阅读说明：本技术 掩膜板坯料、相移掩膜板及半导体器件的制造方法 (Mask blank, phase shift mask and manufacturing method of semiconductor device ) 是由 野泽顺 宍戸博明 酒井和也 于 2014-01-14 设计创作，主要内容包括：一种掩膜板坯料、相移掩膜板及其它们的制造方法,根据该掩膜板坯料,即使在形成相移膜的材料中使用硅类材料的情况下,该相移膜在面内及膜厚方向上的组成以及光学特性的均匀性也较高,多个基板间的相移膜的组成以及光学特性的均匀性也较高,进而缺陷也较少。该掩膜板坯料的特征在于,在透光性基板上设有相移膜,所述相移膜以规定的透光率使ArF曝光光透过,且具有使透过的ArF曝光光产生规定量相移的功能,所述相移膜具有层叠了低透光层及高透光层的结构,所述低透光层及所述高透光层由以硅及氮形成的材料形成,或者由在该材料中含有从半金属元素、非金属元素以及稀有气体中选择的一种以上元素的材料形成,所述低透光层与所述高透光层相比,氮含量相对少。(A mask blank, a phase shift mask and a method for manufacturing the same, according to which even when a silicon-based material is used as a material for forming a phase shift film, the uniformity of the composition and optical characteristics of the phase shift film in the in-plane and film thickness directions is high, the uniformity of the composition and optical characteristics of the phase shift film between a plurality of substrates is high, and further, defects are small. The mask blank is characterized in that a phase shift film is provided on a light-transmitting substrate, the phase shift film transmitting ArF exposure light at a predetermined light transmittance and having a function of phase-shifting the transmitted ArF exposure light by a predetermined amount, the phase shift film having a structure in which a low light-transmitting layer and a high light-transmitting layer are laminated, the low light-transmitting layer and the high light-transmitting layer being formed of a material formed of silicon and nitrogen or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas, and the low light-transmitting layer having a relatively small nitrogen content as compared with the high light-transmitting layer.)

1. A mask blank having a phase shift film on a light-transmitting substrate, characterized in that,

The phase shift film transmits ArF exposure light at a light transmittance of 1% to 30%, and has a function of generating a phase difference of 170 degrees to 190 degrees between the ArF exposure light transmitted through the phase shift film and the ArF exposure light in the air having passed through the same distance as the thickness of the phase shift film,

The phase shift film has a nitrogen-containing layer formed of a material containing silicon and nitrogen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas, silicon, and nitrogen,

The nitrogen-containing layer has an oxygen content of not more than a lower limit value of detection by composition analysis by X-ray photoelectron spectroscopy,

The nitrogen-containing layer has a refractive index n of 2.5 or more with respect to ArF exposure light and an extinction coefficient k of 0.39 or more and less than 1.0.

2. A mask blank according to claim 1, wherein the nitrogen-containing layer is formed of a material formed of silicon and nitrogen.

3. A mask blank according to claim 1 or 2, wherein the phase shift film has an uppermost layer at a position farthest from the light-transmissive substrate, the uppermost layer being formed of a material formed of silicon, nitrogen, and oxygen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas, and silicon, nitrogen, and oxygen.

4. A mask blank according to claim 3, wherein the uppermost layer is formed of a material formed of silicon, nitrogen, and oxygen.

5. A mask blank according to claim 3 or 4, wherein the uppermost layer has a composition inclination in which the oxygen content gradually increases as it gradually moves away from the light transmissive substrate side.

6. A mask blank according to claim 1 or 2, wherein the phase shift film has an uppermost layer formed of a material formed of silicon and oxygen at a position farthest from the light-transmissive substrate.

7. A mask blank according to any one of claims 1 to 6, wherein the nitrogen-containing layer is bonded to the surface of the light-transmissive substrate.

8. A phase shift mask characterized in that the phase shift film of the mask blank according to any one of claims 1 to 7 is provided with a transfer pattern.

9. A method for manufacturing a semiconductor device, comprising the step of exposing and transferring the transfer pattern to a photoresist film on a semiconductor substrate by using the phase shift mask according to claim 8.

Technical Field

The invention relates to a mask blank, a phase shift mask manufactured by using the mask blank and a manufacturing method thereof.

Background

In general, in a manufacturing process of a semiconductor device, a fine pattern is formed by photolithography. The fine pattern is usually formed using a plurality of substrates called a transfer mask. In the process of miniaturizing the pattern of a semiconductor device, it is necessary to shorten the wavelength of an exposure light source used for photolithography in addition to miniaturizing the mask pattern formed on a transfer mask. In recent years, the wavelength of exposure light used for manufacturing semiconductor devices has been reduced from KrF excimer laser (wavelength 248nm) to ArF excimer laser (wavelength 193 nm).

The type of transfer mask is not limited to a binary mask having a light-shielding film pattern formed of a chromium-based material on a conventional translucent substrate, but may be a halftone type phase shift mask. Molybdenum silicide (MoSi) -based materials are widely used in phase shift films of halftone-type phase shift masks. However, as disclosed in patent document 1, in recent years, it has been found that MoSi-based films have low resistance to exposure light of ArF excimer laser light (i.e., ArF light resistance). In patent document 1, plasma treatment, UV irradiation treatment, or heat treatment is performed on the patterned MoSi-based film, and a passivation film is formed on the patterned surface of the MoSi-based film, thereby improving ArF light resistance.

Patent document 2 discloses a photomask blank having a transition metal silicon-based material film formed of a material containing a transition metal, silicon, oxygen, and nitrogen on a transparent substrate. In patent document 2, there is a technical problem that if exposure light of ArF excimer laser light (ArF exposure light) is irradiated to a patterned transition metal silicon material film for a long time, a line width of the pattern changes. On the other hand, patent document 3 discloses a phase shift mask having a phase shift film formed of SiNx.

Disclosure of Invention

Problems to be solved by the invention

In patent document 1, a passivation film is formed on the surface of a pattern formed of a MoSi-based film so as to improve ArF light resistance, but this method does not change the internal structure of the MoSi-based film. That is, the ArF light resistance inside the MoSi-based film was the same as before. Therefore, it is necessary to form a passivation film not only on the surface layer of the pattern upper surface but also on the surface layer of the sidewall of the MoSi-based film. In patent document 1, a passivation film is formed by patterning a MoSi-based film and then performing plasma treatment, UV irradiation treatment, or heat treatment. However, the patterns formed on the MoSi-based film often have large in-plane density differences, and the distance between the sidewalls between adjacent patterns also has large differences. Therefore, there is a problem that it is difficult to form a passivation film of the same thickness on all the pattern sidewalls.

The transition metal silicon-based material film in patent document 2 has an oxygen content of 3 atomic% or more, a silicon content and a transition metal content within a range satisfying a predetermined relational expression, and a surface oxide layer is provided on a surface layer of the transition metal silicon-based material film, whereby ArF light resistance can be improved. The ArF light resistance can be expected to be further improved than that of the conventional transition metal silicon-based material film. There is a strong assumption that the line width variation (broadening) of the pattern formed of the transition metal silicon-based material film, which occurs with the irradiation of the ArF excimer laser light, is caused because the transition metal in the film becomes unstable due to the light excitation caused by the irradiation of the ArF excimer laser light. Therefore, even in the transition metal silicon-based material film of patent document 2, since it contains a transition metal, there is a problem that it is difficult to sufficiently solve the light resistance problem of ArF.

In addition, the transition metal silicon-based material film in patent document 2 must contain oxygen. The inclusion of oxygen in the film results in a significant increase in the transmittance thereof to ArF exposure light over that of nitrogen. Therefore, there is a problem that the thickness of the film is inevitably increased as compared with a transition metal silicon-based material film (for example, MoSiN or the like) containing no oxygen.

on the other hand, patent document 3 describes a film formed of SiNx containing no transition metal, but the inventors of the present invention verified that after forming a pattern on the SiNx film and attempting to irradiate it with ArF excimer laser light for a long time, it was confirmed that the pattern width change (broadening) was very small compared to the transition metal silicon-based material film, and it was a film having high ArF light resistance. However, it is difficult to apply the SiNx film to a halftone type phase shift film suitable for ArF exposure light.

a halftone type phase shift film (hereinafter, simply referred to as a "phase shift film") is required to have a function of transmitting ArF exposure light at a predetermined light transmittance and generating a predetermined phase difference between the ArF exposure light transmitted through the phase shift film and light in the air that has passed a distance equal to the thickness of the phase shift film. When the phase shift film is formed as a single layer, it is necessary to use a material having a refractive index n to some extent higher than that of ArF exposure light and a small extinction coefficient k to some extent. Silicon is a material whose extinction coefficient k to ArF exposure light is somewhat large, but whose refractive index n is small. The transition metal is a material having a large extinction coefficient k to ArF exposure light and a large refractive index n. When oxygen is contained in the film material, the extinction coefficient k to ArF exposure light is greatly reduced, and the refractive index n also tends to be reduced. When the film material contains nitrogen, the extinction coefficient k to ArF exposure light tends to decrease, but the refractive index n tends to increase.

Since each element has the optical characteristics as described above, when a phase shift film is formed from a conventional transition metal silicon-based material, since the transition metal is contained and the transition metal has a large refractive index n and a large extinction coefficient k, even if oxygen is contained to some extent or the nitrogen content is reduced to some extent, predetermined light transmittance and retardation can be ensured. In contrast, in the case of forming a phase shift film from a silicon-based material containing no transition metal, since silicon is a material having a small refractive index n, it is necessary to contain nitrogen, which is an element that increases the refractive index, in a larger amount than the conventional transition metal silicon-based material. Further, the large amount of nitrogen contained means that the light transmittance of the phase shift film is increased, so that it is necessary to reduce the oxygen content in the phase shift film as much as possible. Therefore, if an attempt is made to form a phase shift film having a single-layer structure from a silicon-based material containing no transition metal, such as SiNx, there are many limitations compared to the conventional phase shift film.

In general, not limited to the phase shift film, a thin film for forming a mask blank pattern is formed by a sputtering method. When a thin film is formed on a light-transmissive substrate by a sputtering method, conditions are generally selected under which the film formation can be relatively stable. For example, in the case of forming a SiNx film by a sputtering method, a Si target is disposed in a film forming chamber, a mixed gas of a rare gas such as Ar and nitrogen is continuously circulated, and Si particles sputtered by collision of the rare gas turned into plasma with the Si target trap nitrogen in the middle and deposit on a light-transmissive substrate (such sputtering is generally referred to as "reactive sputtering"). The nitrogen content of the SiNx film is mainly adjusted by increasing or decreasing the mixing ratio of nitrogen in the mixed gas, and thereby SiNx films having various nitrogen contents can be formed on the light-transmissive substrate.

However, when a SiNx film is formed by reactive sputtering, if the mixing ratio of nitrogen in the mixed gas is such that Si ₃ N ₄ which is stable in terms of chemical quantity can be formed or a film having a nitrogen content close to that of Si can be formed (such a film formation condition region is referred to as "poisoning mode" or "reaction mode", see fig. 4), the film can be formed relatively stably.

Accordingly, the present invention has been made to solve the conventional problems, and an object of the present invention is to provide a mask blank in which a phase shift film is provided on a light-transmissive substrate, and in which uniformity of composition and optical characteristics of the phase shift film in an in-plane direction and a film thickness direction is improved, uniformity of composition and optical characteristics of the phase shift film among a plurality of substrates is improved, and defects are reduced, even when a silicon-based material not containing a transition metal that causes deterioration of ArF light resistance is used as a material for forming the phase shift film. The present invention also provides a phase shift mask manufactured using the mask blank. Further, the present invention is directed to a method for manufacturing the mask blank and the phase shift mask.

Means for solving the problems

In order to solve the above problems, the present invention has the following configuration.

A first aspect of the present invention is a mask blank in which a translucent substrate is provided with a phase shift film that transmits ArF exposure light at a predetermined light transmittance and has a function of phase-shifting the transmitted ArF exposure light by a predetermined amount,

The phase shift film has a structure in which a low light-transmitting layer and a high light-transmitting layer are laminated,

The low-light-transmitting layer and the high-light-transmitting layer are formed using a material formed using silicon and nitrogen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas,

The low light-transmitting layer has a relatively small nitrogen content as compared with the high light-transmitting layer.

The second aspect of the present invention is as follows. That is, the mask blank according to the first aspect is characterized in that the low-light-transmitting layer and the high-light-transmitting layer are formed of the same constituent element.

A third aspect of the present invention is as follows. That is, the mask blank according to the first or second aspect, wherein a set of laminated structures is formed by one layer of the low light-transmitting layer and one layer of the high light-transmitting layer, and the phase shift film has two or more sets of the laminated structures.

A fourth aspect of the present invention is as follows. That is, the mask blank according to any one of the first to third aspects is characterized in that the low-light-transmitting layer and the high-light-transmitting layer are formed of a material formed of silicon and nitrogen.

A fifth aspect of the present invention is as follows. That is, the mask blank according to any one of the first to fourth aspects, wherein the low light-transmitting layer is formed of a material having a refractive index n of less than 2.5 with respect to ArF exposure light and an extinction coefficient k of 1.0 or more,

The high-light-transmitting layer is formed of a material having a refractive index n of 2.5 or more with respect to ArF exposure light and an extinction coefficient k of less than 1.0.

A sixth aspect of the present invention is as follows. That is, the mask blank according to any one of the first to fifth aspects is characterized in that the thickness of each of the low-light-transmitting layer and the high-light-transmitting layer is 20nm or less.

The seventh aspect of the present invention is as follows. That is, the mask blank according to any one of the first to sixth aspects is characterized in that the phase shift film has an uppermost layer at a position farthest from the transparent substrate, and the uppermost layer is formed of a material containing silicon, nitrogen, and oxygen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas.

The eighth aspect of the present invention is as follows. That is, the mask blank according to the seventh aspect is characterized in that the uppermost layer is formed of a material formed of silicon, nitrogen, and oxygen.

A ninth aspect of the present invention is a phase shift mask blank according to any one of the first to eighth aspects, wherein a transfer pattern is formed on the phase shift film of the mask blank.

A tenth aspect of the present invention is a method for manufacturing a mask blank having a phase shift film that transmits ArF exposure light at a predetermined light transmittance and has a function of phase-shifting the transmitted ArF exposure light by a predetermined amount, the method being characterized in that,

The phase shift film has a structure in which a low light-transmitting layer and a high light-transmitting layer are laminated,

The manufacturing method comprises the following steps:

a low-light-transmitting layer forming step of forming the low-light-transmitting layer on the light-transmitting substrate by reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas, using a silicon target or a target formed of a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements;

And a high-light-transmitting layer forming step of forming the high-light-transmitting layer on the light-transmitting substrate by reactive sputtering in a sputtering gas using a silicon target or a target formed of a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements, wherein the sputtering gas in the high-light-transmitting layer forming step is a sputtering gas containing a nitrogen-based gas and a rare gas, and the mixing ratio of the nitrogen-based gas is higher than that in the low-light-transmitting layer forming step.

An eleventh aspect of the present invention is as follows. That is, the method of manufacturing a mask blank according to the tenth aspect is characterized in that the nitrogen-based gas mixture ratio selected as the sputtering gas used in the low light-transmitting layer forming step is smaller than the range of the nitrogen-based gas mixture ratio in the transition mode that tends to make film formation unstable;

The mixing ratio of the nitrogen-based gas selected as the sputtering gas used in the high-light-transmitting layer forming step is larger than the range of the mixing ratio of the nitrogen-based gas that causes the transition mode.

The twelfth aspect of the present invention is as follows. That is, the method for manufacturing a mask blank according to the tenth or eleventh aspect is characterized in that the low light-transmitting layer forming step forms the low light-transmitting layer by reactive sputtering in a sputtering gas formed of nitrogen gas and a rare gas using a silicon target; in the high-light-transmission layer forming step, the high-light-transmission layer is formed by reactive sputtering in a sputtering gas containing nitrogen and a rare gas, using a silicon target.

The thirteenth aspect of the present invention is as follows. That is, according to any one of the tenth to twelfth aspects, the low light-transmitting layer is formed of a material having a refractive index n of less than 2.5 with respect to ArF exposure light and an extinction coefficient k of 1.0 or more;

the high-light-transmitting layer is formed of a material having a refractive index n of 2.5 or more with respect to ArF exposure light and an extinction coefficient k of less than 1.0.

A fourteenth aspect of the present invention is the method for manufacturing a mask blank according to any one of the tenth to thirteenth aspects, including an uppermost layer forming step of forming an uppermost layer at a position of the phase shift film farthest from the transparent substrate by sputtering in a sputtering gas containing a rare gas using a silicon target or a target formed of a material containing silicon and containing one or more elements selected from semimetal elements and nonmetallic elements.

A fifteenth aspect of the present invention is the method for manufacturing a mask blank according to the twelfth aspect, including an uppermost layer forming step of forming an uppermost layer at a position of the phase shift film farthest from the transparent substrate by reactive sputtering in a sputtering gas formed of nitrogen gas and a rare gas using a silicon target, and performing oxidation treatment on at least a surface layer of the uppermost layer.

A sixteenth aspect of the present invention is a method for manufacturing a phase shift mask, including a step of forming a transfer pattern on the phase shift film of the mask blank manufactured by the method for manufacturing a mask blank according to any one of the tenth to fifteenth aspects.

