JP2007220845A - Antireflection film and exposure method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an antireflection film capable of sufficiently reducing reflectance on an interface between a resist layer and a silicon semiconductor substrate, even if exposure light is made incident obliquely by one layer in a liquid immersion lithography technique, and to provide an exposure method. <P>SOLUTION: The double layer structure antireflection film is used in exposing a resist layer in an exposure system having a wavelength of 190-195 nm and having a numerical aperture of 1.0-1.1, and is formed between the resist layer and a silicon nitride film formed on the surface of the silicon semiconductor substrate. When complex refractive indexes N<SB>1</SB>, N<SB>2</SB>of upper layer and lower layer constituting the antireflection film are expressed by the expressions; N<SB>1</SB>=n<SB>1</SB>-k<SB>1</SB>i, N<SB>2</SB>=n<SB>2</SB>-k<SB>2</SB>i, film thicknesses of the upper layer and lower layer are d<SB>1</SB>, d<SB>2</SB>and a predetermined combination is selected as the combination of values [n<SB>10</SB>, k<SB>10</SB>, d<SB>10</SB>, n<SB>20</SB>, k<SB>20</SB>, d<SB>20</SB>], n<SB>1</SB>, k<SB>1</SB>, d<SB>1</SB>, n<SB>2</SB>, k<SB>2</SB>, d<SB>2</SB>satisfy the following relation: ä(n<SB>1</SB>-n<SB>10</SB>)/(n<SB>1m</SB>-n<SB>10</SB>)}<SP>2</SP>+ä(k<SB>1</SB>-k<SB>10</SB>)/(k<SB>1m</SB>-k<SB>10</SB>)}<SP>2</SP>+ä(d<SB>1</SB>-d<SB>10</SB>)/(d<SB>1m</SB>-d<SB>10</SB>)}<SP>2</SP>+ä(n<SB>2</SB>-n<SB>20</SB>)/(n<SB>2m</SB>-n<SB>20</SB>)}<SP>2</SP>+ä(k<SB>2</SB>-k<SB>20</SB>)/(k<SB>2m</SB>-k<SB>20</SB>)}<SP>2</SP>+ä(d<SB>2</SB>-d<SB>20</SB>)/(d<SB>2m</SB>-d<SB>20</SB>)}<SP>2</SP>≤1. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an antireflection film used when a resist layer is exposed in a manufacturing process of a semiconductor device, and an exposure method using the antireflection film.

  In the field of semiconductor devices, with the high integration of semiconductor devices, there is an urgent need to establish a new process technology that enables processing of ultrafine patterns of, for example, 65 nm or less. And so-called photolithography technology is indispensable for processing fine patterns, and in order to improve optical resolution by shortening the wavelength of exposure light (illumination light), and to cope with ultra-fine processing, An argon-fluorine (ArF) excimer laser having a wavelength of 193 nm is used as an exposure light source.

  The patterning of the silicon semiconductor substrate is performed using a photosensitive resist layer applied and formed on the surface of the silicon semiconductor substrate covered with the nitride film, but at the interface between the resist layer and the underlying silicon nitride film. When the reflectance of the exposure light (illumination light) is large, a standing wave is remarkably induced in the resist layer. As a result, the side surface of the resist layer patterned by development becomes uneven according to the shape of the standing wave, and there is a problem that a good rectangular pattern cannot be formed on the resist layer. A pattern formed on the resist layer may be called a resist pattern. For example, in the case of exposure light having a wavelength of 193 nm, the reflectivity when a resist layer having a refractive index of 1.70 is provided on a silicon nitride film having a thickness of 100 nm formed on the surface of a silicon semiconductor substrate When normal incidence is assumed, this is a very large value of about 10%.

In order to solve such a problem, in a conventional photolithography technique using exposure light having a wavelength of 193 nm, a single-layer antireflection film is formed between the silicon semiconductor substrate and the resist layer. For example, when a complex refractive index is N ₀ (where N ₀ = n ₀ -k ₀ i) on a silicon nitride film having a thickness of 100 nm formed on the surface of a silicon semiconductor substrate, n ₀ = 1. When an antireflection film having a thickness of 100 nm having a value of 75, k ₀ = 0.30 is formed, and a resist layer having a refractive index of 1.70 is further formed thereon, when exposure light is vertically incident, The reflectance at the interface between the resist layer and the antireflection film is greatly reduced to about 0.3%.

JP 2001-242630 A Buntarika, Ozawa, Someya, Proceedings of the 65th Japan Society of Applied Physics, 2p-R-9

  By the way, the limiting resolution in the photolithography technique is about the product of 0.3 of the wavelength of exposure light. Therefore, in a photolithography technique using an ArF excimer laser with a wavelength of 193 nm as an exposure light source, the limit resolution is about 60 nm.

  As a technique for forming a pattern finer than about 60 nm on a silicon semiconductor substrate, a medium having a higher refractive index than air (for example, an immersion liquid made of water) is provided between the exposure system (illumination system) and the resist layer. ), The development of immersion lithography technology that achieves higher resolution is being promoted.

  By performing exposure through the immersion liquid, the effective wavelength of the exposure light is obtained by dividing the wavelength of the exposure light in vacuum by the refractive index of the immersion liquid, enabling higher resolution performance to be achieved. It becomes. For example, when an ArF excimer laser having a wavelength of 193 nm is used as an exposure light source and water is used as the immersion liquid (refractive index at 193 nm is 1.44), the effective wavelength is about 134 nm, and the limit resolution is 0. Multiply .3 to obtain approximately 40 nm. That is, a pattern finer than about 60 nm can be formed on the silicon semiconductor substrate by the immersion lithography technique using water.

Further, the focus tolerance (DOF) at the time of exposure is given by the following equation. Here, n _Liq is the refractive index of the immersion liquid, K ₂ is a process-dependent constant, λ is the wavelength of exposure light (illumination light) in vacuum, and NA is the numerical aperture of the exposure system (illumination system). .

DOF = n _Liq・ K ₂・ λ / NA ²

Therefore, when the numerical aperture NA is constant, the focus tolerance DOF is n _Liq times in the case of the immersion lithography technique as compared with the case of the conventional photolithography technique by exposure in air. That is, in the case of the immersion lithography technique using water as the immersion liquid, the focus tolerance DOF is 1.44 times, and it is possible to construct a mass production process with more margin.

  However, such an immersion lithography technique has a problem that the conventional single-layer antireflection film does not function effectively.

