JP3166223B2 - Phase shift exposure mask, method for manufacturing phase shift exposure mask, and exposure method using phase shift exposure mask - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a phase shift exposure mask , a method of manufacturing a phase shift exposure mask , and a phase shift exposure mask.
The present invention relates to an exposure method using an exposure mask . According to the present invention
The phase shift exposure technique can be used in various fields. For example, the phase shift exposure mask of the present invention can be used as a mask for forming various patterns, for example, used as an exposure mask for forming various patterns such as a resist pattern in a semiconductor device manufacturing process. be able to.

[0002]

2. Description of the Related Art Processes for forming a pattern using a photomask, for example, a semiconductor device, tend to be miniaturized year by year. Against this background, in a photolithography technique for obtaining a miniaturized semiconductor device, in order to further improve the resolution, a so-called phase shift technique for giving a phase difference to light transmitted through a mask and thereby improving a light intensity profile. Is in the limelight.

A conventional phase shift technique is disclosed in Japanese Patent Application Laid-Open No. 58-173744 and MARC D.A. LEVE
NSON et al. “Improving Resolution
onin Photolithography wit
ha Phase-Shifting Mask "I
EEE TRANSACTIONS ON ELECT
RON DEVICES. Vol. ED-29 No.
12, DECEMBER 1982, P1828-18
36, and MARC D.C. Levenson and others "T
he Phase-Shifting Mask II:
Imaging Simulations and S
ubmicrometer Resist Expos
ures ", the same magazine, Vol. ED-31, No. 6, JU
NE1984, pages 753 to 763.

Japanese Patent Publication No. Sho 62-50811 has a predetermined pattern formed of a transparent portion and an opaque portion.
A phase shift mask is disclosed in which a phase member is provided on at least one of the transparent portions on both sides sandwiching the opaque portion, and a phase difference is generated between the transparent portions on both sides.

A conventionally known phase shift technique will be described with reference to FIG.
For example, when a line and space pattern is to be formed, a conventional mask usually includes Cr (chromium), other metal, and metal oxide on a transparent substrate 1 such as a quartz substrate, as shown in FIG. The light-shielding portion 10 is formed by using a light-shielding material such as an object, thereby forming a line-and-space repetitive pattern to be used as an exposure mask. The intensity distribution of the light transmitted through the exposure mask is shown in FIG.
As shown in FIG.
Is zero at other portions (the transmission portions 12a and 12a).
In b), the light is transmitted. Considering one transmission portion 12a, the transmitted light given to the material to be exposed has a light intensity distribution having small mountain-like maxima on both sides as shown by A2 in FIG. Become. The transmitted light A2 'of the transmission part 12b is shown by a dashed line. Each transmission part 12a, 12
When the light from b is combined, the light intensity distribution loses sharpness as shown in A3, and an image is blurred due to light diffraction.
As a result, sharp exposure cannot be achieved. On the other hand, the light transmitting portions 12a, 12b of the above-described repetitive pattern
As shown in FIG. 7B, every other phase shift portion 11a (referred to as a shifter, which is made of a material such as SiO ₂ or a resist) is provided on the top of each other, and the image is blurred due to light diffraction. Is canceled by the reversal of the phase, a sharp image is transferred, and the resolution and the focus margin are improved. That is, as shown in FIG. 7B, when the phase shift portion 11a is formed in one of the transmission portions 12a, if the phase shift portion 11a gives a phase shift of, for example, 180 °, it passes through the phase shift portion 11a. The light is inverted as shown by the symbol B1. Since the light from the transmission part 12b adjacent thereto does not pass through the phase shift part 11a, such inversion does not occur. In the light applied to the material to be exposed, the lights inverted from each other cancel each other out at the position indicated by B2 in the figure at the bottom of the light intensity distribution, and the distribution of the light applied to the material to be exposed is eventually as shown in FIG. As shown by B3, a sharp ideal shape is obtained.

In the above case, it is most advantageous to invert the phase by 180 ° in order to ensure this effect most. (N is the refractive index of the phase shift portion, and λ is the exposure wavelength).

When a pattern is formed by exposure, an object to be reduced and projected is referred to as a reticle, an object to be projected one-to-one is referred to as a mask, or an equivalent to a master is referred to as a reticle, and a duplicate thereof is referred to as a mask. However, in the present invention, masks and reticles in such various meanings are collectively referred to as masks.

