JP3418187B2 - Phase shift mask, method for manufacturing the phase shift mask, and exposure method using the phase shift mask - Google Patents

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, a method of manufacturing the phase shift mask, and an exposure method using the phase shift mask.

[0002]

2. Description of the Related Art In recent years, there has been a remarkable increase in the degree of integration and miniaturization of semiconductor integrated circuits. with this,
The miniaturization of circuit patterns formed on a semiconductor substrate (hereinafter simply referred to as a wafer) has been rapidly advanced.

Above all, the photolithography technique is widely recognized as a basic technique in pattern formation. Therefore, various developments and improvements have been made to date. However, the miniaturization of patterns has never stopped, and the demand for improving the resolution of patterns has become stronger.

The photolithography technique is a technique in which a mask (original image) pattern is transferred to a photoresist applied on a wafer, and a lower layer to-be-etched film is patterned by using the transferred photoresist. . When this photoresist is transferred, the photoresist is developed, but the type that removes the photoresist in the light-exposed areas is positive type, and the photoresist in the areas not exposed to light is removed. This type is called a negative photoresist. Hereinafter, a conventional exposure method in the photolithography technique will be described.

FIG. 45 is a schematic diagram of an optical system for explaining a conventional exposure method. Referring to FIG. 45, this optical system reduces the pattern on the mask and projects it onto the photoresist on the wafer. The optical system has an illumination optical system from the light source to the photomask pattern and a projection optical system from the photomask pattern to the wafer.

The illumination optical system includes a mercury lamp 11 as a light source.
1, a reflecting mirror 112, a condenser lens 118, a fly-eye lens 113, a diaphragm 114b, and a condenser lens 116.
a, 116b, 116c, the blind diaphragm 115,
And a reflecting mirror 117. The projection optical system has telephoto lenses 119a and 119b and a pupil plane diaphragm 125.

In the exposure operation, the light 111a emitted from the mercury lamp 111 is first reflected by the reflecting mirror 112, for example, only the g-line (wavelength: 436 nm) and becomes light of a single wavelength. Next, the light 111a is transmitted to each fly-eye constituent lens 113a of the fly-eye lens 113.
Respectively, and then passes through the diaphragm 114b.

Here, the light 111b indicates an optical path created by one fly-eye forming lens 113a,
The light 111c indicates the optical path created by the fly-eye lens 113.

The light 111a passing through the diaphragm 114 passes through the condenser lens 116a, the blind diaphragm 115 and the condenser lens 1116b, and is reflected by the reflecting mirror 117 at a predetermined angle.

Light 111a reflected by the reflecting mirror 117
After passing through the condenser lens 116c, uniformly illuminates the entire surface of the photomask 720 on which a predetermined pattern is formed. After this, the light 111a is projected onto the projection lenses 119a and 119.
The photoresist 121a on the semiconductor wafer 121 is exposed by being reduced to a predetermined magnification by b.

In general, the resolution limit R (nm) in the photolithography technique using the reduction exposure method is expressed as R = k ₁ · λ / (NA). Where λ is the wavelength of the light used (nm),
NA is the numerical aperture of the lens, and k ₁ is a constant that depends on the resist process.

As can be seen from the above equation, in order to improve the resolution limit R, that is, in order to obtain a fine pattern, there is a method of decreasing the values of k ₁ and λ and increasing the value of NA. Conceivable. That is, it is only necessary to reduce the constant depending on the resist process, and to shorten the wavelength and increase the NA.

However, it is technically difficult to improve the light source and the lens, and the focal depth δ of the light (δ = k ₂ · λ / (NA) ² ) becomes shallow as the wavelength is shortened and the NA is increased. On the contrary, there is a problem that the resolution is lowered.

Therefore, studies have been conducted to improve the pattern size by improving the photomask instead of the light source and the lens. Recently, a phase shift mask has attracted attention as a photomask for improving pattern resolution. The structure and principle of this phase shift mask will be described below in comparison with an ordinary photomask.
Regarding the phase shift mask, the Levenson method and the halftone method will be described.

FIGS. 46 (a), (b) and (c) are views showing the cross section of the mask, the electric field on the mask and the light intensity on the wafer when a normal photomask is used. Figure 46
Referring to (a), an ordinary photomask has a structure in which a metal mask pattern 503 is formed on a glass substrate 501. In such a normal photomask, the electric field on the mask is an electric field spatially pulse-modulated by the metal mask pattern 503 as shown in FIG.

However, as shown in FIG. 46C, when the pattern is miniaturized, the exposure light transmitted through the photomask is exposed to a non-exposure region on the wafer (the exposure light by the metal mask pattern 503) due to the diffraction effect of the light. Area where transmission of is blocked). Therefore, the non-exposed region on the wafer is also irradiated with the light, and the contrast of the light (the difference in the light intensity between the exposed region and the non-exposed region on the wafer) is reduced. As a result, the resolution is lowered and it becomes difficult to transfer a fine pattern.

47 (a), (b) and (c) are diagrams showing a mask cross section, an electric field on the mask and a light intensity on the wafer when a Levenson type phase shift mask is used. First, with reference to FIG. 47A, in the phase shift mask, an optical member 505 called a phase shifter is provided in a normal photomask.

That is, a chrome mask pattern 503 is formed on a glass substrate 501, an exposure region and a light-shielding region are provided, and the phase shifter 50 is provided every other exposure region.
5 are provided. The phase shifter 505 serves to convert the phase of transmitted light by 180 degrees.

Referring to FIG. 47 (b), since the phase shifters 505 are provided every other exposure area as described above, the electric field on the mask due to the light transmitted through the phase shift mask is
The phases are alternately inverted by 180 degrees. In this way, the phases of the light beams are opposite to each other between the adjacent exposure regions, so that the light beams cancel each other out in the overlapping portions of the light beams having the opposite phases due to the light interference effect.

As a result, as shown in FIG. 47 (c), the light intensity becomes small at the boundary between the exposure regions, and a sufficient light intensity difference between the exposure region and the non-exposure region on the wafer is ensured. You can As a result, the resolution can be improved and a fine pattern can be transferred.

FIGS. 48 (a), (b) and (c) are diagrams showing a mask cross section, an electric field on the mask and a light intensity on the wafer when a halftone type phase shift mask is used. First, with reference to FIG. 48A, also in this halftone type phase shift mask, an optical member 506 called a phase shifter is provided as in the Levenson method described above.

However, the optical member 506 is the glass substrate 5
01 is formed only on the semitransparent film 503, and a two-layer structure of a phase shifter 506 and a semitransparent film 503 is provided. The phase shifter 506 plays a role of converting the phase of the transmitted light by 180 degrees similarly to the above, and the semitransparent film 503 plays a role of attenuating the intensity of the exposure light without completely blocking the exposure light. It is a thing.

Referring to FIG. 48 (b), since the two-layer structure of the phase shifter 506 and the semitransparent film 503 is provided as described above, the electric field on the mask is formed by alternately converting the phases by 180 degrees. And the intensity of one phase is less than the intensity of the other phase. That is, the phase is converted by 180 degrees by being transmitted through the phase shifter 506, and the semi-transmissive film 5 is
The light intensity of the light is attenuated so that the photoresist has a predetermined film thickness after the development because the light passes through 03. Since the phases of the light in the adjacent exposure areas are opposite to each other, the lights cancel each other out in the portion where the lights having the opposite phases overlap each other.

As a result, as shown in FIG. 48C, since the phase is inverted at the edge of the exposure pattern, the light intensity at the edge of the exposure pattern can be reduced.
As a result, the difference in the light intensity of the exposure light between the region that transmits the semi-transparent film 503 and the region that does not transmit becomes large, and the resolution of the pattern image can be increased.

As described above, there are various types of phase shift masks such as the Levenson method and the halftone method. Among these, the so-called Levenson type phase shift mask invented by Marc Levenson has excellent resolution in principle, and is considered to be the best type in terms of resolution.

However, although various techniques have been invented and proposed for the manufacturing method, none have been put to practical use. Among these suggestions Marc D. Levenson et a
l., "Phase-Shifting Mask Strategies: Isolated Dark
Lines "MICROLITHOGRAPHY WORLD, pp.6-12, March / Apri
l The manufacturing technology of Marc Levenson, published in 1992., is considered to be an excellent prior art.

Therefore, first, the Levenson type phase shift mask manufactured by such a technique will be described as a conventional first phase shift mask, and its structure and manufacturing method will be described below.

FIG. 49 is a sectional view schematically showing the structure of a conventional first phase shift mask. Referring to FIG. 49, a conventional first phase shift mask 720 has a quartz substrate 701 and a light shielding film 703. Quartz substrate 70
A groove is formed at a predetermined depth on the main surface of No. 1. The region in which the groove is not formed forms the first light transmitting portion 701a, and the region in which the groove is formed forms the second light transmitting portion 701b. A light shielding film 703 is formed on the quartz substrate 701 so as to cover the side wall of the groove and expose predetermined regions of the first and second light transmitting portions 701a and 701b. The light-shielding film 703 has a transmittance of 1% or less, and the film thickness thereof is 100 when chromium (Cr) is used as the material.
It is about 0Å.

The first light transmitting portion 701a and the second light transmitting portion 701b are constructed so that the phases of the exposure light transmitted therethrough are different by 180 degrees. In this way, the phase of the exposure light transmitted through the light transmitting portions adjacent to each other is converted by 180 degrees, so that the resolution can be improved as described above.

Further, the bottom wall and the side wall of the groove form a substantially right angle. Next, a conventional method of manufacturing the first phase shift mask will be described.

50 to 57 are schematic cross-sectional views showing a method of manufacturing a conventional first phase shift mask in the order of steps.
First, with reference to FIG. 50, the surface 701a of the quartz substrate 701.
A chrome film 705a is formed on top. This chrome film 70
A resist film 707a is applied on 5a. The resist film 707a is exposed and developed.

Referring to FIG. 51, a resist pattern 707 having a desired shape is formed by the above exposure and development. Anisotropic etching is performed using this resist pattern 707 as a mask. After that, the resist pattern 707 is removed.

Referring to FIG. 52, the chromium film pattern 705 to which the shifter pattern is transferred is thus formed.

Referring to FIG. 53, quartz substrate 701 is anisotropically etched using chromium film pattern 705 as a mask. As a result, the surface 701 of the quartz substrate 701 is
A groove is formed in a and the shifter pattern is transferred. Then, the chromium film pattern 705 is removed.

Referring to FIG. 54, in this way, the first light transmitting portion 701a and the second light transmitting portion 701b are formed on the quartz substrate 701.

Referring to FIG. 55, a chromium film 703a is formed on the entire surface on which the first and second light transmitting portions 701a and 701b are formed. A resist film 709a is applied on this chrome film 703a. This resist film 709
a is exposed and developed.

Referring to FIG. 56, a resist pattern 709 having a desired shape is formed by this exposure and development. By performing anisotropic etching using the resist pattern 709 as a mask, the light shielding film 70 that exposes desired regions of the first and second light transmitting portions 701a and 701b.
3 is formed. Thereafter, resist pattern 709 is removed and conventional phase shift mask 720 shown in FIG.
Is formed.

In the above-described conventional method of manufacturing a phase shift mask, the resist films 707a and 709a are not directly coated on the quartz substrate 701. Therefore, a method of manufacturing a phase shift mask in which a resist film is directly coated on a substrate (Japanese Patent Laid-Open No. 4-355758 and Japanese Patent Laid-Open No. 21-21).
1450), it has an advantage that the following defects can be easily corrected.

As an example of a method of directly applying a resist film on a substrate, a method of manufacturing a phase shift mask disclosed in Japanese Patent Laid-Open No. 4-355758 will be described.

58 to 61 are schematic sectional views showing a method of manufacturing the phase shift mask disclosed in the above publication in the order of steps. First, referring to FIG. 58, after the phase change film 803 is formed on the surface of the quartz layer 801, the light blocking film 805 is formed on the surface of the phase change film 803.

Referring to FIG. 59, after a resist pattern 807 is formed on light blocking film 805, light blocking film 805 is patterned by etching using resist pattern 807 as a mask. After that, the resist pattern 807 is removed.

Referring to FIG. 60, photoresist 809 is directly coated on the surfaces of patterned light blocking film 805 and phase change film 803. This photoresist 8
After the pattern 09 is patterned into a desired shape, the phase change film 803 is etched using the resist pattern 809 as a mask.

