JPH05158217A - Exposing mask - Google Patents

Abstract

PURPOSE:To easily provide an exposing mask capable of transferring a highly accurate pattern whose edge is sharp. CONSTITUTION:At least, a part of a mask pattern is constituted of a phase shifter, and a large area pattern in the phase shifter, is constituted of a small pattern R2 whose size is comparatively large, inside a pattern, and simultaneously, a shift pattern R1 whose size is small, in the vicinity of a pattern edge. It is desirable that the small pattern is a pattern like a cross on a board, or a stripe-like pattern.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure mask,
In particular it relates to lithographic masks.

[0002]

2. Description of the Related Art Semiconductor integrated circuits are becoming highly integrated and miniaturized. When manufacturing the semiconductor integrated circuit,
Lithography technology is particularly important as a cornerstone of processing.

In the current lithographic technology, a method of projecting and exposing a mask pattern on an LSI substrate through a reduction optical system is mainly used, but if a high pressure mercury lamp is used as a light source, the minimum line width is about 0.5 μm. Is the limit. 0.5
Direct patterning technology using KrF excimer laser or electron beam and X-ray same-magnification exposure technology are being developed for pattern dimensions of less than μm, but due to reasons such as mass productivity and process versatility, optical lithography Expectations for are very high.

Under these circumstances, g
Of various light sources such as X-ray, i-line, excimer laser, and X-ray is being considered, and development of a new resist and a new resist treatment such as REL are also being considered for the resist. Research is also proceeding on the CEL image reverse method and the like.

On the other hand, regarding the mask manufacturing technique,
Though not fully examined, 1982 IBM
Levenson et al. Of the company proposed a phase shift method and received attention.

The phase shift method is a technique for improving the resolution and contrast of a projected image by manipulating the phase of light passing through a mask.

In this method, as shown in FIG. 8A, chromium (C) formed on the quartz substrate 11 by a sputtering method or the like is used.
r) or chromium oxide (Cr ₂ O ₃ ), the transparent film 1 is formed on one of a pair of adjacent transparent portions of the mask pattern 12.
By using the mask formed with No. 3, the phase of this part is inverted as shown in FIG. 8 (b) so that the light amplitudes cancel each other at the boundary between two transmission parts (FIG. 8 (c)). ). As a result, the light intensity at the boundary between the two transmissive portions becomes 0, and the pattern formed on the wafer can be separated from the two transmissive portions as shown in FIG. 8D. In this way, the 0.7 μm pattern was resolved by the g-line stepper with NA = 0.28, and the resolution was improved by about 40%. At this time, in order to invert the phase, the film thickness d of the phase shifter is d = λ / 2 where n is the refractive index of the shifter material and λ is the exposure wavelength.
The relationship of (n-1) is required.

Furthermore, Nitayama et al. Of Toshiba have proposed a phase shift mask structure in which a phase shifter is provided around the transmissive portion or the light shielding portion.

FIG. 9 shows this principle. As shown in FIG. 9 (a), this mask is projected so as to extend around the mask pattern 12 formed of a laminated film of chromium (Cr) and chromium oxide (Cr ₂ O ₃ ) formed on the quartz substrate 11. Using a mask with a transparent film formed as the formed phase shifter 13, as shown in FIG. 9 (b), the phase of this portion is inverted so that the light amplitude is 0 ° and 180 ° at both ends of the transmitting portion. They cancel each other out to reduce the light intensity and improve the contrast (Fig. 9 (c)). As a result, the light intensity at both ends of the transmissive portion becomes almost zero, and the pattern formed on the wafer can be separated from the two transmissive portions as shown in FIG. 9D.

In this mask, PMMA is used as a phase shifter.
However, PMMA has a high transmittance,
Since the resist profile is sharp, it becomes a good phase shifter. Further, in this method, the pattern of the phase shifter can be formed in a self-aligned manner, so that mask alignment and selective processing steps are unnecessary, and the pattern can be easily formed.

When exposure is performed using such a mask,
The light passing through each opening has their phases inverted as shown by the broken line, so the light intensity at the lower part of the mask layer is significantly reduced, and the light as a whole is shown by the solid line. In view of the intensity distribution, it is possible to resolve up to approximately half the size of the conventional mask.

Further, the shifter edge utilizing type mask has a phase shifter arranged on a transparent substrate as shown in FIG. The mask forming process is simple, the effect of improving resolution is large, and a large focus margin can be expected.

FIG. 11 shows a comparison of resolution improvement effects by simulation of these phase shift methods, and FIG. 12 shows a comparison of focus margins. From these comparisons, it can be said that the Levenson type and the shifter edge utilizing type have a larger effect of improving resolution and a larger focus margin than the self-aligning type.