[0008] The above-described phase shift mask that shifts the phase of light (ideally 180 ° inversion) between adjacent light transmitting portions is called a spatial frequency modulation type (or Levenson type). In addition, the phase shift mask
There are various types such as an edge enhancement type, a light-shielding effect enhancement type, and the like. Among them, there is a type formed without a light-shielding portion (called a chromeless type or the like).
In any case, at least a part of the portion that transmits the exposure light is provided with a phase shift unit that shifts the phase.

[0009]

The problems of the conventional phase shift exposure mask will be described below with reference to FIG. The structure example shown in FIG. 2 is a so-called Cr-less mask having no light-shielding portion. However, the problem to be solved by the present invention occurs regardless of the presence or absence of the light-shielding portion. Can be applied to various phase shift masks regardless of the principle. In FIG. 2, a transparent conductive film for preventing charge-up at the time of EB drawing is omitted.

Referring to FIG. This figure shows how a phase difference occurs due to the difference between the refractive index n of a phase shift unit (hereinafter also referred to as a shifter) 11 for shifting the phase of transmitted light and the refractive index ≒ 1 of air. 1 is a substrate made of quartz or the like. Now, let us consider how much difference occurs in the phase difference of transmitted light when a defect having a depth d occurs in the shifter 11.

Assuming that the original thickness of the shifter 11 is x,
Assuming that the phase is shifted by 180 ° (π) by the shifter 11,

When the shifter becomes thinner by d, the phase becomes

[0013] (Wavelength: 3650 °) In the case of using for lithography, when considering a SiO ₂ based shifter (n ≒ 1.45), it is 1.233 × 10 ^-4 · d.

Usually, the film thickness uniformity of the shifter is such that the phase difference is ±
It is required to control the temperature to within 10 ° (for this, see Table 1 on page 106 of the July 1990 issue of “Nikkei Micro Device” by Nikkei McGraw-Hill).
In the “Permissible phase error” column). That is, ± 18 for both
Only an error for the wavelength is allowed. 1.233 ×
Calculating from 10 ⁻⁴ · d ≦ 1/18, d ≦ 450 °.

In addition, since the above description is based on the assumption that the film is formed uniformly, the actual requirements become more severe. Now, the calculation is made assuming that the defect is a depression, but the same can be said for the case where the defect protrudes.

As described above, it has been pointed out that minute defects on the shifter pose a serious problem. Although it has been proposed that the defect be corrected by using a focused ion beam (FIB), an extra portion is replaced with Ga.
It has been difficult to completely flatten the surface even if it is scraped off with ions or the like, or if the insufficient part is corrected by performing CVD by FIB irradiation using a gas such as SiH ₄ .

[0017]

An object of the present invention is to solve the above-mentioned problems of the prior art and reduce the influence of defects such as minute irregularities on the surface of a phase shift portion (shifter) for shifting the phase of transmitted light and residues on a substrate. It is another object of the present invention to provide a phase shift exposure mask capable of correcting a large defect by using FIB or the like.

[0018]

SUMMARY OF THE INVENTION The present invention relates to a phase shift exposure mask having a phase shift portion for shifting the phase of transmitted light in a part of a light transmitting portion. A transparent film is formed flat at least on a light transmitting portion necessary for pattern transfer, and the refractive index difference between the phase shift portion and the film is set to 0.45 or less.
The phase error is set within ± 10 ° as the phase difference, and the above-mentioned object is achieved by this configuration. In the present invention, the light transmitting portion is a general term for both a portion that transmits light without shifting the phase and a portion that transmits light by shifting the phase (phase shift portion). Further, according to the present invention, in a method for manufacturing a phase shift exposure mask in which a phase shift portion for shifting the phase of transmitted light is provided on a part of a light transmitting portion, a coating optically transparent to the exposure light may be formed in at least a pattern. The film is formed flat on the light transmitting portion necessary for transfer, and the difference in the refractive index between the phase shift portion and the coating is set to 0.45 or less, thereby reducing the phase shift.
The present invention is characterized in that the error is set within ± 10 ° as a phase difference, and the above object is achieved by this configuration. Further, according to the present invention, in a method for manufacturing a phase shift exposure mask in which a phase shift portion for shifting the phase of transmitted light is provided in a part of a light transmitting portion, a phase shift portion formed on a substrate, The defect of the shift portion is corrected, and a coating optically transparent to exposure light is formed flat on at least a portion necessary for pattern transfer on the phase shift portion after the defect correction , and the phase shift Department and
The difference in the refractive index from the film is set to 0.45 or less, and the phase error is reduced.
The phase difference is set within ± 10 °, and the above object is achieved by this configuration. Further, the present invention is an exposure method using a phase shift exposure mask provided with a phase shift portion for shifting the phase of the transmitted light in a part of the light transmission portion, after forming the phase shift portion on the substrate, performs correction of defects in the formed phase shifter, an optically transparent film on the phase shift unit after correction of said defective to exposure light, while flatly formed on the portions necessary for at least the pattern transfer , The phase shift unit
And the refractive index difference between the film and the film is set to 0.45 or less, and the phase error
Is performed using a phase shift exposure mask having a phase difference within ± 10 ° ,
With this configuration, the above object is achieved.

