 90 ° or less, the size of the transparent region having a phase difference of 0 ° is made smaller than the size of the transparent region having a phase difference of 180 °, so that the light transmitted through the light-shielding region is reduced. When the phase of light exceeds 90 °, the size of the transparent region having a phase difference of 0 ° is made larger than the size of the transparent region having a phase difference of 180 °. However, when the correction is performed based on the dimensions of the mask pattern, there is a limitation on the grid size of the mask drawing apparatus. An electron beam drawing apparatus is mainly used for mask drawing. In this electron beam lithography system, the electron beam is focused to a certain size,
A predetermined mask pattern is formed by discriminating whether a point on the grid determined by the beam diameter is exposed or not exposed. For this reason, the dimensions of the mask pattern cannot be changed with a grid smaller than the minimum beam diameter of the exposure apparatus. Usually, since the minimum grid size is 0.01 μm, in the 0.2 μmL / S pattern, the line pattern portion is corrected to 0.01 μm on one side with the minimum correction, and the width is set to 0.18 μm or 0.22 μm. It will be. Therefore, there is no problem if the dimensional difference of the mask is 0.02 μm, but if the dimensional difference is smaller than the grid size, accurate correction cannot be performed. Therefore, in order to eliminate the dimensional difference between adjacent patterns, the phase difference of the transmitted light from
It is desirable to set the angle to 0 °.

However, as described above, the phase difference of the transmitted light is 9
When the focal position is shifted only by setting it to 0 °, as shown in FIG. 3, the distribution of the relative light intensity in the areas corresponding to the transparent area having a phase difference of 0 ° and the transparent area having a phase difference of 180 °, respectively. Will be different. That is, the focus characteristics are different between the transparent region having a phase difference of 0 ° and the transparent region having a phase difference of 180 °. For this reason, the width of the pattern formed in each area differs. Further, the phase difference of the transmitted light from the light-shielding film shown in FIGS.
In the case of 0 °, even if the focal position is shifted, the difference between the peaks of the relative light intensity in the transparent region having a phase difference of 0 ° and the region corresponding to the transparent region having a phase difference of 180 ° does not change much. When the phase difference is 0 °, the difference increases as the focal position shifts.

Next, the phase difference between adjacent transparent regions is set to 18
The effect of intentionally causing a phase error by correcting an appropriate value from 0 ° will be described. In general, the focus characteristics between adjacent patterns that cause a phase error have mutually opposite inclinations, and the inclination is reversed depending on whether the phase error is positive or negative, and the degree of the inclination depends on the magnitude of the phase error. It is known to be proportional. Therefore, by intentionally causing a phase error using this property, it is possible to correct the focus characteristic between adjacent patterns.

Next, this correction method will be described. FIG. 6A is a schematic diagram showing the amplitude of the transmitted light before the phase difference is corrected, and FIG. 6B is a schematic diagram showing the amplitude after the correction. 6A and 6B, θ indicates that the phase difference is 0 °.
, And the phase difference between the transmitted light from the light shielding film and T represents the relative intensity of the transmitted light from the light shielding film. FIG.
As shown in (a), before correcting the phase difference of the transmitted light from the transparent region having a phase difference of 180 °, there is nothing that cancels out the horizontal amplitude of the transmitted light from the light shielding film in the figure. For this reason, when the focal position shifts, a difference occurs in the dimension of the line pattern. On the other hand, after the phase of the transmitted light from the transparent region having a phase difference of 180 ° is corrected by an appropriate value θ1, the amplitude is canceled, so that the influence of this amplitude is eliminated. At this time, the phase difference θ1 is expressed by the following equation (3).

[0036]

Equation ¹ θ1 = sin ⁻¹ (α × √ (T) × sin θ) where α is the area ratio of the light-shielding film to the transparent region for correcting the phase difference. As described above, since the light-shielding film of the phase shift mask has a slight transmittance, if the phase difference of the transmitted light is not an integral multiple of 180 °, the inclination of the focus characteristic is intentionally corrected to correct the inclination. It is necessary to cause a phase error. When the phase difference of the transmitted light is an integral multiple of 180 °, θ1 is 0 ° according to the above equation (3).

In the first embodiment, α is 2 and T is 2
× 10 ^-3 and θ were 91.4 °. When these are substituted into Equation 3, θ1 becomes 5 °. The phase difference of the transmitted light passing through the transparent region 4 was 175 °, which was corrected by 5 ° from the phase difference of 180 °. FIG. 7 shows the focus characteristic at this time, and FIG. 8 shows the focus characteristic when the phase difference remains at 180 °. FIG. 7 is a graph showing the focus characteristic when the phase difference is 175 ° with the focus position taken on the horizontal axis and the line pattern dimension on the vertical axis, and FIG. 8 is also the focus when the phase difference is 180 °. It is a graph which shows a characteristic. As shown in FIG. 7, when the phase difference is corrected by 5 °, the focus characteristics of the transparent region having the phase difference of 0 ° and the focus characteristic of the transparent region having the phase difference of 180 ° are the same. On the other hand, when the phase difference is not shifted, as shown in FIG. 8, the focus characteristics of the two do not match.

