JP3294506B2 - Beam diameter estimation method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for estimating a beam diameter in an exposure apparatus using a beam, such as an electron beam lithography apparatus or a focused ion beam exposure apparatus.

[0002]

2. Description of the Related Art The miniaturization of semiconductor devices has not been limited, and semiconductor devices having a gate length of 0.1 .mu.m or less have been manufactured on a trial basis. Electron beam lithography is mainly used as a means for forming such an element. However, even in the case of electron beam lithography, especially when the pattern size to be formed is close to the beam diameter used for writing, the beam diameter varies depending on the beam diameter and the deflection position in the electron beam writing apparatus. It is necessary to consider how much. This is because, depending on the beam diameter, the beam irradiation portion that can provide the required illuminance to the resist varies, causing problems such as a variation in line pattern dimensions and the inability to resolve repetitive patterns.

[0003]

Here, as a method of grasping the beam diameter of an electron beam lithography apparatus, there is a method called a knife edge method (a method also called a sharp edge method). In this method, a metal whose edge is sharpened thinly is put near the focal point of a focused electron beam, and the change in current at that time is measured to estimate the spot diameter of the electron beam (see “Electron / Ion Beam Handbook, Nikkan Kogyo Shimbun” , Published on December 20, 1973, p. 202. This knife edge method is mainly used for maintenance of an electronic drawing apparatus.
However, there is a problem that the beam diameter estimated by the knife edge method is not very accurate.

[0004]

Therefore, there is a need for a new method capable of more accurately estimating the beam diameter of an exposure apparatus that forms a resist pattern using a beam.

[0006]

According to the invention of the beam diameter estimating method of this application, the beam diameter of an exposure apparatus for forming a resist pattern by using a beam is determined by the distribution of the incident exposure amount and the resist pattern to be actually formed. Is estimated from the comparison result. Specifically, it is preferable to estimate the beam diameter by the following procedure.

(A) A plurality of types of isolated line patterns having different line widths are used as experimental patterns.

(B) One of these isolated line patterns is assumed to be a reference pattern, and an exposure amount (reference exposure amount) for resolving the reference pattern according to the design line width is obtained.

(C) An exposure amount at which the exposure amount per line is the same as when the reference pattern is formed at the reference exposure amount,
A resist pattern for each of the remaining isolated line patterns other than the reference pattern is formed, and the line width of each of the formed resist patterns is measured.

(D) On the other hand, assuming that the beam diameters are arbitrary plural types and the line width is a line width corresponding to the plural types of isolated line patterns, a combination of these beam diameters and line widths is assumed. Each several minutes of the incident exposure amount distribution is obtained.

(E) The design line width of the reference isolated line pattern is given from each of the incident exposure amount distributions obtained assuming that the line width is the line width of the reference pattern. The beam intensity is obtained for each beam diameter.

(F) The obtained beam intensities are applied to the remaining incident light amount distributions for each beam diameter, and the line width obtained at the beam intensity is obtained from each of the incident light amount distributions.

(G) A beam diameter assumed for an incident exposure distribution showing a line width most matching the line width actually measured from the resist pattern among the obtained line widths, and the plurality of types of isolated line patterns are calculated. Estimate the beam diameter of the exposed beam.

Alternatively, the beam diameter may be estimated by the following procedure.

(1). A plurality of types of repetitive patterns having different pitches, each pitch having a size of not more than twice the beam diameter estimated by the knife-edge method, Used as an experimental pattern.

(2) A resist pattern based on these repetitive patterns is actually formed, and up to which pitch of the formed resist pattern is resolved is observed by a known method.

(3) On the other hand, assuming that the plurality of types of repetitive patterns are respectively formed with a plurality of types of arbitrary beam diameters, their incident exposure amount distributions are obtained.

(4) Among the obtained incident exposure amount distributions,
Attention is paid to the incident exposure distribution in which the resolution is determined in the above experiment and the pitch of which is minimum and the ratio of the maximum value to the minimum value of the beam intensity satisfies a predetermined value.

(5) Extraction of the focused incident exposure distribution having the largest assumed beam diameter from the focused exposure dose distribution, and setting the largest beam diameter to the beam of the beam exposed to the plurality of types of repetitive patterns. Estimate the diameter.