The structure and operation of the present invention will be described below with reference to FIG. The present invention is shown in FIG.
As shown in the above example, at least on a portion necessary for pattern transfer (portion contributing to pattern exposure) including on a phase shift portion (shifter) 11 for shifting the phase of transmitted light,
The effect of defects is reduced by forming the coating 2 transparent to exposure light flat.

The coating 2 is required to have a flat surface. In the present invention, flat means that the distance from the coating surface to the substrate is the same. That is, the thickness from the substrate to the surface of the coating 2 is a portion where the shifter 11 is present or not, and in some cases, a portion where a light shielding portion such as Cr is present,
Must be equal. Of course, equality is within a range that does not affect the phase shift effect, and the variation in film thickness due to film formation or the like can be neglected to some extent.

According to the present invention, the presence of the coating 2 as described above and the difference in refractive index between the phase shift portion 11 and the exposure environment (typically air) are 0.45 or less.
The effects of the present invention are provided. For example, an SiO ₂ -based shifter most commonly used in a mask and having a refractive index closest to air (≒ 1) has a refractive index of 1.4 to
Since it is 1.45, in this case, the difference in the refractive index between the coating 2 and the phase shift portion 11 is preferably 0.45 or less.
At this time, the difference in refractive index between the SiO ₂ -based shifter and the exposure environment (air) is about 0.45, and the effect is obtained when the difference is less than 0.45.

The operation of the present invention will be described in more detail as follows. As shown in FIG. 1, the thickness x2, x1 and the refractive index n2, n1 of the coating 2 and the shifter 11 are determined, and the influence of the defect is calculated. First, the following equation is established between a portion having no shifter 11 and a portion having the shifter 11. n ₂ x ₁ + n ₂ x ₂ = aλ... (Where a is a real number, k is an odd number, a> 0, k> 0)

Assuming a depression of depth d as in FIG.
The phase is If there is no depression, the difference is 2πa-πk. Becomes Again, considering the i-line and the SiO ₂ shifter, assuming that n ₂ = 1.65 as the coating, the allowable range of d is ±
1010 °. However, x ₂ = (a / 1.65−k /
0.4) Since λ, if k = 1, a ≧ 4.12
5, k = 1 and a = 5, x ₁ = 9125Å, x ₂
= 1935 °.

Of course, it is not necessary that a = 5 and it is not necessary to be an integer. It is.

[0025] It should be considered as. Calculate based on this, n ₁ =
If 1.65, n ₂ = 1.45, exactly the same result would be obtained. This effect is not limited to shifter defects. For example, even if there is a residue 3 of the shifter material as shown by hatching in FIG. 1, the effect can be reduced.

As described above, by selecting a material and forming a shifter and a coating so that the difference between n ₁ and n ₂ becomes small,
The effects of defects are reduced.

[0027]

Embodiments of the present invention will be described below with reference to the drawings. However, needless to say, the present invention is not limited to the embodiments described below.