The exposure light travels through the incident medium, enters the resist layer, and further enters the antireflection film. Here, the incident angle when the exposure light is incident on the resist layer from the incident medium is θ _in , the refractive index of the incident medium is n _in , and the incident light when the exposure light is incident on the silicon semiconductor substrate or the antireflection film from the resist layer When the angle is θ _IF and the refractive index of the resist material constituting the resist layer is n _Res , the following equation is satisfied. In the conventional photolithography technique, since the incident medium is air, n _in = 1. In the immersion lithography technique, for example, when water is an immersion liquid, the incident medium is water, so n _in = 1.44.

NA = n _in · sin (θ _in ) = n _Res · sin (θ _IF )

From the above equation, it can be seen that when θ _in is constant, the numerical aperture NA is n _Liq times in the immersion lithography technique compared to the conventional photolithography technique (n _in = 1.0). That is, if the refractive index n _Res of the resist material constituting the resist layer is constant, sin (θ _IF ) becomes larger, which means that θ _IF increases in the immersion lithography technique. . That is, in the immersion lithography technique, it can be said that the exposure light is incident from an oblique direction as compared with the conventional photolithography technique.

  On the other hand, when a single-layer antireflection film is used, the reflectance when the exposure light is incident vertically can be sufficiently lowered, but the reflectance when obliquely incident cannot be sufficiently reduced. There's a problem.

An antireflection film having a thickness of 100 nm and having a complex refractive index N ₀ of n ₀ = 1.75 and k ₀ = 0.30 on a silicon nitride film having a thickness of 100 nm formed on the surface of the silicon semiconductor substrate When a resist layer having a refractive index of 1.70 is further formed thereon, the reflectance at the interface between the resist layer and the antireflection film is as follows when the exposure light is perpendicularly incident (that is, the incident angle). θ _IF = 0 degree), which is greatly reduced to about 0.3%. However, when the incident angle θ _IF is about 60 degrees, the reflectance of the s wave greatly increases to about 4%.

  On the other hand, with the progress of miniaturization, the allowable maximum value of the reflectance at the interface between the silicon nitride film formed on the surface of the silicon semiconductor substrate and the resist layer is decreasing year by year. The maximum allowable reflectance in the fine generation as applied is 0.4% for the line and space pattern (Buntarica, Ozawa, Someya, 65th JSAP Scientific Lecture Presentation 2p-R-9).

  In the 45 nm generation where immersion lithography will be applied, the size of the contact hole formed is assumed to be approximately 90 nm (pitch 140 nm). Under these conditions, the numerical aperture NA is set to 1.05 corresponding to immersion lithography, and optical profile calculation software Prolith ver. Manufactured by KLA-Tencor is used. FIG. 1 shows the result of calculating the reflectance dependency of the contact hole diameter variation from the standing wave shape in 8.1.

  For a contact hole diameter of 90 nm, it is necessary to suppress the variation in diameter to 5 nm or less. From FIG. 1, it can be seen that for this purpose, the reflectance needs to be about 0.4% or less.

  That is, in the case of a line and space pattern or a contact hole pattern, the reflectance needs to be suppressed to 0.4% or less.

  However, the single-layer antireflection film that has been used in the conventional photolithography technique may not sufficiently reduce the reflectance because exposure light may be incident more obliquely in the immersion lithography technique. Since the reflectance cannot be lowered sufficiently, the standing wave appears remarkably in the resist layer, so that the problem that a good rectangular pattern cannot be formed on the resist layer cannot be solved.

  Accordingly, an object of the present invention is to increase the numerical aperture of the exposure system (illumination system) so that exposure light (illumination light) is generated in a photolithography technique for achieving a large focus tolerance, such as an immersion lithography technique. An antireflection film capable of sufficiently reducing the reflectance at the interface between the silicon nitride film formed on the surface of the silicon semiconductor substrate and the resist layer even when entering the resist layer at an angle, and An object of the present invention is to provide an exposure method using an antireflection film.

  An antireflection film of the present invention for achieving the above object is used in a manufacturing process of a semiconductor device, has a wavelength of 190 nm to 195 nm, and has an aperture NA of 1.0 <NA ≦ 1.1. Provided between the resist layer used when exposing the resist layer in the system and a silicon nitride film having a thickness T (unit: nm) of the following value formed on the surface of the silicon semiconductor substrate This is an antireflection film having a two-layer structure. The exposure method of the present invention for achieving the above object is used in a manufacturing process of a semiconductor device, has a wavelength of 190 nm to 195 nm, and has a numerical aperture NA of 1.0 <NA ≦ 1.1. In the exposure system, an antireflection film having a two-layer structure is provided between the resist layer and a silicon nitride film having the following thickness T (unit: nm) formed on the surface of the silicon semiconductor substrate. An exposure method for exposing the resist layer in a state.

And the antireflection film of the present invention or the exposure method of the present invention comprises:
The upper-layer complex refractive index N ₁ and the lower-layer complex refractive index N ₂ constituting the antireflection film are respectively expressed as follows:
N ₁ = n ₁ −k ₁ i
N ₂ = n ₂ −k ₂ i
The upper layer thickness is d ₁ (unit: nm), the lower layer thickness is d ₂ (unit: nm),
When one of the following is selected as a combination of values of [n ₁₀ , k ₁₀ , d ₁₀ , n ₂₀ , k ₂₀ , d ₂₀ ] depending on the film thickness T of the silicon nitride film,
n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , and d ₂ satisfy the following relational expression, or use such an antireflection film.

{(N ₁ −n ₁₀ ) / (n _1m −n ₁₀ )} ² + {(k ₁ −k ₁₀ ) / (k _1m −k ₁₀ )} ² + {(d ₁ −d ₁₀ ) / (d _1m −d ₁₀ )} ² + {(n ₂ −n ₂₀ ) / (n _2m −n ₂₀ )} ² + {(k ₂ −k ₂₀ ) / (k _2m −k ₂₀ )} ² + {(d ₂ − d ₂₀ ) / (d _2m −d ₂₀ )} ² ≦ 1

However, the value of n _{1m in} the case is adopted based on the magnitude relationship between n ₁ and n ₁₀ , the value of k _{1m in} the case is adopted based on the magnitude relationship between k ₁ and k ₁₀ , and d ₁ and d ₁₀ The value of d _{1m in} the case is adopted based on the magnitude relationship, the value of n _{2m in} the case is adopted based on the magnitude relationship between n ₂ and n ₂₀ , and the value of k in the case based on the magnitude relationship between k ₂ and k ₂₀ is adopted. adopts a value of _2m, adopts the value of d _2m in the case based on the magnitude relation between d ₂ and d _20.