Referring to FIG. 61, a groove 811 is formed in phase change film 803 by this etching. After that, the resist pattern 809 is removed and a phase shift mask is formed.

In the method of manufacturing the phase shift mask disclosed in the above publication, the shifter pattern is formed on the phase change film 803 after the light blocking film 805 is patterned.
Therefore, in the process shown in FIG. 60, the phase change film (substrate) is
The resist film 809 is directly coated on the surface of 803.

FIG. 62 shows a general resist film coating method.
Pinholes 8 that penetrate the resist film 809 as shown in FIG.
09a will occur. When etching is performed using the resist pattern 809 as a mask with the pinhole 809a formed, the state shown in FIG. 63 is obtained.

Referring to FIG. 63, the etchant enters the pinhole 809a, and even the phase change film 803 at the bottom of the pinhole 809a is removed by etching, resulting in a defect. When a phase shift mask having such a defect is used for exposing a wafer, even in a region other than a desired region,
The phase of the exposure light is changed at this defective portion. Therefore, the resist film applied on the wafer cannot be exposed in a desired shape.

When the resist film is directly coated on the substrate as described above, when the resist pattern is used as a mask and etching or the like is performed, a shifter defect is directly introduced into the substrate.

On the other hand, in the manufacturing method proposed by Marc Levenson, the resist film is not directly formed on the quartz substrate as described above. Therefore, even if the lower layer is etched using the resist film having the pinhole as a mask, the shifter defect is not directly formed on the substrate.

Specifically, referring to FIG. 64, even if a pinhole 707a is formed in resist pattern 707, chromium film 705 is formed below resist pattern 707.
Exists. Therefore, even if the lower layer is etched using the resist pattern 707 as a mask, the pinhole 70
The chromium film 705 at the bottom of 7a is only removed by etching.

That is, the pinhole defect (white defect) 705a generated at the bottom of the pinhole 707a is the quartz substrate 70.
It occurs not in 1, but in the chromium film 705 thereabove. Therefore, the pinhole defect 705a is, as shown in FIG.
Carbon film 7 by FIB (Focussed Ion Beam) method
It can be easily corrected by embedding with the deposition of 05c.

Further, referring to FIG. 64, a residual defect (black defect) in which a chromium film remains where it should be removed by etching.
705b may occur. However, this remaining defect 7
Since 05b is not a defect directly formed on the quartz substrate 701, it can be easily repaired by blowing (melting) and removing it by laser irradiation using a YAG laser as shown in FIG.

As described above, the conventional manufacturing method proposed by Marc Levenson has the advantage that the defect of the shifter is not directly formed on the substrate, so that the defect can be easily corrected.

Next, the structure of the halftone type phase shift mask as the second conventional phase shift mask will be described below.

FIG. 66 is a sectional view schematically showing the structure of a conventional second phase shift mask. 66, the second conventional phase shift mask 920 has a quartz substrate 901 and a semitransparent film 903. Quartz substrate 9
On the main surface of No. 01, a groove is formed with a predetermined depth.

The region where the groove is formed constitutes the first light transmitting portion 901a, and the region where the groove is not formed is the second region.
Of the light transmission part 901b. A semi-transparent film 903 is formed on the quartz substrate 901 so as to cover the side wall of the groove and expose a predetermined region of the first light transmitting portion 901a. The semitransparent film 903 plays a role of attenuating the intensity of the exposure light transmitted through the semitransparent film 903 to such an extent that the exposure light does not expose the photoresist on the wafer to light or leaves a predetermined film thickness after development. I am doing it.

The first light transmitting portion 901a and the second light transmitting portion 901b are constructed so that the phases of the exposure light transmitted therethrough are different by 180 degrees. In this way, the phase of the exposure light transmitted through the light transmitting portions adjacent to each other is converted by 180 degrees, so that the resolution can be improved as described above.

Further, the bottom wall and the side wall of the groove form a substantially right angle.

[0058]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
As described above, the phase shift mask of (1) can obtain a higher resolution than the ordinary photomask. However, the conventional first and second phase shift masks also have a problem that it is difficult to obtain a desired pattern shape due to complication of the [I] circuit pattern and generation of defects during formation of the [II] shift mask. It was Below, regarding the problem,
The details will be described.

[I] Complicated circuit pattern In recent years, in order to obtain a large-capacity and multifunctional semiconductor integrated circuit, the circuit pattern of the semiconductor integrated circuit has been miniaturized and complicated. For example, in a DRAM (Dynamic Random Access Memory), a periodic pattern is densely formed in its memory cell area (hereinafter, simply referred to as a dense pattern), and a circuit pattern having individual functions is isolated in its peripheral circuit area. (Hereinafter, simply referred to as an isolated pattern).

When a circuit pattern in which a dense pattern and an isolated pattern are mixed as described above is to be formed by a Levenson type phase shift mask, a high resolution can be obtained in the dense pattern, but not so high in the isolated pattern. . On the other hand, if an attempt is made to form a halftone phase shift mask, a high resolution can be obtained with an isolated pattern, but not so high with a dense pattern.

FIG. 67 is a diagram for explaining that an isolated pattern cannot be formed with high resolution by a Levenson-type phase shift mask. FIG. 67A shows a sectional view of a Levenson-type phase shift mask, and FIG. ) Indicates the light intensity on the wafer when exposed using the phase shift mask of (a).

Referring to FIG. 67, the isolated pattern exists apart from other circuit patterns. Therefore, the exposure area 701b forming the isolated pattern (hereinafter, referred to as an isolated exposure area) is arranged considerably away from the exposure areas forming the other circuit patterns. Therefore, there is no portion where the exposure light that passes through the isolated exposure region 701b and is irradiated onto the wafer overlaps the exposure light of the opposite phase that passes through the other exposure regions. Therefore, it is not possible to obtain the effect of the phase shift mask in which high-resolution is obtained by canceling lights having opposite phases from each other due to the interference effect of light.

On the other hand, in the dense pattern, FIG.
As shown in FIG. 5, the light beams of opposite phases overlap each other between the adjacent exposure regions. For this reason, the lights cancel each other out at the overlapping portions, so that the resolution can be improved.

Next, FIG. 68 is a diagram for explaining that a dense pattern cannot be formed with a high resolution by a halftone phase shift mask, and FIG. 68A is a sectional view of the halftone phase shift mask. And (b) shows the light intensity on the wafer when exposed using the phase shift mask of (a).

Referring to FIG. 68, since the circuit patterns are dense in the dense pattern, a plurality of exposure areas 901 are formed.
a and 901a are arranged close to each other. For this reason, there occurs a portion P _{3 in} which lights of the same phase overlap each other between the adjacent exposure regions 901a. When the lights of the same phase overlap each other, the intensity of the lights of the opposite phase transmitted through the exposure region 901b is small, and therefore the edge of the exposure pattern (P ₃
The situation occurs that the light intensity does not decrease in some areas. In this case, the difference in light intensity between the exposed area and the non-exposed area becomes insufficient, so that the resolution cannot be improved.

On the other hand, in the isolated pattern, as shown in FIG.
As shown in, there is no overlap of in-phase light.
Therefore, at the edge of the exposure pattern, due to the overlap with the light of the opposite phase, the difference in light intensity is sufficient,
The resolution can be improved.

As described above, in the pattern in which the dense pattern and the isolated pattern are mixed, even if the conventional first or second phase shift mask is used, one of the patterns cannot be formed with high resolution. Therefore, it becomes difficult to obtain a desired pattern shape with a complicated circuit pattern having a dense pattern and an isolated pattern.

[II] Occurrence of Defects During Shift Mask Formation In the conventional manufacturing method proposed by Marc Levenson, the quartz substrate 701 is anisotropically etched to form a shifter pattern. Therefore, in this manufacturing method, it is necessary to obtain a predetermined pattern shape because the adhesion of the light-shielding film to the quartz substrate is poor, the foreign matter is likely to remain on the quartz substrate, and the harmful effects due to residual defects are likely to occur. There was a problem that it was difficult.

Adhesiveness of Light-Shielding Film to Quartz Substrate In the conventional method of manufacturing a phase shift mask, as shown in FIGS.
In process 3, the quartz substrate 701 is anisotropically etched. Therefore, the side wall and the bottom wall of the groove formed by this etching form a substantially right angle. Next, in the process shown in FIG. 55, a chrome film 703a is formed so as to cover this groove as well. However, it is difficult to normally form a film on the side wall of the groove having the above-described shape. In particular, the chromium film 703a is formed by a method such as a sputtering method that has a poor step coverage.
When forming a film, it is more difficult to form a normal film.

Therefore, as shown in FIG. 69, the chromium film 703a is formed on the side wall of the groove (the area indicated by D ₃ in the drawing).
Adhesion to the quartz substrate 701 deteriorates. If the adhesion of the chrome film 703a is not good as described above, the chrome film 703a is easily peeled off in the cleaning step in the manufacturing process of the phase shift mask. Further, the light shielding film 703 is easily peeled off even by cleaning after the completion of the phase shift mask.

The portion where the light-shielding film 703 is peeled off in this way becomes a so-called white defect. When a wafer is exposed using the phase shift mask in which the white defect has occurred, even an area on the wafer that should not be exposed is exposed,
It becomes impossible to pattern into a desired shape.

Residue of Foreign Particles on Quartz Substrate As described above, in the conventional method of manufacturing a phase shift mask, the quartz substrate 701 is anisotropically etched by the process of FIGS. As shown, the bottom wall and side wall of the groove formed by this etching form a substantially right angle. Thus, since the step edge (the area indicated by D ₄ in the drawing) has a substantially right angle, the foreign matter 750a is easily trapped (captured) at the step edge.

Specifically, after the groove is formed on the surface of the quartz substrate 701 by the process shown in FIGS. 53 and 54, the chromium film pattern 705 is removed. At the time of this removal, the foreign matter is trapped at the edge of the step. Further, foreign matter generated from the inside of the etching apparatus and foreign matter in the etching solution are also trapped at the step edge portion.

When the light-shielding film is formed in a state where the foreign matter is trapped, the etchant passes through the foreign matter (or, when the light-shielding film 703 shown in FIGS. 71 and 72 is patterned).
After the foreign matter is melted at high speed, it goes under the resist 709. As a result, the light-shielding film 703 is excessively removed by etching as shown in FIG. The excessively removed portion becomes a so-called white defect. This white defect exposes a region on the wafer that should not be exposed, so that the pattern cannot be patterned into a desired shape.

Further, the light shielding film 70 shown by the arrow k in FIG.
The portion 3 does not contact the side wall of the step, and is in a state of protruding in the hollow. Therefore, the portion k of the light-shielding film 703 is easily peeled off by cleaning after removing the resist, cleaning after repairing the defects of the light-shielding film 703, and a white defect is generated as described above.

Further, as shown in FIG. 74, the surface of the light shielding film 703 formed on the step edge is also reflected by the step shape of the base. Therefore, the foreign matter 750b is easily trapped in the portion along the step edge of the light-shielding film 703 (the area indicated by D ₅ in the drawing) as in the above case.

When the trapped foreign matter 750b has a large shape, the foreign matter 750b may remain outside the light-shielding portion. This foreign object 7
If 50b is a material that does not transmit the exposure light, it will be a so-called black defect. Even if the foreign matter 750b allows the exposure light to pass therethrough, when the material that changes the phase of the transmitted light by a considerable amount (usually 10 to 20 degrees) or more, the light shielding mask 703 is a normal phase shift mask. Will prevent you from working as. For this reason, the shape transferred to the resist on the wafer is distorted, resulting in defects.

When foreign matter is trapped in the stepped portion of the quartz substrate 701 in the phase shift mask process as described above, it becomes difficult to expose the resist on the wafer into a desired shape.

Harmful Defects Due to Residual Defects In patterning the chromium film 705a shown in FIGS. 50 and 51, residual defects 705b may occur as shown in FIG. This remaining defect 705b can be repaired by a method such as the laser blow described above. But,
Not all the remaining defects 705b that have occurred can be detected and corrected. In addition, in order to simplify the manufacturing process, it may be desirable to omit the correction process. In such a case, a state in which the remaining defects 705b still remain occurs.