Example 1 Here, a phase shift exposure mask was formed by the steps shown in FIG. The phase shift exposure mask of this example is of a so-called Levenson type, and is SOG (Spin).
or glass) SiO ₂ shifter 11
To form a coating 2 of a resist. Initially, the same steps as those for forming a normal phase shift mask are performed, that is, the transparent conductive film 6 (here, ITO, ie, Indium) is formed on the quartz substrate 1.
An EB resist is applied to a mask formed in the order of a light-shielding material film made of Cr, and the light-shielding portion 10 is patterned by EB drawing.
In this case, the transparent conductive film 6 of ITO has a thickness of 300
The thickness was 800 mm, and ERR-9 (Toray Industries, Inc.) was used as the EB resist.

After the light-shielding portion 10 is formed by performing the Cr pattern etching, SOG serving as a shifter is applied. However, since it becomes considerably thick as calculated previously, here, Tokyo Oka OCD TYPE which can be applied thickly as SOG is used. VII was used. After coating (spin coating) at 2500 rpm for 15 seconds, baking is performed at 200 ° C. As a result, almost 7
A film of 300 ° (± 50 °) was formed. EB on this
A resist (SEL-N552, Somar Co.) was applied, EB lithography for forming a shifter was performed, and after development, etching was performed with a 2% hydrofluoric acid solution to form a shifter 11. The EB resist was removed to obtain the structure shown in FIG.

Thereafter, a resist SAL-601 (manufactured by Shipley Co., Ltd., a chemically amplified resist using novolak resin as a base resin, a cross-linking agent as methylol melamine, and DDT as a photoacid generator) was applied at 2000 rpm for 1 minute.
Since a level difference is left as it is, coating was further performed twice under the same conditions, and baking was performed at 85 ° C. for 1 minute as two layers to finish. The coating 2 was formed by the resist layers 41 and 42. Although the film thickness became approximately 1.7 μm from the substrate,
AL-601 has high transparency for i-rays having a wavelength of 365 nm, and has a transmittance of less than 90% in this state. If the transmittance is desired to be improved, the resist SAL-601 may be etched back using an O ₂ asher to reduce the thickness. As described above, the thickness of this coating is arbitrary as long as it is flat. SAL-601 has a refractive index of 1.69 with respect to the i-line, and OCD TYPE-VII as a shifter material.
Since SiO ₂ is 1.44 derived from, from the equation, as k = 1, the x ₁ = 7300Å. In order for the phase difference to fall within ± 10 °, the shifter film thickness must be ± 810.
Within Å, the tolerance was wider than that of a normal shifter.

Embodiment 2 In this embodiment, a self-aligned phase shift exposure mask is formed by the method shown in FIG.
Was used as a shifter material, and an exposure mask having a resist film (CMS film) as a film was formed.

The difference from the first embodiment is that an SiNx layer 5 is formed on the substrate 1 by sputtering at 6300 ° instead of ITO. The Cr layer 7 for forming the light shielding portion functions as a conductive film. After patterning Cr in the same manner as in Example 1, the SiNx layer 5 is removed by dry etching with the resist remaining. Etching is O ₂ / CHF
A mixed gas of ₃ was used. The SiNx layer after the etching is indicated by 51. Thereafter, the Cr layer 7 is scraped off by side etching as shown in FIG. 6 to form a shifter portion 11 (see also FIG. 4D). Thus, the structure shown in FIG. 4B is obtained. The side-etched Cr layer becomes the light shielding portion 10. The side etching at this time may be either dry or wet, but here, wet etching using an etching solution for Cr was performed. After resist removal, CMS (chromium methyl styrene) is applied at 1000 rpm.
(FIG. 4 (c)), applied twice as in Example 1, and baked at 120 ° C. for 2 minutes to finish. As a result, the coating 2 composed of the resist layers 41 and 42
Was obtained, and the structure of FIG. 4D was obtained.

Since the refractive index of SiNx is 1.95 and the CMS is 1.66 (depending on the degree of polymerization, a material of 1.66 is used here), x ₁ = 6290 °. The allowable range was also ± 700 °, which was a sufficiently wide value.

Embodiment 3 Here, a description will be given of defect correction when obtaining the phase shift exposure mask of the present invention. As shown in FIG.
The correction of the defects 3 and 31 of 1 is most effective by the FIB. As shown in FIG. 5, the defects 3 and 31 composed of the convex portions can be removed by, for example, Ga + ions. As a result, the defect can be kept within the allowable range d (FIG. 5B).