  The exposure method of the present invention is applied to, for example, processing of an extremely fine pattern in a semiconductor device. Specifically, the antireflection film of the present invention is formed on a silicon nitride film formed on the surface of a silicon semiconductor substrate. A step of forming, a step of coating and forming a resist layer having a photosensitive action on the antireflection film, a step of selectively exposing the resist layer with exposure light (ultraviolet light) to sensitize, and developing the resist layer And obtaining a predetermined resist pattern.

  In the antireflection film of the present invention or the exposure method of the present invention (hereinafter, these may be collectively referred to simply as the present invention), the exposure light (ultraviolet rays) has a wavelength of 190 nm to 195 nm, preferably Preferably has a wavelength of 192 nm to 194 nm, and more specifically, it is more desirable to use an ArF excimer laser with a wavelength of 193 nm as an exposure light source.

Further, d ₁ ≦ 250 and d ₂ ≦ 250 are satisfied, in other words, the thickness of the upper layer of the antireflection film does not exceed 250 nm, and the thickness of the lower layer does not exceed 250 nm. It is preferable. If the upper layer thickness exceeds 250 nm or the lower layer thickness exceeds 250 nm, the resist layer is exposed with exposure light and developed, and then the silicon semiconductor substrate is etched using the resist layer as an etching mask. In the process, a so-called process conversion difference (also referred to as a dimension conversion amount or a dimension shift), which is a difference between the resist pattern dimension of the resist layer and the actual etching processing dimension in the silicon semiconductor substrate, becomes too large. There is a possibility that a pattern having a size cannot be obtained.

Further, the refractive index of the resist layer is desirably 1.60 to 1.80. When a resist layer made of a resist material not in this range is used, an antireflection film that satisfies any of the various combinations of the above-described conditions (n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , d ₂ ) Even in the entire region from the incident angle (maximum incident angle θ _in-max ) to vertical incident (minimum incident angle θ _in-min , specifically 0 degree) corresponding to the corresponding numerical aperture. Therefore, it becomes difficult to reduce the reflectance at the interface between the silicon nitride film formed on the surface of the silicon semiconductor substrate and the resist layer to 0.4% or less, and a good resist pattern shape may not be obtained. .

Furthermore, the fact that the space between the resist layer and the exposure system is filled with a medium having a refractive index of 1.44 ± 0.02, in other words, the refractive index as an immersion liquid is 1.44 ± 0.02. It is desirable to use the above medium, and it is more desirable to use water as the medium. When the refractive index deviates greatly from this value, the antireflection film satisfies any of the various sets of conditions (n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , d ₂ ) described above. In addition, over the entire region from the incident angle (maximum incident angle θ _in-max ) to vertical incident (minimum incident angle θ _in-min , specifically 0 degree) corresponding to the corresponding numerical aperture. The reflectance at the interface between the silicon nitride film formed on the surface of the silicon semiconductor substrate and the resist layer becomes difficult to be 0.4% or less, and a good resist pattern shape may not be obtained.

  In the exposure method of the present invention, it is desirable to form a topcoat layer on the resist layer (specifically, on the upper layer). If the topcoat layer is not present, the interaction between the immersion liquid and the resist layer (for example, a phenomenon in which a defect occurs in the resist layer when the resist layer comes into contact with the immersion liquid) can be suppressed. Therefore, a good resist pattern shape may not be obtained. Here, examples of the material constituting the top coat layer include organic substances and inorganic substances, specifically, for example, polyvinyl alcohol, amorphous fluoropolymer, and NaCl.

In the following description, the entire area from the incident angle (maximum incident angle θ _in-max ) to the normal incident angle (minimum incident angle θ _in-min ) of the exposure light corresponding to the corresponding numerical aperture is referred to as “corresponding exposure. It may be referred to as “the entire incident angle corresponding to the numerical aperture NA of the system” or simply “the entire incident angle”.

  Further, the film thickness of the resist layer is preferably about 2 to 5 times the minimum resist pattern size to be formed. When the thickness of the resist layer is less than twice the minimum resist pattern size, it is possible to pattern the resist layer into a predetermined pattern, but good when etching the silicon semiconductor substrate after patterning the resist layer There is a possibility that the etching cannot be performed. In addition, the number of film defects in the resist layer may increase. On the other hand, when the film thickness of the resist layer exceeds 5 times the minimum resist pattern size, the patterned resist layer may fall down and good patterning of the silicon semiconductor substrate may not be possible.

Any material can be used as the material constituting the upper and lower layers of the antireflection film as long as it satisfies various combinations of the above-mentioned conditions (n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , d ₂ ). It may be a material. For example, examples of the material constituting the upper layer and the lower layer include polymer materials, inorganic oxide materials, metal materials, and hybrid materials thereof. Specifically, for example, polyimide, SiCH film, SiCHN film, Examples thereof include a SiCOH film, an epoxy thermosetting resin, an acrylic thermosetting resin, an epoxy ultraviolet curing resin, and an acrylic ultraviolet curing resin.

  Further, in order to improve adhesion between the resist layer formed on the antireflection film and the antireflection film, the surface of the upper layer constituting the antireflection film is subjected to silane coupling treatment or the like. A surface modification treatment may be performed.

  In general, a single-layer antireflection film has a reflectivity of 0 over the entire incident angle region corresponding to the numerical aperture NA of the exposure system, regardless of how the film thickness and complex refractive index are changed. It cannot be less than 4%. On the other hand, if the thickness of the antireflection film is too thick, a problem of processing conversion difference occurs in the step of etching the silicon semiconductor substrate after exposing the resist layer with exposure light and developing it.

  On the other hand, in the present invention, a two-layer antireflection film having a film thickness and a complex refractive index within a predetermined range is formed between the silicon nitride film formed on the surface of the silicon semiconductor substrate and the resist layer. As a result, the reflectance can be reduced to 0.4% or less over the entire incident angle region corresponding to the numerical aperture NA of the corresponding exposure system, and a resist pattern having a more excellent shape can be obtained. This makes it possible to perform ultrafine processing more than ever. In other words, when the numerical aperture NA of the exposure system is 1.0 <NA ≦ 1.1, the film thickness and the complex refractive index in the antireflection film having a two-layer structure satisfy the above conditions, so that the corresponding aperture is obtained. As a result of the reflectivity being 0.4% or less over the entire region from the incident angle of exposure light corresponding to several NA to the normal incidence, a good resist pattern shape can be obtained. In addition, the processing conversion difference can be reduced.

  Hereinafter, the present invention will be described based on embodiments with reference to the drawings. First, when the present invention is applied, the entire incident angle corresponding to the numerical aperture NA of the corresponding exposure system (the numerical aperture of the exposure system). The reason why the reflectance is 0.4% or less over the entire range from the incident angle of the exposure light corresponding to NA to the vertical incidence) will be described.