In the conventional method of manufacturing a phase shift mask,
Grooves are formed on the surface of the quartz substrate 701 by anisotropic etching in the process shown in FIGS. However, if anisotropic etching is performed in the state where the residual defects 705b are generated, a region that is not to be etched even though it is a region to be etched as shown in FIG. 76 (region shown by D ₆ in the drawing). Occurs.

Referring to FIG. 77, in the phase shift mask thus formed, the phase of the exposure light transmitted between the adjacent light transmitting portions is not converted by 180 degrees. Specifically, the phases of the exposure light transmitted through the light transmitting portion 701a and the light transmitting portion 701ab are the same. Therefore, both transmission parts 701a,
The exposure lights transmitted through the 701ab are mutually strengthened in the overlapping portion of the exposure lights. As a result, the difference in light intensity between the exposed region and the light-shielded region on the wafer is reduced, the resolution is reduced, and it becomes difficult to form a desired pattern shape.

Further, as shown in FIG. 78, when the non-etched region 701ab due to the residual defect partially occurs in the light transmitting portion, the shape of the transfer pattern is destroyed.

That is, the region 701a which is not etched
b and the region 701bb to be removed by etching have opposite phases of the transmitted exposure light. Therefore, at the boundary portion P between the region 701ab that is not removed by etching and the region 701bb that is removed by etching, the exposure lights cancel each other, so that a portion having a light intensity of 0 is generated.

The resist on the wafer is not exposed in the portion where the light intensity is 0. Therefore, a region which is to be exposed but is not exposed occurs in the resist, and a resist pattern having a desired shape cannot be obtained. When the lower layer is patterned using such a resist pattern,
The patterning shape of the lower layer becomes defective.

In particular, when a negative resist in which a portion not exposed to light is removed by a developing solution is used, the light intensity is 0.
The resist does not remain in the portion (region P). That is,
Although there is an area where the resist should be left, there is an area where the resist is not left. When such a resist pattern is used to form, for example, a wiring layer by patterning, the wiring is broken in a region P where the light intensity is 0 as shown in FIG.

In this way, when the remaining defect occurs,
Not only the resolution is lowered, but also the defective pattern shape occurs.

The present invention has been made in view of the above items [I] and [II], and an object of the present invention is to provide a phase shift mask which can easily form a desired pattern shape, A method of manufacturing a phase shift mask and an exposure method using the phase shift mask.

[0088]

The phase shift mask of the present invention comprises a substrate and a semi-light-shielding film. The substrate has a first light transmitting portion that transmits the exposure light and a second light transmitting portion. The second light transmission portion is adjacent to the first light transmission portion and transmits the exposure light at a phase different from the phase of the exposure light transmitted through the first light transmission portion. The semi-light-shielding film is located at the boundary between the first and second light transmitting portions adjacent to each other and is formed in a partial region of the first and second light transmitting portions. First
The light transmission part of has a first transmission region and a first attenuated transmission region on which a semi-light-shielding film is formed. The intensity of the exposure light transmitted through the first transmission region is higher than the intensity of the exposure light transmitted through the first attenuated transmission region. The second light transmitting portion is the second
And a second attenuating and transmitting region having a semi-light-shielding film formed thereon. The intensity of the exposure light transmitted through the second transmission region is higher than the intensity of the exposure light transmitted through the second attenuated transmission region. The transmissivity of this semi-light-shielding film is 3% or more and 30% or less.

In the phase shift mask according to a preferred aspect of the present invention, the substrate has a step having a predetermined height formed by the surface of the first light transmitting portion and the surface of the second light transmitting portion. There is. The semi-light-shielding film covers the steps of this substrate. The side wall of the step of the substrate has a height substantially to concave shape such curvature of the same radius of the step.

In the phase shift mask according to another preferred aspect of the present invention, the substrate is made of a first film and a second film made of a material having an etching property different from that of the first film formed on the first film. A film, and the surface of the first light transmitting portion is the first
And the surface of the second light transmitting portion is the surface of the second film.

In the phase shift mask according to still another preferred aspect of the present invention, in the semi-shielding film, the difference between the phase of the exposure light before the transmission of the semi-shielding film and the phase of the exposure light after the transmission is 3
It is formed to be 60 ° × n (n is 0 or a positive integer).

The exposure method using the phase shift mask of the present invention includes the following steps. First, exposure light is emitted from the light source. Then, the exposure light is applied to the phase shift mask. Then, the exposure light transmitted through the phase shift mask is projected onto the photoresist on the film to be etched, and the desired region of the photoresist is exposed. This phase shift mask includes a substrate and a semi-shield film. The substrate has a first light transmitting portion that transmits the exposure light and a second light transmitting portion. The second light transmitting portion is adjacent to the first light transmitting portion and transmits the exposure light at a phase different from the phase of the exposure light passing through the first light transmitting portion. The semi-light-shielding film is located at the boundary between the first and second light transmitting portions adjacent to each other and is formed in a partial area of the first and second light transmitting portions. First
The light transmission part of has a first transmission region and a first attenuated transmission region on which a semi-light-shielding film is formed. The intensity of the exposure light transmitted through the first transmission region is higher than the intensity of the exposure light transmitted through the first attenuated transmission region. The second light transmitting portion is the second
And a second attenuating and transmitting region having a semi-light-shielding film formed thereon. The intensity of the exposure light transmitted through the second transmission region is higher than the intensity of the exposure light transmitted through the second attenuated transmission region. The transmittance of the semi-light-shielding film is 3% or more and 30% or less.

The method of manufacturing the phase shift mask of the present invention includes the following steps. First, a first light transmitting portion that transmits the exposure light and a second light transmitting portion that is adjacent to the first light transmitting portion and that transmits the exposure light at a phase different from the phase of the exposure light that transmits the first light transmitting portion. And a light transmitting portion of the substrate are formed.
Then, a semi-light-shielding film that is located at the boundary between the first and second light transmitting portions adjacent to each other and extends in a partial region of the first and second light transmitting portions is formed. As a result, the first light transmission portion has the first transmission region and the first attenuated transmission region where the semi-shield film is formed, and the intensity of the exposure light transmitted through the first transmission region is the first. The intensity of the exposure light transmitted through the first attenuated transmission region is larger than that of the second transmission region,
A second attenuated transmission region having a semi-shielding film formed thereon, and the intensity of the exposure light transmitted through the second transmission region is higher than the intensity of the exposure light transmitted through the second attenuation transmission region. This semi-light-shielding film is formed so that its transmittance is 3% or more and 30% or less.

[0094]

In the phase shift mask of the present invention, when the distance between the first and second transmissive regions separated by the semi-light-shielding film is small and dense (that is, a dense pattern), the first and second The exposure lights transmitted through the two transmission regions overlap each other at the edge portion of the exposure pattern. Since the overlapping exposure lights have different phases from each other, that is, the phases are opposite to each other, the exposure lights cancel each other at the overlapping portions. For this reason, a portion having a light intensity of 0 always occurs at the edge portion of the exposure pattern, so that the shape of the exposure pattern becomes sharp and the resolution can be improved.

The transmittance of the semi-light-shielding film is set to 3% or more and 30% or less, and the exposure light is transmitted to some extent. That is, the exposure light also passes through the first and second attenuated transmission regions. Therefore, the distance between the first and second transmissive regions separated by the semi-light-shielding film becomes large, and even when one of the transmissive regions (for example, the first transmissive region) is isolated and becomes an isolated pattern, The exposure light transmitted through the first transmission region and the exposure light transmitted through the second attenuated transmission region overlap each other at the edge portion of the exposure pattern. Since the overlapping exposure lights have different phases, that is, the phases are opposite to each other, the exposure lights cancel each other at the overlapping portions. For this reason, a portion having a light intensity of 0 always occurs at the edge portion of the exposure pattern, so that the shape of the exposure pattern becomes sharp and the resolution can be improved.

As described above, since a high resolution can be obtained in both the dense pattern and the isolated pattern, a desired pattern shape can be easily formed even if the circuit pattern is complicated.

However, the intensity of light transmitted through the semi-shielding film is
It must be adjusted so that the photoresist is not exposed or that the photoresist remains exposed to the desired film thickness. If the transmissivity of the semi-shield film exceeds 30%, the light transmitted through the semi-shield film may sensitize the photoresist. Therefore, the transmissivity of the semi-shield film must be 30% or less.

If the intensity of the light transmitted through the semi-shielding film is too weak, the effect of obtaining a sharp exposure pattern cannot be obtained due to the overlapping of the light transmitted through the semi-shielding film and the light having a different phase. When the transmissivity of the semi-shielding film is smaller than 3%, the light transmitted through the semi-shielding film becomes too weak and the above effect cannot be obtained. Therefore, the transmissivity of the semi-light-shielding film must be 3% or more.

In the phase shift mask according to a preferred aspect of the present invention, the side wall of the step of the substrate has a shape having a radius of curvature substantially the same as the height of the step. That is,
The step sidewall is formed gently. Therefore, even if the semi-light-shielding film is formed on the step portion, the adhesion of the semi-light-shielding film to the substrate becomes good.

Further, since the side wall of the step is formed gently, it is possible to prevent foreign matter generated in the cleaning step from being trapped at the bottom of the step.

As described above, since the phase shift mask is less likely to have defects, a desired pattern shape can be easily formed.

In a phase shift mask according to another preferred aspect of the present invention, the substrate has first and second films made of materials having different etching characteristics. Therefore, the first film functions as an etching stopper layer during etching of the second film for forming a step on the substrate surface. Therefore, the controllability of the height of the step formed on the surface of the substrate becomes good, and the step having a predetermined height can be easily formed. Therefore, the first light transmitting portion and the second light transmitting portion
The controllability of the phase shift angle with respect to the light transmitting portion of is extremely excellent.

In the exposure method using the phase shift mask of the present invention, the phase shift mask which can obtain high resolution in both the dense pattern and the isolated pattern is used. Therefore, a desired pattern shape can be easily formed even with a circuit pattern in which a dense pattern and an isolated pattern are mixed.

Since both the dense pattern and the isolated pattern can be formed by the phase shift mask, the coherency σ can be set to an appropriate value. Therefore, the effect of the phase shift can be more remarkably obtained, so that the resolution is improved and a desired pattern shape can be easily formed.

According to the method of manufacturing a phase shift mask of the present invention, it is possible to manufacture a phase shift mask in which high resolution can be obtained in both a dense pattern and an isolated pattern, and thereby a desired pattern shape can be easily obtained. .

[0106]

Embodiments of the present invention will be described below with reference to the drawings. Example 1 FIG. 1 is a sectional view schematically showing the structure of a phase shift mask in a first example of the present invention.

With reference to FIG. 1, the phase shift mask 20 in the first embodiment has a quartz substrate 1 and a semi-shielding film 3. Grooves are formed on the surface of the quartz substrate 1. The region where the groove is not formed is the first light transmitting portion 1
a and the region where the groove is formed is the second light transmitting portion 1
b. The first light transmissive portion 1a and the second light transmissive portion 1b are configured such that the phases of the exposure light transmitted therethrough are different by 180 degrees.

The first light transmitting portion 1a has a first transmitting region 1L which is not covered with the semi-shielding film 3 and a first attenuated transmitting region 1N ₁ which is covered with the semi-shielding film 3. . The second light transmitting portion 1b has a second light transmitting region 1M not covered with the semi-shielding film 3 and a second attenuated transmitting region 1N ₂ covered with the semi-shielding film 3.

The side wall and the bottom wall of this groove have a substantially right-angled shape. The side wall of this groove constitutes a step. This step is covered and the first and second light transmitting portions 1a, 1
The semi-light-shielding film 3 is formed on the quartz substrate 1 so as to expose a predetermined region of b
Is formed on the surface of.

The semi-light-shielding film 3 has a transmittance of 3% or more and 30% or less. Further, the semi-shielding film 3 makes the phase of the exposure light after passing through 0 ° with respect to the phase of the exposure light before passing,
360 °, 360 ° × 2, ..., 360 × n °, ... That is, the semi-shielding film 3 does not substantially change the phase of the transmitted light, and the phase of the exposure light before passing through the semi-shielding film 3 and the phase of the exposure light after passing through the semi-shielding film 3 are substantially the same.

As a material of the semi-light-shielding film 3, for example, a chromium (Cr) film is used. Further, the specific thickness of the chromium film is i when the exposure light is transmitted by 10%.
It is about 200Å when a line is used and about 150Å when a KrF excimer laser beam is used. The thickness of the chrome film varies depending on the wavelength of the exposure light used and the transmittance set, and therefore may be set within the range of 100 to 300 Å.

