JPH0855775A - Pattern for drawing electron beam, and proximity effect evaluation method using it, and electron beam drawing method - Google Patents

Abstract

PURPOSE:To improve the accuracy of proximity effect by providing a drawing pattern for detecting that the accumulated energy by the electron beam at the position of arrangement has reached a certain value, having such a size as not to affect the drawing area density and being arranged approximately at the center of a pattern. CONSTITUTION:An electron beam drawing pattern is composed of a pattern 1, which decides drawing area density, and a pattern 2 for detection, which does not affect the drawing area density but detects that the accumulated energy of an electron beam at the point has reached a certain value from the pattern after development. The area pattern 1 is a uniform pattern where the lines of drawing parts L in width are arranged intervals S apart. Accordingly, defining that the drawing area density is A, A=L/(L+S). For the amount of irradiation given to the area pattern 1, the accumulated energy at the center of the detection pattern is sought by the specified formula. Hereby, the drawing pattern can be selected larger enough than the range of rear scattering, and reflection coefficient can be sought independent of other proximity effect pattern.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electron beam drawing method,
In particular, the present invention relates to a proximity effect parameter evaluation pattern that is important when a proximity effect is corrected and drawing is performed, and an evaluation method using the same.

[0002]

2. Description of the Related Art In electron beam drawing, there are problems that the pattern size becomes thicker than the design size when the intervals between a plurality of drawing patterns are close to each other, and the isolated fine pattern becomes thinner than the design size. It is widely known as an effect phenomenon.

A method for solving this problem is called proximity effect correction, and several methods have been proposed. In general, a representative point is determined for each drawing pattern, and the computer calculates the optimum electron beam irradiation dose given so that the energy of the electron beam deposited at this point is constant. At this time, what is important is the deposition energy distribution function that expresses how much energy is deposited at a distance r from the irradiation point when the electron beam is irradiated and the parameter (proximity effect parameter) used there. .

As the deposition energy distribution function, an approximate expression (double Gaussian approximation) of the sum of two Gaussian functions represented by the following equation 1 is often used.

[0005]

F (r) = (2 / √π) [(1 / α ² ) {EXP (−r ² / α ² )} + (η / β ² ) {EXP (−r ² / β ² )) }] (Equation 1) where f (r) is a deposition energy distribution function, α, β, η
Are proximity effect parameters and are called forward scattering coefficient, back scattering coefficient, and reflection coefficient, respectively.

Many methods are known for the evaluation method (method of obtaining) of this proximity effect parameter. For example, technical paper (Microelectronic Engineering 5 Vol. 141 (Microelectronic Engineering 5 (1986)
p.141) and the Proceedings of Micro Process
Conference page 64 (Proceeding of MicroProcess
Conference p.64 (1993)). However, as shown in the deposition energy distribution function formula (Equation 1), the three proximity effect parameters are related to each other, and it was difficult to obtain each parameter independently from the drawing result.

On the other hand, as known from Japanese Patent Laid-Open No. 3-225816, there is a method of correcting the proximity effect by calculating the drawing area ratio in a small area and obtaining the electron beam irradiation amount from the drawing area ratio to perform drawing. Proposed. In this case, of the proximity effect parameters, in particular, the reflection coefficient η determines the correction accuracy, and therefore precise evaluation thereof is important. However, it is difficult to measure the reflection coefficient, and especially it is difficult to obtain it independently of the other two parameters.

[0008]

The problem to be solved by the present invention is to obtain the reflection coefficient η which is an indispensable parameter for the proximity effect correction in electron beam drawing independently of other parameters.

[0009]

In order to solve the above problems, in the present invention, as shown in FIG. 1, a pattern (hereinafter abbreviated as area pattern) 1 and a drawing area density for determining a drawing area density of an electron beam drawing pattern. However, the detection pattern 2 is used to detect from the pattern after development that the deposition energy of the electron beam at that point has a constant value. The area pattern 1 in FIG. 1 is a uniform pattern in which lines having a width L of a drawing portion are arranged with a space S therebetween.
Therefore, if the drawing area density is A, then A = L / (L + S). Assuming that the dose given to the area pattern 1 is D1 and the dose given to the detection pattern 2 is D2, the deposition energy E at the center of the detection pattern is obtained from the equation 1, and becomes the equation 2.