ADVANTAGEOUS EFFECTS OF INVENTION

The mask blank of the present invention is a mask blank having a structure in which a low-light-transmitting layer and a high-light-transmitting layer are laminated, the low-light-transmitting layer and the high-light-transmitting layer being formed of a material formed of silicon and nitrogen or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas in the material, and the low-light-transmitting layer having a relatively small nitrogen content as compared with the high-light-transmitting layer. By forming the mask blank having the above configuration, it is possible to form a low light-transmitting layer made of a material having a small nitrogen content under a film-forming condition capable of forming a film under a stable film-forming condition by using a mixed gas having a small mixing ratio of nitrogen in a sputtering gas in reactive sputtering, and to form a high light-transmitting layer made of a material having a high nitrogen content under a film-forming condition capable of forming a film under a stable film-forming condition by using a mixed gas having a high mixing ratio of nitrogen in a sputtering gas in reactive sputtering. This can improve the uniformity of the composition and optical characteristics of the phase shift film in the in-plane and film thickness directions, and also can improve the uniformity of the composition and optical characteristics of the phase shift films between the plurality of substrates, thereby realizing a mask blank with fewer defects.

The present invention also relates to a method for producing a mask blank having a phase shift film provided on a light-transmissive substrate, the phase shift film transmitting ArF exposure light at a predetermined light transmittance and having a function of phase-shifting the ArF exposure light transmitted by a predetermined amount, the method comprising a low-light-transmissive-layer forming step of forming a low-light-transmissive layer on the light-transmissive substrate by reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas, using a silicon target or a target formed of a material containing one or more elements selected from semimetal elements and nonmetal elements in silicon, and a high-light-transmissive-layer forming step of forming a low-light-transmissive layer on the light-transmissive substrate by using the silicon target or the target formed of a material containing one or more elements selected from semimetal elements and nonmetal elements in silicon, the high-light-transmitting layer is formed on the light-transmitting substrate by reactive sputtering in a sputtering gas, the sputtering gas in the high-light-transmitting layer forming step is a sputtering gas containing a nitrogen-based gas and a rare gas, and the mixing ratio of the nitrogen-based gas is higher than that in the low-light-transmitting layer forming step. The above-described method for manufacturing a mask blank can form a low-light-transmitting layer made of a material having a small nitrogen content under a stable film-forming condition by using a mixed gas having a small mixing ratio of a nitrogen-based gas in a sputtering gas in reactive sputtering, and can form a high-light-transmitting layer made of a material having a high nitrogen content under a stable film-forming condition by using a mixed gas having a high mixing ratio of a nitrogen-based gas in a sputtering gas in reactive sputtering. This makes it possible to improve the uniformity of the composition and optical characteristics of the phase shift film in the in-plane and film thickness directions, and also to improve the uniformity of the composition and optical characteristics of the phase shift film between the plurality of substrates, thereby making it possible to manufacture a mask blank having fewer defects.

Drawings

Fig. 1 is a sectional view showing the structure of a mask blank according to the first and second embodiments;

Fig. 2 is a cross-sectional view showing another structure of the mask blank according to the first and second embodiments;

Fig. 3 is a sectional view showing a manufacturing process of a mask plate for transfer according to the first and second embodiments;

FIG. 4 is a schematic view for explaining a film formation mode in the case of forming a thin film in reactive sputtering;

Fig. 5 is a flowchart showing a method of manufacturing a mask blank according to a second embodiment;

FIG. 6 is a schematic view showing an example of an RF sputtering apparatus used in the manufacturing method of the second embodiment;

Fig. 7 is a sectional view showing the structure of a mask blank according to a third embodiment;

Fig. 8 is a schematic view showing an example of a film formation apparatus used in the method for manufacturing a mask blank according to the third embodiment;

Fig. 9 is a schematic view showing a positional relationship between a substrate and a sputtering target in a film formation apparatus used in the method for manufacturing a mask blank according to the third embodiment;

Fig. 10 is a sectional view showing the structure of a phase shift mask according to a third embodiment.

Detailed Description

First embodiment

Next, a first embodiment of the present invention will be explained.

The present inventors have intensively studied a method for forming a phase shift film from a silicon-based material film containing silicon and nitrogen and containing no transition metal, thereby realizing a film having high uniformity of composition and optical characteristics in the film thickness direction and having few defects. In the conventional film formation technology, it is necessary to use a film formation technology by reactive sputtering for forming a silicon-based material film containing silicon and nitrogen and containing no transition metal on a substrate in a state where uniformity of composition and optical characteristics is high. However, in the film formation of a thin film by reactive sputtering, a film formation rate and a voltage of the thin film often fluctuate due to a mixing ratio of a reactive gas in a film formation chamber.

Fig. 4 schematically shows a general trend of a curve of a change in a film forming rate caused when a mixing ratio of a reactive gas in a mixed gas of a rare gas and a reactive gas in a film forming chamber (or a flow rate ratio of the reactive gas in the mixed gas) is changed in the case of forming a thin film by reactive sputtering. Fig. 4 shows a change curve I of the film formation rate when the mixing ratio of the reactive gas in the mixed gas is gradually increased (increase mode), and a change curve D of the film formation rate when the mixing ratio of the reactive gas in the mixed gas is gradually decreased (decrease mode). Generally, in a region where the mixing ratio of the reactive gas in the mixed gas is low (a region of the metal mode M in fig. 4) and a region where the mixing ratio of the reactive gas in the mixed gas is high (a region of the reaction mode R in fig. 4), the width of change in the film formation rate is small with a change in the mixing ratio of the reactive gas in the mixed gas in both the increase mode and the decrease mode. Further, in the case where the mixing ratio of the reactive gas in the mixed gas is the same, the difference in film formation rate between the increase mode and the decrease mode is also small. Therefore, the thin film can be stably formed in the metal mode M region and the reaction mode R region. That is, the metal mode M region and the reaction mode R region can be said to be regions where a thin film having high uniformity of composition and optical characteristics and few defects can be formed.

On the other hand, in the transition mode T region sandwiched between the metal mode M region and the reaction mode R region in fig. 4, the width of change in film forming speed is large in both the increase mode and the decrease mode with a change in the mixing ratio of the reactive gas in the mixed gas. In addition, when the mixing ratio of the reactive gas in the mixed gas is the same, the difference in film formation rate between the increase mode and the decrease mode is also large. In the transition mode T region, a variation in the film formation rate due to a slight change in the mixing ratio of the reactive gas in the mixed gas in the film formation chamber increases, and the slight change in the mixing ratio causes a transition from the increase mode to the decrease mode, which also causes a variation in the film formation rate. Therefore, the thin film is formed in a state where the film forming rate is unstable. The variation in the film forming rate affects the amount of the reactive gas component contained in the thin film. That is, the transition mode T region can be said to be a region where it is difficult to form a thin film having high uniformity of composition and optical characteristics and few defects.

When a phase shift film having a single-layer structure formed of a silicon-based material film containing no transition metal is formed by reactive sputtering, the necessity of forming a film in the transition mode T region is high due to the limitation of required optical characteristics. There is also a method of searching for a combination of reactive gases that makes a difference in film formation rate between the increase mode and the decrease mode smaller in the transition mode T in the case where the mixing ratio of the reactive gases in the mixed gas is the same. However, even if such a combination of reactive gases is found, it is impossible to solve the problem that the range of variation in the film forming rate is large in the transition mode T according to the change in the mixing ratio of the reactive gases in the mixed gas.

In the case where a silicon-based material film containing silicon and nitrogen and containing no transition metal is formed by reactive sputtering in the metal mode region, if an attempt is made to ensure a film thickness capable of obtaining a phase difference required as a phase shift film, the extinction coefficient k of the formed film material is high, and therefore the light transmittance to ArF exposure light is lower than the required light transmittance to ArF exposure light. Such a film hardly produces a phase shift effect and is not suitable for a phase shift film. On the other hand, in the case where a silicon-based material film containing silicon and nitrogen and not containing a transition metal is formed by reactive sputtering in the reaction mode region, if an attempt is made to ensure a film thickness capable of obtaining a phase difference required as a phase shift film, the extinction coefficient k of the formed film material is low, and therefore the light transmittance to ArF exposure light is higher than that required. Although such a film can obtain a phase shift effect, the film is not suitable for use as a phase shift film because transmitted light from a pattern portion other than a region where the phase shift effect is generated may be sensitive to a photoresist film on a semiconductor wafer.

As a result of intensive studies on means for solving various technical problems occurring in the process of realizing a phase shift film suitable for ArF exposure light from a silicon-based material film containing silicon and nitrogen and not containing a transition metal, it has been concluded that the aforementioned technical problems can be solved if such a structure is adopted in which a phase shift film has a structure in which a low light-transmitting layer formed by reactive sputtering of a metal mode region and a high light-transmitting layer formed by reactive sputtering of a reaction mode region are laminated.

That is, the present invention relates to a mask blank having a phase shift film that transmits ArF exposure light at a predetermined light transmittance and has a function of phase-shifting the transmitted ArF exposure light by a predetermined amount, on a light-transmissive substrate. The mask blank of the present invention has the following features. That is, the phase shift film has a structure in which a low-light-transmitting layer and a high-light-transmitting layer are stacked. The low-light-transmitting layer and the high-light-transmitting layer are formed using a material formed using silicon and nitrogen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas. In addition, the low light-transmitting layer has a relatively small nitrogen content as compared with the high light-transmitting layer.

The present invention also relates to a method for manufacturing a mask blank having a phase shift film that transmits ArF exposure light at a predetermined light transmittance and has a function of phase-shifting the transmitted ArF exposure light by a predetermined amount, the phase shift film being provided on a light-transmissive substrate. The phase shift film has a structure in which a low light-transmitting layer and a high light-transmitting layer are laminated. The manufacturing method of the mask plate blank comprises a low light-transmitting layer forming process and a high light-transmitting layer forming process. In the low light-transmitting layer forming step, the low light-transmitting layer is formed on the light-transmitting substrate by reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas using a silicon target or a target formed of a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements. In the high-light-transmitting layer forming step, the high-light-transmitting layer is formed on the light-transmitting substrate by reactive sputtering in a sputtering gas using a silicon target or a target formed of a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements, the sputtering gas in the high-light-transmitting layer forming step is a sputtering gas containing a nitrogen-based gas and a rare gas, and the mixing ratio of the nitrogen-based gas is higher than that in the low-light-transmitting layer forming step.

In the method for manufacturing a mask blank, the nitrogen-based gas mixing ratio selected as the sputtering gas used in the low light-transmitting layer forming step is preferably smaller than the range of the nitrogen-based gas mixing ratio in the transition mode that tends to cause film formation instability. In the method for manufacturing a mask blank, the mixing ratio of nitrogen gas selected as the sputtering gas used in the high light-transmitting layer forming step is preferably larger than the range of the mixing ratio of nitrogen gas causing the transition mode.

The mask blank and the manufacturing method thereof form the phase shift film into a laminated structure of a low light-transmitting layer and a high light-transmitting layer instead of a single-layer structure. By forming such a laminated structure, the low light-transmitting layer can be formed by reactive sputtering in the metal pattern region where a film with a small nitrogen content tends to be formed, and the high light-transmitting layer can be formed by reactive sputtering in the reaction pattern region where a film with a large nitrogen content tends to be formed. Accordingly, both the low-light-transmitting layer and the high-light-transmitting layer can be formed by reactive sputtering under film formation conditions in which the film formation rate and the voltage variation are small during film formation, and as a result, a phase shift film having high uniformity of composition and optical characteristics and few defects can be formed.

The low-light-transmitting layer and the high-light-transmitting layer are formed using a material formed using silicon and nitrogen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas. The low-light-transmitting layer and the high-light-transmitting layer do not contain a transition metal which reduces the light resistance to ArF exposure light. It is preferable that the low-light-transmitting layer and the high-light-transmitting layer do not contain any metal element other than the transition metal. This is because the possibility that the light resistance to ArF exposure light is lowered by other metal elements than the transition metal cannot be denied. The low-light-transmitting layer and the high-light-transmitting layer may contain any semimetal element other than silicon. Among these semimetal elements, one or more elements selected from boron, germanium, antimony, and tellurium are preferably contained. This is because, by adding these elements to the sputtering target, the conductivity of the sputtering target can be expected to be improved.

₂The low-light-transmitting layer and the high-light-transmitting layer may contain any non-metal element, among which one or more elements selected from carbon, fluorine, and hydrogen are preferably contained, the oxygen content in the low-light-transmitting layer and the high-light-transmitting layer is preferably 10 atomic% or less, more preferably 5 atomic% or less, and oxygen is not actively contained (the detection lower limit value or less as a result of composition analysis of RBS, XPS, and the like).

In the target formed of a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements, the semimetal element preferably contains one or more elements selected from boron, germanium, antimony, and tellurium. Because these semimetal elements can be expected to improve the conductivity of the target material. Particularly in the case of forming the low-light-transmitting layer and the high-light-transmitting layer by the DC sputtering method, it is preferable to contain these semimetal elements in the target material in order to improve the conductivity of the target material.

The low-light-transmitting layer and the high-light-transmitting layer may contain a rare gas. The rare gas is an element which exists in the film forming chamber when a thin film is formed by reactive sputtering, and thereby the film forming speed can be increased and the productivity can be improved. The rare gas is converted into plasma and collides with the target, thereby sputtering the target constituent elements from the target. The sputtered target constituent elements trap the reactive gas halfway, and are deposited on the transparent substrate to form a thin film. A small amount of rare gas is incorporated into the film forming chamber until the target constituent elements are sputtered from the target and adhere to the transparent substrate. As the rare gas required for the reactive sputtering, argon, krypton, and xenon are preferably used. In addition, helium or neon with a small atomic weight is actively doped into the thin film in order to relax the stress of the thin film.

₂In the low light-transmitting layer forming step of forming the low light-transmitting layer of the phase shift film and the high light-transmitting layer forming step of forming the high light-transmitting layer, a nitrogen-based gas is contained in the sputtering gas, and any nitrogen-containing gas can be used as the nitrogen-based gas.

The present invention relates to a method for manufacturing a mask blank, wherein a phase shift film is provided on a light-transmitting substrate, the phase shift film transmits ArF exposure light at a predetermined light transmittance, and has a function of phase-shifting the transmitted ArF exposure light by a predetermined amount. The phase shift film has a structure in which a low light-transmitting layer and a high light-transmitting layer are laminated. The manufacturing method of the mask plate blank comprises a low light-transmitting layer forming process and a high light-transmitting layer forming process. In the low-light-transmitting layer forming step, the low-light-transmitting layer is formed on the light-transmitting substrate by reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas, using a silicon target or a target formed of a material in which silicon contains one or more elements selected from a semimetal element and a nonmetal element. The high-light-transmission layer forming step is a step of forming the high-light-transmission layer on the light-transmitting substrate by reactive sputtering in a sputtering gas using a silicon target or a target formed of a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements, wherein the sputtering gas in the high-light-transmission layer forming step is a sputtering gas containing a nitrogen-based gas and a rare gas, and the mixing ratio of the nitrogen-based gas is higher than that in the low-light-transmission layer forming step.

Preferably, the low-light-transmitting layer and the high-light-transmitting layer of the phase shift film are stacked in direct contact with each other without providing another film therebetween. In the mask blank of the present invention, it is preferable that any one of the low-light-transmitting layer and the high-light-transmitting layer has a film structure in which the film formed of a material containing a metal element is not in contact with the film. Because the metal element tends to easily diffuse into the silicon-containing thin film if heat treatment or irradiation with ArF exposure light is performed in a state where the silicon-containing film is in contact with the film containing the metal element.

Preferably, the low-light-transmitting layer and the high-light-transmitting layer are formed of the same constituent element. If any one of the low-light-transmitting layer and the high-light-transmitting layer contains a different constituent element, and if heat treatment or irradiation with ArF exposure light is performed in a state where these layers are in contact and stacked, the different constituent element may migrate and diffuse into the layer on the side not containing the constituent element. Therefore, the optical characteristics of the low-light-transmitting layer and the high-light-transmitting layer may be greatly changed from the initial stage of film formation. In particular, when the different constituent element is a semimetal element, it is necessary to form the low-light-transmitting layer and the high-light-transmitting layer using different targets.

In the mask blank of the present invention, as the material of the light-transmissive substrate, there can be mentioned, in addition to synthetic quartz glass, aluminosilicate glass, soda-lime glass, and low thermal expansion glass (SiO ₂ -TiO ₂ glass, etc.), synthetic quartz glass having a high light transmittance to ArF excimer laser light (wavelength of 193nm), and is particularly suitable as a material of the light-transmissive substrate for forming the mask blank.

In the case where the phase shift film is configured such that the low-light-transmitting layer and the high-light-transmitting layer are laminated in this order from the light-transmitting substrate side, since the low-light-transmitting layer is a silicon-containing film containing a small amount of nitrogen, there is an effect that etching selectivity is more easily obtained between the low-light-transmitting layer and the light-transmitting substrate formed of a material containing SiO ₂ as a main component, and further, when a pattern is formed in the silicon-containing film by dry etching, although an etching gas used in the dry etching is usually a fluorine-based gas, a chlorine-based gas may be used as an etching gas for the silicon-containing film containing a small amount of nitrogen.

On the other hand, in the case where the phase shift film is configured by laminating a high-light-transmitting layer and a low-light-transmitting layer in this order in contact with the light-transmitting substrate, the high-light-transmitting layer is a silicon-containing film containing a large amount of nitrogen, and therefore, in the case where the high-light-transmitting layer is formed in contact with the light-transmitting substrate made of a material containing SiO ₂ as a main component, there is an effect that high adhesion is easily obtained between the surface of the light-transmitting substrate and the high-light-transmitting layer.