The depth of the groove (height of the step) is about 405 when the i-line is used as the exposure light because the phase difference between the first and second light transmitting portions 1a and 1b is provided.
It is 0 Å, and when KrF excimer laser light is used, it is about 2720 Å.

Next, a method of manufacturing the phase shift mask of this embodiment will be described. FIG. 2 is a schematic cross-sectional view showing the method for manufacturing the phase shift mask in the first embodiment of the present invention. Referring to FIG. 2, after a quartz substrate having a thickness of 6.35 mm is prepared, a groove is formed on the surface of quartz substrate 1 by the same process as the conventional process. As a result, the first light transmitting portion 1a and the second light transmitting portion 1b are formed on the quartz substrate 1.

A chrome film 3 having a film thickness of 100 to 300 Å is formed on the entire surface so as to cover the groove. A resist pattern 9 having a desired shape on the surface of the chromium film 3
Is formed. By anisotropically etching the chromium film 3 using the resist pattern 9 as a mask,
Covering the side wall of the groove, and the first and second light transmitting portions 1a,
A semi-light-shielding film 3 exposing the desired region of 1b is formed. Then, resist pattern 9 is removed to form phase shift mask 20 shown in FIG.

In the phase shift mask of this embodiment, high resolution can be obtained in both the dense pattern and the isolated pattern. Hereinafter, this will be described in detail.

FIG. 3 is a diagram for explaining that the phase shift mask of this embodiment can obtain high resolution in a dense pattern, and FIG. 3A is a phase shift of this embodiment applied to a dense pattern. It is a schematic cross-sectional view of the mask,
(B) is a figure which shows the light intensity on a wafer when the phase shift mask of (a) is used.

Referring to FIG. 3, in the case of a dense pattern, adjacent transmissive regions (regions not covered by the semi-light-shielding film 3) 1L and 1M are arranged close to each other. For this reason, the exposure light transmitted through each of the adjacent transmission regions 1L and 1M overlaps with each other at the edge portion of the exposure pattern as shown by the dashed line in the figure. The overlapping exposure lights cancel each other out because they are out of phase with each other, that is, they are out of phase with each other. Therefore, as shown by the solid line in the figure, the light intensity of the exposure pattern always has a portion of light intensity 0 at the edge portion (between adjacent exposure patterns). Therefore, the shape of the edge portion of the exposure pattern becomes sharp and a sufficient light intensity difference can be provided, so that the resolution can be improved.

In the region k ₁ shown in the figure, the light intensity of the exposure pattern is locally high because the light shielding film 3 does not completely block the transmission of the exposure light but transmits the exposure light to some extent. This is because.

FIG. 4 is a diagram for explaining that the phase shift mask of this embodiment can obtain high resolution even in an isolated pattern, and FIG. 4A shows the phase shift of this embodiment applied to an isolated pattern. It is a schematic cross-sectional view of the mask,
(B) is a figure which shows the light intensity on a wafer when the phase shift mask of (a) is used.

Referring to FIG. 4, in the case of an isolated pattern, adjacent transmissive regions are arranged with a considerable distance therebetween. Therefore, the exposure lights transmitted through the adjacent transmissive regions 1M and 1M (or 1M and 1L, or 1L and 1L) do not overlap with each other. However, in this embodiment, the light-shielding film 3 is set so that the transmittance thereof is 3% or more and 30% or less, and the exposure light is transmitted to some extent.
Therefore, the attenuated transmission area (area covered with the light shielding film 3) 1
N also transmits the exposure light to some extent. Therefore, the transparent area 1M
And the exposure light transmitted through the attenuated transmission region 1N overlap each other at the edge portion of the exposure pattern. Since the overlapping exposure lights have different phases, they cancel each other out. Therefore, as shown by the solid line in the figure, the light intensity of the exposure pattern always has a portion with the light intensity of 0 at the edge portion, whereby the shape of the edge portion of the exposure pattern becomes sharp and the resolution can be improved. .

As described above, in the phase shift mask of this embodiment, high resolution can be obtained in both the dense pattern and the isolated pattern. Therefore, even when dense and isolated patterns are mixed on the same mask, high resolution can be obtained by applying this embodiment. Therefore, even if the circuit pattern is further miniaturized and complicated, a desired pattern shape can be easily formed.

However, in the phase shift mask of this embodiment, the exposure light transmitted through the semi-shielding film 3 does not expose the photoresist, or even if the photoresist is exposed, the exposure light is left to a certain thickness after development. The intensity of must be adjusted.

FIG. 5 is a diagram showing the relationship between the exposure amount at the time of exposure for forming a hole pattern and the resist residual film after development. Referring to FIG. 5, in the exposure of a normal hole pattern, the exposure amount at which the residual film of the resist becomes 0 after development (R in the figure
The exposure light is applied to the mask with a light amount about 3 to 4 times as large as that of (dot). Therefore, when the transmissivity of the semi-shielding film 3 exceeds 30%, the exposure light transmitted through the semi-shielding film 3 reduces the film thickness of the photoresist to 0, or the photoresist is reduced in film thickness. It cannot be used. Therefore, the transmittance of the semi-light-shielding film 3 must be 30% or less.

If the intensity of the light transmitted through the semi-shielding film 3 is too weak, the effect of obtaining a sharp exposure pattern cannot be obtained due to the overlapping of the exposure lights having different phases as described with reference to FIG. If the transmittance of the semi-light-shielding film 3 is smaller than 3%, the intensity of the exposure light transmitted through the semi-light-shielding film 3 becomes too small, and the above effect cannot be obtained. Therefore, the transmittance of the semi-light-shielding film 3 must be 3% or more.

Next, a specific structure of the phase shift mask in the case where the phase shift mask of this embodiment is applied to a pattern in which a dense pattern and an isolated pattern are mixed, and a mixed pattern formed thereby will be described.

FIGS. 6 and 7 show a specific structure (a) in the case where the phase shift mask of this embodiment is applied to a mixed pattern and a wafer cross-sectional structure (b) having a resist pattern formed thereby. It is a figure.

Referring to FIG. 6 (a) or FIG. 7 (a), a plurality of grooves are formed on the surface of quartz substrate 1 at desired intervals. The region where the groove is not formed becomes the first light transmitting portion 1a, and the region where the groove is formed becomes the second light transmitting portion 1b. The light-shielding film 3 is patterned so as to cover the side walls of the groove and expose predetermined regions of the first and second light transmitting portions 1a and 1b.

Referring to FIG. 6B or FIG. 7B, when the photoresist on the wafer 121 is exposed and developed using the above phase shift mask, resist patterns 121b and 121c are obtained. A negative type photoresist is used as the photoresist. The area E _{11 of the} photoresist 121b and the area E ₁₂ of the resist pattern 121c are dense patterns. The region F ₁₁ of the resist pattern 121b is a so-called isolated pattern, and the resist pattern 121b
The regions F ₁₂ and F ₁₃ of c are so-called isolated remaining patterns.

Next, in FIG. 1, the transmissive regions 1 adjacent to each other
The relationship between the distance S ₁ between L and 1M and the degree of overlap of the exposure patterns of the exposure light transmitted through both regions 1L and 1M will be considered.

FIG. 8 shows J. Miyazaki et al. "The effect.
of duty ratio of line and space in phase-shifting l
ithography "SPIE vol.1927,55, pp.677 to 685. This figure shows the Levenson type phase shift mask and a normal mask. It is a graph which shows the relationship between and a depth of focus (DOF).

Referring to FIG. 8, the vertical axis represents the depth of focus, and the horizontal axis represents the distance between adjacent transmission regions on the wafer (that is, the size of the light-shielding portion irradiated on the wafer).
As can be seen from this figure, when the size of the light shielding portion of the mask is 0.7 μm or less on the wafer, the Levenson type phase shift mask has a larger depth of focus than the ordinary mask. However, when the dimension is 0.7 μm or more, the Levenson type phase shift mask and the normal mask have almost the same depth of focus, and the difference between them disappears.

In this experiment, i-line is used,
Since the wavelength of this i-line is 0.365 μm, the dimension of 0.
7 μm corresponds to about twice the wavelength λ of the i-line (2λ). That is, the Levenson-type phase shift effect cannot be obtained unless the size of the light-shielding portion irradiated on the wafer is 2λ or less.

To convert the size of the light-shielding portion irradiated on this wafer into the size of the light-shielding portion on the phase shift mask,
It is necessary to consider the magnification of the projection optical system shown in FIG.
Referring to FIG. 45, usually, the circuit pattern of the phase shift mask 720 is reduced at a predetermined magnification by the projection optical system and is applied to the photoresist 121a. If the circuit pattern of the phase shift mask 720 is reduced by a factor of 5 and then the photoresist 121a is irradiated, a dimension of 5 μm on the phase shift mask 720 is illuminated by 1 μm on the photoresist 121a. Becomes Therefore, if the size of the light-shielding portion that is shrunk to n times by the projection optical system and irradiated on the wafer 121 is 2λ,
The size of the light shielding portion on the phase shift mask is 2λ × n.

That is, in the ordinary Levenson type phase shift mask as shown in FIG. 49, the effect of the phase shift cannot be obtained unless the size of the light shielding portion on the phase shift mask is about 2λ × n or less. This is explained as follows.

FIG. 9 is a diagram for explaining the relationship between the size of the light-shielding portion on the phase shift mask and the effect as the phase shift mask. FIG. 9A is a schematic cross section of an ordinary Levenson type phase shift mask. It is a figure, (b) shows each light intensity when the dimension of a light-shielding part is larger than 2 (lambda) xn, and (c) shows each light intensity when the dimension of a light-shielding part is smaller than 2 (lambda) xn.

Referring to FIGS. 9A and 9B, when the dimension S ₁₀₀ of the attenuated transmission region 1N exceeds 2λ × n, the exposure light beams passing through the adjacent transmission regions 1L and 1M overlap each other. Absent. Therefore, when the dimension S ₁₀₀ of the attenuated transmission region 1N is larger than 2λ × n, it is considered that the effect of the phase shift mask cannot be obtained and the depth of focus becomes similar to that of a normal mask.

On the other hand, referring to FIGS. 9A and 9C, when the dimensions of the attenuated transmission region 1N are smaller than 2λ × n, the exposure lights transmitted through the adjacent transmission regions 1L and 1M overlap each other. . Therefore, it is considered that the effect of the phase shift mask is obtained and the depth of focus becomes larger than that of a normal mask.

From the above, whether or not the exposure lights transmitted through the adjacent transmission regions 1L and 1M overlap each other is determined by the size of the attenuation transmission region 1N provided between the transmission regions 1L and 1M.
It is considered to be determined by whether it is larger or smaller than λ × n. The following conclusions can be obtained by applying this consideration to the present embodiment.

Referring to FIG. 3, if the dimension of the attenuated transmission region 1N is smaller than 2λ × n, the adjacent transmission regions 1L, 1
The exposure lights transmitted through M overlap with each other. Therefore, the Levenson type phase shift effect can be obtained.

On the other hand, referring to FIG. 4, if the dimensions of the attenuated transmission region 1N are larger than 2λ × n, the exposure lights transmitted through the adjacent transmission regions 1L and 1M do not overlap each other. However, the exposure lights transmitted through the transmission region 1M and the attenuated transmission region 1N overlap each other. Therefore, in this case, a halftone phase shift effect can be obtained.

In conclusion, in the phase shift mask of this embodiment, if the dimension S _{10 0} of the attenuated transmission region 1N is smaller than 2λ × n, the Levenson method is used, and the dimension S ₁₀₀ thereof is 2.
If it is larger than λ × n, the halftone phase shift effect can be obtained.

Therefore, when forming a circuit pattern in which a dense pattern and an isolated pattern are mixed as shown in FIG. 7, the dimension S _{3 of the} semi-light-shielding film 3 is formed in the dense pattern area E _12.
Is set to be smaller than 2λ × n, and the dimension S ₂ of the semi-shield film 3 in the isolated pattern region may be set to be larger than 2λ × n.