The upper layer of the thickness d ₁ of the antireflection film from 10nm to 200 nm, at 10nm increments, As for the lower layer of thickness d _2, from 10nm to 200 nm, at 10nm increments, in each combination, a numerical aperture NA of the exposure system Is 1.0 <NA ≦ 1.1 (specifically 1.1), the corresponding incident angle from the most oblique angle (maximum incident angle θ _in-max ) to normal incidence (minimum incident angle θ _{in- min} = 0 degree), that is, optimization simulation of the complex refractive indexes N ₁ and N ₂ of the upper layer and the lower layer to minimize the reflectance in the entire incident angle region, the film thickness of the silicon nitride film is 2 nm. This was performed for the case of 205 nm or less.

  In this calculation, the reflectance of the two-layer antireflection film was calculated using a calculation method for calculating the Fresnel coefficient in each layer (reference: basic theory of optical thin film, Mitsunobu Koisoyama, published by Optronics, 2003). For optimization of the complex refractive index of the upper and lower layers, Fletcher-Reeves optimization method (reference: nonlinear optimization problem, written by J. Korick, MR Osborne, Yoshiyuki Yamamoto, Takeo Koyama, Showa 45 Year, issued by Baifukan).

  In the optimization, the entire incident angle region was divided into 22 equal parts. Then, the reflectance at each incident angle was calculated, and the sum of squares of the reflectance at each incident angle was minimized.

In this way, the numerical aperture NA of the exposure system is 1.1, the upper layer thickness d ₁ is from 10 nm to 250 nm in increments of 10 nm, and the lower layer thickness d ₂ is also from 10 nm to 250 nm. For each case when changing in increments of 10 nm, the complex refractive indexes N ₁ and N ₂ of the upper and lower layers are obtained by optimizing the complex refractive indexes of the upper and lower layers by the above method. I was able to.

  The reason why the calculation in the case of a film thickness thicker than 250 nm was not performed is that when the film thickness is larger than this, the processing conversion difference in the etching process becomes large, and it becomes impossible to process a good silicon semiconductor substrate. .

Then, from the above calculation results, the numerical aperture NA of the exposure system is 1.1, the upper layer thickness d ₁ is from 10 nm to 250 nm in increments of 10 nm, and the lower layer thickness d ₂ is also from 10 nm to 250 nm, 10 nm. For each case when changing the increment, the film thickness conditions are set so that the sum of squares of reflectivity, which is an evaluation function for optimization, is minimized after optimization of the complex refractive index of the upper and lower layers. Asked.

Based on this film thickness condition, the upper layer film thickness d ₁ and the lower layer film thickness d ₂ are further finely divided, and the complex refractive index of the upper layer and the lower layer corresponding to the film thickness condition is the most suitable complex refraction. The details of the rate were sought.

As a result, the following most preferable combination of complex refractive index and film thickness was obtained. That is, the complex refractive index of the upper layer (layer located far from the surface of the silicon semiconductor substrate having the silicon nitride film on the surface) in the two-layer antireflection film is N ₁ , and the lower layer (silicon semiconductor substrate having the silicon nitride film on the surface) N ₁₀ , n ₂₀ , k ₁₀ , and k ₂₀ , where the complex refractive index of the layer located near the surface) is N ₂ ,
N ₁ = n ₁₀ −k ₁₀ i
N ₂ = n ₂₀ −k ₂₀ i
The following Table 1-A to Table 1-T (numerical aperture) where d ₁₀ is the upper layer thickness (unit: nm) and d ₂₀ is the lower layer thickness (unit: nm) The combination of the values of [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ] shown in NA: 1.0 <NA ≦ 1.1) is the most preferable combination, that is, the minimum of reflectance. A combination that yields a value.

[Case A-01] to [Case A-26] in Table 1-A or Table 2-A to be described later are values when the thickness of the silicon nitride film is 2 nm or more and 15 nm or less.
[Case B-01] to [Case B-22] in Table 1-B or Table 2-B to be described later are values when the film thickness of the silicon nitride film is greater than 15 nm and less than 25 nm.
[Case C-01] to [Case C-24] in Table 1-C or Table 2-C described later are values when the thickness of the silicon nitride film is greater than 25 nm and not greater than 35 nm.
[Case D-01] to [Case D-25] in Table 1-D or Table 2-D to be described later are values when the thickness of the silicon nitride film is greater than 35 nm and less than 45 nm.
[Case E-01] to [Case E-22] in Table 1-E or Table 2-E described later are values when the thickness of the silicon nitride film is greater than 45 nm and less than or equal to 55 nm.
[Case F-01] to [Case F-20] in Table 1-F or Table 2-F to be described later are values when the thickness of the silicon nitride film is greater than 55 nm and less than or equal to 65 nm.
[Case G-01] to [Case G-24] in Table 1-G or Table 2-G to be described later are values when the film thickness of the silicon nitride film is greater than 65 nm and less than or equal to 75 nm.
[Case H-01] to [Case H-21] in Table 1-H or Table 2-H to be described later are values when the film thickness of the silicon nitride film is greater than 75 nm and less than 85 nm.
[Case I-01] to [Case I-19] in Table 1-I or Table 2-I described later are values when the thickness of the silicon nitride film is greater than 85 nm and less than or equal to 95 nm.
[Case J-01] to [Case J-21] in Table 1-J or Table 2-J to be described later are values when the thickness of the silicon nitride film is greater than 95 nm and less than or equal to 105 nm.
[Case K-01] to [Case K-23] in Table 1-K or Table 2-K to be described later are values when the thickness of the silicon nitride film is greater than 105 nm and less than or equal to 115 nm.
[Case L-01] to [Case L-20] in Table 1-L or Table 2-L to be described later are values when the thickness of the silicon nitride film is greater than 115 nm and less than or equal to 125 nm.
[Case M-01] to [Case M-20] in Table 1-M or Table 2-M to be described later are values when the thickness of the silicon nitride film is greater than 125 nm and less than or equal to 135 nm.
[Case N-01] to [Case N-23] in Table 1-N or Table 2-N to be described later are values when the film thickness of the silicon nitride film is greater than 135 nm and less than or equal to 145 nm.
[Case O-01] to [Case O-23] in Table 1-O or Table 2-O to be described later are values when the thickness of the silicon nitride film is greater than 145 nm and less than or equal to 155 nm.
[Case P-01] to [Case P-21] in Table 1-P or Table 2-P to be described later are values when the film thickness of the silicon nitride film is larger than 155 nm and not larger than 165 nm.
[Case Q-01] to [Case Q-19] in Table 1-Q or Table 2-Q to be described later are values when the thickness of the silicon nitride film is larger than 165 nm and not larger than 175 nm.
[Case R-01] to [Case R-23] in Table 1-R or Table 2-R described later are values when the thickness of the silicon nitride film is greater than 175 nm and less than 185 nm.
[Case S-01] to [Case S-21] in Table 1-S or Table 2-S to be described later are values when the thickness of the silicon nitride film is larger than 185 nm and not larger than 195 nm.
[Case T-01] to [Case T-20] in Table 1-T or Table 2T described later are values when the thickness of the silicon nitride film is greater than 195 nm and equal to or less than 205 nm.