In the first embodiment, the g / i of the mercury lamp is used.
Although the phase shift mask is used for ultraviolet light such as line and excimer laser light, the present invention can be applied to a transmission type mask for reduced X-ray exposure with a shorter wavelength.

Next, a phase shift mask according to a second embodiment of the present invention will be described. FIG. 2 is a sectional view showing a phase shift mask according to a second embodiment of the present invention. The phase shift mask according to the present embodiment is a reflective mask for X-ray exposure. In this embodiment, a multilayer coating mirror 13 is formed on a silicon substrate 11. The multilayer coating mirror 13 has a low refractive index layer 13a made of a low refractive index material and a high refractive index layer 13b made of a high refractive index material.
Have a structure in which the X-rays are stacked alternately with a predetermined film thickness. By controlling the phase of the reflected light at the boundary of each layer so that the reflected lights strengthen each other, X-rays are efficiently emitted. It can reflect well. The material of the multilayer coating mirror 13 is selected according to the wavelength of the X-ray. For example, a combination of a silicon layer and a molybdenum layer and a combination of a ruthenium layer and a silicon oxide layer are used.
Absorbing layers 12 having a predetermined width are formed on the multilayer coating mirror 13 at the same interval as the width.
Thereby, the reflection regions 14 and 15 are alternately formed between the absorption layers 12. The material of the absorption layer 12 is, for example, gold or tungsten. Furthermore, multi-layer coated mirror 1
A phase shifter 16 is formed between 3 and the absorption layer 12. Note that the phase shifter 16 is formed in the reflection region 14 but is not formed in the reflection region 15. Further, below the absorption layer 12 adjacent to the reflection region 15, there is a region where the phase shifter 16 is not formed by side etching. The phase shifter 16 has S as a transmission material.
iC or diamond is used.

In this embodiment, when the thickness of the absorbing layer 12 is 0.5 μm or less and the intensity of the X-ray transmitted through the reflecting region is 100%, the intensity of the X-ray transmitted through the absorbing layer (reflecting Rate) is 2%. Further, the phase difference of the reflected light is 90 °. The phase difference between the reflection areas 14 and 15 caused by the phase shifter 13 is 164 ° and 180 °.
Is corrected by 16 °.

In an X-ray exposure mask, even if a heavy metal such as gold is used as the absorbing material of the absorbing layer, the absorbing layer cannot completely absorb X-rays, and the absorbing region on which the absorbing layer is formed is formed. However, X-ray reflection occurs. This reflection can be prevented by increasing the thickness of the absorbing layer, but in this case, the processing of the absorbing layer becomes extremely difficult, and is not practical. For this reason, the thickness of the absorption layer is usually set so that the reflectance of the image pattern can be ignored so that the influence on the contrast is negligible.

In this embodiment, the reflectivity of the absorption region where the absorption layer 12 is formed is 0.02, the phase difference of the reflected light is 90 °, and the phase difference between the reflection regions 14 and 15 is 164 °.
Therefore, Equation 3 is satisfied. For this reason, the focus characteristics of the reflected light reflected from the reflective region 14 and the reflected light reflected from the reflective region 15 match, and the line width does not change even if the focal position of the X-rays is not shifted or shifts. A uniform pattern can be formed. Further, since the thickness of the absorbing layer is 0.5 μm or less, the processing is not difficult.

In this embodiment, similarly to the first embodiment, the size of the mask pattern is adjusted in accordance with the phase of the transmitted light passing through the absorption region, so that the pattern formed by each reflection region is adjusted. Can be made equal.

[0044]

As described in detail above, according to the present invention,
In a transmission type phase shift mask, even when a small amount of transmitted light is generated in a light-shielded region, a phase error is additionally generated so as to cancel the effect, so that a dimensional difference between adjacent patterns can be eliminated. it can. Even in the reflection type phase shift mask, a pattern having a uniform line width can be obtained even when a slight transmitted light is generated in the absorption region.

[Brief description of the drawings]

FIG. 1 is a sectional view showing a phase shift mask according to a first embodiment of the present invention.

FIG. 2 is a sectional view showing a phase shift mask according to a second embodiment of the present invention.

FIG. 3 shows the position on the semiconductor substrate on the horizontal axis and the relative light intensity on the vertical axis, and the phase difference of the transmitted light leaking from the light shielding film is 9
It is a graph which shows the relative light intensity distribution at the time of 0 degree.

FIG. 4 is a graph showing a relative light intensity distribution when the phase difference is 0 °.

FIG. 5 is a graph showing a relative light intensity distribution when the phase difference is 180 °.