According to these methods, an unprecedented beam diameter estimating method is realized. In addition, the beam diameter can be estimated more accurately than the knife edge method.

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

Embodiments of the present invention will be described below with reference to the accompanying drawings. In addition,
The drawings used in the description only schematically show the dimensions, shapes, and arrangements of the components so that these inventions can be understood. Numerical conditions such as line width, pitch, exposure amount, film thickness, temperature and time exemplified in the following description are merely examples for facilitating understanding of the present invention. In each of the drawings, the same components are denoted by the same reference numerals or symbols, and overlapping description may be omitted.

1. DESCRIPTION OF THE INVENTION OF THE BEAM DIAMETER ESTIMATION METHOD First, an embodiment of a method for estimating a beam diameter of an exposure apparatus that forms a resist pattern using a beam from a comparison result between an incident exposure amount distribution and a resist pattern actually formed will be described. I do. Here, an example in which the exposure apparatus using a beam is an electron beam lithography apparatus is considered. The incident exposure amount distribution indicates the intensity distribution of the energy of the electron beam incident on the resist. Assuming that the beam spreads substantially in a normal distribution (actually spreads in a normal distribution), this incident exposure amount distribution can be easily obtained as an addition of the normal distribution.

1-1. First Embodiment First, as shown in FIGS. 1A to 1D, the line width is w1.
Consider an example in which a plurality of different isolated line patterns 11a to 11d such as .about.w4 are used as experimental patterns. Here, w1 = 20 nm, w2 = 30n
m, w3 = 40 nm and w4 = 60 nm. Next, one of the isolated line patterns 11a to 11d
Here, here, the line width w4, that is, the isolated line pattern 11d having the largest line width is set as the reference pattern.

Next, the reference pattern is drawn on the wafer to which the electron beam resist has been applied. However, each image is drawn under various exposure conditions. Then, the wafer is developed. As a result, resist patterns having different exposure amounts can be obtained. Next, from these resist patterns, a resist pattern having a design line width w4 (that is, a line width of 60 nm) is found by SEM or the like. Then, the exposure amount when the resist pattern is formed is defined as a reference exposure amount. Specifically, the reference exposure amount is obtained by the following procedure.

A positive type electron beam resist called SAL601 manufactured by Shipley Co., Ltd.
and then heat it to 120 ° C on a hot plate.
Bake for 2 minutes at this temperature. Next, a reference pattern is drawn on this wafer by an electron beam drawing apparatus HL-700F2 (manufactured by Hitachi, Ltd.). However, drawing is performed with various exposure amounts. Next, this wafer is transferred to MF622 (SA).
Developer for L601. (Shipley Co., Ltd.) for 120 seconds to obtain a resist pattern. From each of the formed resist patterns, a design line width w4 (that is, a line width of 60
nm) of the resist pattern. The exposure amount when this resist pattern is formed is defined as a reference exposure amount.
The reference exposure amount thus obtained is 80 μC /
cm ² .

Next, the remaining isolated line pattern other than the reference pattern is drawn with the same exposure amount per line as when the reference pattern is formed with the reference exposure amount. Here, the reference exposure amount when w4 = 60 nm is 8
Since it was 0 μC / cm ² , writing of an isolated line pattern of w1 = 20 nm was performed with an exposure amount of 240 μC / cm ² , and writing of an isolated line pattern of w2 = 30 nm was 160 μm.
The exposure is performed at an exposure of C / cm ² , and the drawing of an isolated line pattern of w3 = 40 nm is performed at an exposure of 120 μC / cm ² . In this case, since the exposure amount per one line becomes the same value at each level, the amount of backscattering can be made the same, so that the influence of backscattering in beam diameter estimation can be eliminated. next,
Each exposed pattern is developed for 120 seconds using MF622 in the same manner as described above, thereby forming each resist pattern.