That is, under the conditions described above, when the numerical aperture NA is 1.0 <NA ≦ 1.1,
When the thickness of the silicon nitride film is 2 nm or more and 15 nm or less, the minimum value of the reflectance at 26 points (a combination of 26 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). Can get the
When the thickness of the silicon nitride film is greater than 15 nm and less than or equal to 25 nm, the minimum value of reflectance is obtained at 22 points (a combination of 22 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 25 nm and less than or equal to 35 nm, the minimum value of reflectance is obtained at 24 points (a combination of 24 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 35 nm and less than or equal to 45 nm, the minimum value of the reflectance is obtained at 25 points (a combination of 25 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 45 nm and less than or equal to 55 nm, the minimum value of reflectance is obtained at 22 points (a combination of 22 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 55 nm and less than or equal to 65 nm, the minimum value of reflectance is obtained at 20 points (a combination of 20 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 65 nm and less than or equal to 75 nm, the minimum value of the reflectance is obtained at 24 points (a combination of 24 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 75 nm and less than or equal to 85 nm, the minimum value of reflectance is obtained at 21 points (a combination of 21 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 85 nm and less than or equal to 95 nm, the minimum value of reflectance is obtained at 19 points (a combination of 19 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 95 nm and less than or equal to 105 nm, the minimum value of reflectance is obtained at 21 points (a combination of 21 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 105 nm and less than or equal to 115 nm, the minimum value of reflectance is obtained at 23 points (a combination of 23 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 115 nm and less than or equal to 125 nm, the minimum value of reflectance is obtained at 20 points (a combination of 20 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 125 nm and less than or equal to 135 nm, the minimum value of reflectance is obtained at 20 points (a combination of 20 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 135 nm and less than or equal to 145 nm, the minimum value of reflectance is obtained at 23 points (a combination of 23 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 145 nm and less than or equal to 155 nm, the minimum value of reflectance is obtained at 23 points (a combination of 23 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 155 nm and less than or equal to 165 nm, the minimum value of reflectivity is obtained at 21 points (a combination of 21 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is larger than 165 nm and smaller than 175 nm, the minimum value of the reflectance is obtained at 19 points (a combination of 19 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 175 nm and less than 185 nm, the minimum value of the reflectance is obtained at 23 points (a combination of 23 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the film thickness of the silicon nitride film is greater than 185 nm and less than or equal to 195 nm, the minimum value of the reflectance is obtained at 21 points (a combination of 21 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). It is possible,
When the thickness of the silicon nitride film is greater than 195 nm and less than or equal to 205 nm, the minimum value of the reflectance is obtained at 20 points (a combination of 20 [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ]). I was able to.

If a combination of the above values of [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ] that can obtain the minimum value of reflectance is used, a predetermined numerical aperture NA can be obtained. In the silicon film thickness, the reflectivity can be 0.4% or less over the entire incident angle region corresponding to the numerical aperture NA of the corresponding exposure system. That is, the above combination is an effective combination from the viewpoint that the reflectance can be reduced to 0.4% or less even when the numerical aperture NA is smaller than the corresponding numerical aperture NA.

In each case of the above combinations of values of [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ], when 5 of these 6 variables are fixed, the remaining 1 A simulation was performed as to how far the variable fluctuated and the reflectivity exceeded 0.4%. As a result, the allowable fluctuation range shown in the following Table 2-A to Table 2-T (numerical aperture NA: 1.0 <NA ≦ 1.1) was obtained.

In the following,
n _1-min : Minimum value of n ₁₀ when reflectance does not exceed 0.4% n _1-max : Maximum value of n ₁₀ when reflectance does not exceed 0.4% k _1-min : Reflection the minimum value k _1-max of k ₁₀ when the rate does not exceed 0.4% maximum value d _1-min of k ₁₀ when the reflectance does not exceed 0.4%: reflectance of 0.4% D ₁₀ -minimum value d _{1 -max} when the reflectance does not exceed 0.4% d _{2 -min} maximum value of d ₁₀ when the reflectance does not exceed 0.4% n when the reflectance does not exceed 0.4% ₂₀ minimum value n _2-max : Maximum value of n ₂₀ when the reflectance does not exceed 0.4% k _2-min : Minimum value of k ₂₀ when the reflectance does not exceed 0.4% k _{2 -max:} maximum value d _2-min of k ₂₀ when the reflectance does not exceed 0.4%: minimum d _2-max of d ₂₀ when the reflectance does not exceed 0.4%: reflectance The maximum value of d ₂₀ when 0.4% is not exceeded.

The combination of the values of [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ] minimizes (minimizes) the reflectance in the entire incident angle region corresponding to the numerical aperture NA of the corresponding exposure system. It is a combination. That is, when the evaluation function for minimization (minimization) is f, the evaluation function f is a function of n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , and d ₂₀ , and f ( n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ) is a combination that minimizes (minimizes) [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d _10]. , D ₂₀ ]. That is, the evaluation function f is minimized by a combination of the values [n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ] as described above.

In general, the multivariable function f (x _i ) (i = 0, 1, 2,... N) is
x _i = x _i-min
In the vicinity of this minimum value, f (x _i ) can be expressed by the following formula (1). That is, it can be approximated by a quadratic function. However, the symbol “Σ” means the sum in i = 0, 1, 2,... N. The same applies to equation (2).

f (x _i ) = Σa _i (x _i −x _i−min ) ² + b (1)

In this case, the condition for f (x _i ) to be equal to or smaller than a certain number c greater than b can be expressed by an elliptic function type of the following equation (2). Here, x _ic is a value of x _i where f (x _i ) = c when only x _i is varied with all other variables fixed.

Σ (x _i −x _1−min ) ² / (x _ic −x _i−min ) ² ≦ 1 (2)

Therefore, only one variable is changed, the other variables are fixed, and n _1-max , n _1-min , k _1-max , k ₁ , which are values of each variable with a reflectance of 0.4%. _{-min, d 1-max, d} 1-min, n 2-max, n 2-min, k 2-max, k 2-min, d 2-max, by using a value of d _2-min, or less If n ₁₀ , n ₂₀ , k ₁₀ , n ₂₀ , k ₂₀ , and d ₂₀ satisfying the expression (3) are adopted, the reflectance does not exceed 0.4%.