Although the Levenson type phase shift mask can be applied to an isolated pattern by using an auxiliary pattern, the mask forming process requires forming a shifter after patterning a light shielding material on a transparent substrate. There are complex elements such as

On the other hand, the shifter edge type mask has a problem that the mask forming process is simple and the effect of improving the resolution is large and a large focus margin can be expected, but the resist pattern size after exposure is uniformly determined. That is, as shown in FIG. 13, since the light transmitted through the mask shown in (a) exposes the resist with the intensity distribution shown in (b), the resist pattern when the negative resist is used is ( It has a constant size F as shown in c). That is, the pattern size F of the resist formed using this mask is determined only by the exposure amount.

Therefore, when a pattern other than a fine pattern is formed by using the edge-using phase shift method, the edge-using phase shift method as shown in FIG. 14 is used. In this case, when the pattern width d is large, the dark portions of the transmitted light formed by the respective edges are combined. Here, since the light transmitted through the mask shown in (a) exposes the resist with the intensity distribution shown in (b), the resist pattern when a negative resist is used is as shown in (c). It is slightly larger than the pattern width d. This method is called a double edge type. By this method, a small pattern other than a fine pattern can be formed.

In order to form such a phase shifter using both edges, it is necessary that the size is such that the dark portion of the transmitted light formed at each edge is properly combined. A method of combining patterns of a predetermined size to form a large pattern has been proposed, but if one shifter size is too large as shown in FIG. 15 (a), then FIG.
As shown in FIG. 3, there is a problem that unevenness is formed at the end of the resist pattern.

Therefore, it is preferable to make the size sufficiently small, but if one shifter size is small as shown in FIG. 16 (a), the resist pattern end portion is formed smoothly as shown in FIG. 16 (b), but the shifter is formed. However, there is a problem that the number of CAD data becomes large and the CAD data becomes huge when computer processing is performed, which is not practical.

[0019]

As described above, when a large area pattern is formed on the same surface as a fine pattern by using the conventional shifter edge utilizing type phase shift method, it is attempted to arrange phase shifters of a small size. There has been a problem that the number of shifters increases and the CAD data becomes enormous, which makes practical application difficult.

The present invention has been made in view of the above circumstances, and it is an object of the present invention to easily provide an exposure mask which has a sharp edge and which can perform highly accurate pattern transfer.

[0021]

Therefore, in the exposure mask of the present invention, at least a part of the mask pattern is composed of a phase shifter, and a large area pattern of the phase shifter has a relatively large internal area of the pattern. And a small size shifter pattern in the region near the pattern edge.

This small pattern is preferably a grid pattern or a stripe pattern.

[0023]

According to the above construction, in the case of a large area pattern, the inner area of the pattern is composed of a relatively large size small pattern and the area near the pattern edge is composed of a small size shifter pattern. Reduce the number of CAD data to reduce the CAD data,
An exposure image with smooth edges can be formed.

[0024]

Embodiments of the present invention will now be described in detail with reference to the drawings.

Example 1 FIG. 1 is a diagram showing a cross section of an exposure mask of Example 1 of the present invention.

This exposure mask is shown in FIGS. 1 (a) and 1 (b).
As shown in the plan view and the cross-sectional view in FIG. 5, the mask is a quintuple mask, and the film thickness of 0.25 μm is formed on the surface of the transparent quartz substrate 1.
An internal region R1 made of a 0.35 μm × 0.35 μm square (on wafer) grid-like first small pattern 2s made of a semitransparent film 2 (having a transmittance almost the same as that of a quartz substrate);
0.25 μm made of the same semi-transparent film as the small pattern 2s
It has a phase shifter pattern composed of an edge region R2 composed of a second minute pattern 2b having a grid pattern of 0.25 μm square (on the wafer). In other areas, a fine pattern portion (not shown) formed of only a fine pattern is also formed.

As the semitransparent film, a silicon oxide film having an optical constant substantially equal to that of the quartz substrate 1 is selected.

This film thickness is configured to have a thickness of λ / 2 (n-1), and the phase of the light transmitted through this semitransparent film and the normal phase transmitted through the transmissive region adjacent to that part. 180 degrees different from the phase of the light. The allowable range of phase shift due to film thickness variation is about ± 10%. Here, λ is the wavelength of the exposure light, and n is the refractive index of the shifter material (semitransparent film) at the exposure wavelength.

The exposure mask thus formed is mounted on a KrF excimer laser stepper having a projection lens with NA = 0.42, and a pattern is transferred (λ) onto a wafer coated with a negative resist on a silicon substrate. = 248
nm, coherence factor σ = 0.5) and developed with a dedicated developer.

This pattern is exposed and transferred onto the resist pattern to form a large resist pattern having smooth edges as shown in FIG. 1 (c).

According to this exposure mask, even when a minute pattern and a large area pattern are mixed, it can be formed at the same time, and a pattern can be drawn with a smaller amount of data, and the edge has a smooth and highly accurate pattern. Can be obtained.