Preferably, the low-light-transmitting layer and the high-light-transmitting layer of the phase shift film are stacked in direct contact with each other without providing another film therebetween. For the foregoing reason, it is preferable that the silicon containing film is not in a state of being in contact with a film formed of a material containing a metal element.

Preferably, a set of laminated structures is formed by one low-light-transmitting layer and one high-light-transmitting layer, and the phase shift film has more than two sets of the laminated structures. Further, it is preferable that both the low-light-transmitting layer and the high-light-transmitting layer have a thickness of 20nm or less. Since the difference in the required optical characteristics is large, the difference in the nitrogen content in the films of the low-light-transmitting layer and the high-light-transmitting layer is also large. Therefore, the difference in etching rate in dry etching by a fluorine-based gas increases between the low-light-transmitting layer and the high-light-transmitting layer. In the case where the phase shift film has a two-layer structure including one low-light-transmitting layer and one high-light-transmitting layer, when a pattern is formed by dry etching using a fluorine-based gas, a step is likely to occur in a cross section of the etched phase shift film. By providing the phase shift film with two or more sets of a laminated structure formed of one low-light-transmitting layer and one high-light-transmitting layer, the thickness of each layer (one layer) of the low-light-transmitting layer and the high-light-transmitting layer is thinner than that in the case of the above-described double-layer structure (one set of laminated structure). Therefore, the step generated on the pattern cross section of the phase shift film after etching can be reduced. Further, by limiting the thickness of each layer (one layer) of the low-light-transmitting layer and the high-light-transmitting layer to 20nm or less, steps occurring in the pattern cross section of the phase shift film after etching can be further suppressed.

In recent years, when a transfer pattern is formed on a thin film (phase shift film) of a mask blank by dry etching and a mask for transfer (phase shift mask) is manufactured, a black defect portion generated is often repaired by defect repair (EB defect repair) using electron beam irradiation, in which a substance in an unexcited state such as XeF ₂ is vaporized and supplied to the black defect portion, and an electron beam is irradiated to the black defect portion, thereby changing the thin film of the black defect portion into a volatile fluoride and removing the volatile fluoride.

The low-light-transmitting layer of the phase-shift film of the present invention is a silicon-based material film containing a small amount of nitrogen and not actively containing oxygen, and therefore, the low-light-transmitting layer has a tendency to be easily etched by a fluorine-based gas in a non-excited state such as XeF ₂ at the time of EB defect repair, and in order to avoid etching of the low-light-transmitting layer, the low-light-transmitting layer is preferably provided in a state of being hardly affected by a fluorine-based gas in a non-excited state such as XeF ₂, and on the other hand, since the high-light-transmitting layer is a silicon-based material film containing a large amount of nitrogen, the low-light-transmitting layer has a tendency of being hardly affected by a fluorine-based gas in a non-excited state such as XeF ₂, as described above, by combining a laminated structure in which the phase-shift film has two or more sets of low-light-transmitting layer and high-light-transmitting layer, the low-light-transmitting layer can be formed in a structure sandwiched between the two high-light-transmitting layer, and the low-light-transmitting layer can be etched by being in a state where a fluorine-based gas in which is hardly trapped in a state of being initially contacted with the low-light-transmitting layer (because the surface of the side wall of the low-light-transmitting layer is more easily trapped than a side wall of the low-light-transmitting layer, and the side wall of the low-light-transmitting layer ₂, and the side wall of the low-light-transmitting layer, the low-light-transmitting layer is etched.

Preferably, the low-light-transmitting layer and the high-light-transmitting layer are formed using a material formed using silicon and nitrogen. In the method for manufacturing a mask blank, in the step of forming the low light-transmitting layer, the low light-transmitting layer is preferably formed by reactive sputtering in a sputtering gas containing nitrogen gas and a rare gas using a silicon target. In the high light-transmitting layer forming step, the high light-transmitting layer is preferably formed by reactive sputtering in a sputtering gas containing nitrogen gas and a rare gas using a silicon target.

As described above, the inclusion of a transition metal in the low-light-transmitting layer and the high-light-transmitting layer results in a decrease in light resistance to ArF exposure light. When the low-light-transmitting layer and the high-light-transmitting layer contain a metal other than a transition metal or a semimetal element other than silicon, there is a possibility that the optical characteristics of the low-light-transmitting layer and the high-light-transmitting layer change as the contained metal or semimetal element moves between the low-light-transmitting layer and the high-light-transmitting layer. Further, if the low-light-transmitting layer and the high-light-transmitting layer contain a non-metal element, and further if the low-light-transmitting layer and the high-light-transmitting layer contain oxygen, the light transmittance to ArF exposure light is greatly reduced. In consideration of these factors, the low-light-transmitting layer and the high-light-transmitting layer are preferably formed of a material formed of silicon and nitrogen. The rare gas is an element that is difficult to detect even when the composition of the thin film is analyzed by RBS or XPS. Therefore, the material formed of silicon and nitrogen can be regarded as including a material further containing a rare gas.

Preferably, the low-light-transmitting layer is formed of a material having a refractive index n of less than 2.5 (preferably 2.4 or less, more preferably 2.2 or less, and even more preferably 2.0 or less) against ArF exposure light and an extinction coefficient k of 1.0 or more (preferably 1.1 or more, more preferably 1.4 or more, and even more preferably 1.6 or more). Preferably, the high-light-transmitting layer is formed of a material having a refractive index n of 2.5 or more (preferably 2.6 or more) with respect to ArF exposure light and an extinction coefficient k of less than 1.0 (preferably 0.9 or less, more preferably 0.7 or less, and still more preferably 0.4 or less). In the case where the phase shift film is formed of a laminated structure of two or more layers, the refractive index n and the extinction coefficient k of the low-light-transmitting layer and the high-light-transmitting layer must be within the above ranges, respectively. As a result, the phase shift film having a laminated structure of two or more layers satisfies the characteristics required as a phase shift film, that is, the characteristics of a predetermined phase difference and a predetermined light transmittance with respect to ArF exposure light.

the refractive index n and the extinction coefficient k of the film are not determined solely by the composition of the film. The film density, the crystal state, and the like of the thin film are also factors that affect the refractive index n and the extinction coefficient k. Therefore, various conditions for forming a thin film by reactive sputtering are adjusted so that the thin film has a desired refractive index n and extinction coefficient k after the film is formed. In order to make the low-light-transmitting layer and the high-light-transmitting layer fall within the above ranges of the refractive index n and the extinction coefficient k, the ratio of the mixed gas of the rare gas and the reactive gas may be adjusted in the case of film formation by reactive sputtering, but is not limited thereto. The film formation conditions that affect the refractive index n and the extinction coefficient k relate to various aspects such as the pressure in the film formation chamber, the power applied to the target, and the positional relationship such as the distance between the target and the transparent substrate when film formation is performed by reactive sputtering. These film formation conditions need to be optimized for each film formation apparatus and appropriately adjusted so that the thin film to be formed has a desired refractive index n and extinction coefficient k.

Preferably, the phase shift film has an uppermost layer at a position farthest from the light-transmissive substrate, and the uppermost layer is formed of a material including silicon, nitrogen, and oxygen, or a material including one or more elements selected from a semimetal element, a nonmetal element, and a rare gas. In the method for manufacturing a mask blank, it is preferable that the method includes an uppermost layer forming step of forming an uppermost layer at a position of the phase shift film farthest from the transparent substrate by sputtering in a sputtering gas containing a rare gas, using a silicon target or a target formed of a material containing silicon and containing at least one element selected from a semimetal element and a nonmetallic element. Further, in the method for manufacturing a mask blank, it is more preferable that the method further includes an uppermost layer forming step of forming an uppermost layer at a position of the phase shift film farthest from the transparent substrate by reactive sputtering in a sputtering gas containing nitrogen gas and a rare gas using a silicon target, and performing oxidation treatment on at least a surface layer of the uppermost layer.

Although the silicon-based material film containing nitrogen without actively containing oxygen has high light resistance against ArF exposure light, the silicon-based material film tends to have lower chemical resistance than the silicon-based material film containing oxygen actively. Further, when a mask blank having a high light-transmitting layer or a low light-transmitting layer which does not actively contain oxygen and contains nitrogen is disposed as the uppermost layer of the phase shift film on the side opposite to the light-transmitting substrate, the mask blank is cleaned and stored in the atmosphere, and therefore the surface layer of the phase shift film is sometimes oxidized, which is difficult to avoid. Since the surface layer of the phase shift film is oxidized, the optical characteristics of the phase shift film are largely changed from those at the time of film formation of the thin film. In particular, when a low-light-transmitting layer is provided as the uppermost layer of the phase shift film, the low-light-transmitting layer is oxidized, and thus the increase in light transmittance increases. The phase-shift film is further provided with an uppermost layer formed of a material formed of silicon, nitrogen, and oxygen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas, over the stacked structure of the low-light-transmitting layer and the high-light-transmitting layer, whereby surface oxidation of the low-light-transmitting layer and the high-light-transmitting layer can be suppressed.

_{2 2}The uppermost layer formed of a material composed of silicon, nitrogen, and oxygen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas in the material may have a structure having a substantially same composition in the layer thickness direction, or may have a structure having a composition inclined in the layer thickness direction (a structure having a composition inclined such that the oxygen content in the layer gradually increases as the uppermost layer gradually moves away from the light-transmissive substrate).

The uppermost layer forming step suitable for forming the uppermost layer may be performed by using a silicon target or a target formed of a material containing silicon containing one or more elements selected from a semimetal element and a nonmetal element, and forming the uppermost layer by reactive sputtering in a sputtering gas containing nitrogen, oxygen, and a rare gas, and may be performed by using a silicon target or a target formed of a material containing silicon dioxide (SiO ₂) or a material containing one or more elements selected from a semimetal element and a nonmetal element in silicon dioxide (SiO ₂) and performing reactive sputtering in a sputtering gas containing a rare gas.

In the uppermost layer forming step suitable for forming the uppermost layer, a silicon target or a target formed of a material containing silicon and at least one element selected from semimetal elements and nonmetallic elements is used, and the uppermost layer is formed by reactive sputtering in a sputtering gas containing nitrogen gas and a rare gas, and at least the surface layer of the uppermost layer is subjected to an oxidation treatment. The uppermost layer forming step can be applied to the formation of an uppermost layer having a composition inclined in the layer thickness direction in principle. In this case, the treatment for oxidizing the uppermost layer surface layer may be a heat treatment in an oxygen-containing gas such as air, or a treatment in which ozone or oxygen plasma is brought into contact with the uppermost layer.

The low light-transmitting layer, the high light-transmitting layer, and the uppermost layer in the phase shift film are formed by sputtering, and any sputtering method such as DC sputtering, RF sputtering, and ion beam sputtering can be used. When a target having low conductivity (a silicon target, a target containing no semimetal element or a silicon compound in a small amount, or the like) is used, RF sputtering or ion beam sputtering is preferably used. When a target having a low conductivity is used, RF sputtering is more preferably used in consideration of the film formation rate.

In the step of forming the low light-transmitting layer and the high light-transmitting layer in the phase-shift film by sputtering, the low light-transmitting layer and the high light-transmitting layer may be formed in the same film forming chamber, or the low light-transmitting layer and the high light-transmitting layer may be formed in different film forming chambers. In the case where the low-light-transmitting layer and the high-light-transmitting layer are formed in the same film forming chamber, the low-light-transmitting layer and the high-light-transmitting layer may be formed using the same target material, or the low-light-transmitting layer and the high-light-transmitting layer may be formed using different target materials, which may be used. In the case where the low-light-transmitting layer and the high-light-transmitting layer are formed in different film forming chambers, the film forming chambers are preferably connected to each other via another vacuum chamber, for example. In this case, it is preferable that the vacuum chamber is connected to a sample lock chamber through which the translucent substrate in the atmosphere is introduced into the vacuum chamber. Further, it is preferable that a transfer device (robot) for transferring the transparent substrate between the sample lock chamber, the vacuum chamber, and each film forming chamber is provided.

In order to effectively exert the phase shift effect, the transmittance of the phase shift film of the mask blank of the present invention to ArF exposure light is preferably 1% or more, and more preferably 2% or more. The transmittance of the phase shift film to ArF exposure light is preferably adjusted to 30% or less, more preferably 20% or less, and still more preferably 18% or less. Preferably, the phase shift film is adjusted so that a phase difference between the transmitted ArF exposure light and the light passing through the air corresponding to the thickness distance of the phase shift film is in a range of 170 to 190 degrees.

In the mask blank of the present invention, a light-shielding film is preferably laminated on the phase shift film. In general, in a transfer mask, it is required to ensure an Optical Density (OD) of a predetermined value or more in an outer peripheral region of a region where a transfer pattern is formed (transfer pattern forming region). This is to prevent the resist film from being affected by the exposure light transmitted through the outer peripheral region when the resist film on the semiconductor wafer is subjected to exposure transfer by the exposure apparatus. The same applies to the phase shift mask in this regard. Generally, on the outer peripheral region of the transfer mask including the phase shift mask, an OD of 3.0 or more is preferable, and an OD of at least 2.8 or more is necessary. As described above, the phase shift film has a function of transmitting exposure light at a predetermined light transmittance, and it is difficult to ensure an optical density of a predetermined value only by the phase shift film. Therefore, it is desirable to stack a light-shielding film for ensuring insufficient optical density on the phase shift film at the stage of manufacturing a mask blank. By forming the above-described mask blank, it is possible to manufacture a phase shift mask in which a predetermined optical density is secured in the outer peripheral region by removing the light shielding film in the region where the phase shift effect is used (basically, the transfer pattern forming region) in the process of manufacturing the phase shift film.

The light-shielding film may have a single-layer structure or a stacked structure of two or more layers. Each of the light-shielding films having a single-layer structure and a stacked structure of two or more layers may have substantially the same composition in the thickness direction of the film or layer, or may have a composition inclined in the layer thickness direction.

In the case where no other film exists between the light-shielding film and the phase-shift film, the light-shielding film must be made of a material having sufficient etching selectivity to an etching gas used when forming a pattern on the phase-shift film. In this case, the light-shielding film is preferably formed of a material containing chromium. As a material containing chromium for forming the light-shielding film, in addition to chromium metal, a material containing chromium and at least one element selected from oxygen, nitrogen, carbon, boron, and fluorine can be given. Generally, although the chromium-based material is etched in a mixed gas of a chlorine-based gas and oxygen, the etching rate of chromium metal by the etching gas is not so high. In consideration of increasing the etching rate of a mixed gas of a chlorine-based gas and an oxygen gas as an etching gas, it is preferable to use chromium containing at least one element selected from oxygen, nitrogen, carbon, boron, and fluorine as a material for forming the light-shielding film. The material containing chromium for forming the light-shielding film may contain one or more elements selected from molybdenum, indium, and tin. The etching rate of the mixed gas of the chlorine-based gas and the oxygen gas can be further increased by containing one or more elements selected from molybdenum, indium, and tin.

On the other hand, in the mask blank according to the present invention, when another film exists between the light-shielding film and the phase shift film, it is preferable that the other film is formed of the material containing chromium (etching stopper and etching mask) and the light-shielding film is formed of the material containing silicon. Although a material containing chromium is etched by a mixed gas of a chlorine-based gas and oxygen, a photoresist film made of an organic material is easily etched by the mixed gas. The silicon-containing material is generally etched in a fluorine-based gas or a chlorine-based gas. Since these etching gases do not contain oxygen, the film reduction amount of the resist film formed of an organic material can be reduced as compared with the case of etching with a mixed gas of a chlorine-based gas and oxygen. Therefore, the film thickness of the resist film can be reduced.

The material containing silicon which forms the light-shielding film may contain a transition metal, or may contain a metal element other than a transition metal. This is because, when a phase shift mask is produced from this mask blank, since the pattern formed on the light-shielding film is substantially a light-shielding band pattern in the outer peripheral region, the cumulative amount of irradiation with ArF exposure light is small compared with the transfer pattern formation region, and since the light-shielding film is left in a fine pattern relatively rarely, it is difficult to cause a practical problem even if ArF light resistance is low. The reason is that when the light-shielding film contains a transition metal, the light-shielding performance of the light-shielding film can be greatly improved and the thickness of the light-shielding film can be reduced as compared with the case where the light-shielding film does not contain a transition metal. Examples of the transition metal contained in the light-shielding film include any one of molybdenum (Mo), tantalum (Ta), tungsten (W), titanium (Ti), chromium (Cr), hafnium (Hf), nickel (Ni), vanadium (V), zirconium (Zr), ruthenium (Ru), rhodium (Rh), niobium (Nb), palladium (Pd), and the like, or an alloy of these metals.

the light-shielding film may be formed of a material containing tantalum (Ta) and one or more elements selected from hafnium (Hf) and zirconium (Zr), and containing no oxygen except for a surface layer thereof. In this case, the light-shielding film can be dry-etched by an etching gas containing a chlorine-based gas and not containing an oxygen gas, and has an etching selectivity with respect to a material for forming the phase-shift film in dry etching in an etching gas containing a fluorine-based gas.