Next, an exposure method using the phase shift mask of this embodiment will be described. FIG. 10 shows the first aspect of the present invention.
3 is a schematic diagram of an optical system that realizes an exposure method using a phase shift mask in the example of FIG. With reference to FIG. 10, the exposure method of the present embodiment is almost the same as the conventional exposure method except for the mask 20 and diaphragm 114 used. The diaphragm 114 is different from the conventional one because the coherency σ of the exposure light used is different by using the phase shift mask of the present embodiment as compared with the case of using the conventional mask. Hereinafter, this will be described in detail.

In the Levenson-type and halftone-type phase shift masks, coherency σ
Is preferably small. Here, coherency σ is
It represents the degree of light interference. The smaller the coherency σ, the larger the degree of light interference, and the larger the coherency σ, the smaller the degree of light interference. That is, the phase shift mask is a technique for improving the resolution by the interference of the light of the opposite phase, and therefore the larger the degree of the light interference is, the more preferable. Therefore, when the phase shift mask is used, it is preferable that the coherency σ be small.
However, due to practical restrictions such as reduction of exposure amount, the coherency σ is set to about 0.3 in the phase shift mask.

On the other hand, since it is preferable that light does not interfere with an ordinary mask, it is preferable that the coherency σ be large. Therefore, in a normal mask, the coherency σ is set to about 0.6 due to facility restrictions.

A memory cell area 121 as shown in FIG.
A dense pattern such as M and its peripheral circuit area 121P
In the case of forming a circuit pattern including an isolated pattern as described above, conventionally, a phase shift mask is used for the dense pattern and a normal mask is used for the isolated pattern. That is, the phase shift mask and the normal mask are formed in a mixed manner on the same mask.

In this case, if the coherency σ of the exposure light applied to this mask is set to 0.3, the resolution of the ordinary mask is lowered, and if the coherency σ is set to 0.6, the resolution of the phase shift mask is lowered. . For this reason, a coherency σ of about 0.4 to 0.5 is used in the middle of both masks. However, this coherency σ
At (0.4 to 0.5), a high resolution cannot be obtained with both the phase shift mask and the normal mask. Therefore, when the circuit pattern is further miniaturized, it is difficult to form a desired pattern shape.

On the other hand, in this embodiment, both the dense pattern and the isolated pattern can be formed by using the phase shift effect. For this reason, in this embodiment, even if the pattern has both dense and isolated patterns, a high resolution can be obtained by setting the coherency σ of the exposure light applied to the mask to 0.3, which is suitable for the phase shift mask. Therefore, it is possible to easily obtain a desired pattern shape.

This coherency σ is determined by the aperture diameter of the diaphragm 114 in FIG. Therefore, in this embodiment having a different coherency σ from the conventional example, the configuration of the diaphragm 114 is
Especially, the opening diameter is different from the conventional one. The diaphragm 114 of this embodiment will be described below.

FIG. 12 is a plan view of the diaphragm shown in FIG.
Referring to FIG. 12, the diaphragm 114 is a light-shielding disc 114a.
And an opening 114b. The exposure light is the diaphragm 114
Through the opening 114b. By reducing the opening diameter r of the opening 114b, the coherency σ can be reduced. Further, the coherency σ increases as the opening diameter r increases. Second Embodiment FIG. 13 is a sectional view schematically showing the structure of a phase shift mask according to a second embodiment of the present invention. With reference to FIG. 13, a plurality of semi-light-shielding films 203 are formed on the surface of a quartz substrate 201 at predetermined intervals. Semi-shading film 20
A quartz layer 205 serving as a shifter is formed on every other surface of the quartz substrate 201 exposed from No. 3. This shifter 2
The portion where 05 is formed becomes the first light transmitting portion 201a, and the portion where the shifter is not formed is the second light transmitting portion 2a.
It becomes 01b. The first and second light transmitting portions 201
In a and 201b, the phase of the exposure light that is transmitted differs by about 180 degrees.

The first light transmitting portion 201a is made up of the semi-shielding film 20.
No. 3 of the first transmission region 201L and the semi-light-shielding film 203 of the first attenuation transmission region 201N ₁
And have. In addition, the second light transmitting portion 201b is the second light transmitting region 201 in which the semi-shielding film 203 is not formed.
M and the second attenuated transmission region 201N ₂ having the semi-shield film 203 formed therein.

A chromium film, for example, is used for this semi-light-shielding film 203. The thickness of the chrome film is 10% of the exposure light.
If it is transmitted, it is about 200 Å when the i-line is used as the exposure light, and about 150 Å when the KrF excimer laser light is used.

The thickness of the chrome film is 100 to 100 because it depends on the wavelength of the exposure light used and the transmittance set.
It may be set within the range of 300Å.

Next, the manufacturing method of this embodiment will be described. 14 to 16 are schematic cross-sectional views showing the method of manufacturing the phase shift mask in the second embodiment of the present invention in the order of steps. First, referring to FIG. 14, a chromium film having a film thickness of, for example, 200 Å is formed on the surface of quartz substrate 201.
By patterning this chrome film by photolithography, the semi-light-shielding film 203 having a desired shape is formed.

Referring to FIG. 15, a quartz layer 205a is formed on the entire surface of the quartz substrate 201 so as to cover the semi-shielding film 203.

Referring to FIG. 16, a resist pattern 209 having a desired shape is formed on the surface of quartz layer 205a. Using this resist pattern 209 as a mask, the quartz layer 205 is anisotropically etched. The quartz substrate 201 exposed from the semi-light-shielding film 203 by this etching
Shifters 205 are formed of quartz on every other surface of the. The area where the shifter 205 is formed is the first light transmitting portion 201a, and the area where the shifter is not formed is the second light transmitting portion 201b. Then, resist pattern 209 is removed to obtain the phase shift mask shown in FIG.

In the phase shift mask of this embodiment, the semi-shielding film 203 is formed in a partial area of the first and second light transmitting portions and has a transmittance of 3% or more and 30% or less. ing. Therefore, similar to the first embodiment, high resolution can be obtained in both the dense pattern and the isolated pattern. Therefore, it is possible to easily form a desired pattern shape even with a complicated circuit pattern.

When the phase shift mask of this embodiment is applied to a pattern in which a dense pattern and an isolated pattern are mixed, the structure of the phase shift mask shown in FIGS. 17 (a) and 18 (a) can be obtained, for example.

Referring to FIGS. 17A and 18A, a plurality of semi-light-shielding films 203 are formed on the surface of quartz substrate 201 so as to have desired intervals. A shifter 205 is formed on the surface of the quartz substrate 201 or the surface of the semi-shielding film 205 exposed from the semi-shielding film 203. The area where the shifter 205 is formed becomes the first light transmitting portion 201a, and the area where the shifter 205 is not formed becomes the second light transmitting portion 201b. The first and second light transmitting portions 201a and 201b differ from each other in the phase of the exposure light transmitted by about 180 degrees.

Referring to FIGS. 17B and 18B, resist patterns 121d and 121e are formed on wafer 121 using the above phase shift mask. The resist pattern 121d region E ₂₁ and the resist pattern 121e region E ₂₁ is a dense pattern area. The region F ₂₁ of the resist pattern 121d is a so-called isolated removal pattern, and the regions F ₂₂ and F ₂₃ of the resist pattern 121e are so-called isolated remaining patterns.

Also in the phase shift mask of this embodiment, as in the first embodiment, the dimension S ₅ of the semi-shielding film 203 shown in FIG. 13 is smaller than 2λ × n (n: magnification of the projection optical system),
Alternatively, by increasing the value, either the Levenson method or the halftone method phase shift effect can be selected.

Specifically, in FIG. 17, the dimension S ₆ of the semi-light-shielding film 203 is 2λ in the dense pattern area E ₂₁ .
The dimension S ₇ of the semi-shielding film 203 is set to be larger than 2λ × n in the isolated pattern region. As a result, in the dense pattern area E ₂₁ , the phase shift effect of the Levenson method is obtained, and the isolated pattern area E ₂₁ is obtained.
_{In 22} , the halftone phase shift effect is obtained. Therefore, high resolution can be obtained in both the dense pattern and the isolated pattern.

Furthermore, in the phase shift mask of this embodiment,
Similar to the first embodiment, the phase shift effect can be used to form both a dense pattern and an isolated pattern. Therefore, by setting the coherency σ of the exposure light to a value (0.3) suitable for the phase shift mask, high resolution can be obtained even in a mixed pattern. Third Embodiment FIG. 19 is a sectional view schematically showing the structure of a phase shift mask according to a third embodiment of the present invention. Also in FIG.
FIG. 20 is a partial sectional view showing a region D _{1 of} FIG. 19 in an enlarged manner.

Referring to FIGS. 19 and 20, the phase shift mask 320 in the third embodiment has a quartz substrate 301 and a light shielding film 303. Grooves are formed on the surface of the quartz substrate 301. The region in which the groove is not formed forms the first light transmitting portion 301a, and the region in which the groove is formed forms the second light transmitting portion 301b. The first light transmitting portion 301a and the second light transmitting portion 301b are
The phases of the exposure light beams that pass through are different by 180 degrees.

The surface of the first light transmitting portion 301a is the substrate 3
It is formed at the first height position from the bottom surface of 01. On the other hand, the surface of the second light transmitting portion 301b has the quartz substrate 3
It is formed at a second height position lower than the first height from the bottom surface of 01. Therefore, there is a step between the first light transmitting portion 301a and the second light transmitting portion 301b. The shape of this step has a radius R that is substantially the same as the height R _{31 of the} step.
_It has a curvature of ₃₂ .

A light-shielding film 303 is formed on the surface of the quartz substrate 301 so as to cover the step and expose the predetermined regions of the first and second light transmitting portions 301a and 301b.

Next, a method of manufacturing the phase shift mask of this embodiment will be described. 21 to 28 are schematic cross-sectional views showing the method of manufacturing the phase shift mask in the third embodiment of the present invention in the order of steps.

First, with reference to FIG. 21, a quartz substrate 301 made of synthetic quartz and having a required shape and flatness is subjected to necessary cleaning and pretreatment. After this, the quartz substrate 30
A doped amorphous silicon film 305a having a film thickness of 300 to 1000 Å is formed on the substrate 1 by a normal DC discharge type sputtering apparatus. An electron beam resist film 307a having a film thickness of 3000 Å or less is formed on the entire surface of the doped amorphous silicon film 305a by an ordinary spin coater. A shifter pattern of a Levenson type phase shift mask is drawn on this electron beam resist film 307a by an electron beam drawing apparatus.

Referring to FIG. 22, after this drawing, the resist pattern 307 is formed by developing with a normal developing device. Using the resist pattern 307 as a mask, the doped amorphous silicon film 305a is selectively etched with respect to the quartz substrate 301 by plasma of a gas in which CF ₄ and O ₂ are mixed at a ratio of 95: 5 by a plasma etching apparatus.

The etching selection ratio at this time is about 60. Therefore, the doped amorphous silicon film 305
The etching amount of the quartz substrate when a is over-etched 100% is 5 to 16Å, which is about 10Å. Therefore, the effect of this etching on the phase angle of the shifter is negligible. After that, the electron beam resist pattern 7 is removed by normal ashing with oxygen plasma. After that, cleaning is further performed with an alkali cleaning solution.

With reference to FIG. 23, the shifter pattern is formed on the doped amorphous silicon film 305 in this manner. The defect inspection of this shifter pattern is carried out by an ordinary optical defect inspection machine. Although common in such a mask process, white defects are rarely generated, and only black defects are often detected and detected. This black defect can be repaired by the above-mentioned ordinary laser repair device, and by doing this, it is made defect-free.

With reference to FIG. 24, in order to improve wettability in wet etching (to make it easier to wet),
Surface treatment with oxygen plasma is performed. After that, a solution in which ammonium fluoride and hydrofluoric acid are mixed at a ratio of 50: 1 and a surfactant added thereto is used to form a desired amount on the quartz substrate 301 (about 410 in the i-line exposure mask).
0Å) Wet etching is applied. By this wet etching, a shifter pattern is formed on the quartz substrate 301.

In this wet etching, the doped amorphous silicon film 305 serves as an etching mask material for the quartz substrate 301. At this time, since the adhesion between the doped amorphous silicon film 305 and the quartz substrate 301 is sufficient, it is difficult for the etchant to enter the interface between the doped amorphous silicon film 305 and the quartz substrate 301.