[0010]

## EQU00002 ## E = (1 + A..eta.). Multidot.D1 + D2 (Expression 2) However, in FIG. 1, the size W of the drawing pattern is sufficiently larger than the backscattering coefficient .beta. And the size of the detection pattern is sufficiently smaller than .beta .. Different areal density A (different L in FIG. 1
(Combination of S and S), the drawing energy is changed by changing the electron beam irradiation amount, and if the shapes or thicknesses of the detection patterns 2 are selected, the deposition energy in (Equation 2) is equal. Become. Therefore, if there are patterns having a plurality of area densities Aa and Ab and the electron beam doses D1a, D2a, D1b, and D2b given to them, respectively, the expression 2 becomes the expression 3 and the reflection coefficient becomes the expression 4.

[0011]

[Equation 3] E = (1 + Aa · η) · D1a + D2a = (1 + Ab · η) D1b + D2b (Equation 3)

[0012]

[Equation 4]

[0013]

In this drawing pattern, a pattern having a uniform area density such as a line / space pattern is selected as the area pattern 1 which determines the area density, or a radial distance from the evaluation point such as a fan shape. Regardless of this, by selecting a pattern having a constant area density, the drawing pattern can be selected sufficiently larger than the range of backscattering. As a result, it became possible to obtain the reflection coefficient η independently of other proximity effect parameters.

Further, in the conventional method, it was necessary to measure the dimensions of the pattern and observe it with an electron microscope. However, since the shape or the film thickness of the detection pattern 2 can be observed with a metallographic microscope, rapid evaluation is possible. Become.

[0015]

【Example】

(Embodiment 1) FIG. 2 is a view showing an embodiment of the present invention, which is an example in the case where the area pattern is a fan shape. Radius r centered on point O
A fan-shaped area pattern 3 having a μm and a central angle of θ ° and a detection pattern 4 are arranged substantially at the center thereof. The characteristic of the sector is that the drawing area density does not depend on the distance r from the observation point O but only on the central angle θ. When the accelerating voltage of the electron beam is 50 kV, the range of backscattering is about 20 μm, so if the radius r is selected to be about 20 μm or more, all the effects of backscattering can be captured. That is, the correct reflection coefficient can be obtained regardless of the value of the backscattering coefficient β.

In this embodiment, as shown in FIG. 3, the radius r of the sector is 25 μm and the central angle θ is 0 °, 45 °, 90 °.
°, 135 °, 180 °, 225 °, 270 °, 315
Nine kinds of patterns of 90 ° and 360 ° were designed. The set of 9 types of patterns was defined as a pattern group. The detection pattern 4 was a square with one side of 1 μm. The dose given to the area pattern 3 is D1, and the dose given to the detection pattern 4 is D
When k is 2, k is defined by Equation 5.

[0017]

[Equation 5] k = D2 / D1 (Equation 5) When electron beam writing is performed by changing the irradiation amount,
Although the irradiation dose of the entire pattern group was changed, the irradiation doses of D1 and D2 were changed while keeping k constant. At this time, the equation 2 is transformed into the equation 6.

[0018]

[Equation 6] E = {(1 + k) + A · η} · D1 (Equation 6) When the central angle θ = 0 °, that is, when the drawing area density A = 0, the minimum irradiation amount that the detection pattern resolves is D0. Then the number 7
Becomes

[0019]

[Equation 7] E = (1 + k) · D1 = D0 (Equation 7) Next, if the function f (A) of the drawing area density A is defined as Equation 8 as in the following equation, Equation 6 and Equation 7 9 and f (A)
Is proportional to the drawing area density A, and the reflection coefficient η is obtained from the slope of the straight line.

[0020]

F (A) = D0 / D1- (1 + k) (Equation 8)

[0021]

F (A) = A · η (Equation 9) Negative electron beam resist SAL601 on a silicon substrate
1% increments from dose 2μC / cm ² the pattern shown (the ship lay Far East trade name) to 1.0μm coated view 3 by electron beam at an acceleration voltage 50 kV 20 [mu] C / c
It was drawn by changing to m ² . Also, the value of k in Equation 5 is 0.
It was 2. After developing with MF312 (trade name of Shipleigh Far East Co., Ltd.) diluted to 70% for 2 minutes, the pattern was observed with a metallurgical microscope. The minimum irradiation dose at which the detection pattern was just resolved was selected for each drawing area density.