If the above etching characteristics are taken into consideration, it is preferable to form the light-shielding film by a tantalum-hafnium alloy, a tantalum-zirconium alloy, a tantalum-hafnium-zirconium alloy, or a compound containing an element other than oxygen in these alloys. As other elements than oxygen contained in the light-shielding film in this case, elements such as nitrogen (N), carbon (C), hydrogen (H), and boron (B) can be exemplified. In this case, the material of the light-shielding film may contain an inert gas such as helium (He), argon (Ar), krypton (Kr), or xenon (Xe). In this case, the light-shielding film is made of a material that can be dry-etched in an etching gas containing a chlorine-based gas and not containing oxygen, and the etching gas is less likely to etch the phase shift film.

in the mask blank having the light-shielding film laminated on the phase shift film, preferably, an etching mask formed of a material having etching selectivity to an etching gas used when etching the light-shielding film is further laminated on the light-shielding film. Since the light-shielding film is required to have a function of securing a predetermined optical density, there is a limit to reduction in thickness thereof. The thickness of the etching mask is not limited to optical terms, as long as the etching mask can function until the dry etching for forming a pattern on the light-shielding film immediately below the etching mask is completed. Therefore, the thickness of the etching mask can be made thinner than the thickness of the light-shielding film. Further, the thickness of the photoresist film made of an organic material may be reduced to a larger thickness than the conventional one, as long as the thickness of the photoresist film can function as an etching mask during the time until the dry etching for forming a pattern on the etching mask is completed.

In addition, it is preferable that the etching mask in this case is formed of SiO ₂, SiN, SiON, or the like, and further, as a material of the etching mask in the case where the light-shielding film is formed of a material containing chromium, a material containing tantalum may be used in addition to the above-described material.

In the mask blank of the present invention, an etching stopper film formed of a material having etching selectivity to both the transparent substrate and the phase shift film (the material containing chromium, for example, Cr, CrN, CrC, CrO, CrON, CrC, etc.) may be formed between the transparent substrate and the phase shift film.

In the mask blank according to the present invention, it is preferable that a photoresist film of an organic material is formed in a film thickness of 100nm or less in contact with the surface of the etching mask. In the case of a fine pattern corresponding to DRAM hp32 generation, there is a case where an SRAF (Sub-resolution assist Feature) having a line width of 40nm is provided on a transfer pattern (phase shift pattern) to be formed on an etching mask. However, in this case, since the cross-sectional aspect ratio of the photoresist pattern can be reduced to 1: 2.5, therefore, the photoresist pattern can be inhibited from collapsing or falling off when the photoresist film is developed or cleaned. The thickness of the resist film is preferably 80nm or less.

the phase shift mask plate is characterized in that a transfer printing pattern is formed on a phase shift film of the mask plate blank. The method for manufacturing a phase shift mask according to the present invention includes a step of forming a transfer pattern on a phase shift film of a mask blank manufactured by the above-described manufacturing method. According to the phase shift mask plate of the present invention, since the material itself constituting the phase shift film is a material having high light resistance against ArF exposure light, it is possible to omit the processing for improving the light resistance of ArF after forming a pattern on the phase shift film.

Fig. 1 is a sectional view showing the structure of a mask blank 100 according to an embodiment of the present invention. The mask blank 100 has a structure in which a phase shift film 2, a light shielding film 3, and an etching mask 4 are laminated on a light-transmissive substrate 1, wherein the phase shift film 2 is formed by laminating a low light-transmissive layer 21, a high light-transmissive layer 22, and an uppermost layer 23 in this order. On the other hand, fig. 2 is a cross-sectional view showing the structure of a mask blank 101 according to another embodiment of the present invention. The mask blank 101 has a structure in which a phase shift film 2, a light shielding film 3, and an etching mask 4 are laminated on a light-transmissive substrate 1, wherein the phase shift film 2 is formed by laminating a high light-transmissive layer 22, a low light-transmissive layer 21, and an uppermost layer 23 in this order. The detailed description of each structure of the mask blanks 100 and 101 is as described above. Next, an example of a method for manufacturing a phase shift mask according to the present invention will be described based on the manufacturing process shown in fig. 3. In this example, a material containing chromium is used for the light-shielding film, and a material containing silicon is used for the etching mask.

First, a photoresist film is formed by a spin coating method next to the etching mask 4 of the mask blank 100. Then, the photoresist film is exposed to light to form a first pattern, which is a transfer pattern (phase shift pattern) to be formed on the phase shift film, and a predetermined process such as a development process is further performed to form a first photoresist pattern 5a having a phase shift pattern (see fig. 3 a). Next, dry etching using a fluorine-based gas is performed using the first photoresist pattern 5a as a mask, and a first pattern (etching mask pattern 4a) is formed on the etching mask 4 (see fig. 3 b).

Next, after the photoresist pattern 5a is removed, dry etching using a mixed gas of a chlorine-based gas and an oxygen gas is performed using the etching mask pattern 4a as a mask, and a first pattern (light-shielding film pattern 3a) is formed on the light-shielding film 3 (see fig. 3 c). Next, dry etching using a fluorine-based gas is performed using the light-shielding pattern 3a as a mask to form a first pattern (phase shift pattern 2a) on the phase shift film 2 and remove the etching mask pattern 4a at the same time (see fig. 3 d).

Next, a photoresist film is formed on the mask blank 100 by a spin coating method. Then, the photoresist film is exposed to light to form a second pattern that is a pattern (light-shielding pattern) to be formed on the light-shielding film, and a predetermined process such as a development process is further performed to form a second photoresist pattern 6b having the light-shielding pattern. Next, dry etching using a mixed gas of a chlorine-based gas and an oxygen gas is performed using the second photoresist pattern 6b as a mask, and a second pattern (light-shielding pattern 3b) is formed on the light-shielding film 3 (see fig. 3 (e)). Further, the second photoresist pattern 6b is removed, and a predetermined process such as cleaning is performed to obtain the phase shift mask 200 (see fig. 3 f).

The chlorine-based gas used in the dry etching is not particularly limited as long as it contains Cl, and examples thereof include Cl ₂, SiCl ₂, CHCl ₃, CH ₂ Cl ₂, CCl ₄, and BCl ₃, and examples thereof include F, and examples thereof include a fluorine-based gas used in the dry etching, and examples thereof include CHF ₃, CF ₄, C ₂ F ₆, C ₄ F ₈, and SF ₆.

The phase shift mask of the present invention can suppress the CD variation (broadening) of a phase shift pattern within a small range even after exposure light of ArF excimer laser light is cumulatively irradiated. Therefore, even if the phase shift mask after the cumulative irradiation is placed on the mask stage of the exposure apparatus using ArF excimer laser light as exposure light and the phase shift pattern is transferred by exposing the mask to the resist film on the semiconductor device, the pattern can be transferred to the resist film on the semiconductor device with a precision that sufficiently satisfies the design specification.

Further, a method for manufacturing a semiconductor device according to the present invention is characterized in that a pattern is transferred by exposing a photoresist film on a semiconductor substrate using the phase shift mask or a phase shift mask manufactured using the mask blank. Since the phase shift mask and the mask blank according to the present invention have the effects as described above, even when the phase shift mask of the present invention, which is obtained by accumulating exposure light for irradiation of ArF excimer laser light, is placed on the mask stage of an exposure apparatus using ArF excimer laser light as exposure light, and a phase shift pattern is transferred by exposure to a resist film on a semiconductor device, the pattern can be transferred onto the resist film on the semiconductor device with a precision that sufficiently satisfies the design specifications. Therefore, when the circuit pattern is formed by dry etching the lower layer film using the pattern of the photoresist film as a mask, a high-precision circuit pattern can be formed without causing short-circuit and disconnection of the wiring due to insufficient precision.

Second embodiment

Next, a second embodiment of the present invention will be described.

As described above, in the case of forming a thin film by reactive sputtering, two stable sputtering modes occur depending on the flow rate of the reactive gas. These two modes are a metal mode in which the surface of the target is sputtered in the state of the target and a poisoning mode in which the surface of the target is sputtered in the state of reacting with a reactive gas, respectively. In the poisoning mode, the cathode applied voltage applied to the target is higher and the film formation rate is slower than in the metal mode. Therefore, if the flow rate of the reactive gas in the sputtering gas is increased or decreased, the cathode applied voltage and the film formation rate change rapidly at a certain flow rate, and an unstable transition occurs between the metal mode and the poisoning mode. Further, the transient flow rate value of the reactive gas is different between the increase (I in fig. 4) and the decrease (D in fig. 4), and if the relationship between the reactive gas concentration and the film formation rate is plotted, hysteresis (a phenomenon in which the curve does not match between the increase and the decrease in the flow rate) occurs (fig. 4). If the film is formed at a reactive gas flow rate corresponding to the hysteresis range, the voltage applied to the target and the film forming rate become unstable, and a film having stable physical properties cannot be formed.

As in patent document 3, when a film of a silicon compound is formed on a substrate by reactive sputtering using a silicon target material containing silicon as a target material, a film containing silicon at a high ratio in the composition of the film is formed in the metal mode, and a film containing a silicon compound in an amount close to the theoretical amount is formed in the poisoning mode. For example, when a silicon compound is formed on a mask blank substrate in the formation of a phase shift film for a halftone phase shift mask for ArF excimer laser light (wavelength λ 193nm), if the film is formed in a metal mode in which the nitrogen flow rate of the reactive gas is low, a film having a large extinction coefficient (k) is formed. Therefore, the phase shift film formed in the metal mode has too low light transmittance, and thus interference effect due to phase difference or the like cannot be obtained in practice.

On the other hand, if the film is formed in a poisoning mode in which the nitrogen flow rate of the reactive gas is high, a film having a low extinction coefficient (k) is formed on the substrate. Therefore, the phase shift film formed in the poisoning mode has too high light transmittance, and thus cannot obtain a necessary light-shielding property. In order to obtain the function of the phase shift mask, it is required that the extinction coefficient (k) of the phase shift film is located between the film formed in the poisoning mode and the film formed in the metal mode. However, in order to obtain a film having an extinction coefficient (k) in between, it is necessary to form the film at an unstable transient gas flow rate, which is a problem.

An object of the second embodiment is to provide a method for manufacturing a mask blank and the like that can solve the above-described problems and the problems described later related to the problems. In addition, an object of the second embodiment is to provide a method for manufacturing a mask blank, and particularly, a mask blank having a phase shift film on a translucent substrate, in which even when a silicon-based material containing no transition metal is used as the phase shift film, the uniformity of the composition and the optical characteristics of the phase shift film in the in-plane direction and in the film thickness direction is high, the uniformity of the composition and the optical characteristics of the phase shift film between a plurality of substrates is high, and defects are less. In addition, the second embodiment is also intended to provide a method for manufacturing a phase shift mask manufactured by using the mask blank obtained by the above-described manufacturing method.

In order to solve the above problem (problem), the invention of the second embodiment has the following technical means.

(technical solution 1A)

A method for manufacturing a mask blank, comprising a step of introducing an inert gas and a reactive gas into a vacuum chamber and forming a phase shift film having a laminated structure on a light-transmissive substrate by sputtering,

Disposing two or more target materials containing at least silicon in a vacuum container;

The step of forming the phase shift film includes: a step of forming a low-light-transmitting layer by sputtering one of the targets in a metal mode; a step of forming a high-transmittance layer by sputtering the other of the target materials in a poisoning mode; the method also includes a step of forming a phase shift film having a laminated structure in which a low-light-transmitting layer and a high-light-transmitting layer are laminated in different order through the above-described steps.

(technical solution 2A)

the method of manufacturing a mask blank according to claim 1A, wherein the reactive gas is the same gas as the reactive gas introduced when forming the low light-transmitting layer.

(technical solution 3A)

The method for manufacturing a mask blank according to claim 1A or 2A, wherein the target material is silicon.

(technical solution 4A)

the method according to any one of claims 1A to 3A, wherein the reactive gas is nitrogen.

(technical solution 5A)

The method for manufacturing a mask blank according to any one of claims 1A to 4A, wherein the mask blank is a mask blank,

The low-light-transmitting layer is a layer having a refractive index n of less than 2.4 and an extinction coefficient k of 1.0 or more with respect to exposure light of an ArF excimer laser;

The high-transmittance layer is a layer having a refractive index n of 2.4 or more with respect to exposure light of an ArF excimer laser and an extinction coefficient k of less than 1.0.

(technical solution 6A)

The method of manufacturing a mask blank according to any one of claims 1A to 5A, wherein a laminated structure in which the low-light-transmitting layer and the high-light-transmitting layer are laminated in different order is formed by alternately laminating the low-light-transmitting layer and the high-light-transmitting layer by three or more layers.

(technical solution 7A)

The method of manufacturing a mask blank according to any one of claims 1A to 6A, wherein an oxide layer is formed on a surface of a phase shift film having a laminated structure in which the low light-transmitting layer and the high light-transmitting layer are laminated in different order.

(technical solution 8A)

A method for manufacturing a phase shift mask, characterized in that a transfer pattern is formed on the phase shift film of a mask blank manufactured by the mask blank manufacturing method according to any one of claims 1A to 7A.

According to the method of manufacturing a mask blank according to each of the second embodiments, it is possible to manufacture a mask blank having high uniformity of composition and optical characteristics of the phase shift film in the in-plane direction and in the film thickness direction, high uniformity of composition and optical characteristics of the phase shift film between the plurality of substrates, and less defects. Further, a phase shift mask is manufactured using the mask blank manufactured by the mask blank manufacturing method according to each aspect of the second embodiment, whereby a phase shift mask having excellent transferability can be obtained.

In each of the above-described embodiments of the second embodiment, the "target material" is not limited to containing other components as long as it contains silicon. As the target material used for forming the phase shift film of the halftone phase shift mask, a material containing silicon can be selected and used as appropriate. Examples of the target used for forming the phase shift film include a silicon target and a mixed target containing silicon. As the "silicon-containing mixed target" mentioned herein, a mixed target containing silicon and a transition metal (for example, silicon and molybdenum) can be cited. Examples of the transition metal in the mixed target material containing silicon and a transition metal include molybdenum, tantalum, tungsten, titanium, hafnium, nickel, vanadium, zirconium, niobium, palladium, ruthenium, rhodium, chromium, and the like. The target material preferably used in the second embodiment is a silicon target or a mixed target containing silicon, and the target material preferably used is silicon.

In the second embodiment, the "silicon" of the target material is actually a target material made of silicon. A silicon target material whose surface is naturally oxidized, a silicon target material containing an impurity element mixed at the time of manufacturing the target material, and a silicon target material containing a component such as an element added for the purpose of improving stability and sputtering efficiency are included in the concept of the target material made of "silicon" mentioned herein. Examples of the element to be added for the purpose of improving the sputtering yield include boron, germanium, antimony, and tellurium. Since the silicon target to which these elements are added can maintain the conductivity of the target surface during sputtering to a high level, it is preferably used when the film is formed by a DC sputtering method.

In each of the embodiments of the second embodiment, the "two targets" may be any target as long as both targets contain silicon, and the content of silicon and other factors such as other substances than silicon may be different. For example, a mixed target of metal and silicon may be used as the target used in the metal mode, and a silicon target may be used as the target used in the poisoning mode, and both may be disposed in the chamber. In the case where a silicon target is used for both targets, a substance imparting conductivity may be added only to the silicon target used in the medium mode.

Among the films formed by sputtering using these target materials, the films formed in the metal mode and the films formed in the poisoning mode have optical characteristics such as extinction coefficient and refractive index different from each other. For example, a plurality of targets made of the same material are prepared and used as a target for a poisoning mode and a target for a metal mode, respectively, and films formed in different modes are stacked in different order, whereby a thin film having a stacked structure with desired optical characteristics can be formed.

In each of the embodiments of the second embodiment, the "reactive gas" is not limited to a specific gas type as long as it is a reactive gas generally used in reactive sputtering in which a film formed of a reaction product is formed by reacting with a target material. Specific examples of the reactive gas include nitrogen, oxygen, nitrogen oxide, methane, and carbon dioxide.

In each of the embodiments of the second embodiment, examples of the gas that can be used as the "inert gas" include rare gases such as helium, argon, krypton, and xenon. The inert gas is converted into plasma and collides with the target, whereby the constituent elements of the target are sputtered from the target and react with the reactive gas. A thin film is formed on a light-transmitting substrate from a reaction product of a reactive gas and a constituent element of a target.

In each of the above-described embodiments, the "transparent substrate" may be, in addition to the synthetic quartz glass, aluminosilicate glass, soda-lime glass, low thermal expansion glass (SiO ₂ — TiO ₂ glass, etc.), and the like, and the synthetic quartz glass has high light transmittance to ArF excimer laser light (wavelength of 193nm), and is particularly suitable as a material for forming a transparent substrate of a mask blank.

In each of the above-described embodiments of the second embodiment, the "sputtering method" may be any of DC sputtering and RF sputtering. When a silicon target having low conductivity or a mixed target containing silicon is used, RF sputtering is preferably used. In consideration of the film formation rate, RF sputtering is also preferably employed.

In the phase shift film according to the second embodiment, the low-light-transmitting layer and the high-light-transmitting layer may be laminated in this order from the light-transmitting substrate side, and any layer may be formed first as long as a more suitable layer is a layer on the light-transmitting substrate side in accordance with the characteristics of the light-transmitting substrate and the layer in contact therewith. For example, when the low-light-transmitting layer can obtain better etching selectivity with respect to the substrate than the high-light-transmitting layer, the low-light-transmitting layer may be selected as a layer in contact with the substrate. In addition, in the case where the high light-transmitting layer has better adhesion to the substrate than the low light-transmitting layer, the high light-transmitting layer may be selected as a layer in contact with the substrate.

One of the features of each of the second embodiments is that two or more targets containing at least silicon are arranged and used separately as a target for film formation in the metal mode and a target for film formation in the poisoning mode. Thus, only one target is sputtered when forming either one of the low-light-transmitting layer and the high-light-transmitting layer. When a voltage is applied to one target, no voltage is applied to the other target.