{(N ₁ −n ₁₀ ) / (n _1m −n ₁₀ )} ² + {(k ₁ −k ₁₀ ) / (k _1m −k ₁₀ )} ² + {(d ₁ −d ₁₀ ) / (d _1m −d ₁₀ )} ² + {(n ₂ −n ₂₀ ) / (n _2m −n ₂₀ )} ² + {(k ₂ −k ₂₀ ) / (k _2m −k ₂₀ )} ² + {(d ₂ − d ₂₀ ) / (d _2m −d ₂₀ )} ² ≦ 1 (3)

However, n _1m , k _1m , d _1m , n _2m , k _2m , and d _2m take the following values.

n _1m: n _1-max when the n ₁ ≧ n _10, when _{n 1 <n 10 n 1-} min
k _1m: k _1-max when the k ₁ ≧ k _10, when the _{k 1 <k 10 k 1-} min
d _1m: d _1-max when d ₁ ≧ d _10, when _{d 1 <d 10 d 1-} min
n _2m: n _2-max when the n ₂ ≧ n _20, when _{n 2 <n 20 n 2-} min
k _2m: k _2-max when k ₂ ≧ k _20, when _{k 2 <k 20 k 2-} min
d _2m: when _{d 2 ≧ d 20 d 2-} max, when _{d 2 <d 20 d 2-} min

Note that the values of n _1m , k _1m , d _1m , n _2m , k _2m , and d _2m are the same as n ₁ , n ₂ , k ₁ , n ₂ , k ₂ , and d ₂ , respectively, n ₁₀ , n ₂₀ , k ₁₀ , n ₂₀ , k ₂₀ , and d ₂₀ when compared with d ₂₀
n _1−max −n ₁₀ = n ₁₀ −n _1−min
k _1-max −k ₁₀ = k ₁₀ −k _1-min
d _1-max −d ₁₀ = d ₁₀ −d _1-min
n _2-max -n ₂₀ = n ₂₀ -n _2-min
k _2-max -k ₂₀ = k ₂₀ -k _2-min
d _2−max −d ₂₀ = d ₂₀ −d _2−min
In the definition of the ellipsoid diameter defined by the equation (2), n ₁ , n ₂ , k ₁ , n ₂ , k ₂ , and d ₂ are n ₁₀ , n ₂₀ , respectively. , K ₁₀ , n ₂₀ , k ₂₀ , and d ₂₀ . In other words, the curvature of the ellipsoid is different when, for example, n ₁ ≧ n ₁₀ and n ₁ <n ₁₀ , and k ₁₀ , d ₁₀ , n ₂₀ , k ₂₀ , and d ₂₀ are also different. This is because the same can be said.

Thus, as a combination of values of [n ₁₀ , k ₁₀ , d ₁₀ , n ₂₀ , k ₂₀ , d ₂₀ ], the numerical aperture NA is 1.0 <NA ≦ 1.1,
In the case of 2 (nm) ≦ T ≦ 15 (nm), any one of case [A-01] to case [A-26],
In the case of 15 (nm) <T ≦ 25 (nm), any one of case [B-01] to case [B-22],
In the case of 25 (nm) <T ≦ 35 (nm), any of case [C-01] to case [C-24],
When 35 (nm) <T ≦ 45 (nm), any of case [D-01] to case [D-25],
When 45 (nm) <T ≦ 55 (nm), any of case [E-01] to case [E-22],
When 55 (nm) <T ≦ 65 (nm), any of case [F-01] to case [F-20],
When 65 (nm) <T ≦ 75 (nm), any of case [G-01] to case [G-24],
When 75 (nm) <T ≦ 85 (nm), any one of case [H-01] to case [H-21],
When 85 (nm) <T ≦ 95 (nm), any of case [I-01] to case [I-19],
When 95 (nm) <T ≦ 105 (nm), any of case [J-01] to case [J-21],
In the case of 105 (nm) <T ≦ 115 (nm), any of the case [K-01] to the case [K-23],
In the case of 115 (nm) <T ≦ 125 (nm), any of case [L-01] to case [L-20],
In the case of 125 (nm) <T ≦ 135 (nm), any of the case [M-01] to the case [M-20],
In the case of 135 (nm) <T ≦ 145 (nm), any one of case [N-01] to case [N-23],
In the case of 145 (nm) <T ≦ 155 (nm), any one of case [O-01] to case [O-23],
In the case of 155 (nm) <T ≦ 165 (nm), any one of case [P-01] to case [P-21],
In the case of 165 (nm) <T ≦ 175 (nm), any one of case [Q-01] to case [Q-19],
In the case of 175 (nm) <T ≦ 185 (nm), any one of case [R-01] to case [R-23],
In the case of 185 (nm) <T ≦ 195 (nm), any one of case [S-01] to case [S-21],
In the case of 195 (nm) <T ≦ 205 (nm), any one of case [T-01] to case [T-20],
To ensure that the reflectivity does not exceed 0.4%,
Based on the magnitude relationship between n ₁ and n ₁₀ , the value of n _1m in this case is adopted,
the value of k _{1 m} in the case is adopted based on the magnitude relationship between k ₁ and k _10,
Based on the magnitude relationship between d ₁ and d ₁₀ , the value of d _{1m in} the case is adopted,
Based on the magnitude relationship between n ₂ and n ₂₀ , the value of n _2m in this case is adopted,
The value of k _2m in the case is adopted based on the magnitude relation of k ₂ and k _20,
By adopting the value of d _2m in this case based on the magnitude relationship between d ₂ and d ₂₀ ,
N ₁ is within the range of the maximum value (n _1-max ) and the minimum value (n _1-min ) of n _{1m in} the case,
K ₁ is within the range of the maximum value (k _1-max ) and the minimum value (k _1-min ) of k _{1m in} the case,
D ₁ is within the range of the maximum value (d _1-max ) and the minimum value (d _1-min ) of d _{1m in} the case,
N ₂ is within the range of the maximum value (n _2-max ) and the minimum value (n _2-min ) of n _{2m in} the case,
K ₂ is within the range of the maximum value (k _2-max ) and the minimum value (k _2-min ) of k _{2m in} the case,
When d ₂ is within the range of the maximum value (d _2−max ) and the minimum value (d _2−min ) of d _{2m in} the case, the antireflection film has a reflectivity of 0. As a result of ensuring that it does not exceed 4%, a good resist pattern can be obtained.