If the etchant easily enters this interface, the quartz substrate 301 is removed along the interface as shown by the dotted line in the figure. Therefore, the amorphous silicon film 305 is easily peeled off. However, as described above, since the adhesiveness is good and the etchant is difficult to enter into this interface, there is no fear that such peeling of the doped amorphous silicon film 305 occurs.

In addition, the etching rate of the quartz substrate 301 during the wet etching is sufficiently low. Therefore, the controllability of the phase shift angle determined by the etching amount is within ± 5 degrees.

After that, the doped amorphous silicon film 305 is removed by plasma of a gas in which CF ₄ and O ₂ are mixed at a ratio of 95: 5. At this time, the etching selection ratio between the doped amorphous silicon film 305 and the quartz substrate 301 is about 60. Therefore, when the doped amorphous silicon film 305 is 100% over-etched, the etching amount of the quartz substrate 301 is 5 to 16Å.
And about 10Å. Therefore, the effect of this etching on the phase angle of the shifter is negligible.

Referring to FIG. 25, the first light transmitting portion 301a and the second light transmitting portion 301b are thus formed on the quartz substrate 301 in this manner. After that, cleaning with an alkaline cleaning liquid is performed.

Referring to FIG. 26, a chromium film 303a serving as a light-shielding film material for a phase shift mask is formed on the entire surface with a film thickness of about 1000Å by a normal DC discharge type sputtering apparatus. An electron beam resist film 309a is formed on the entire surface of the chromium film 303a by an ordinary spin coater to a thickness of about 3000.
It is formed with a film thickness of Å. This electron beam resist film 309a
Then, a desired shape is drawn by the electron beam drawing apparatus.

Referring to FIG. 27, an electron beam resist pattern 309 is formed by developing with a normal developing device after this drawing.

At the time of electron beam drawing in this process, alignment with the underlying shifter pattern is necessary.
The accuracy of the ordinary electron beam drawing apparatus is sufficient for this accuracy. That is, since the accuracy of the ordinary electron beam drawing apparatus is about 0.1 μm, which is sufficiently smaller than the pattern shape of the underlayer, the accuracy of alignment with the underlying pattern is sufficient with the accuracy of this electron beam drawing apparatus.

After that, the electron beam resist pattern 309 is removed. Further, ordinary defect inspection / repair is performed, thereby forming a desired Levenson-type phase shift mask 320 as shown in FIG.

The phase shift mask of this embodiment has the structure shown in FIG.
As described with reference to FIG. 20 and FIG.
The shape of the step between 1a and the second light transmitting portion 301b is
The radius of curvature R ₃₂ is substantially the same as the height R _{31 of the} step. Therefore, the step in this embodiment has a gentler shape than the step of the conventional phase shift mask 720 shown in FIG. Therefore, the adhesion of the light shielding film 303 formed so as to cover the step portion to the quartz substrate 301 is better than that of the conventional phase shift mask 720.

Also, the phase shift mask 320 of this embodiment.
Then, the steps of the quartz substrate 301 are formed gently.
Therefore, it is possible to prevent foreign matters and the like generated when the doped amorphous silicon film 305 is removed by the process shown in FIGS. 24 and 25 from being trapped at the bottom of the step. Therefore, foreign matter is prevented from remaining on the bottom of the step.

In the manufacturing method of this embodiment, the quartz substrate 301 is isotropically etched by the process shown in FIGS. Therefore, as shown in FIG. 29, the remaining defects 305b of the doped amorphous silicon film 305 are temporarily assumed.
Even if there is, the lower layer of the residual defect 305b is also etched away.

That is, referring to FIG. 30, in the isotropic etching, the wrapping property of the etchant is good. Therefore, the etchant wraps around under the mask 305 and the remaining defects 305b. Therefore, the quartz substrate 301 distributed in the lower region of the mask 305 and the lower region of the residual defect 305b is also removed by etching. When removed by etching in this way, the residual defect 305b loses its lower layer and falls off from the quartz substrate 301. Therefore, the remaining defect 305
Since there is b, no unremoved part occurs in the region to be removed. Therefore, when the resist film on the wafer is exposed by using the phase shift mask manufactured as described above, good resolution is obtained, and the pattern shape is not defective.

In the phase shift mask of this embodiment, the semi-light-shielding film used in the first and second embodiments may be used instead of the light-shielding film 303. For example, a chromium film is used as the semi-light-shielding film. The chrome film thickness is about 200 Å when the i-line is used for the exposure light and about 150 Å when the KrF excimer laser light is used, assuming that the exposure light transmits 10%. The thickness of the chrome film varies depending on the wavelength of the exposure light used and the transmittance set, and therefore may be set within the range of 100 to 300 Å.

If a semi-light-shielding film is used instead of the light-shielding film in this embodiment, the same effects as those obtained in the first and second embodiments can be obtained.

That is, referring to FIG. 19, the semi-shield film 3
03 is the first and second light transmitting portions 301a and 301b
Is formed in a partial region of and has a transmittance of 3% or more and 30% or less. Therefore, similar to the first and second embodiments, high resolution can be obtained with both the dense pattern and the isolated pattern. Therefore, it is easy to form a desired pattern shape even with a complicated circuit pattern.

When the phase shift mask of this embodiment is applied to a pattern in which a dense pattern and an isolated pattern are mixed, a concrete configuration of the phase shift mask is shown in FIG. 31, for example.
(A) and FIG. 32 (a).

Referring to FIGS. 31 (a) and 32 (a),
A plurality of grooves are formed on the surface of the quartz substrate 301 at desired intervals. The sidewalls of the groove have a radius of curvature that is substantially the same as the depth of the groove. The region where the groove is not formed becomes the first light transmitting portion 301a, and the region where the groove is formed becomes the second light transmitting portion 301b.

The first and second light transmitting portions 301a,
In 301b, the phases of the exposure light transmitted are different from each other by 180 degrees. A semi-light-shielding film 303 is formed so as to cover the side wall of the groove and expose predetermined regions of the first and second light transmitting portions 301a and 301b.

Also in the phase shift mask of this embodiment, as shown in FIG. 19, the dimension S ₁₀ of the semi-light-shielding film 303 is 2λ × n.
The phase shift effect of the Levenson method mask halftone method can be selected by making the value smaller or larger than (n: magnification of projection optical system).

Specifically, as shown in FIG. 31, a dense pattern
Region E₃₁, The dimension S of the semi-light-shielding film 303₁₁2
Less than λ × n and semi-shading in isolated pattern area
Dimension S of membrane 303₁₂Is larger than 2λ × n. this
The dense pattern area E ₃₁In Levenson
Equation phase shift effect is obtained, and the isolated pattern region F
₃₁, The halftone phase shift effect was obtained.
Be done. Therefore, both dense and isolated patterns
High resolution can be obtained with.

Further, in the phase shift mask of this embodiment, both the dense pattern and the isolated pattern can be formed by using the phase shift effect as in the first embodiment. Therefore, high resolution can be obtained by setting the coherency σ of the exposure light to a value (0.3) suitable for the phase shift mask. Fourth Embodiment FIG. 33 is a sectional view schematically showing the structure of a phase shift mask according to a fourth embodiment of the present invention. Also, FIG.
FIG. 34 is a partial cross-sectional view showing a region D _{2 of} FIG. 33 in an enlarged manner.

33 and 34, the phase shift mask 420 according to the fourth embodiment has a quartz substrate 411.
And an etching stopper film 413, a silicon oxide film (SOG: Spin On Glass) 415, and a light shielding film 403. An etching stopper film 413 and a silicon oxide film 415 are sequentially stacked on the surface of the quartz substrate 411. Further, the quartz substrate 411, the etching stopper film 413, and the silicon oxide film 415 constitute a substrate 401 that transmits exposure light.

The etching stopper film 413 is made of, for example, SnO, Al ₂ O ₃ or the like, and has a thickness of 100 to 1000 Å.
It has a film thickness of. The silicon oxide film 415 has about 4
It is formed with a film thickness of 000Å. A groove is formed in the silicon oxide film 415, and a part of the surface of the etching stopper film 413 is exposed at the bottom of the groove. The region where the groove is not formed is the first light transmitting portion 4
01a, and the region where the groove is formed constitutes the second light transmitting portion 401b.

The upper surface of the etching stopper film 413 and the upper surface of the silicon oxide film 415 differ in height position by the thickness of the silicon oxide film 415. Therefore, a step is formed by the upper surfaces of the etching stopper film 413 and the silicon oxide film 415. That is, a step is formed between the surface of the first light transmitting portion 401a and the surface of the second light transmitting portion 401b. The shape of this step has a curvature with a radius R ₄₂ that is substantially the same as the height R _{41 of the} step.

Chromium (C) is formed so as to cover the step portion formed by the side wall of the groove and to expose the desired regions of the first light transmitting portion 401a and the second light transmitting portion 401b.
A light-shielding film 403 made of r) is formed.

Next, a method of manufacturing the phase shift mask according to the fourth embodiment of the present invention will be described.

35 to 42 are schematic cross-sectional views showing the method of manufacturing the phase shift mask in the fourth embodiment of the present invention in the order of steps.

First, referring to FIG. 35, an etching stopper film 413, a silicon oxide film 415, and a doped amorphous silicon film 405a are formed on the entire surface of the quartz substrate 411 in the range of 100 to 1000 Å, 4000 Å or less, 300 to Å, respectively.
It is formed by sequentially laminating it with a film thickness of 1000Å. An electron beam resist film 407a is formed on the entire surface of the doped amorphous silicon film 405a by an ordinary spin coater 3000.
It is formed with a film thickness of Å. This electron beam resist film 407a
Then, the shifter pattern of the Levenson type phase shift mask is drawn by the electron beam drawing apparatus.

Referring to FIG. 36, after drawing, the resist pattern 40 is developed by developing with a normal developing device.
7 is formed. Using the resist pattern 407 as a mask, the doped amorphous silicon film 405 is converted into a silicon oxide film 41 by plasma of a gas in which CF ₄ and O ₂ are mixed at a ratio of 95: 5 by an isotropic plasma etching apparatus.
5 is etched selectively.

At this time, the etching selection ratio of the doped amorphous silicon film 405 to the silicon oxide film 415 is about 60. Therefore, when the doped amorphous silicon film 405 is overetched 100%, the etching amount of the silicon oxide film 415 is 5 to 16Å, which is about 10Å. Therefore, the effect of this etching on the phase angle of the shifter is negligible. After that, the electron beam resist pattern 407 is removed by normal ashing with oxygen plasma, and cleaning with an alkaline cleaning liquid is performed.

Referring to FIG. 37, the shifter pattern is formed on the doped amorphous silicon film 405 in this manner. The defect inspection of this shifter pattern is performed by a normal optical defect inspection machine. Although common in such a mask process, white defects are rarely generated, and only black defects are often detected and detected. This black defect can be repaired by an ordinary laser repairing device, and by doing this, it is made defect-free.

Next, in order to improve wettability in wet etching (to make it easier to wet), surface treatment is carried out by oxygen plasma. After that, wet etching is performed until the surface of the etching stopper film 413 is exposed, by adding a surfactant to a solution in which ammonium fluoride and hydrofluoric acid are mixed at a ratio of 50: 1. The substrate 4 is formed by this wet etching.
A shifter pattern is formed at 01.

At this time, the doped amorphous silicon film 405 and the silicon oxide film 41, which are etching masks, are used.
Adhesion with 5 is sufficient. Therefore, there is no concern that the doped amorphous silicon film 405 will peel off. The etching stopper film 413 serves as an etching stopper during the wet etching of the silicon oxide film 415. Therefore, the controllability of the phase shift angle is extremely good.

Referring to FIG. 38, a part of silicon oxide film 415 is removed by the above wet etching,
A part of the surface of the etching stopper film 413 is exposed. After that, the doped amorphous silicon film 405 is removed by plasma of a gas in which CF ₄ and O ₂ are mixed at a ratio of 95: 5 by an isotropic plasma etching apparatus.

In this way, referring to FIG. 39, the first
A substrate 401 (including a quartz substrate 411, an etching stopper film 413, and a silicon oxide film 415) having a light transmitting portion 401a and a second light transmitting portion 401b is obtained. After that, cleaning with an alkaline cleaning liquid is performed.