As shown in FIG. 6, the defects 32 and 33 in the concave portions are caused to flow by a gas such as SiH ₄ + O ₂ and irradiate with a Si + ion beam to cause a CVD reaction to cause a SiO 2 reaction.
By depositing ₂ , a correction that falls within the allowable range d can be made. (If the film is SiN, the gas may be changed). In practice, this method is used for defect correction of Cr. In general, in the case of a shifter, since the film thickness is an important factor, bulges and dents which are inevitable to remain after correction are obstructive factors, but as in the present invention, a film having a small difference in refractive index is formed on the upper part By doing so, the influence of the irregularities can be reduced, so that some irregularities can be tolerated more than ever.

For the processing by FIB, regarding the embedding of the concave portion, the 6th Annual Meeting of the Japan Society of Applied Physics, Spring 1989
There is an article of 4p-L-6 on page 21, and there is an article of 29a-ZG-8 on page 538 of the Transactions of the Japan Society of Applied Physics in the fall of 1990 regarding processing of a convex defect. From these articles, it can be seen that it is possible to keep the processing accuracy sufficiently within the required range.

[0037]

According to the present invention, the influence of defects such as minute irregularities on the surface of the phase shift portion and residues on the substrate can be reduced, and large defects can be corrected by using FIB or the like. It is possible to provide a certain phase shift exposure mask, for example, it is difficult to flatten the shifter by FIB, and it is said that a phase difference remains between the uncorrected portion, but according to the present invention, By reducing the influence, it became possible to have sufficient practicality.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an example of the configuration of a phase shift exposure mask of the present invention, and is a diagram for describing the operation thereof.

FIG. 2 is a structural example of a conventional phase shift exposure mask;
It is a figure for explaining a problem.

FIG. 3 is an example of a method for producing a phase shift exposure mask of the present invention, showing a manufacturing process in Example 1.

FIG. 4 shows another example of a method for producing a phase shift exposure mask of the present invention, showing a manufacturing process in Example 2.

FIG. 5 is a diagram showing a defect repair method.

FIG. 6 is a diagram showing a defect correction method.

FIG. 7 is a diagram illustrating the principle of a phase shift exposure mask.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Substrate 2 Coating 11 Phase shift part (shifter)

Claims (4)

(57) [Claims] 1. A phase shift exposure mask in which a phase shift portion for shifting the phase of transmitted light is provided in a part of a light transmitting portion, wherein a coating optically transparent to exposure light is required at least for pattern transfer. And a film having a refractive index difference between the phase shift portion and the coating film of 0.
45 or less, and phase error within ± 10 ° of phase difference
Phase shift exposure mask, wherein the meta. 2. A method of manufacturing a phase shift exposure mask in which a phase shift portion for shifting the phase of transmitted light is provided in a part of a light transmitting portion, wherein at least a pattern optically transparent to exposure light is provided on a pattern. It is formed flat on the light transmitting part necessary for transfer, and the refractive index difference between the phase shift part and the coating is 0.45 or less.
A method of manufacturing a phase shift exposure mask, wherein the phase error is set within ± 10 ° as a phase difference . 3. A method of manufacturing a phase shift exposure mask in which a phase shift portion for shifting the phase of transmitted light is provided in a part of a light transmitting portion, wherein a phase shift portion is formed on a substrate, A defect in the shift portion is corrected, and a coating optically transparent to exposure light is formed flat on at least a portion necessary for pattern transfer on the phase shift portion after the defect correction , and the phase shift Part and the cover
The phase error is set to be less than 0.45
A method for manufacturing a phase shift exposure mask, wherein the difference is set within ± 10 ° . 4. An exposure method using a phase shift exposure mask provided with a phase shift portion for shifting the phase of transmitted light in a part of a light transmission portion, wherein the phase shift portion is formed on a substrate. performs correction of defects in the formed phase shifter, an optically transparent film on the phase shift unit after correction of said defective to exposure light, while flatly formed on the portions necessary for at least the pattern transfer , The phase shift unit and the
Set the phase difference to the phase difference by setting the refractive index difference with the film to 0.45 or less.
An exposure method using a phase shift exposure mask, wherein the exposure is performed using a phase shift exposure mask within ± 10 ° .