When film formation in the metal mode and film formation in the poisoning mode are performed using the same target, it is necessary to carefully perform a target adjustment process before changing the physical properties (mode) of the target surface. However, the conditioning process of the target material may cause impurities to appear in the vacuum chamber. Impurities are one of the causes of film defects. Specific examples of such impurities include impurities deposited on the surface of the target material during mode change and impurities generated by mechanical operation of a shutter or the like used during adjustment processing. Considering these impurities, the number of target material adjustment processes is as small as possible to produce a mask blank with few defects.

According to the respective aspects of the second embodiment, since the target for film formation in the metal mode and the target for film formation in the poisoning mode are used separately, it is not necessary to frequently perform the target adjustment process in accordance with the mode change. According to the second embodiment, the generation of impurities as in the above example can be effectively suppressed at the time of forming the phase shift film. Therefore, a mask blank with a reduced number of defects due to impurities and a reduced number of defects can be manufactured.

One of the characteristics of the manufacturing method according to each of the second embodiments is that the low light-transmitting layer is formed in the metal mode and the high light-transmitting layer is formed in the poisoning mode. Since the target can stably perform reactive sputtering in both the poisoning mode and the metal mode, the uniformity of the composition and the optical characteristics of the phase-shift film in the in-plane direction and the film thickness direction can be improved, and the uniformity of the composition and the optical characteristics of the phase-shift film among the plurality of substrates can also be improved. Further, by combining the layer formed in the poisoning mode with the layer formed in the metal mode and adjusting the thickness of each layer, the degree of freedom in optical design of the phase shift film can be widened.

In the method for manufacturing a mask blank according to the second embodiment, the reactive gas is preferably the same as the reactive gas introduced in the formation of the low light-transmitting layer and the reactive gas introduced in the formation of the high light-transmitting layer. That is, the reactive gas is preferably the same kind of gas in forming the low light-transmitting layer and in forming the high light-transmitting layer.

If the kind of the reactive gas is the same when the film is formed in the poisoning mode and when the film is formed in the metal mode, the constituent element of the high light-transmitting layer obtained in the poisoning mode and the constituent element of the low light-transmitting layer obtained in the metal mode are also the same. Therefore, side reactions such as exchange reaction occurring at the interface between the high-light-transmitting layer and the low-light-transmitting layer are not complicated. Therefore, even in the case of a film having a laminated structure, the overall composition of the film can be uniformly secured, and unpredictable variations in the phase shift film function are unlikely to occur. In this embodiment, the reactive gas introduced when the low light-transmitting layer is formed in the metal mode and the reactive gas introduced when the high light-transmitting layer is formed in the poisoning mode may be the same type of gas, and the flow rate and the partial pressure of the introduced reactive gas and the pressure in the vacuum chamber may be different from each other.

In the method for manufacturing a mask blank according to the second embodiment, the material of the target is preferably silicon. I.e. the target is preferably a silicon target. According to this embodiment, the phase shift film having a laminated structure obtained is composed of a reaction product of silicon and a reactive gas, and does not contain a transition metal or the like. The phase-shifting portion obtained from the phase-shifting film formed in this manner is less likely to change the pattern size even when irradiated with short-wavelength exposure light for a long period of time, specifically, even when irradiated with exposure light of ArF excimer laser light having a wavelength of 193nm for a long period of time.

In the method for manufacturing a mask blank according to the second embodiment, the reactive gas is preferably nitrogen gas, "nitrogen gas" referred to herein means a gas capable of forming a nitride layer by reactive sputtering in practice, and specifically, "nitrogen gas" referred to herein is a gas in which nitrogen (N ₂) contained in the reactive gas is 90% by volume or more, preferably 99% by volume or more.

When a phase shift film which is translucent to exposure light of ArF excimer laser light is formed, an oxide, a nitride, an oxynitride, or the like can be selected. In the case of a silicon-based phase shift film, an oxide, a nitride, an oxynitride, or the like can be selected, but if an oxide or an oxynitride is selected, the refractive index (n) tends to decrease. Therefore, when oxygen is contained in the composition of the silicon-based phase shift film, the phase shift film needs to be formed thick. Further, when a silicon-based phase shift film containing oxygen is formed on a light-transmitting substrate made of synthetic quartz glass, there is a problem that it is difficult to obtain etching selectivity between the silicon-based phase shift film and the substrate in dry etching performed when forming a mask pattern. According to this aspect, a mask blank having a phase shift film which has excellent optical characteristics and excellent etching selectivity to a translucent substrate made of quartz glass can be manufactured.

In the method for producing a mask blank according to the second embodiment, the low-light-transmitting layer is preferably formed to have a refractive index (n) of less than 2.4 and an extinction coefficient (k) of 1.0 or more with respect to exposure light of ArF excimer laser light, and the high-light-transmitting layer is preferably formed to have a refractive index (n) of 2.4 or more and an extinction coefficient (k) of less than 1.0 with respect to exposure light of ArF excimer laser light. By laminating layers in which the refractive index (n) and extinction coefficient (k) of the low-light-transmitting layer and the high-light-transmitting layer are adjusted to be within the above ranges, a phase shift film having desired optical characteristics can be formed.

The refractive index (n) of the low-light-transmitting layer is preferably 2.3 or less, more preferably 2.2 or less, and still more preferably 2.0 or less. The extinction coefficient (k) of the low-light-transmitting layer is preferably 1.1 or more, more preferably 1.4 or more, and even more preferably 1.6 or more. The refractive index (n) of the high-light-transmitting layer is preferably 2.5 or more. The extinction coefficient (k) of the high-light-transmitting layer is preferably 0.9 or less, more preferably 0.7 or less, and still more preferably 0.4 or less. The refractive index (n) and the extinction coefficient (k) of the thin film are determined not only by the composition of the thin film but also vary depending on the film density and the crystal state of the thin film. Therefore, it is preferable to form a thin film with a desired refractive index (n) and extinction coefficient (k) by adjusting various conditions for forming the thin film by reactive sputtering.

In the method for manufacturing a mask blank according to the second embodiment, the laminated structure in which the low-light-transmitting layer and the high-light-transmitting layer are laminated in different order is preferably a structure in which the low-light-transmitting layer and the high-light-transmitting layer are alternately laminated by three or more layers. The low-light-transmitting layer and the high-light-transmitting layer have different composition ratios even when the elements constituting each layer are the same. Even if the films are composed of the same element, the etching rate in dry etching will be different if the composition ratio is different. For example, when a mask pattern is formed by dry etching using a fluorine-based gas or the like, if layers having different etching rates are stacked, a step is generated on a wall surface of the pattern. In order to suppress this phenomenon, it is effective to reduce the thickness of the stacked layers. According to this aspect, since the thickness of the layer constituting the phase shift film can be reduced (for example, 20nm or less), the step due to the difference in etching rate of each layer can be reduced.

In the method for manufacturing a mask blank according to the second embodiment, it is preferable that an oxide layer be formed on a surface of the phase shift film having a laminated structure in which the low-light-transmitting layer and the high-light-transmitting layer are laminated in different order. The "formation of an oxide layer" referred to herein means, for example, formation of an oxide layer by performing oxidation treatment after forming a phase shift film having a laminated structure in which a low-light-transmitting layer and a high-light-transmitting layer are laminated in different order by sputtering; and additionally forming a thin film of, for example, silicon oxide on a surface of the phase shift film of the laminated structure in which the low-light-transmitting layer and the high-light-transmitting layer are laminated in different order. Examples of the method of oxidizing the surface include a method of heating in the atmosphere or in an oxygen atmosphere, a method of bringing ozone or oxygen plasma into contact with the surface, and the like. As a method for separately forming an oxide layer on the surface of the phase shift film having the above-described laminated structure, sputtering deposition using silicon dioxide as a target material, or the like can be given.

The substrate having the thin film whose surface is not oxidized exposed is cleaned and stored in the atmosphere, so that the surface layer is easily oxidized. In the case of a phase shift film, the optical characteristics of the film may be largely changed by oxidation of the surface during film formation. Particularly in the case of a structure in which a low light-transmitting layer is provided on the uppermost layer of the phase shift film, the increase in light transmittance is increased by oxidation of the low light-transmitting layer. By forming an oxide layer that does not affect the optical characteristics of the phase shift film on (on) the upper surface (surface) of the laminated structure (different in order) of the low-light-transmitting layer and the high-light-transmitting layer, the surface oxidation of the low-light-transmitting layer and the high-light-transmitting layer can be suppressed.

next, a preferred embodiment is described as a method for manufacturing a mask blank and a method for manufacturing a phase shift mask according to the example of the second embodiment.

(mode 1)

The mask blank is provided with a phase shift film having a function of generating a prescribed amount of phase shift to exposure light of the ArF excimer laser. The transparent substrate is a synthetic quartz glass substrate.

(mode 2)

The layer of the phase shift film in contact with the light-transmissive substrate is a silicon nitride layer (low light-transmissive layer) sputtered in a metal mode. Although the synthetic quartz glass substrate can be dry-etched in the same etching gas as the silicon nitride layer, the silicon nitride layer having a small nitrogen content formed in the metal mode has a good etching selectivity because it has different etching characteristics from the synthetic quartz glass.

(mode 3)

The layer of the phase shift film in contact with the light-transmissive substrate is a high light-transmissive layer formed of silicon nitride sputtered in the poisoning mode. When a nitride film or an oxide film is formed on a synthetic quartz glass substrate by reactive sputtering, a film having a high degree of nitridation or oxidation has a characteristic of good adhesion. By forming a silicon nitride layer (high light-transmitting layer) having a high degree of nitridation, which is formed in the poisoning mode, as a layer in contact with the substrate, the adhesion between the phase shift film and the light-transmitting substrate can be improved.

(mode 4)

The low-light-transmitting layer and the high-light-transmitting layer are formed using silicon nitride films having different degrees of nitridation using respective targets. The degree of nitridation of the silicon nitride layer (high light-transmitting layer) is relatively greater than that of the silicon nitride layer (low light-transmitting layer). The degree of nitridation of the silicon nitride layer (high-light-transmitting layer) includes a case where the nitrogen content is more than 50 at%, and the remaining element is silicon. The degree of nitridation of the silicon nitride layer (low-light-transmitting layer) includes a case where the nitrogen content is less than 50 at%, and the remaining element is silicon.

(mode 5)

The sputtering apparatus used for forming the phase shift film is an RF sputtering apparatus. In DC sputtering, although it depends on the type of target, if reactive sputtering is performed in the poisoning mode, the conductivity of the target surface portion deteriorates and it is difficult to apply a voltage. Therefore, discharge may become unstable, and the high-light-transmitting layer may not be stably formed. In addition, ion beam sputtering has a problem of slow film formation rate. When the film is formed by RF sputtering in the poisoning mode, even a silicon target having a significantly reduced conductivity on the target surface can be stably formed at a relatively high film formation rate.

(mode 6)

As the conditions of the sputtering gas at the time of film formation in the metal mode and the poisoning mode, the relationship between the flow rate ratio of the inert gas and the reactive gas in the sputtering gas and the film formation rate was examined in advance in the single-wafer RF sputtering apparatus used, and film formation conditions such as the flow rate ratio capable of stable film formation in the two mode regions were selected.

(mode 7)

And adjusting the phase shift film of the phase shift mask blank to ensure that the transmittance of the ArF excimer laser exposure light is within the range of more than 1% and less than 30%. The phase shift film is adjusted so that the phase difference between the exposure light of the transmitted ArF excimer laser and the light passing through the air by a distance corresponding to the thickness of the phase shift film is within the range of 170 to 190 degrees. By manufacturing a mask blank in which the transmittance and the phase difference of the phase shift film are adjusted within the above ranges, a phase shift mask with good transfer accuracy can be manufactured.

(mode 8)

The phase shift film has a laminated structure of two or more sets of low light-transmitting layers and high light-transmitting layers. The thickness of each layer is 30nm or less. As described above, the low-light-transmitting layer and the high-light-transmitting layer have different element composition ratios of the constituent layers, and a difference in sputtering rate in dry sputtering is also increased. Therefore, if the mask pattern is formed by isotropic dry etching using a fluorine-based gas or the like, a step is formed on the pattern sidewall. By limiting the thickness of each of the low-light-transmitting layer and the high-light-transmitting layer to 30nm or less, preferably 20nm or less, steps formed on the side walls of the phase shift film pattern after etching can be further suppressed.

(mode 9)

In the production of a mask blank, a light-shielding film is formed on a phase shift film. The transfer mask is required to ensure an Optical Density (OD) of a predetermined value or more in an outer peripheral region of a region where a transfer pattern is formed (transfer pattern forming region). This is to prevent the resist film from being affected by the exposure light transmitted through the outer peripheral region when the resist film on the semiconductor wafer is subjected to exposure transfer by the exposure apparatus. The same applies to the phase shift mask in this regard. In general, in the outer peripheral region of a transfer mask including a phase shift mask, it is desirable that the OD be 3.0 or more and the OD must be at least 2.8 or more. As described above, the phase shift film has a function of transmitting exposure light at a predetermined light transmittance, and it is difficult to ensure an optical density of a predetermined value only by the phase shift film. Therefore, in order to secure an insufficient optical density, a light-shielding film is preferably laminated on the phase shift film at the stage of manufacturing the mask blank. By forming such a mask blank structure, a phase shift mask in which a predetermined optical density is secured in the peripheral region can be manufactured by removing the light shielding film in the region where the phase shift effect is used (basically, the transfer pattern forming region) in the process of manufacturing the phase shift film. The light-shielding film may have a single-layer structure or a stacked structure of two or more layers.

(mode 10)

The light-shielding film is formed of a material containing chromium, which is selective to an etching gas used when forming a pattern on the phase shift film. As a material containing chromium for forming the light-shielding film, in addition to metal chromium, a material containing chromium and at least one element selected from oxygen, nitrogen, carbon, boron, and fluorine may be mentioned. Generally, although a chromium-based material is etched in a mixed gas of a chlorine-based gas and oxygen, the etching rate of metallic chromium by the etching gas is not so high. In view of increasing the etching rate of an etching gas formed from a mixed gas of a chlorine-based gas and an oxygen gas, a material containing chromium and one or more elements selected from oxygen, nitrogen, carbon, boron, and fluorine is preferable as a material for forming the light-shielding film. The material containing chromium which forms the light-shielding film may contain one or more elements selected from tin, indium, and molybdenum.

(mode 11)

In the production of a mask blank, an etching mask is formed on a light-shielding film. The "etching mask" mentioned here is a thin film formed of a material having etching selectivity to an etching gas used when etching the light-shielding film. Since the light-shielding film must have a function of ensuring a prescribed optical density, there is a limit to reducing the thickness thereof. The thickness of the etching mask is not limited to optical terms, as long as the etching mask can function until the dry etching for forming a pattern on the light-shielding film immediately below the etching mask is completed. Therefore, the thickness of the etching mask can be made thinner than the thickness of the light-shielding film. Further, the thickness of the organic photoresist film described later can be reduced to a larger thickness than in the conventional case, as long as the film can function as an etching mask during the time from the end of dry etching for forming a pattern on the etching mask.

As a material of the etching mask in the case where the light-shielding film is formed of a material containing chromium, a silicon-based material can be given. The description of the silicon-based material will be described later. In the case where the light-shielding film is formed of a material containing chromium, a material containing tantalum may be used. In this case, the material containing tantalum may include, in addition to metal tantalum, a material in which tantalum contains one or more elements selected from nitrogen, oxygen, boron, silicon, and carbon. Examples thereof include Ta, TaN, TaON, TaBN, TaBON, TaCN, TaCON, TaBCN, TaBOCN, TaSi, TaSiN and TaSiON. Further, in the case where the light-shielding film is formed of a material containing silicon, the etching mask is preferably formed of a material containing chromium. Other matters related to the light-shielding film (including matters related to the laminated structure of the phase shift film) are the same as those of the light-shielding film of the first embodiment.

(mode 12)

Other matters of the etching mask are the same as those of the etching mask of the first embodiment.

(mode 13)

And forming a photoresist film made of an organic material on the surface of the etching mask. Other matters regarding the photoresist film are the same as those of the photoresist film of the first embodiment described above.

Third embodiment

Next, a third embodiment of the present invention will be described.

In the production of a general mask blank, a thin film such as a phase shift film is formed on the main surface of a transparent substrate by a single-wafer sputtering apparatus. In particular, in a mask blank and a phase shift mask having a phase shift film on a transparent substrate, it is important to ensure uniformity of the light transmittance and the phase shift amount of the phase shift film in the surface of the transparent substrate. Therefore, as disclosed in japanese patent application laid-open No. 2002-90978 (japanese patent application laid-open No.), when a phase shift film is formed on the main surface of a light-transmissive substrate, film formation is performed by an oblique incidence spin sputtering method in which the sputtering surface of a sputtering target is arranged to be inclined with respect to the main surface of the rotating light-transmissive substrate.

However, in the production of the mask blank, even if the main surface of the transparent substrate has minute scratches (defective portions), if the positions of the defective portions do not significantly hinder the formation of the transfer pattern, it is sometimes determined that the transparent substrate can be used to form a thin film and then prepare a mask blank. This is because, if information on the plane coordinates of the defective portion of the light-transmissive substrate and the type of defect (convex defect or concave defect) is added to the corresponding mask blank and transferred, the defective portion can be adjusted to a position that does not affect the exposure transfer when the transfer pattern is formed on the thin film.

However, it is known that when the phase shift film is formed by the oblique incidence spin sputtering method, if a minute recessed defect portion exists on the main surface of the transparent substrate, a low density region having a density lower than that of the surrounding region is locally formed on the phase shift film formed in the recessed defect portion. This is presumably caused by the fact that, in the deposition by the oblique incidence spin sputtering method, sputtered particles scattered from a sputtering target intrude from a direction inclined with respect to the perpendicular direction to the main surface of the transparent substrate, and the cross-sectional shape of the concave defect portion is often rounded, and it is considered that the particles reaching the transparent substrate have a difference in intrusion angle and speed, which results in the above-described result.