Referring to FIG. 40, after cleaning, a chromium film 403a serving as a light-shielding film material is formed with a thickness of about 1000 Å by an ordinary DC discharge type sputtering device so as to cover the entire surface. An electron beam resist film 409a is formed on the entire surface of the chrome film 403a by an ordinary spin coater to a film thickness of about 3000Å. A pattern having a desired shape is drawn on the electron beam resist film 409a by an electron beam drawing apparatus.

Referring to FIG. 41, after this drawing, the resist pattern 409 is formed by developing with a normal developing device.

At the time of drawing an electron beam in this process, alignment with the underlying shifter pattern is necessary.
This alignment accuracy is sufficient with the accuracy of an ordinary electron beam drawing apparatus. That is, the accuracy of the ordinary electron beam drawing apparatus is about 0.1 μm, which is sufficiently smaller than the pattern shape of the base, and therefore the accuracy of the electron beam drawing apparatus is sufficient as the alignment accuracy.

By selectively etching the chrome film 403a using the electron beam resist pattern 409 as a mask, the light shielding film pattern 403 of the chrome film is formed. After that, the electron beam resist pattern 409 is removed by the usual ashing with oxygen plasma. After this, a normal defect inspection / correction is further performed, as shown in FIG.
A desired Levenson-type phase shift mask 420 shown in FIG.

In the phase shift mask of this embodiment, as shown in FIG.
3, the shape of the step formed between the surface of the first light transmitting section 401a and the surface of the second light transmitting section 401b is substantially the same as the step height R _41. It has a radius of curvature R ₄₂ . That is, the side wall of the step in this embodiment has a gentler shape than the side wall of the step of the conventional phase shift mask 720 shown in FIG. Therefore, even if the light shielding film 403 is formed on the stepped portion of this embodiment, the adhesion of the light shielding film 403 to the silicon oxide film 415 becomes good.

Further, the step is formed gently.
For this reason, it is possible to prevent foreign substances generated when the doped amorphous silicon film 405 is removed in the processes of FIGS. 38 and 39 from being trapped at the bottom of the step.

Further, in the method of manufacturing the phase shift mask of this embodiment, the silicon oxide film 115 is isotropically etched by the process shown in FIGS. In this isotropic etching, the etchant wraps around the mask 405 and the lower region of the remaining defect (not shown). Therefore, as described with reference to FIGS. 29 and 30 in the third embodiment, even in the present embodiment, even if a residual defect occurs, a portion which is not removed by etching due to this is a portion which is not removed by etching. It never happens. Therefore, when the resist film on the wafer is exposed by using the phase shift mask manufactured as described above, good resolution is obtained, and the pattern shape is not defective.

In addition, in this embodiment, the substrate 401 is the third
It does not have a single-layer structure as in the embodiment. That is, the substrate 4
01 is a quartz substrate 411 and an etching stopper film 413.
And a silicon oxide film 415 have a three-layer structure.
The etching stopper film 413 is the silicon oxide film 4
It acts as an etching stopper when etching 15. Therefore, the etching stopper film 413 is hardly etched when the silicon oxide film 415 is etched. Therefore, as shown in FIG. 39, the first
The distance d ₁ between the surface of the light transmitting portion 401a and the surface of the second light transmitting portion 401b can be easily controlled.

That is, since the etching stopper film 413 is provided in the lower layer of the silicon oxide film 415, the first etching can be performed without strictly controlling the etching amount of the silicon oxide film 415.
Also, the distance between the surfaces of the second light transmitting portions 401a and 401b can be easily controlled to the desired distance d ₁ . The difference between the phase shift angles of the first and second light transmitting portions 401a and 401b is determined by this distance d ₁ . Therefore, the substrate 401 having the above-described three-layer structure makes it possible to easily control the difference in phase shift angle between the first light transmitting portion 401a and the second light transmitting portion 401b. Therefore, the controllability of the phase shift angle becomes extremely good.

In the phase shift mask of this embodiment, the semi-light-shielding film used in the first and second embodiments can be used instead of the light-shielding film 403. A chromium film, for example, is used as the semi-light-shielding film. Further, the thickness of the chromium film is such that if the exposure light is transmitted by 10%,
When i line is used as exposure light, it is about 200Å,
When KrF excimer laser light is used, it is about 150Å.

Since the thickness of the chromium film varies depending on the wavelength of the exposure light used and the transmittance to be set,
It may be set within the range of 300Å.

When the phase shift mask of this embodiment using a semi-light-shielding film is applied to a pattern in which a dense pattern and an isolated pattern are mixed, for example, the specific configuration of the phase shift mask is shown in FIGS. 43 (a) and 44. As shown in (a).

43 (a) and 44 (a), the etching stopper film 4 is formed on the entire surface of the quartz substrate 411.
13 is formed. This etching stopper film 41
A silicon oxide film 415 is formed on a predetermined region of No. 3. A groove reaching the etching stopper film 413 is formed in the silicon oxide film 415. The sidewall of this groove has a radius of curvature substantially the same as the film thickness of the silicon oxide film 415.

The portion where the silicon oxide film 415 is formed becomes the first light transmitting portion 401a, and the silicon oxide film 415 is formed.
The portion where no is formed becomes the second light transmitting portion 401b. A semi-light-shielding film 403 is formed so as to cover the side wall of the silicon oxide film 415 having a curvature and to expose predetermined regions of the first and second light transmitting portions 401a and 401b.

The resist pattern formed by using the phase shift mask shown in FIGS. 43 (a) and 44 (a) is as shown in FIG.
3 (b) and FIG. 44 (b), respectively. Region E ₄₁ of resist pattern 121i and resist pattern 1
The area E ₄₂ of 21h is a dense pattern area, and the area F ₄₁ of the resist pattern 121i is a so-called isolated pattern. Further, the region F _{42 of} the resist pattern 121h,
F ₄₃ is a so-called isolated leaving pattern.

In the phase shift mask of this embodiment using the semi-shielding film, the dimension S _{15 of the} semi-shielding film 403 shown in FIG. 33 is used.
Is smaller or larger than 2λ × n, the phase shift effect of the Levenson mask halftone system can be selected.

Specifically, the dense pattern area E of FIG.
The size S ₁₆ of the semi-shielding film 413 of ₄₁ is smaller than 2λ × n, and the size S ₁₇ of the semi-shielding film 403 in the isolated pattern region is 2
It is larger than λ × n. As a result, the Levenson type phase shift effect is obtained in the dense pattern region E ₄₁ , and the halftone type phase shift effect is obtained in the isolated pattern region. Therefore, high resolution can be obtained in both the dense pattern and the isolated pattern.

Further, in the phase shift mask of this embodiment using the semi-light-shielding film, both the dense pattern and the isolated pattern can be formed by using the phase shift effect. Therefore, the coherency σ of the exposure light can be set to a value (0.3) suitable for the phase shift mask. Therefore, high resolution can be obtained for both the dense pattern and the isolated pattern.

In the first to fourth embodiments, chromium is used as the semi-light-shielding films 3, 203, 303 and 403, but it is not particularly limited to chromium and only the transmittance can be controlled. Any material that does not substantially change the phase can be applied.

In this embodiment, the case where the substrate 401 has the three-layer structure of the quartz substrate 411, the etching stopper film 413 and the silicon oxide film 415 has been described, but the present invention is not limited to this. Specifically, it may be a laminated structure of at least two layers having mutually different etching characteristics, and may be a laminated structure of three or more layers.

In this embodiment, the etching strike is
The upper film 413 is SnO or Al ₂ O₃ If
However, the present invention is not limited to this. Ingredient
Physically, it has a transmittance of 90% or more, and has a silicon oxide film.
Any material that is resistant to the wet etching of
Good.

The silicon oxide film 415 is not limited to this material, and any material having at least a transmittance of 90% or more may be used. The thickness of the silicon oxide film is 400
Although described as 0Å, this film thickness varies depending on the material. That is, the film thickness d is

[0231]

[Equation 1]

Any value may be used as long as it satisfies the relationship represented by (λ ₀ : wavelength of exposure light, n: refractive index of shifter material).

In the first to fourth embodiments, the quartz substrates 1, 201, 301 and 411 do not have to be quartz, and may be made of a material having a transmittance of 90% or more.

In the third and fourth embodiments,
The doped amorphous silicon films 305 and 405 may be made of other materials. Specifically, MoS
Any material having conductivity and excellent hydrofluoric acid resistance such as i ₂ may be used. The reason why the material is required to have conductivity is that the resist film coated on the surface of this film is exposed by an electron beam drawing method. Therefore, when the resist film formed on the surface of this film is exposed by a method other than the electron beam drawing method, conductivity is not required as a characteristic of the film.

The light shielding films 303 and 403 are made of chromium (C
Although the case of r) has been described, the present invention is not limited to this. Specifically, it may be a multilayer film such as CrO / Cr / CrO. Light-shielding film 303, 40
The properties required for No. 3 are that it does not transmit exposure light, it has excellent chemical resistance (in consideration of cleaning), and that it has adhesion. Therefore, a material that satisfies these characteristics can be applied to the light shielding films 303 and 403.

In addition, the dimensions, materials, etc. described in the first to fourth embodiments are not limited to these and can be arbitrarily selected.

[0237]

In the phase shift mask of the present invention, the semi-shielding film is located between the first and second light transmitting portions and is formed in a partial area of the first and second light transmitting portions. The semi-light-shielding film has a transmittance of 3% or more and 30% or less. Therefore, high resolution can be obtained in the dense pattern and the isolated pattern. Therefore, a desired pattern shape can be easily obtained.

In the phase shift mask according to a preferred aspect of the present invention, the side wall of the step of the substrate has a shape having a radius of curvature substantially the same as the height of the step. For this reason,
When the semi-light-shielding film is formed on this step, the adhesion of the semi-light-shielding film to the substrate is improved. Further, since the step is formed gently, it is possible to prevent foreign matter generated in the cleaning process from being trapped at the bottom of the step. Therefore, it is possible to easily obtain a pattern having a desired shape.

In a phase shift mask according to another preferred aspect of the present invention, the substrate has first and second films made of materials having different etching characteristics. Therefore, the controllability of the phase shift angle between the first light transmitting portion and the second light transmitting portion becomes extremely good. Therefore, a desired pattern shape can be easily obtained.

In the exposure method using the phase shift mask of the present invention, the phase shift mask capable of obtaining high resolution in both the dense pattern and the isolated pattern is used.
Therefore, a desired pattern shape can be easily obtained even with a circuit pattern in which a dense pattern and an isolated pattern are mixed.

Since both the dense pattern and the isolated pattern can be formed by the phase shift effect, the coherency σ of the exposure light can be set to an appropriate value for the phase shift mask. Therefore, a desired pattern shape can be easily obtained.

In the method of manufacturing the phase shift mask of the present invention, high resolution can be obtained for both the dense pattern and the isolated pattern. Therefore, it is possible to manufacture a phase shift mask that can easily obtain a desired pattern shape.

[Brief description of drawings]

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

FIG. 2 is a schematic cross-sectional view showing the method of manufacturing the phase shift mask in the first embodiment of the present invention.

FIG. 3 is a diagram for explaining that the phase shift mask according to the first embodiment of the present invention can obtain high resolution in a dense pattern, and FIG. 3A is a schematic sectional view of the phase shift mask according to the present embodiment. And (b) is a diagram showing the light intensity on the wafer.

FIG. 4 is a diagram for explaining that the phase shift mask according to the first embodiment of the present invention can obtain high resolution in an isolated pattern, and FIG. 4A is a schematic sectional view of the phase shift mask according to the present embodiment. And (b) is a diagram showing the light intensity on the wafer.

FIG. 5 is a graph showing a relationship between an exposure amount when exposing a resist and a resist residual film after the resist is developed.

FIG. 6 is a cross-sectional view schematically showing a configuration and a resist pattern formed by applying the phase shift mask according to the first embodiment of the present invention to a mixed pattern of a dense pattern and an isolated pattern.

FIG. 7 is a cross sectional view schematically showing a configuration and a resist pattern formed by applying the phase shift mask in the first embodiment of the present invention to a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 8 is a graph showing the relationship between the size of the light-shielding film on the wafer and the depth of focus for a Levenson-type phase shift mask and a normal mask.

FIG. 9 is a diagram for explaining the relationship between the size of the light shielding film and the exposure pattern.

FIG. 10 is a schematic diagram of an optical system that realizes an exposure method using the phase shift mask in the first embodiment of the present invention.