That is, the center is (n ₁₀ , n ₂₀ , k ₁₀ , k ₂₀ , d ₁₀ , d ₂₀ ), and the diameters are (n _1m −n ₁₀ ), (k _1m −k ₁₀ ), (d _1m −d ₁₀ ). ), (N _2m −n ₂₀ ), (k _2m −k ₂₀ ), (d _2m −d ₂₀ ) (n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , d ₂ ) Assuming a 6-dimensional ellipse with these six variables, a combination of arbitrary values of (n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , d ₂ ) within the range that falls inside this ellipse If selected, the reflectance of the antireflection film can be 0.4% or less.

Antireflection of two layers having a refractive index and a film thickness shown in Table 3 to be described later by plasma enhanced CVD method described in JP-A-2001-242630 and Proceedings of SPIE 2003, 5039, 152 (K. Babich et al.) The film was formed on the surface of a silicon semiconductor substrate on which a silicon nitride film was formed. The silicon nitride film was formed on the silicon substrate by a thermal CVD method. Specifically, SiH ₄ + NH ₃ was used as a source gas in the CVD method, and the reaction temperature was set to 400 ° C.

  The plasma enhanced CVD method is a film forming method performed in a parallel electrode reactor as described in detail in the above-mentioned reference, and a silicon semiconductor substrate is placed on one electrode. A negative bias is applied to the silicon semiconductor substrate by this electrode, the pressure in the reactor, the type of reaction precursor introduced into the reactor (tetramethylsilane, trimethylsilane, tetramethyltetrasiloxane, tetramethylgermane, oxygen Etc.), flow rate, and substrate temperature can be controlled to form and form layers having various values of complex refractive index.

  In Table 3, Examples are antireflection films that satisfy the conditions of the present invention, while Comparative Examples are antireflection films for comparison that do not satisfy the conditions of the present invention.

  More specifically, in Table 3 showing the evaluation results in 1.0 <NA ≦ 1.1, in Example 1-1, the film thickness (T) of the silicon nitride film is 45 (nm) <T ≦ 55 (nm ), And Example 1-2 is an antireflection film satisfying the conditions of the present invention when 55 (nm) <T ≦ 65 (nm). Example 1-3 is an antireflection film that satisfies the conditions of the present invention when 95 (nm) <T ≦ 105 (nm). Example 1-4 is 145 (nm) <T ≦ 155 (nm). In Example 1, the antireflection film satisfies the conditions of the present invention when 175 (nm) <T ≦ 185 (nm). On the other hand, Comparative Example 1-1 is an antireflection film that does not satisfy the conditions of the present invention in the case of 45 (nm) <T ≦ 55 (nm). Comparative Example 1-2 has 55 (nm) <T ≦ It is an antireflection film that does not satisfy the conditions of the present invention in the case of 65 (nm), and Comparative Examples 1-3 and 1-4 are those of the present invention in the case of 95 (nm) <T ≦ 105 (nm). It is an antireflection film that does not satisfy the conditions, and Comparative Example 1-5 is an antireflection film that does not satisfy the conditions of the present invention when 145 (nm) <T ≦ 155 (nm). 175 (nm) <T ≦ 185 (nm) An antireflection film that does not satisfy the conditions of the present invention.

  Here, the material constituting the upper and lower layers of the antireflection film in the examples and comparative examples in Table 3 is SiCOH. The complex refractive index of each film was measured with an ellipsometer manufactured by SOPRA.

  On these two antireflection films, as a resist layer, a photoresist ARX2014J made by JSR Corporation was spin-coated so as to have a film thickness of 100 nm, followed by baking at 105 ° C. for 60 seconds, As a top coat layer, a top coat material TCX001 manufactured by the same company was spin-coated so as to have a film thickness of 30 nm. Thereafter, the entire film was baked at 100 ° C. for 30 seconds.

  The sample thus prepared was exposed using a two-beam interference exposure apparatus. In this two-beam interference exposure apparatus, an ArF excimer laser is used as a light source, and a prism having a triangular or pentagonal cross section is installed on the optical path of the laser. The sample was placed on the lower surface of the prism so that the distance between the sample and the lower surface of the prism was 1 mm. For example, when a prism having a triangular cross section is used, the apex of the prism is arranged at the center of the optical path of the laser, and the surface facing the apex is the lower surface of the prism. When laser light is irradiated onto the prism from above the prism toward the lower surface of the prism, the laser light incident on the two side surfaces of the prism is refracted depending on the angle between each side surface and the incident laser light, and the optical path The direction changes. A laser light having a different traveling direction from each of the two side surfaces interferes with the lower surface of the prism, so that a periodic optical intensity distribution can be obtained on the sample, whereby the resist layer can be exposed. The converted numerical aperture NA of the prism used in the evaluation in 1.0 <NA ≦ 1.1 is 1.05.

  Immersion exposure using water as an immersion liquid was performed by introducing water into the 1 mm gap between the sample and the lower surface of the prism using a capillary phenomenon.

  The exposed sample was baked at 120 ° C for 90 seconds, then developed with a standard developer solution of 2.38% TMAH (tetramethylammonium hydroxide), and a resist pattern observation sample did. The shape of the resist layer is observed by dividing the silicon semiconductor substrate, observing the cross section with a scanning electron microscope, and scattering in the range of 40 μm × 40 μm using optical scatterometry (SCD-100 manufactured by KLA-Tencor). The accuracy of the machining shape due to the light distribution was measured.

  As a result of the observation, a resist pattern having a good rectangular cross section was obtained as a circle “◯”, and otherwise a good rectangular cross section was obtained as a cross “X” in Table 3. I summarized it.

  As is clear from Table 3, the antireflection film having the two-layer structure to which the present invention is applied can obtain a better resist cross-sectional shape than the antireflection film having the two-layer structure to which the present invention is not applied. I understand that I can do it.

  As described above, by applying the present invention, a two-layer antireflection film having a complex refractive index and a film thickness in a specific range is formed between a resist layer and the surface of a silicon semiconductor substrate on which a silicon nitride film is formed. By forming in between, in the antireflection film corresponding to the numerical aperture NA of the exposure system in a certain range, the reflectance from the silicon semiconductor substrate can be reduced and a good resist pattern can be obtained.

  In the above-described embodiment, the antireflection film having a two-layer structure formed by the plasma enhanced CVD method has been described as an example. However, the present invention is not limited to this, for example, a spin coating method or the like. A two-layer antireflection film formed by other methods can also be used.