Such a low-density region has higher light transmittance than the surrounding region, and has different optical characteristics. When a phase shift mask is produced from such a mask blank having a concave defect portion on a transparent substrate, it has been thought that the concave defect portion is covered with a large-area phase shift film not using the phase difference, and thus the pattern design can prevent the concave defect portion from affecting the exposure transfer of the fine pattern. However, since it is difficult to reduce the light transmittance of the recessed defect portion to the same level as the light transmittance of the portion around which the transfer pattern is present, when the pattern is transferred by exposure to a transfer target (such as a resist film on a semiconductor substrate) using a completed phase shift mask, the portion may be exposed to the transfer target due to the high light transmittance of the phase shift film on the recessed defect portion, which may cause transfer failure.

An object of the third embodiment is to provide a mask blank and a phase shift mask in which even when a translucent substrate has a minute concave defect portion, white defects having higher light transmittance than the surroundings are not generated, and thus occurrence of transfer failure can be reduced. Further, the third embodiment provides a method for manufacturing the mask blank, a method for manufacturing a phase shift mask, and a method for manufacturing a semiconductor device.

As a method for solving the above problem, the third embodiment has the following means.

(technical solution 1B)

A mask plate blank is characterized in that,

A phase shift film is provided on a main surface of the light-transmissive substrate,

The light-transmitting substrate has a concave defect portion on a main surface on a side where the phase shift film is formed,

The phase shift film has a structure in which a high light-transmitting layer and a low light-transmitting layer having a light transmittance lower than that of the high light-transmitting layer are sequentially stacked from the light-transmitting substrate side,

an inner region of the high-light-transmitting layer formed in a portion above the recessed defect portion has a low-density region,

The density of the low-density region is relatively lower than the density of an inner region of the high-light-transmitting layer formed in a portion on the main surface without the recessed defect portion.

(technical means 2B)

The mask blank according to claim 1B, wherein the high light-transmitting layer and the low light-transmitting layer are formed of a material containing silicon and nitrogen, and the high light-transmitting layer has a relatively higher nitrogen content than the low light-transmitting layer.

(technical means 3B)

The mask blank according to claim 1B or 2B, wherein the high-light-transmitting layer and the low-light-transmitting layer are formed of a material formed of silicon and nitrogen, or a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas.

(technical means 4B)

A mask blank according to any one of claims 1B to 3B, wherein the high-light-transmitting layer and the low-light-transmitting layer are formed of the same constituent elements.

(technical solution 5B)

A mask blank according to any one of claims 1B to 4B, wherein the phase shift film has a combination of two or more sets of a laminated structure of the high light-transmitting layer and the low light-transmitting layer.

(technical solution 6B)

a mask blank according to any one of claims 1B to 5B, wherein the high-light-transmitting layer and the low-light-transmitting layer are formed of a material formed of silicon and nitrogen.

(technical solution 7B)

A mask blank according to any one of claims 1B to 6B, wherein the high light-transmitting layer has a film thickness greater than that of the low light-transmitting layer.

(technical solution 8B)

a mask blank according to any one of claims 1B to 7B, wherein the phase shift film has an uppermost layer formed of a material containing silicon and oxygen at a position farthest from the transparent substrate.

(technical solution 9B)

A mask blank according to any one of claims 1B to 8B, wherein the phase shift film has an uppermost layer at a position farthest from the transparent substrate, the uppermost layer being formed of any one of a material formed of silicon and oxygen, a material formed of silicon, nitrogen, and oxygen, and a material containing one or more elements selected from a semimetal element, a nonmetal element, and a rare gas in the materials.

(technical solution 10B)

The mask blank according to claim 8B or 9B, wherein the phase shift film formed on the recessed defect portion remains after the phase shift film is washed with warm water for 5 minutes or more.

(technical means 11B)

A phase shift mask characterized in that a transfer pattern is formed on the phase shift film of the mask blank according to any one of claims 1B to 10B.

(technical solution 12B)

A method for manufacturing a mask blank having a phase shift film on a main surface of a light-transmissive substrate, characterized in that,

A step of forming the phase shift film on the main surface of the transparent substrate having the recessed defect portion by a sputtering method in which the transparent substrate is rotated by a rotation axis passing through a center of the main surface and a sputtering surface of a sputtering target is arranged so as to be inclined with respect to the main surface of the transparent substrate having the recessed defect portion;

The step of forming the phase shift film includes: a high-light-transmitting layer forming step of forming a high-light-transmitting layer on the main surface of the light-transmitting substrate having the recessed defect portion, and a low-light-transmitting layer forming step of forming a low-light-transmitting layer having a lower light transmittance than the high-light-transmitting layer on the high-light-transmitting layer;

The high-light-transmitting layer formed in the high-light-transmitting layer forming step has a low-density region in an inner region of the high-light-transmitting layer in a portion formed on the recessed defect portion, the density of the low-density region being relatively lower than the density of an inner region of the high-light-transmitting layer in a portion formed on the main surface without the recessed defect portion.

(technical solution 13B)

The method for manufacturing a mask blank according to claim 12B, wherein the mask blank is a mask blank,

The high-light-transmission layer forming step of forming the high-light-transmission layer by reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas, using the sputtering target made of a material containing silicon;

The low-light-transmitting layer forming step is a step of forming the low-light-transmitting layer by reactive sputtering in a sputtering gas using the sputtering target made of a material containing silicon, wherein the sputtering gas in the low-light-transmitting layer forming step contains a nitrogen-based gas and a rare gas, and the mixing ratio of the nitrogen-based gas is lower than that in the high-light-transmitting layer forming step.

(technical solution 14B)

the method for manufacturing a mask blank according to claim 12B or 13B, wherein the mask blank is a mask blank,

The high-light-transmitting layer forming step of forming the high-light-transmitting layer by reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas, using the sputtering target formed of silicon or a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements;

The low-light-transmitting layer forming step is a step of forming the low-light-transmitting layer by reactive sputtering in a sputtering gas using the sputtering target formed of silicon or a material in which silicon contains one or more elements selected from semimetal elements and nonmetal elements, wherein the sputtering gas in the low-light-transmitting layer forming step contains a nitrogen-based gas and a rare gas, and the mixing ratio of the nitrogen-based gas is lower than that in the high-light-transmitting layer forming step.

(technical means 15B)

The method for manufacturing a mask blank according to any one of claims 12B to 14B, wherein,

The high-light-transmission layer forming step of forming the high-light-transmission layer by reactive sputtering in a sputtering gas made of nitrogen gas and a rare gas using the sputtering target made of silicon;

The low-light-transmitting layer forming step is a step of forming the low-light-transmitting layer by reactive sputtering in a sputtering gas using the sputtering target made of silicon, the sputtering gas in the low-light-transmitting layer forming step is made of nitrogen gas and a rare gas, and the mixing ratio of the nitrogen-based gas is lower than that in the high-light-transmitting layer forming step.

(technical solution 16B)

The method for manufacturing a mask blank according to any one of claims 12B to 15B, wherein the mask blank is a mask blank,

The high-light-transmitting layer forming step forms the high-light-transmitting layer by reactive sputtering in a poisoning mode;

The low-light-transmitting layer forming step forms the low-light-transmitting layer by reactive sputtering in a metal mode.

(technical solution 17B)

The method for manufacturing a mask blank according to any one of claims 12B to 16B, wherein,

The method includes an uppermost layer forming step of forming an uppermost layer made of a material containing silicon and oxygen at a position of the phase shift film farthest from the light-transmissive substrate.

(technical solution 18B)

The method for manufacturing a mask blank according to any one of claims 12B to 17B, wherein,

The method includes an uppermost layer forming step of forming an uppermost layer of a material including silicon and oxygen, a material including silicon, nitrogen, and oxygen, or a material including one or more elements selected from a semimetal element, a nonmetal element, and a rare gas in a position of the phase shift film farthest from the transparent substrate.

(technical means 19B)

A method for manufacturing a phase shift mask, comprising a step of forming a transfer pattern on the phase shift film of a mask blank manufactured by the method for manufacturing a mask blank according to any one of claims 12B to 18B.

(technical means 20B)

A method for manufacturing a semiconductor device, characterized by comprising a step of exposing a photoresist film on a substrate to transfer the transfer pattern of the phase shift mask plate using the phase shift mask plate according to claim 11B.

(technical solution 21B)

A method for manufacturing a semiconductor device, comprising a step of exposing a resist film on a substrate to transfer the transfer pattern of the phase shift mask manufactured by the method for manufacturing a phase shift mask according to claim 19B.

according to the aspects of the third embodiment, the low-density region on the recessed defect portion is formed in the high-light-transmitting layer provided on the light-transmitting substrate side, whereby the light transmittance of the phase shift film can be controlled mainly by the low-light-transmitting layer not including the low-density region. Therefore, the uniformity of the light transmittance of the phase shift film on the light-transmitting substrate having the concave defect portion in the plane can be improved, and thereby the formation of white defects in the mask blank and the phase shift mask can be suppressed, and the occurrence of transfer failure using the phase shift mask can be prevented.

In the production of a mask blank, in the process of forming a thin film on the main surface of a transparent substrate, film formation is performed by an oblique incidence rotary sputtering method in which a sputtering surface of a sputtering target is arranged to face the main surface of the rotating transparent substrate while being inclined.

However, in a mask blank obtained by forming a thin film of a transition metal silicide-based material typified by MoSi-based material by such an oblique incidence spin sputtering method, it has been found that pinhole defects having a minute size (about 100 nm) may occur on the main surface of the thin film in the process of producing a transfer mask using the mask blank. Further, it has been found that, although defects are not detected in the mask defect inspection immediately after the transfer mask is manufactured, if the transfer mask is set in an exposure apparatus and the mask defects are inspected after the mask is cleaned after a prescribed number of times of use, pinhole defects of minute sizes are sometimes found to be generated on the main surface of the thin film as well. In either case, such pinhole defects are not detected when a defect inspection is performed on a mask blank manufactured by forming a thin film.

The inventors of the present invention conducted intensive studies on the cause of the pinhole defect. As a result, it was found that, in all mask blanks in which the pinhole defects were generated, scratches (concave defects) having a minute depth (40nm or less) were present on the main surface of the transparent substrate directly below the pinhole defect portion. Next, a cross-sectional TEM image of the concave defect portion was observed with respect to a mask blank produced using a translucent substrate having a minute concave defect. As a result, it was confirmed that film formation unevenness occurred in the thin film on the concave defect portion.

Further, as a result of examining the film formation unevenness, it was found that the density of the film formation unevenness was lower than that of the other portions of the thin film. Further, it was confirmed that a region having a low density (low density region) in which the film formation unevenness occurred had a shape extending from the transparent substrate side to the front surface side of the thin film in cross section, and further extended from the vicinity of the outer periphery of the concave defect portion to the center side in plan view.

The reason why the low density region is formed is presumably that, in the film formation of a thin film by the oblique incidence spin sputtering method, since sputtered particles scattered from a sputtering target material intrude from a direction inclined with respect to the perpendicular direction to the main surface of the transparent substrate and the cross-sectional shape of the concave defect portion is often rounded, the particles reaching the transparent substrate have a difference in intrusion angle or speed.

it is problematic here that the low-density region of the above-described film becomes a region having higher light transmittance than the surrounding film portion. Therefore, even if the thin film on the light-transmissive substrate is made of a material that does not cause a pinhole defect, for example, it is difficult to suppress the light transmittance of the concave defect portion to the same degree as the light transmittance of the surroundings thereof while securing a predetermined phase shift amount, as long as the thin film is a phase shift film. Therefore, in a phase shift mask plate having a concave defect portion on a light-transmitting substrate, even in a pattern design in which the concave defect portion is covered with a large-area phase shift film not using a phase difference, for example, it is difficult to suppress the light transmittance of the concave defect portion to be as low as the light transmittance of a portion around where a transfer pattern is present, and there is a risk of occurrence of a white defect.

Therefore, in each mask blank according to the third embodiment, the phase shift film on the main surface of the light-transmissive substrate has a structure in which a high light-transmissive layer and a low light-transmissive layer are stacked in this order from the light-transmissive substrate side. Since the high light-transmitting layer is originally a layer having a high light transmittance, even if a local low-density region is formed in the high light-transmitting layer, the degree of increase in the light transmittance of the phase shift film as a whole in the film thickness direction is low, and even if a phase shift mask is produced using a mask blank having the phase shift film, the influence of the low-density region on the exposure transfer of the pattern can be suppressed to a small degree.

When a phase shift mask is produced from a mask blank, pattern arrangement is generally performed so that a phase shift film on a concave defect portion of a transparent substrate does not become a region where a fine transfer pattern is to be formed. Therefore, even if the amount of phase shift of the phase shift film on the concave defect portion is not within a desired range, no practical problem is caused. Therefore, even if a local low-density region is formed in the high-transmittance layer mainly for adjusting the phase shift amount of the phase shift film, it is not likely to become an obstacle in manufacturing the phase shift mask.

The detailed structure of the present invention will be described below with reference to the accompanying drawings. In the drawings, the same components are denoted by the same reference numerals.

mask blank

fig. 7 is a sectional view of a main part of a mask blank 300 showing a configuration example of the third embodiment. As shown in the drawing, the mask blank 300 has a structure in which a halftone phase shift film (hereinafter, referred to as a phase shift film) 320 is provided on the main surface S on the translucent substrate 301 side. The phase shift film 320 has a laminated structure 324 in which a high-light-transmitting layer 322 and a low-light-transmitting layer 321 having a lower light transmittance than the high-light-transmitting layer 322 are laminated in this order from the light-transmitting substrate 301 side. The phase shift film 320 may have an uppermost layer 323 at a position farthest from the transparent substrate 301. Such a phase shift film 320 is formed by oblique incidence spin sputtering, which will be described in detail later. The mask blank 300 may have a structure in which a light shielding film 303, an etching mask 304, and a photoresist film 305 are sequentially stacked on the phase shift film 320 as necessary. Next, details of the main components of the mask blank 300 will be described.

Translucent substrate 301

the light-transmitting substrate 301 has a flaw having a minute depth as a recessed defect portion F on the main surface S on the side where the phase shift film 320 is provided. The depth of the recessed defect portion F is, for example, 40nm or less.

_{2 2}The light-transmitting substrate 301 is formed of a material having a good light-transmitting property with respect to exposure light used in an exposure step in photolithography, and if ArF excimer laser light (wavelength: about 193nm) is used as the exposure light, it may be formed of a material having a light-transmitting property according thereto.

The exposure step in photolithography mentioned here refers to an exposure step in photolithography using a phase shift mask produced from the mask blank 1, and the exposure light mentioned in the following description refers to exposure light used in the exposure step. As the exposure light, any of ArF excimer laser (wavelength: 193nm), KrF excimer laser (wavelength: 248nm) and i-ray light (wavelength: 365nm) can be used, but from the viewpoint of the miniaturization of the transferred pattern in the exposure step, it is preferable to use ArF excimer laser as the exposure light. Therefore, in the following description, an embodiment in which ArF excimer laser light is used as exposure light will be described.

High light transmission layer 322, low light transmission layer 321

of the high-light-transmitting layer 322 and the low-light-transmitting layer 321 that constitute the phase shift film 320, the high-light-transmitting layer 322 provided on the light-transmitting substrate 301 side is a layer mainly for adjusting the amount of phase shift in the phase shift film 320. In the high-light-transmitting layer 322, the inner region of the portion formed on the recessed defect portion F of the light-transmitting substrate 301 has the low-density region D. The density of the low-density region D in the high-light-transmitting layer 322 is relatively lower than the density of the region inside the high-light-transmitting layer 322 in the portion of the light-transmitting substrate 301 formed on the main surface S without the recessed portion F.

The low-light-transmitting layer 321 is a layer provided on the main surface S side of the light-transmitting substrate 301 with the high-light-transmitting layer 322 interposed therebetween, and has lower light transmittance than the high-light-transmitting layer 322. The low-light-transmitting layer 321 is a layer mainly for adjusting light transmittance of the phase-shift film 320. In addition, the low-density region D formed in the recessed defect portion F does not reach the low-light-transmitting layer 321.

The high light-transmitting layer 322 and the low light-transmitting layer 321 are formed using a material containing silicon and nitrogen, and the nitrogen content in the high light-transmitting layer 322 is relatively higher than that in the low light-transmitting layer 321. The high light-transmitting layer 322 and the low light-transmitting layer 321 may be formed using a material (1) formed using silicon and nitrogen, or may be formed using a material (2) in which one or more elements selected from a semimetal element, a nonmetal element, and a rare gas are contained in the material (1).

The semimetal element contained in the high light-transmitting layer 322 and the low light-transmitting layer 321 may be any semimetal element other than silicon, and one or more elements selected from boron, germanium, antimony, and tellurium may be used. These elements are elements contained in the high light-transmitting layer 322 and the low light-transmitting layer 321 in the process of forming the phase-shift film 320 by a sputtering method described later, and are elements introduced to improve the conductivity of a sputtering target using silicon.