FIG. 11 is a plan view of a specific circuit pattern having a dense pattern and an isolated pattern.

FIG. 12 is a plan view schematically showing the configuration of a diaphragm.

FIG. 13 is a sectional view schematically showing the structure of a phase shift mask according to a second embodiment of the present invention.

FIG. 14 is a schematic cross-sectional view showing a first step of a method for manufacturing a phase shift mask in the second embodiment of the present invention.

FIG. 15 is a schematic cross sectional view showing a second step of the method for manufacturing the phase shift mask in the second embodiment of the present invention.

FIG. 16 is a schematic cross sectional view showing a third step of the method for manufacturing the phase shift mask in the second embodiment of the present invention.

FIG. 17 is a cross sectional view schematically showing a configuration and a resist pattern formed by using the phase shift mask according to the second embodiment of the present invention in a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 18 is a cross sectional view schematically showing a configuration and a resist pattern formed by using the phase shift mask according to the second embodiment of the present invention in a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 19 is a sectional view schematically showing the structure of a phase shift mask according to a third embodiment of the present invention.

20 is a partial cross-sectional view showing a region D ₁ of FIG. 19 in an enlarged manner.

FIG. 21 is a schematic cross sectional view showing a first step of a method for manufacturing a phase shift mask in the third embodiment of the present invention.

FIG. 22 is a schematic cross sectional view showing a second step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 23 is a schematic cross sectional view showing a third step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 24 is a schematic sectional view showing a fourth step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 25 is a schematic cross sectional view showing a fifth step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 26 is a schematic cross sectional view showing a sixth step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 27 is a schematic cross sectional view showing a seventh step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 28 is a schematic cross sectional view showing an eighth step of the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 29 is a first step diagram for explaining a case where a residual defect occurs in the method for manufacturing the phase shift mask in the third example of the present invention.

FIG. 30 is a second step diagram for explaining the case where the remaining defect occurs in the method for manufacturing the phase shift mask in the third embodiment of the present invention.

FIG. 31 is a cross sectional view schematically showing a configuration and a resist pattern formed by using the phase shift mask according to the third embodiment of the present invention in a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 32 is a cross sectional view schematically showing a configuration and a resist pattern formed by using the phase shift mask according to the third embodiment of the present invention in a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 33 is a cross sectional view schematically showing a configuration of a phase shift mask according to a fourth embodiment of the present invention.

FIG. 34 is a partial cross-sectional view showing a region D ₂ of FIG. 33 in an enlarged manner.

FIG. 35 is a schematic sectional view showing a first step of a method for manufacturing a phase shift mask in the fourth example of the present invention.

FIG. 36 is a schematic cross sectional view showing a second step of the method for manufacturing the phase shift mask in the fourth embodiment of the present invention.

FIG. 37 is a schematic cross sectional view showing a third step of the method for manufacturing the phase shift mask in the fourth embodiment of the present invention.

FIG. 38 is a schematic cross sectional view showing a fourth step of the method for manufacturing the phase shift mask in the fourth embodiment of the present invention.

FIG. 39 is a schematic cross sectional view showing a fifth step of the method for manufacturing the phase shift mask in the fourth embodiment of the present invention.

FIG. 40 is a schematic cross sectional view showing a sixth step of the method for manufacturing the phase shift mask in the fourth embodiment of the present invention.

FIG. 41 is a schematic cross sectional view showing a seventh step of the method for manufacturing the phase shift mask in the fourth example of the present invention.

FIG. 42 is a schematic sectional view showing an eighth step of the method for manufacturing the phase shift mask in the fourth embodiment of the present invention.

FIG. 43 is a cross sectional view schematically showing a configuration and a resist pattern formed by using the phase shift mask according to the fourth embodiment of the present invention in a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 44 is a cross sectional view schematically showing a configuration and a resist pattern formed by using the phase shift mask according to the fourth embodiment of the present invention in a pattern in which a dense pattern and an isolated pattern are mixed.

FIG. 45 is a schematic diagram of an exposure method using a conventional phase shift mask.

FIG. 46 is a diagram for explaining a mask cross section, an electric field on the mask, and a light intensity on the wafer when a normal photomask is used.

FIG. 47 is a diagram for explaining a mask cross section, an electric field on the mask, and a light intensity on the wafer when a Levenson-type phase shift mask is used.

FIG. 48 is a diagram for explaining a mask cross section, an electric field on the mask, and a light intensity on the wafer when a halftone phase shift mask is used.

FIG. 49 is a sectional view schematically showing a configuration of a conventional phase shift mask.

FIG. 50 is a first method of manufacturing a conventional phase shift mask.
It is an outline sectional view showing a process.

FIG. 51 is a second conventional manufacturing method of a phase shift mask.
It is a schematic sectional drawing which shows a process.

FIG. 52 is a third conventional method for manufacturing a phase shift mask.
It is a schematic sectional drawing which shows a process.

FIG. 53 is a fourth conventional method for manufacturing a phase shift mask.
It is an outline sectional view showing a process.

FIG. 54 is a fifth conventional manufacturing method of a phase shift mask.
It is an outline sectional view showing a process.

FIG. 55 is a sixth method of manufacturing a conventional phase shift mask.
It is a schematic sectional drawing which shows a process.

FIG. 56 is a seventh conventional manufacturing method of a phase shift mask.
It is an outline sectional view showing a process.

FIG. 57 is an eighth conventional method for manufacturing a phase shift mask.
It is an outline sectional view showing a process.

FIG. 58 is a schematic cross-sectional view showing the first step of the method for manufacturing the phase shift mask shown in the prior art document.

FIG. 59 is a schematic cross sectional view showing a second step of the method for manufacturing the phase shift mask shown in the prior art document.

FIG. 60 is a schematic sectional view showing a third step of the method for manufacturing the phase shift mask shown in the prior art document.

FIG. 61 is a schematic sectional view showing a fourth step of the method for manufacturing the phase shift mask shown in the prior art document.

FIG. 62 is a first process chart for explaining an adverse effect in the manufacturing method of the phase shift mask shown in the prior art document.

FIG. 63 is a second process chart for explaining an adverse effect in the method of manufacturing the phase shift mask shown in the prior art document.

FIG. 64 is a first process chart for explaining the advantage of the conventional method of manufacturing a phase shift mask.

FIG. 65 is a second process chart for explaining the advantage of the conventional method of manufacturing a phase shift mask.

FIG. 66 is a sectional view schematically showing a configuration of a conventional halftone phase shift mask.

FIG. 67 is a diagram for explaining a harmful effect when an isolated pattern is formed by a conventional Levenson-type phase shift mask.

FIG. 68 is a diagram for explaining an adverse effect when a dense pattern is formed using a conventional halftone phase shift mask.

FIG. 69 is a schematic cross-sectional view for explaining that the adhesion of the light shielding film is reduced in the conventional phase shift mask.

FIG. 70 is a first step diagram for explaining an adverse effect caused by residual foreign matter in the conventional phase shift mask.

71A and 71B are second process charts for explaining an adverse effect caused by the residual foreign matter in the conventional phase shift mask.

FIG. 72 is a third process diagram for explaining the adverse effect of foreign matter remaining on the conventional phase shift mask.

FIG. 73 is a fourth step diagram for explaining a harmful effect caused by residual foreign matter in the conventional phase shift mask.

FIG. 74 is a schematic cross-sectional view for explaining an adverse effect of a conventional phase shift mask when foreign matter remains on the stepped portion of the light shielding film.

FIG. 75 is a first step diagram for explaining an adverse effect caused by a residual defect in the conventional phase shift mask.

FIG. 76 is a second step diagram for explaining a problem caused by a residual defect in the conventional phase shift mask.

FIG. 77 is a third process diagram for explaining the adverse effect of the remaining defects on the conventional phase shift mask.

FIG. 78 is a schematic cross-sectional view for explaining an adverse effect when a remaining defect partially occurs in the conventional phase shift mask.

FIG. 79 is a schematic plan view showing a configuration of a wiring layer patterned by using a phase shift mask in which residual defects are partially generated.

[Explanation of symbols]

1, 201, 301 Quartz substrates, 1a, 201a, 30
1a, 401a 1st light transmission part, 1b, 201b, 3
01b, 401b Second light transmitting portion, 3,203 Semi-light-shielding film, 303,403 Light-shielding film, 20, 220, 32
0, 420 phase shift mask, 401 substrate, 411
Quartz substrate, 413 Etching stopper film, 415
Silicon oxide film.

Claims (6)

(57) [Claims] 1. A first light transmitting portion that transmits exposure light, and an exposure light that is adjacent to the first light transmitting portion and has a phase different from the phase of the exposure light that transmits the first light transmitting portion. A substrate having a second light transmitting portion that transmits light, and a partial region of the first and second light transmitting portions that is located at a boundary portion between the first and second light transmitting portions that are adjacent to each other. The first light transmission part has a first transmission region and a first attenuated transmission region on which the semi-shielding film is formed. The intensity of the exposure light transmitted through the transmission region is the first
Intensity of the exposure light transmitted through the attenuated transmission region of, the second light transmission portion has a second transmission region and a second attenuated transmission region in which the semi-light-shielding film is formed, The intensity of the exposure light transmitted through the second transmission region is the second
Intensity of the exposure light transmitted through the attenuated transmission region of, and the transmittance of the semi-shielding film is 3% or more and 30% or less,
Phase shift mask. 2. The substrate has a step having a predetermined height formed by the surface of the first light transmitting portion and the surface of the second light transmitting portion, and the semi-shield film is the substrate. covering the stepped side walls of the step of the substrate has a height substantially the same radius of curvature such to concave shape of the step, according to claim 1
The phase shift mask described in 1. 3. The substrate has a first film, and a second film formed on the first film and made of a material having etching characteristics different from those of the first film, The phase shift mask according to claim 2, wherein the surface of the first light transmitting portion is formed of the surface of the first film, and the surface of the second light transmitting portion is formed of the surface of the second film. 4. The semi-shielding film has a difference between the phase of the exposure light before passing through the semi-shielding film and the phase of the exposing light after passing through the semi-shielding film of 36.
The phase shift mask according to claim 1, wherein the phase shift mask is formed to have an angle of 0 ° × n (n is 0 or a positive integer). 5. A step of emitting exposure light from a light source, a step of irradiating the phase shift mask with the exposure light, and a step of exposing the exposure light transmitted through the phase shift mask to a photoresist on a film to be etched to expose it. The phase shift mask includes a first light transmitting portion that transmits the exposure light, an exposure light that is adjacent to the first light transmitting portion, and that is transmitted through the first light transmitting portion. A substrate having a second light transmitting portion that transmits the exposure light in a phase different from the phase, and the first and second light transmitting portions that are located at the boundary between the first and second light transmitting portions that are adjacent to each other. A semi-light-shielding film formed in a partial region of the light-transmitting portion, wherein the first light-transmitting portion includes a first transmission region and a first attenuated transmission region in which the semi-light-shielding film is formed. And the intensity of the exposure light transmitted through the first transmission region is equal to the first
Intensity of the exposure light transmitted through the attenuated transmission region of, the second light transmission portion has a second transmission region and a second attenuated transmission region in which the semi-light-shielding film is formed, The intensity of the exposure light transmitted through the second transmission region is the second
The exposure method is greater than the intensity of the exposure light transmitted through the attenuated transmission region, and the transmissivity of the semi-shielding film is 3% or more and 30% or less. 6. A first light transmitting portion which transmits the exposure light, and an exposure which is adjacent to the first light transmitting portion and which has a phase different from the phase of the exposure light which transmits the first light transmitting portion. A step of forming a substrate having a second light transmitting portion that transmits light; and a step of forming the substrate at a boundary portion between the first and second light transmitting portions adjacent to each other, and the first and second light transmitting portions. And a step of forming a semi-light-shielding film extending in a partial region of the first light transmitting portion,
A first attenuating / transmitting region on which the semi-light-shielding film is formed, and the intensity of exposure light transmitted through the first transmitting region is higher than the intensity of exposure light passing through the first attenuating transmitting region. The second light transmission portion has a second transmission region and a second attenuated transmission region on which a semi-shield film is formed, and the intensity of the exposure light transmitted through the second transmission region is the second. Which is higher than the intensity of the exposure light transmitted through the attenuated transmission region, and the semi-light-shielding film is formed so that the transmittance is 3% or more and 30% or less.