  A semiconductor device was manufactured using the antireflection film having a two-layer structure of the present invention. A phase shift mask was used as the exposure mask, an ArF excimer laser (wavelength λ: 193 nm) was used as the exposure light source, and an annular illumination method was adopted. The surface of the resist layer was covered with an aqueous layer. And it verified whether a desired pattern could be formed in a resist layer, without the fluctuation | variation of a line width or a shape. As a result, it has been found that in any case, a desired pattern can be formed on the resist layer without variation in line width or shape. Moreover, in any case, the reflectance was 0.4% or less.

  Specifically, an element isolation region having a trench structure was formed. That is, an antireflection film having a two-layer structure is formed on a silicon semiconductor substrate having a silicon nitride film formed on the surface, a resist layer is formed thereon, and the resist layer is exposed and developed to be patterned. A resist layer was obtained. Next, using the patterned resist layer as an etching mask, the silicon semiconductor substrate having the silicon nitride film formed on the surface was etched to a predetermined depth by RIE to form a trench in the silicon semiconductor substrate. Thereafter, an insulating film is formed on the entire surface of the silicon semiconductor substrate including the trench, and the insulating film on the surface of the silicon semiconductor substrate is removed, thereby having a trench structure in which the insulating film is embedded in the trench formed in the silicon semiconductor substrate. An element isolation region was obtained.

  As mentioned above, although this invention was demonstrated based on the preferable Example, this invention is not limited to these Examples. The configuration of the antireflection film and the film thickness and complex refractive index of each layer constituting the antireflection film in the examples are examples, and can be appropriately changed.

FIG. 1 is a graph showing the result of calculating the dependency of contact hole diameter variation on reflectance.

Claims (9)

A resist layer used in a manufacturing process of a semiconductor device and used when exposing a resist layer in an exposure system having a wavelength of 190 nm to 195 nm and a numerical aperture NA of 1.0 <NA ≦ 1.1 An antireflection film having a two-layer structure provided between a silicon nitride film having a thickness T (unit: nm) formed on the surface of a silicon semiconductor substrate and having the following value:
The upper-layer complex refractive index N ₁ and the lower-layer complex refractive index N ₂ constituting the antireflection film are respectively expressed as follows:
N ₁ = n ₁ −k ₁ i
N ₂ = n ₂ −k ₂ i
The upper layer thickness is d ₁ (unit: nm), the lower layer thickness is d ₂ (unit: nm),
When one of the following is selected as a combination of values of [n ₁₀ , k ₁₀ , d ₁₀ , n ₂₀ , k ₂₀ , d ₂₀ ] depending on the film thickness T of the silicon nitride film,
An antireflection film, wherein n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , and d ₂ satisfy the following relational expression.
{(N ₁ −n ₁₀ ) / (n _1m −n ₁₀ )} ² + {(k ₁ −k ₁₀ ) / (k _1m −k ₁₀ )} ² + {(d ₁ −d ₁₀ ) / (d _1m −d ₁₀ )} ² + {(n ₂ −n ₂₀ ) / (n _2m −n ₂₀ )} ² + {(k ₂ −k ₂₀ ) / (k _2m −k ₂₀ )} ² + {(d ₂ − d ₂₀ ) / (d _2m −d ₂₀ )} ² ≦ 1
However, the value of n _{1m in} the case is adopted based on the magnitude relationship between n ₁ and n ₁₀ , the value of k _{1m in} the case is adopted based on the magnitude relationship between k ₁ and k ₁₀ , and d ₁ and d ₁₀ The value of d _{1m in} the case is adopted based on the magnitude relationship, the value of n _{2m in} the case is adopted based on the magnitude relationship between n ₂ and n ₂₀ , and the value of k in the case based on the magnitude relationship between k ₂ and k ₂₀ is adopted. adopts a value of _2m, adopts the value of d _2m in the case based on the magnitude relation between d ₂ and d _20.

The antireflection film according to claim 1, wherein d ₁ ≦ 250 is satisfied and d ₂ ≦ 250 is satisfied.   The antireflective film according to claim 1, wherein the resist layer has a refractive index of 1.60 to 1.80. It is used in the manufacturing process of a semiconductor device, and is formed on the surface of a resist layer and a silicon semiconductor substrate in an exposure system having a wavelength of 190 nm to 195 nm and a numerical aperture NA of 1.0 <NA ≦ 1.1. An exposure method in which the resist layer is exposed in a state where an antireflection film having a two-layer structure is provided between a silicon nitride film having a film thickness T (unit: nm) having the following value:
The upper-layer complex refractive index N ₁ and the lower-layer complex refractive index N ₂ constituting the antireflection film are respectively expressed as follows:
N ₁ = n ₁ −k ₁ i
N ₂ = n ₂ −k ₂ i
The upper layer thickness is d ₁ (unit: nm), the lower layer thickness is d ₂ (unit: nm),
When one of the following is selected as a combination of values of [n ₁₀ , k ₁₀ , d ₁₀ , n ₂₀ , k ₂₀ , d ₂₀ ] depending on the film thickness T of the silicon nitride film,
An exposure method wherein an antireflection film in which n ₁ , k ₁ , d ₁ , n ₂ , k ₂ , and d ₂ satisfy the following relational expressions is used.
{(N ₁ −n ₁₀ ) / (n _1m −n ₁₀ )} ² + {(k ₁ −k ₁₀ ) / (k _1m −k ₁₀ )} ² + {(d ₁ −d ₁₀ ) / (d _1m −d ₁₀ )} ² + {(n ₂ −n ₂₀ ) / (n _2m −n ₂₀ )} ² + {(k ₂ −k ₂₀ ) / (k _2m −k ₂₀ )} ² + {(d ₂ − d ₂₀ ) / (d _2m −d ₂₀ )} ² ≦ 1
However, the value of n _{1m in} the case is adopted based on the magnitude relationship between n ₁ and n ₁₀ , the value of k _{1m in} the case is adopted based on the magnitude relationship between k ₁ and k ₁₀ , and d ₁ and d ₁₀ The value of d _{1m in} the case is adopted based on the magnitude relationship, the value of n _{2m in} the case is adopted based on the magnitude relationship between n ₂ and n ₂₀ , and the value of k in the case based on the magnitude relationship between k ₂ and k ₂₀ is adopted. adopts a value of _2m, adopts the value of d _2m in the case based on the magnitude relation between d ₂ and d _20.

The exposure method according to claim 4, wherein d ₁ ≦ 250 is satisfied and d ₂ ≦ 250 is satisfied.   The exposure method according to claim 4, wherein the refractive index of the resist layer is 1.60 to 1.80.   The exposure method according to claim 4, wherein a topcoat layer is formed on the resist layer.   5. The exposure method according to claim 4, wherein a space between the resist layer and the exposure system is filled with a medium having a refractive index of 1.44 ± 0.02. The exposure method according to claim 8, wherein the medium is water.

Images