As the phase shifter, a semitransparent film such as a resist or a silicon oxide film may be formed on a transparent substrate, but as shown in a modified example in FIG. A part of the flexible substrate 1 may be removed by etching to a thickness of λ / 2 (n-1). Here, n is the refractive index of the substrate at the exposure wavelength. The phase of the light transmitted through the removed portion and the phase of the normal non-removed light adjacent to the portion are 18
0 degrees different. The allowable range of phase shift due to variations in etching amount is about ± 10%.

Next, the dependence of shifter size on the exposure amount is shown in FIG. The shifter size on the vertical axis also changes depending on the stepper and the resist. The lower limit of the exposure amount is determined by the residual film characteristics (D curve) of the resist as shown in the figure. Here, the lower limit value of the exposure amount is shown as a curve with a residual film ratio of 80% as an example, but the residual film ratio changes to 90% or 70%, for example, in the subsequent process. The size of the shifter used in the central portion of the pattern is as large as possible in order to reduce the data in the portion on the right side of the D curve below the exposure amount dependency (A curve or B curve) of the shifter size. For example, when the shape of the shifter is a grid, the size is increased as much as possible in the shaded area in the figure. The shifter size may be selected to be as large as possible on the right side of the D curve below the C curve in the figure so as to smooth the end of the pattern.

In the above-mentioned embodiment, the internal region is formed in a grid pattern, but it may be formed in a stripe pattern as shown in FIGS. At this time, a minute pattern may be formed on the longitudinal side of the stripe as shown in FIG. 4, or a minute pattern may not be formed on the longitudinal side as shown in FIG. You may make it form a pattern. In both FIGS. 4 and 5, (b) is a pattern formed on the resist pattern.

The structure of the phase shifter is shown in FIG.
A semitransparent film may be formed on the translucent substrate as shown in FIG. 2 or a structure in which the translucent substrate is etched by a predetermined thickness as shown in FIG.

In addition, FIG. 6 (a) shows the corner portion of the pattern.
It is desirable to form so that a minute pattern comes as shown in FIG. The exposure pattern image on the resist at this time is shown in FIG.
Shown in (b). When the blanks are placed at the corners as shown in FIG. 6 (c), the exposure pattern image on the resist at this time has a pattern chip at the corners as shown in FIG. 6 (d). The desired pattern cannot be obtained.

Another example will be described.

When it is desired to obtain a resist pattern having a complicated edge as shown in FIG. 7A, the internal region R1 is composed of a phase shifter having a large stripe and the edge region R2 is minute as shown in FIG. 7B. You may make it comprised by the phase shifter which consists of a pattern.

Further, as shown in FIG. 7 (c), the internal region R1 may be constituted by a phase shifter having a large grid pattern and the edges may be constituted by a phase shifter having a minute pattern.

[0040]

As described above, according to the present invention, the number of shifter small patterns can be reduced to reduce the CAD data, and at the same time, the edge can be sharp and the pattern transfer can be performed with high accuracy. A mask can be provided.

[0041]

[Description of Drawings]

[0042]

FIG. 1 is a diagram showing an exposure mask according to a first embodiment of the present invention.

[0043]

FIG. 2 is a view showing a modified example of the exposure mask.

[0044]

FIG. 3 is a diagram showing exposure amount dependency of shifter size.

[0045]

FIG. 4 is a diagram showing another embodiment of the present invention.

[0046]

FIG. 5 is a diagram showing another embodiment of the present invention.

[0047]

FIG. 6 is a diagram showing another embodiment of the present invention.

[0048]

FIG. 7 is a diagram showing another embodiment of the present invention.

[0049]

FIG. 8 is a diagram showing a Levenson-type phase shift method.

[0050]

FIG. 9 is a diagram showing a self-aligned phase shift method.

[0051]

FIG. 10 is a diagram showing an edge-based phase shifter.

[0052]

FIG. 11 is a comparison diagram of the resolution improving effect by simulation of each phase shift method of the conventional example.

[0053]

FIG. 12 is a diagram showing comparison of focus margins by simulation of phase shift methods of a conventional example.

[0054]

FIG. 13 is a diagram showing a phase shift method using a single edge.

[0055]

FIG. 14 is a diagram showing a phase shift method using both edges.

[0056]

FIG. 15 is a diagram showing an example of a conventional phase shift pattern.

[0057]

FIG. 16 is a diagram showing an example of a conventional phase shift pattern.

[0058]

[Explanation of symbols]

 11 Quartz substrate 12 Light-shielding film 13 Phase shifter (semi-transparent film)

Claims (2)

[Claims] 1. An exposure mask comprising a translucent substrate and a mask pattern provided on the translucent substrate, wherein a so-called mask pattern having different optical path lengths with respect to exposure light is used. Including a semi-transparent film pattern as a phase shifter, of the semi-transparent film pattern, for a large-area pattern, the internal area of the pattern is composed of a small shifter pattern of a relatively large size, and a small size of a small size is formed near the pattern edge. An exposure mask characterized by being configured with a shifter pattern. 2. The exposure mask according to claim 1, wherein the small shifter pattern is formed in a grid pattern or a stripe pattern.