FIG. 4 shows this irradiation dose as f (A)
Is plotted and plotted as a function of A. When the measurement points were linearly approximated by the method of least squares, the result was Equation 10, and the reflection coefficient η was obtained from Equation 11 from the slope of the straight line.

[0023]

F (A) = 0.0945 + 0.817A (Equation 10)

[0024]

[Equation 11] η = 0.817 (Equation 11) (Embodiment 2) FIG. 5 shows an embodiment of a different reflection coefficient measurement drawing pattern. In the present embodiment, the entire shape is square instead of the fan shape in FIG. When a variable-shaped electron beam drawing apparatus is used, it takes a drawing time to draw a circular or fan-shaped pattern, so a rectangular or triangular pattern is preferable. If one side of the square is selected to be at least twice the range of the backscattering, only the reflection coefficient can be obtained regardless of the value of the backscattering coefficient as in the case of the fan shape. In this embodiment, one side of the square is 50 μm, and the detection pattern 4 is a rectangle of 0.5 μm square.

The results measured by the same process as in Example 1 are shown in FIG. However, in this example, the value of k of the equation 5 was evaluated for four ways of 0.5, 1.0, 1.5, and 2.0. The value of the reflection coefficient η is 0.84 for each value of k.
The values were in good agreement with 2, 0.847, 0.847 and 0.845.

(Embodiment 3) FIG. 7 shows a further different drawing pattern for evaluating the reflection coefficient. FIG. 7 shows a design in which the pattern for determining the drawing area density is a fine rectangle and the drawing area density is changed by changing the distance between the rectangles. The overall shape is 60 μm, which is twice W1 (= 30 μm)
Is a rectangular pattern. One side W3 of the area density determination rectangle 5 is 0.1 μm, and one side W3 of the detection rectangular pattern 6 is 1.
μm. By changing the distance S1 between the rectangles 5, the drawing area density A can be changed to Eq.

[0027]

[Equation 12] A = W3 ² / {2 · (W3 + S1) ² } (Equation 12) Instead of the rectangle 5, a dot pattern can be used in a point beam type electron beam drawing apparatus.

(Embodiment 4) A thickness of 0.25 μm is formed on a silicon substrate.
The effectiveness of proximity effect correction was confirmed using the proximity effect parameters determined by the present invention using a substrate on which m tungsten was vapor-deposited. The rectangular drawing pattern shown in Example 2 was used for the measurement of the reflection coefficient, and the reflection coefficient η = 1.08.
Got the value of. Next, drawing was performed using a variable shaping electron beam drawing device HL800D (Hitachi Ltd. product name). The electron beam irradiation amount D for each drawing pattern was corrected according to the drawing area density A within 5 μm square around the drawing pattern, which was given by the equation 13.

[0029]

[Equation 13] D = Ds / (1 + 2.A. [Eta]) (Equation 13) Here, Ds is a standard irradiation amount when the drawing area density is 50%. The pattern dimension measurement results after drawing are shown in FIGS. 8A and 8B together with the case where no correction is made. With the correction, the dimension shift from the design value is constant regardless of the design dimension, which shows that the parameter evaluation pattern and the evaluation method used in the present invention are correct.

[0030]

According to the drawing pattern of the present invention, since the internal reflection coefficient η of the proximity effect parameter can be accurately determined independently of other parameters, the accuracy of the proximity effect can be improved. Further, as shown in Examples 2 and 3, since the parameters can be evaluated only by observing with a metallographic microscope, inexpensive and quick measurement can be performed.

[Brief description of drawings]

FIG. 1 is an explanatory diagram of a drawing pattern for evaluation of proximity effect parameters according to the present invention.

FIG. 2 is an explanatory diagram of a drawing pattern for evaluation of proximity effect parameters shown in the first embodiment of the present invention.

FIG. 3 is an explanatory diagram of a drawing pattern for evaluation of proximity effect parameters shown in the first embodiment of the present invention.