The non-metal element included in the high light-transmitting layer 322 and the low light-transmitting layer 321 may be any non-metal element other than nitrogen. Among them, it is preferable that the high light-transmitting layer 322 and the low light-transmitting layer 321 contain one or more elements selected from carbon, fluorine, and hydrogen. On the other hand, the high light-transmitting layer 322 and the low light-transmitting layer 321 preferably have an oxygen content of 10 atomic% or less, more preferably have an oxygen content of 5 atomic% or less, and further preferably do not actively contain oxygen, and have a lower detection limit value or less as a result of composition analysis by RBS, XPS, or the like, for example. By controlling the oxygen content of the high light-transmitting layer 322 and the low light-transmitting layer 321, which are silicon-based material films, to be low in this way, the value of the extinction coefficient k can be secured to some extent, and the thickness of the entire phase shift film 320 can be controlled to be thin.

Further, when the light-transmissive substrate 301 is formed of a material containing SiO ₂ as a main component, such as synthetic quartz glass, the composition difference between the light-transmissive substrate 301 and the high light-transmissive layer 322 can be secured by suppressing the oxygen content in the high light-transmissive layer 322 provided in contact with the light-transmissive substrate 301 to a low level, and thus, the etching selectivity between the phase shift film 320 and the light-transmissive substrate 301 can be increased in the etching for forming a phase shift mask by patterning the phase shift film 320.

The rare gas contained in the high light-transmitting layer 322 and the low light-transmitting layer 321 is an element contained in the high light-transmitting layer 322 and the low light-transmitting layer 321 due to a process of forming the phase-shift film 320 by a sputtering method, which will be described later. The rare gas is an element which exists in the film forming chamber when the phase shift film 320 is formed by reactive sputtering, and thereby the film forming speed can be increased and the productivity can be improved. In the film formation by reactive sputtering, the rare gas is converted into plasma and collides with the target, whereby the target constituent elements are sputtered from the target, and the reactive gas is trapped in the middle and stacked on the transparent substrate to form a thin film. During the time when the target constituent elements are sputtered from the target and adhere to the transparent substrate, a small amount of rare gas is incorporated into the film forming chamber. Preferred examples of the rare gas required for the reactive sputtering include argon, krypton, and xenon. In addition, in order to relax the stress of the phase shift film 320 including the high light-transmitting layer 322 and the low light-transmitting layer 321, helium and/or neon having a small atomic weight may be actively doped into the thin film.

The high light-transmitting layer 322 and the low light-transmitting layer 321 are preferably stacked in direct contact with each other without any other film therebetween. In addition, the mask blank 300 of the present invention is preferably a film structure in which neither the high-light-transmitting layer 322 nor the low-light-transmitting layer 321 is in contact with a film formed of a material containing a metal element. In a state where the film containing silicon and the film containing a metal element are in contact with each other, if heat treatment or irradiation with ArF exposure light is performed, the metal element diffuses into the film containing silicon, and this phenomenon can be prevented by forming the above-described structure, and a change in composition with time of the high light-transmitting layer 322 and the low light-transmitting layer 321 can be suppressed to be small.

Further, the high light-transmitting layer 322 and the low light-transmitting layer 321 are preferably formed using the same constituent element. In a state where any one of the high light-transmitting layer 322 and the low light-transmitting layer 321 contains different constituent elements and these light-transmitting layers are stacked in contact with each other, if heat treatment or irradiation with ArF exposure light is performed, the different constituent element may migrate and diffuse into a layer on the side not containing the constituent element. Therefore, the optical characteristics of the high-light-transmitting layer 322 and the low-light-transmitting layer 321 may change greatly from the initial stage of film formation. However, by constituting the high light-transmitting layer 322 and the low light-transmitting layer 321 with the same constituent element, the above-described variation can be suppressed to be small. In particular, when the different constituent elements are semimetal elements, it is necessary to form the high-light-transmitting layer 322 and the low-light-transmitting layer 321 using different targets, but if the high-light-transmitting layer 322 and the low-light-transmitting layer 321 are formed using the same constituent elements, such is not necessary.

In addition, the low-light-transmitting layer 321 of the high-light-transmitting layer 322 and the low-light-transmitting layer 321 may contain a transition metal for the purpose of adjusting the light transmittance of the phase-shift film 320. However, since the transition metal is a cause of lowering of light resistance against ArF exposure light, the transition metal is contained in the low light-transmitting layer 321 in a case where the lowering of light resistance does not cause a problem. The low-light-transmitting layer 321 may be formed of a material containing a transition metal and not containing silicon. Examples of the transition metal contained in the low light-transmitting layer 321 include a material containing one or more elements selected from titanium (Ti), tantalum (Ta), aluminum (Al), zirconium (Zr), and chromium (Cr). The low-light-transmitting layer 321 containing the transition metal is more preferably any one of titanium nitride (TiN), tantalum nitride (TaN), aluminum nitride (AlN), zirconium nitride (ZrN), and chromium nitride (CrN).

Further, the high-light-transmitting layer 322 and the low-light-transmitting layer 321 are preferably formed using a material formed using silicon and nitrogen. Accordingly, the high-light-transmitting layer 322 and the low-light-transmitting layer 321 can be prevented from being deteriorated in light resistance due to the inclusion of the transition metal, and the high-light-transmitting layer 322 and the low-light-transmitting layer 321 can be prevented from being changed in optical characteristics due to the inclusion of the metal other than the transition metal and the semimetal element other than silicon between the high-light-transmitting layer 322 and the low-light-transmitting layer 31. Further, if the non-metal element, for example, oxygen is contained in the high light-transmitting layer 322 and the low light-transmitting layer 321, the light transmittance to ArF exposure light is greatly increased. In view of these circumstances, the high light-transmitting layer 322 and the low light-transmitting layer 321 are preferably formed using a material formed using silicon and nitrogen. The rare gas is an element that is difficult to detect even when a composition analysis such as RBS or XPS is performed on a thin film such as the high light-transmitting layer 322 and the low light-transmitting layer 321. Therefore, the material formed of silicon and nitrogen described above is considered to include a material containing a rare gas.

In the phase shift film 320 including the high-light-transmitting layer 322 and the low-light-transmitting layer 321 each made of the above-described material, two or more sets of the laminated structures 324 each including the one high-light-transmitting layer 322 and the one low-light-transmitting layer 321 may be laminated in the order of lamination of the high-light-transmitting layer 322 and the low-light-transmitting layer 321. In this case, it is preferable that both of the high light-transmitting layer 322 and the low light-transmitting layer 321 have a thickness of 20nm or less. Since the difference in optical characteristics required for the high-light-transmitting layer 322 and the low-light-transmitting layer 321 is large, the difference in the nitrogen content in the film between the two layers is large. Therefore, when patterning the high-light-transmitting layer 322 and the low-light-transmitting layer 321, the difference in etching rate between the dry etching by the fluorine-based gas increases. In the case where the phase shift film 320 has only one stack structure 324, when a transfer pattern is formed on the phase shift film 320 by dry etching using a fluorine-based gas, a step is likely to occur in the pattern cross section of the phase shift film 320 after etching. By configuring the phase shift film 320 to have two or more sets of the layered structures 324, the thickness of each layer (one layer) of the high-light-transmitting layer 322 and the low-light-transmitting layer 321 is reduced as compared with the case of having only one set of the layered structures 324, and therefore, the step formed in the pattern cross section of the phase shift film 320 after etching can be reduced. Further, by limiting the thickness of each layer (one layer) of the high light-transmitting layer 322 and the low light-transmitting layer 321 to 20nm or less, the step formed in the pattern cross section of the phase-shift film 320 after etching can be further suppressed to be small.

However, the high-light-transmitting layer 322 disposed closest to the light-transmitting substrate 301 preferably has a film thickness that fills the recessed defect portion F of the light-transmitting substrate 301 and covers the low-density region D formed on the recessed defect portion F.

As a modification of the mask blank 300, the mask blank may be configured such that a single high-light-transmitting layer 322 is further provided on the phase shift film, the high-light-transmitting layer 322 is laminated on the low-light-transmitting layer 321 side of a laminated structure 324 formed of the single high-light-transmitting layer 322 and the single low-light-transmitting layer 321, and the thus-provided phase shift film is disposed on the light-transmitting substrate 301. As in the case described above, the multilayer structure 324 may be configured such that two or more sets of layers are stacked, and in this case, one high-light-transmitting layer 322 is stacked over the low-light-transmitting layer 321 provided at the farthest position from the light-transmitting substrate 301. Thus, the mask blank has a phase shift film having a structure in which one high-light-transmitting layer 322 is sandwiched between the stacked structure 324 and the uppermost layer 232 described later.

The refractive index n and the extinction coefficient k of each of the high-light-transmitting layer 322 and the low-light-transmitting layer 321 are set within predetermined ranges so that the phase-shift film 320 configured therefrom satisfies a predetermined phase difference and a predetermined light transmittance with respect to exposure light (e.g., ArF excimer laser light).

The high light-transmitting layer 322 is a layer mainly for adjusting the phase shift amount of the phase shift film 320. Such a high-light-transmitting layer 322 is preferably formed of a material having a refractive index n of 2.5 or more (preferably 2.6 or more) and an extinction coefficient k of less than 1.0 (preferably 0.9 or less, more preferably 0.7 or less, and still more preferably 0.4 or less) with respect to an ArF excimer laser light, for example.

In addition, the low-light-transmitting layer 321 is a layer mainly for adjusting light transmittance of the phase-shift film 320. Such a low-light-transmitting layer 321 is preferably formed of a material having a refractive index n of less than 2.5 (preferably 2.4 or less, more preferably 2.2 or less, and even more preferably 2.0 or less) and an extinction coefficient k of 1.0 or more (preferably 1.1 or more, more preferably 1.4 or more, and even more preferably 1.6 or more) with respect to ArF excimer laser light, for example.

Here, the refractive index n and the extinction coefficient k of the thin film are not always determined by the composition of the thin film. The film density, crystal state, and the like of the thin film are also factors that affect the refractive index n and the extinction coefficient k. Therefore, by adjusting various conditions for forming the high light-transmitting layer 322 and the low light-transmitting layer 321 which constitute the phase-shift film 320, each layer can have a desired refractive index n and extinction coefficient k.

In the high-light-transmitting layer 322 and the low-light-transmitting layer 321 in which the refractive index n and the extinction coefficient k are set as described above, the thickness of the high-light-transmitting layer 322 is larger than the thickness of the low-light-transmitting layer 321. In addition, if the multilayer structure 324 formed of one high-light-transmitting layer 322 and one low-light-transmitting layer 321 is a multilayer structure in which a plurality of groups are stacked, the total thickness of the plurality of high-light-transmitting layers 322 is larger than the total thickness of the plurality of low-light-transmitting layers 321. Further, with the single or plurality of laminated structures 324, even in a structure in which one layer of the high light-transmitting layer 322 is laminated, the total of the film thicknesses of the plurality of high light-transmitting layers 322 is larger than the total of the film thicknesses of the plurality of low light-transmitting layers 321.

Uppermost layer 323

The phase shift film 320 is preferably provided with an uppermost layer 323 made of a material containing silicon and oxygen at a position farthest from the transparent substrate 301. The uppermost layer 323 is formed of a material (3) formed of silicon and oxygen, a material (4) formed of silicon, nitrogen, and oxygen, or any one of materials (5) containing one or more elements selected from semimetallic elements, nonmetallic elements, and rare gases in the material (3) or the material (4).

The semimetal element, the nonmetal element, and the rare gas contained in the uppermost layer 323 are the same as those contained in the high light-transmitting layer 322 and the low light-transmitting layer 321.

in the uppermost layer 323 having such a structure, SiO ₂ and sion are mentioned as examples of materials suitable for a structure having substantially the same composition in the layer thickness direction, and further, as an example of a structure suitable for a structure having a composition inclined in the layer thickness direction, a structure in which the surface layer is SiO ₂ or sion is preferable since SiN is present on the light-transmissive substrate 301 side and the oxygen content increases as the layer gradually moves away from the light-transmissive substrate 301, and in this structure, a layer having an inclined composition may be formed by oxidizing the surface of the high light-transmissive layer 322 or the low light-transmissive layer 321 as the uppermost layer 323.

by providing the phase shift film 320 with the uppermost layer 323, the resistance of the phase shift film 320 can be secured. That is, the high-light-transmitting layer 322 and the low-light-transmitting layer 321 are formed using a material containing silicon and nitrogen, and the oxygen content is suppressed to be low. As described above, although the silicon-based material film containing nitrogen without actively containing oxygen has high light resistance against ArF excimer laser light, the silicon-based material film tends to have lower chemical resistance than the silicon-based material film containing oxygen actively. Therefore, the phase-shift film 320 has the uppermost layer 323 formed of a material containing silicon and oxygen, and thus the chemical resistance of the phase-shift film 320 can be improved.

By providing such an uppermost layer 323, the mask blank 300 does not cause pinhole defects in the phase shift film 320 on the recessed defect portion F even after the phase shift film 320 is washed with warm water for 5 minutes or more, and the phase shift film 320 remains on the recessed defect portion F.

The phase shift film 320 (including the phase shift film of the modified example, the same applies hereinafter) composed of the above layers preferably has a light transmittance of 1% or more, more preferably 2% or more, to ArF excimer laser light in order to effectively exhibit the phase shift effect. The phase shift film 320 is preferably adjusted so that the light transmittance to the ArF excimer laser light is 30% or less, more preferably 20% or less, and still more preferably 10% or less. The phase shift film 320 is adjusted so that the phase difference between the transmitted ArF excimer laser beam and the light passing through the air by a distance corresponding to the thickness of the phase shift film 320 is within a range of 170 to 190 degrees.

Light shielding film 303

The mask blank 300 according to the third embodiment (including the mask blank according to the modification, the same applies hereinafter) is preferably formed by laminating the light-shielding film 303 on the phase shift film 320. Other matters (including matters of the laminated structure with the phase shift film 320) of the light shielding film 303 are the same as those of the light shielding film 3 of the first embodiment.

Etching mask 304

In the mask blank 300 having the light-shielding film 303 stacked on the phase shift film 320, it is further preferable that an etching mask 304 formed of a material having etching selectivity to an etching gas used when etching the light-shielding film 303 is further stacked on the light-shielding film 303. Other matters of the etching mask 304 are the same as those of the etching mask 4 of the first embodiment.

Resist film 305

In the mask blank 300 according to the third embodiment, the photoresist film 305 of the organic material is preferably formed to have a film thickness of 100nm or less in contact with the surface of the etching mask 304. Other matters about the photoresist film 305 are the same as those of the photoresist film of the first embodiment described above.

Method for manufacturing mask blank

Next, a method for manufacturing a mask blank according to the third embodiment will be described.

film Forming apparatus 500

Fig. 8 is a schematic view showing an example of a film formation apparatus 500 applied to the method for manufacturing a mask blank 300 according to the third embodiment. The film forming apparatus 500 shown in the figure is a sputtering film forming apparatus for forming a film by an oblique incidence rotary sputtering method. The film deposition apparatus 500 includes a turntable 531 on which the transparent substrate 301 is placed, and a sputtering target 533 disposed in a predetermined state with respect to the turntable 531, and is capable of forming a thin film on the main surface S of the transparent substrate 301 by a sputtering method. In the film formation apparatus 500 that can be used in the manufacturing method of the present invention, the transparent substrate 301 placed on the turntable 531 and the sputtering target 533 have a predetermined positional relationship.

that is, the film formation apparatus 500 is characterized in that the sputtering surface 533S of the sputtering target 533 is disposed diagonally upward facing the main surface S of the transparent substrate 301 on the turntable 531. The film deposition apparatus 500 includes a film deposition chamber 535 housing the turntable 531 and the sputtering target 533, and performs sputtering film deposition on the transparent substrate 301 in the film deposition chamber 535. The film forming chamber 535 has a gas inlet 537 and an exhaust 539, and functions as a vacuum chamber for forming a film.

Fig. 9 is a schematic view showing a positional relationship between the transparent substrate 301 and the sputtering target 533 in the film formation apparatus 500 that can be used in the manufacturing method of the present invention. During the film formation process, the light-transmissive substrate 301 is placed on the turntable 531. By the rotation of the rotating table 531, the transparent substrate 301 rotates around a rotation axis Φ passing through the center of the main surface S and perpendicular to the main surface S.

The sputtering target 533 is arranged at a predetermined inclination angle θ (target inclination angle θ) with respect to the main surface S of the transparent substrate 301. That is, a central axis Φ 1 passing through the center O of the sputtering target 533 and perpendicular to the sputtering surface 533s of the sputtering target 533 is inclined with respect to the rotation axis Φ of the transparent substrate 301.

The sputtering target 533 is disposed such that the rotation axis Φ of the transparent substrate 301 and a straight line Φ 2 passing through the center O of the sputtering surface 533s and parallel to the rotation axis Φ of the transparent substrate 301 are located at positions deviated by a predetermined offset distance Doff.

Here, in order to improve the in-plane uniformity of the thickness of the thin film formed on the main surface S of the transparent substrate 301, it is necessary to make the positional relationship between the transparent substrate 301 and the sputtering target 533 appropriate. The positional relationship between the transparent substrate 301 and the sputtering target 533 in the film formation apparatus 500 according to the present invention will be further described below.

As shown in fig. 8 and 9, in the film formation apparatus 500 used in the present invention, the target inclination angle θ of the sputtering surface 533S of the sputtering target 533 with respect to the main surface S of the transparent substrate 301 has an influence on the film formation rate as well as the in-plane uniformity of the film thickness of the thin film formed on the main surface S of the transparent substrate 301. Specifically, the target tilt angle θ is preferably 0 to 45 degrees in order to obtain good in-plane uniformity of the film thickness and a large film deposition rate. Further, the target inclination angle θ is preferably 10 degrees to 30 degrees, whereby the in-plane uniformity of the film thickness of the thin film formed on the main surface S of the transparent substrate 301 can be improved.