FIG. 6A is a schematic diagram showing the amplitude of transmitted light before correcting a phase difference, and FIG. 6B is a schematic diagram showing the amplitude after correction.

FIG. 7 is a graph showing a focus characteristic in a case where a horizontal axis indicates a focal position and a vertical axis indicates a line pattern dimension and a phase difference is 175 °.

FIG. 8 is a graph showing focus characteristics when the phase difference is 180 °.

9A is a plan view showing a first Shibuya-Levenson type phase shift mask, and FIG. 9B is a plan view of AA in FIG. 9A.
FIG. 4C is a cross-sectional view taken along a line, and FIG. 4C is a graph showing the amplitude distribution of transmitted light, with the horizontal axis representing the position of the phase shift mask and the vertical axis representing the amplitude of transmitted light.

FIG. 10A is a cross-sectional view showing a second Shibuya-Levenson type phase shift mask, and FIG.
It is sectional drawing which shows a Levenson-type phase shift mask.

[Explanation of symbols]

 1, 21, 31a, 31b; transparent substrate 2, 23, 33a, 33b; light-shielding film 4, 5, 24, 25; transparent region 11; silicon substrate 12; absorption layer 13; multi-layer coating mirror 13a; High refractive index layers 14, 15; reflection regions 16, 26, 36; phase shifters 22, 32; etching stopper

Claims (8)

[Claims] A first transparent region, a second transparent region,
A mask pattern having a light-shielding region provided between the first transparent region and the second transparent region, wherein the transmitted light passing through the first transparent region and the second transparent region A phase shift mask having a phase difference between transmitted light and a transmitted light, wherein the phase difference is less than 180 °. 2. An area ratio between the light-shielding region and the first transparent region is α, and an intensity of light transmitted through the transparent region is 1
Assuming that the intensity of the transmitted light transmitted through the light-shielding region when T is defined as T and the difference between the second phase and the third phase of the transmitted light transmitted through the light-shielded region is θ, the phase difference is ( 18
2. The phase shift mask according to claim 1, wherein 0-sin ⁻¹ (α × √ (T) × sin θ)) °. 3. A first transparent area, a second transparent area,
A mask pattern having a light-shielding region provided between the first transparent region and the second transparent region, wherein the transmitted light passing through the first transparent region and the second transparent region In a phase shift mask having a phase difference between transmitted light and transmitted light, the phase of transmitted light transmitted through the first transmission region is 0 °, and the phase of transmitted light transmitted through the second transmission region is 0 °. The phase shift mask is 180 °, the phase of the transmitted light passing through the light-shielding region is 90 ° or less, and the width of the first transparent region is smaller than the width of the second transparent region. 4. A first transparent area, a second transparent area,
A mask pattern having a light-shielding region provided between the first transparent region and the second transparent region, wherein the transmitted light passing through the first transparent region and the second transparent region In a phase shift mask having a phase difference between transmitted light and transmitted light, the phase of transmitted light transmitted through the first transmission region is 0 °, and the phase of transmitted light transmitted through the second transmission region is 0 °. 180 °, wherein the phase of transmitted light transmitted through the light-shielding region exceeds 90 °, and the width of the first transparent region is larger than the width of the second transparent region. . 5. A first reflection area, a second reflection area,
A mask pattern having an absorption area provided between the first reflection area and the second reflection area, wherein the reflected light that reflects the first reflection area and the second reflection area A phase shift mask having a phase difference between reflected light and reflected light, wherein the phase difference is less than 180 °. 6. An area ratio between the absorption region and the first reflection region is α, and an intensity of light reflected by the reflection region is 1
Assuming that the intensity of the reflected light that reflects the absorption region when T is defined as T and the difference between the second phase and the third phase of the reflected light that reflects the absorption region is θ, the phase difference is ( 18
The phase shift mask according to claim 5, wherein 0-sin ⁻¹ (α × √ (T) × sin θ)) °. 7. A first reflection area, a second reflection area,
A mask pattern having an absorption area provided between the first reflection area and the second reflection area, wherein the reflected light that reflects the first reflection area and the second reflection area In the phase shift mask having a phase difference between the reflected light and the reflected light, the phase of the reflected light that reflects the first reflection area is 0 °, and the phase of the reflected light that reflects the second reflection area is A phase shift mask, wherein the phase of the reflected light is 180 ° or less, and the phase of the reflected light that reflects the absorption region is 90 ° or less, and the width of the first reflection region is smaller than the width of the second reflection region. 8. A first reflection area, a second reflection area,
A mask pattern having an absorption area provided between the first reflection area and the second reflection area, wherein the reflected light that reflects the first reflection area and the second reflection area In the phase shift mask having a phase difference between the reflected light and the reflected light, the phase of the reflected light that reflects the first reflection area is 0 °, and the phase of the reflected light that reflects the second reflection area is 180 °, wherein the phase of the reflected light reflecting the absorption region exceeds 90 °, and the width of the first reflection region is larger than the width of the second reflection region. .