FIG. 4 is a characteristic diagram showing a relationship between a drawing area density A and a function f (A) when evaluated using the drawing pattern for evaluation of proximity effect parameters shown in the first embodiment of the present invention.

FIG. 5 is an explanatory diagram of a drawing pattern for evaluation of proximity effect parameters shown in the second embodiment of the present invention.

FIG. 6 is a characteristic diagram showing the relationship between the drawing area density A and the function f (A) when evaluated using the drawing pattern for evaluation of proximity effect parameters shown in the second embodiment of the present invention.

FIG. 7 is an explanatory view of a drawing pattern for evaluation of proximity effect parameters shown in the third embodiment of the present invention.

FIG. 8 is a characteristic diagram showing pattern dimension measurement results with and without the proximity effect correction shown in Example 4 of the present invention.

[Explanation of symbols]

1 ... Area pattern for determining drawing area density, 2 ... Detection pattern for confirming that the deposition energy has reached a certain level, L ... Length of line portion of line / space pattern, S ... Line / space pattern The length of the space portion of W, the dimension of the evaluation pattern.

Claims (10)

[Claims] 1. A drawing pattern for evaluating a proximity effect in electron beam drawing, wherein the drawing area density is a ratio of the drawing pattern to the entire area of the constant area, the drawing pattern is the constant area. A drawing pattern for mainly determining the drawing area density in the inside and a size that does not affect the drawing area density and is arranged at about the center of the drawing pattern and the deposition energy by the electron beam at the arrangement position becomes a constant value. An electron beam drawing pattern, comprising: a detection drawing pattern for detecting the arrival. 2. The electron beam drawing pattern according to claim 1, wherein the drawing pattern is composed of a plurality of the drawing patterns having different drawing area densities and a plurality of the detection drawing patterns having the same shape. 3. The electron beam drawing pattern according to claim 1, wherein the size of the drawing pattern having each drawing area density is larger than the range of influence of backscattering of the electron beam. 4. The electron according to claim 1, wherein the drawing pattern for determining the drawing area density in a certain region is fan-shaped, and the drawing patterns giving different drawing area densities are fan-shaped having different central angles. Line drawing pattern. 5. The electron beam drawing pattern according to claim 3, wherein the drawing pattern for determining the drawing area density in a certain area is a shape obtained by cutting out a central area of a fan-shaped pattern with a rectangle. 6. The drawing pattern for determining the drawing area density in a certain area according to claim 1, wherein the drawing patterns are linear patterns having a constant width and a certain interval, and different drawing area densities are obtained. An electron beam drawing pattern that uses different line widths or line intervals. 7. The drawing pattern for determining the drawing area density in a certain area according to claim 1, wherein the drawing patterns are rectangular or circular pattern groups having a constant area and a certain interval, and different drawing area densities are obtained. Is an electron beam drawing pattern that uses rectangular patterns or circular patterns with different areas or uses different pattern intervals. 8. A drawing pattern for mainly determining the drawing area density in a fixed area and a size that does not affect the drawing area density, where the ratio of the drawing pattern to the entire area in the fixed area is the drawing area density. When an electron beam drawing pattern is formed from the detection drawing pattern that is provided, a step of applying an electron beam resist on the substrate, a plurality of electron beam drawing patterns having different drawing area densities Step of drawing by electron beam irradiation amount, step of developing electron beam resist, step of setting one judgment criterion for resist shape or resist film thickness of the drawing pattern for detection, for the drawing pattern of one drawing area density The process of selecting the electron beam dose that gives the resist pattern shape or resist film thickness that matches the judgment criteria, and the selected electron beam dose Proximity effect evaluation method characterized by determining the parameters of the more proximity effect and areal density. 9. The approach according to claim 8, wherein the electron beam irradiation amount of the entire drawing pattern is changed while keeping the ratio of the electron beam irradiation amount given to the drawing pattern and the electron beam irradiation amount given to the detection drawing pattern constant. A proximity effect evaluation method for obtaining effect parameters. 10. The proximity effect is corrected by changing the electron beam irradiation dose depending on the drawing area density near the drawing position using the proximity effect parameter obtained by the electron beam drawing pattern. An electron beam drawing method for performing electron beam drawing.