The offset distance Doff in the film forming apparatus 500 can be adjusted according to the area of the thin film formed on the main surface S of the transparent substrate 301, which area is to ensure the in-plane uniformity of the film thickness. In general, when the area to be ensured with good in-plane uniformity is large, the necessary offset distance Doff increases. For example, in the case of a light-transmissive substrate 301 having a size of 152mm square, the region in which the transfer pattern is formed on the thin film is usually the inner region of 132mm square with respect to the center of the light-transmissive substrate 301. In the 132mm square inner region, in order to achieve a film thickness distribution of the thin film with accuracy within ± 1nm, it is necessary to set the offset distance Doff to about 240mm to 400mm, and preferably set the offset distance Doff to 300mm to 380 mm.

by providing the offset distance Doff in this manner, it is possible to prevent the impurity particles from falling from the sputtering surface 533S of the sputtering target 533 to the main surface S of the transparent substrate 301 during sputtering film formation. This prevents the impurity particles from being mixed into the thin film formed on the main surface S of the transparent substrate 301 to form defects, and thus the defect generation rate during film formation can be kept low.

The optimum range of the vertical distance H between the sputtering target 533 and the transparent substrate 301 in the film formation apparatus 500 varies depending on the offset distance Doff. The vertical distance H is a distance from the height position of the transparent substrate 301 to the height position of the center O of the sputtering target 533. For example, in order to ensure good in-plane uniformity in the light-transmissive substrate 301 having a square width of 152mm, the vertical distance H must be about 200mm to 380mm, and preferably 210mm to 300 mm.

In addition, for example, in the case where the film deposition apparatus 500 is a magnetron sputtering apparatus, the sputtering target 533 is attached to the magnetron electrode 541 through the back plate 543. In the method for manufacturing a mask blank according to the third embodiment, the phase shift film is formed by the oblique incidence rotary sputtering method, but the method is not limited to the magnetron sputtering apparatus and may be an ion beam sputtering apparatus as long as the sputtering method can employ the oblique incidence rotary sputtering method. In the case of magnetron sputtering, a Direct Current (DC) power supply or a high frequency (RF) power supply may be used as appropriate depending on the material of the sputtering target 533. For example, when a target 533 having low conductivity (a silicon target, a target containing no semimetal element or a silicon compound having a small content thereof, or the like) is used, RF sputtering or ion beam sputtering is preferably used. In the case of using the target 533 having a low conductivity, if the film formation rate is taken into consideration, RF sputtering is more preferably applied.

Manufacturing process of mask blank

The mask blank 300 according to the present invention (including the mask blank according to the modification, the same applies hereinafter) described above with reference to fig. 7 to 9 is manufactured by the film formation device 500 having the above-described configuration according to the third embodiment.

that is, the method of manufacturing a mask blank according to the present invention is a method of manufacturing a mask blank 300 in which the phase shift film 320 is provided on the main surface S of the translucent substrate 301 having the concave defective portion F. In the step of forming the phase shift film 320, the film is formed by the oblique incidence rotary sputtering method in which the transparent substrate 301 is rotated about the rotation axis passing through the center of the main surface S and the sputtering surface of the sputtering target 533 is disposed so as to face the main surface S of the transparent substrate 301 having the concave defect portion F obliquely.

In the step of forming the phase shift film 320, a high-light-transmission layer forming step of forming the high-light-transmission layer 322 on the main surface S of the light-transmitting substrate 301 having the recessed portion F and a low-light-transmission layer forming step of forming the low-light-transmission layer 321 having a lower light-transmission rate than the high-light-transmission layer 322 on the high-light-transmission layer 322 are performed by oblique incidence spin sputtering. The high-light-transmitting layer 322 formed in the high-light-transmitting layer forming step has a low-density region D in an inner region of the high-light-transmitting layer 322 in a portion formed on the recessed portion F, the density of the low-density region D being relatively lower than the density of an inner region of the high-light-transmitting layer 322 in a portion formed on the main surface S without the recessed portion F.

High light-transmitting layer forming step and low light-transmitting layer forming step

In the high-light-transmission layer forming step and the low-light-transmission layer forming step, reactive sputtering in a sputtering gas containing a nitrogen-based gas and a rare gas is performed using a sputtering target 533 made of a material containing silicon. Here, the mixing ratio of the nitrogen-based gas in the sputtering gas in the low light-transmitting layer forming step is lower than that in the high light-transmitting layer forming step. Thus, in the low-light-transmitting layer forming step, the nitrogen content is relatively lower than that of the high-light-transmitting layer 322 formed in the high-light-transmitting layer forming step, and thus the low-light-transmitting layer 321 having a lower light transmittance than that of the high-light-transmitting layer 322 is formed.

as described above with respect to the structure of the mask blank 300, it is preferable that the high light-transmitting layer 322 and the low light-transmitting layer 321 have a low oxygen content, and therefore, it is preferable to use a nitrogen-based gas containing no oxygen, and further, it is preferable to use nitrogen gas (N ₂ gas) for the formation of the high light-transmitting layer 322 and the low light-transmitting layer 321.

As the sputtering target 533 containing silicon used in the high-light-transmission layer forming step and the low-light-transmission layer forming step, a silicon target made of silicon can be used. In addition, as another example, a material containing silicon containing one or more elements selected from semimetal elements and nonmetal elements may be used. Thus, the high light-transmitting layer 322 and the low light-transmitting layer 321 in which silicon and nitrogen contain one or more elements selected from a semimetal element and a nonmetal element are formed. In this case, the semimetal element preferably contains one or more elements selected from boron, germanium, antimony, and tellurium. These semimetal elements can be expected to improve the conductivity of the target 533. Therefore, in particular, when the high-light-transmitting layer 322 and the low-light-transmitting layer 321 are formed by the DC sputtering method, it is preferable that the target 533 contain these semimetal elements. The non-metallic element may contain one or more elements selected from carbon, fluorine, and hydrogen.

In the above-described high-light-transmission layer forming step and low-light-transmission layer forming step, the same sputtering target 533 may be used, or sputtering targets 533 having different compositions may be used. However, in the high-light-transmission layer forming step and the low-light-transmission layer forming step, different sputtering targets 533 are preferentially used for the formation of the low-light-transmission layer 321 and the formation of the high-light-transmission layer 322. The low light-transmitting layer 321 and the high light-transmitting layer 322 have a large difference in nitrogen content, and the mixing ratio of the nitrogen-based gas in the sputtering gas during sputtering film formation is also greatly different. Therefore, the surface state of the sputtering target 533 when the low-light-transmitting layer 321 is sputter-formed is greatly different from the surface state of the sputtering target 533 when the high-light-transmitting layer 322 is sputter-formed. After the high-light-transmitting layer 322 is formed, if an attempt is made to form the low-light-transmitting layer 321 by using the sputtering target 533 in the original surface state and changing the nitrogen-based gas mixing ratio of the sputtering gas, many impurities may be generated from the sputtering target 533 or the like, and many defects may occur in the formed low-light-transmitting layer 321. If different targets 533 are used for formation of the low-light-transmitting layer 321 and formation of the high-light-transmitting layer 322, there is no fear of this, and the defect quality of the formed film can be improved.

When two different sputtering targets 533 are used for formation of the low-light-transmitting layer 321 and formation of the high-light-transmitting layer 322, a film formation apparatus in which two sputtering targets 533 are arranged in one film formation chamber may be used, or a film formation apparatus in which two film formation chambers are prepared and one sputtering target 533 is arranged in each film formation chamber may be used. When the high light-transmitting layer forming step and the low light-transmitting layer forming step are performed in different film forming chambers, the film forming chambers are preferably connected to each other via, for example, a transport chamber of a film forming apparatus. Thus, the high-light-transmitting layer forming step and the low-light-transmitting layer forming step can be performed in different film forming chambers without requiring the light-transmitting substrate to be placed in the atmosphere for this purpose.

On the other hand, when both the high-light-transmitting layer 322 and the low-light-transmitting layer 321 are formed using a film forming apparatus in which one sputtering target 533 is disposed in one film forming chamber, it is preferable that after the first layer is formed, a sputtering gas for forming the second layer is adjusted, and then a film is formed on the dummy substrate, and after the surface state of the sputtering target 533 becomes the optimum state, the second layer is formed on the first layer. According to such a flow, the defect quality of the second layer can be improved.

As the sputtering target 533 containing silicon used in the high-light-transmission layer forming step and the low-light-transmission layer forming step, particularly in the low-light-transmission layer forming step, a material containing a transition metal in silicon can be used. Thus, the low light-transmitting layer 321 containing a transition metal in silicon and nitrogen is formed.

Further, as the sputtering target 533 containing silicon used in the high-light-transmission layer forming step and the low-light-transmission layer forming step, a silicon target is preferably used in any of the steps. This makes it possible to form the high-light-transmitting layer 322 and the low-light-transmitting layer 321 formed of silicon and nitrogen, which have excellent light resistance against exposure light of ArF excimer laser light and can prevent deterioration of optical characteristics with time.

In the high light-transmitting layer forming step, in particular, in the process of forming the high light-transmitting layer 322 having a higher nitrogen content than the low light-transmitting layer 321, film formation is preferably performed by a reactive sputtering method in a poisoning mode. In the film formation in the poisoning mode, the film formation is performed by a reactive sputtering method in which the nitrogen-based gas mixture ratio is set to be larger than the nitrogen-based gas mixture ratio range in the transition mode in which the film formation tends to be unstable. On the other hand, in the low-light-transmitting layer forming step, in the process of forming the low-light-transmitting layer 321 having a nitrogen content smaller than that of the high-light-transmitting layer 322, the film is preferably formed by a reactive sputtering method in a metal mode. In the film formation in the metal mode, the film formation is performed by a reactive sputtering method in which the nitrogen-based gas mixture ratio is set to be smaller than the nitrogen-based gas mixture ratio range in the transition mode in which the film formation tends to be unstable.

Thus, both the high light-transmitting layer 322 and the low light-transmitting layer 321 can be formed by reactive sputtering under film formation conditions in which the film formation rate and voltage fluctuation during film formation are small. As a result, the phase shift film 320 having high uniformity of composition and optical characteristics and few defects can be formed.

In the case where a plurality of sets of the multilayer structure 324 including the one high-light-transmitting layer 322 and the one low-light-transmitting layer 321 are laminated on the light-transmitting substrate 301, the high-light-transmitting layer forming step and the low-light-transmitting layer forming step may be sequentially repeated. In addition, in the case of manufacturing the mask blank 300 according to the modification described above, the formation of the high-light-transmitting layer 322 in the high-light-transmitting layer forming step and the formation of the low-light-transmitting layer 321 in the low-light-transmitting layer forming step are sequentially performed on the main surface S of the light-transmitting substrate 301, and after repeating the steps as many times as necessary, the high-light-transmitting layer 322 is formed in the high-light-transmitting layer forming step.

in the high-light-transmission layer forming step and the low-light-transmission layer forming step, in order to satisfy a predetermined phase difference and a predetermined light transmittance with respect to exposure light (for example, ArF excimer laser light) of the phase shift film 320 obtained in these steps, it is necessary to form the high-light-transmission layer 322 and the low-light-transmission layer 321 in which the refractive index n and the extinction coefficient k are set within predetermined ranges, respectively. Therefore, in the high light-transmitting layer forming step and the low light-transmitting layer forming step, the mixed gas ratio of the nitrogen-based gas and the rare gas is adjusted. In addition, although the pressure of the film forming atmosphere, the power applied to the sputtering target 533, the positional relationship such as the distance between the sputtering target 533 and the light-transmissive substrate 301, and the like are also adjusted, these film forming conditions are inherent to the film forming apparatus 500, and the high light-transmissive layer 322 and the low light-transmissive layer 321 that are formed are appropriately adjusted to have the desired refractive index n and extinction coefficient k.

Uppermost layer Forming Process

in the case where the uppermost layer 323 made of a material containing silicon and oxygen is provided at the farthest position of the phase shift film 320 from the transparent substrate 301, the uppermost layer forming step is performed after the high-light-transmission layer forming step and the low-light-transmission layer forming step are completed. In the uppermost layer forming step, the uppermost layer 323 is formed by oblique incidence spin sputtering, continuously to the high light-transmitting layer forming step and the low light-transmitting layer forming step.

_{2 2}In the uppermost layer forming step, the uppermost layer 323 is formed by reactive sputtering in a sputtering gas containing nitrogen, oxygen, and a rare gas, using a silicon target 533 or a sputtering target 533 made of a material containing silicon containing one or more elements selected from a semimetal element and a nonmetal element, and the uppermost layer 323 is formed by reactive sputtering in a sputtering gas containing nitrogen, oxygen, and a rare gas.

In the case of forming the uppermost layer 323 with a tilted composition, a material layer containing silicon and nitrogen is formed by the reactive sputtering, and then a treatment for oxidizing at least a surface layer of the material layer is performed. Thereby, the uppermost layer 323 constituting an inclination in the thickness direction of the layer is formed. The surface layer of the material layer in this case is oxidized by, for example, heating in an atmosphere containing oxygen, or bringing ozone or oxygen plasma into contact with the uppermost layer. Accordingly, the uppermost layer 323 having a composition inclined such that the oxygen content in the layer gradually increases with increasing distance from the transparent substrate 301 can also be obtained. The material layer containing silicon and nitrogen may be formed by reactive sputtering under the same film formation conditions as those for the high-light-transmitting layer 322 and the low-light-transmitting layer 321.

In addition, as another example of the uppermost layer forming step of forming the uppermost layer 323 constituting the inclined structure, reactive sputtering in which the mixing ratio of each gas component contained in the sputtering gas is changed may be applied in the uppermost layer forming step. This enables formation of the uppermost layer 323 that is inclined in the layer thickness direction. In this case, for example, as the film formation progresses, the mixing ratio of oxygen in the sputtering gas is increased. Thereby, the oxygen content in the layer gradually increases as the distance from the transparent substrate 301 gradually increases, and the uppermost layer 323 having a composition gradient is obtained.

In the mask blank 300 according to the third embodiment of the present invention, in the case where the light shielding film 303 is laminated on the phase shift film 320, the light shielding film forming step is performed after the phase shift film 320 is formed. In the case where the mask blank 300 has a structure in which the etching mask 304 is stacked on the light shielding film 303, the etching mask forming step is performed after the light shielding film forming step. The light-shielding film formation step and the etching mask formation step are not limited to the respective film formation methods, and for example, a sputtering method may be applied. Further, in the case where the mask blank 300 has a structure in which the photoresist film 305 is laminated on the etching mask 304, the photoresist film forming step is performed after the etching mask forming step. The photoresist film forming step is, for example, spin coating.

phase shift mask

fig. 10 is a sectional view showing the structure of a phase shift mask according to a third embodiment. As shown in this figure, the phase shift mask 600 is characterized in that a transfer pattern 350 is formed on the phase shift film 320 of the mask blank 300 according to the third embodiment. Here, although the configuration in which the transfer pattern 350 is formed on the phase shift film 320 of the mask blank 300 described with reference to fig. 7 is described by taking fig. 10 as an example, the mask blank of the above-described modification example also forms the transfer pattern on the phase shift film in the same manner.

That is, the phase shift mask plate 600 has the transfer pattern 350 formed by patterning the phase shift film 320 in the transfer pattern forming region 303a in the central portion of the transparent substrate 301. And the outer peripheral region 303b of the transfer pattern forming region 303a is covered with the light shielding film 303 remaining on the phase shift film 320. In particular, the concave defect portion F of the transparent substrate 301 is not disposed in a region where a fine transfer pattern is formed by phase shift in the transfer pattern 350. That is, the recessed defect F is covered with the transfer pattern 320a formed of a large-area phase shift film.

Here, as described above, in the phase shift film 320 on the upper portion of the recessed defect portion F, the high-light-transmitting layer 322 closest to the light-transmitting substrate 301 has a low-density region (not shown) having a lower density than the surrounding region. However, since the high light-transmitting layer 322 is originally a layer having high light transmittance, even if the high light-transmitting layer 322 is formed at low density, the degree of increase in the light transmittance of the phase-shift film as a whole in the film thickness direction is low. Therefore, in the phase shift mask 600 having the phase shift film 320, the influence of the low density region on the exposure transfer of the pattern can be suppressed to be small. Even if the transfer pattern 320a on the recessed defect portion F deviates from the set phase shift amount due to the presence of the low density region, the formation of a fine exposure pattern by pattern exposure using the phase shift mask 600 is not affected.

Method for manufacturing phase shift mask

The method of manufacturing the phase shift mask 600 according to the third embodiment includes a step of forming the transfer pattern 350 on the phase shift film 320 of the mask blank 300 manufactured by the manufacturing method. The method for manufacturing the phase shift mask 600 according to the third embodiment is different from the method for manufacturing the phase shift mask according to the first embodiment in that, when a first pattern, which is a transfer pattern (phase shift pattern) to be formed on the phase shift film 320, is etched by exposing a photoresist film, the pattern is designed such that the concave defect portion F of the transparent substrate 301 is covered with the first photoresist pattern without using a phase shift. Otherwise, the method is the same as the method for manufacturing the phase shift mask plate according to the first embodiment.

Method for manufacturing semiconductor device

A method for manufacturing a semiconductor device according to a third embodiment is characterized in that a phase shift mask according to the third embodiment or a phase shift mask manufactured using the mask blank according to the third embodiment is used to transfer a transfer pattern of the phase shift mask to a photoresist film on a substrate by exposure. Otherwise, the same procedure as in the method for manufacturing the semiconductor device according to the first embodiment is used.