Next, the dimensions of each of the formed resist patterns are measured by a conventionally known arbitrary measuring method such as an SEM length measuring machine. Here, the isolated line pattern 11a having a design size of 20 nm is finished at about 45 nm, and the isolated line pattern 11b having a design size of 30 nm is finished at about 5 nm.
The isolated line pattern 11c having a design dimension of 40 nm is finished at about 50 nm. These measurement results are plotted in FIG. 2 by black squares. Note that the horizontal axis in FIG. 2 is the design dimension, and the vertical axis is the line width measured above (hereinafter also referred to as an actually measured value) or the line width obtained from the incident light exposure distribution described later.

On the other hand, if the beam diameter is arbitrary plural kinds, here 2
0 nm, 30 nm, and 40 nm, and the line widths W1 to W1 corresponding to the plurality of types of isolated line patterns.
Assuming that the number of incident light exposures is 4, the number of incident light exposure distributions corresponding to the number of combinations of the beam diameter and the line width, that is, 12 kinds of incident light amount distributions are obtained here. Here, the intensity of the beam is considered to be normally distributed, and the distribution width defined by σ is considered to be the beam diameter.

Next, from these incident exposure amount distributions, the beam intensity giving the design line width of the reference pattern is calculated from each of the incident exposure amount distributions obtained by assuming the line width as the line width of the reference pattern. Obtained for each beam diameter. That is, as shown in FIG. 3A, the assumed line width is w4 and the assumed beam diameter is b _n (b _n is 20 nm, 30 nm, and 40 nm).
), The beam intensity _Xn giving the line width W4 is obtained for each beam diameter (that is, 20 nm, 30 nm, and 40 nm).

Next, the obtained beam intensity X _{n is used} to calculate the remaining incident exposure amount distribution for each beam diameter, that is, the line width to w1, w
The line widths W _n obtained at the beam intensity X _n are obtained from these incident light amount distributions by respectively applying the incident light amount distributions assumed as 2, w3. That is, as shown in FIG. 3B, the assumed beam diameter is b _n and the assumed line width is w _n (where w _n is 20 nm, 30 nm, and 40 _n ).
m)), X _n is applied to each of the incident exposure amount distributions obtained as (m), and the line widths W _n obtained at the beam intensity X _n are obtained. The line widths thus obtained are plotted in FIG. In FIG.
Each point of the white circles is a line width obtained from an incident exposure distribution assuming a beam diameter of 20 nm and a line width of 20, 30, and 40 nm, respectively. Is the line width obtained from the incident light dose distribution assuming the beam diameter of 30 nm and the line widths of 20, 30, and 40 nm, respectively. Each point of the open triangle indicates that the beam diameter is 40 nm. These are the line widths respectively obtained from the incident exposure dose distributions assuming the line widths of 20, 30, and 40 nm, respectively.

A beam diameter assumed for the incident light amount distribution showing the line width most matching the line width actually measured from the resist pattern among the obtained line widths is examined. Then, as is clear from FIG.
It can be seen that each line width obtained from the incident exposure amount distribution for m agrees well with the actually measured value. Therefore, in this case, the beam diameter of the beam that has exposed a plurality of types of isolated line patterns is 2
It is estimated to be 0 nm.

1-2. Second Embodiment Next, an example will be considered in which a plurality of types of repetitive patterns each having a pitch equal to or smaller than twice the beam diameter estimated by the knife edge method are used as experimental patterns. In addition, consider an example in which the diameter of the beam used is estimated to be 60 nm by the knife edge method. Therefore, the pitch is selected from dimensions of 60 nm × 2 = 120 nm or less. Therefore, here, as a plurality of types of repetitive patterns, line patterns having a line width of 20 nm are set to 50 nm and 60 nm.
Consider an example in which first to sixth repetitive patterns having pitches of nm, 70 nm, 80 nm, 90 nm, and 100 nm are used.

The first to sixth repetitive patterns are drawn on a resist to actually form resist patterns. Here, a silicon wafer coated with a 100 nm-thick positive electron beam resist manufactured by Zeon Corporation called ZEP520 is prepared. 200 ° C
Bake for 2 minutes at this temperature. Then, 150 μC was applied to this wafer using an electron beam lithography system HL-700F2.
Each of the above-described first to sixth repeating patterns is drawn at an exposure amount of / cm ² . The drawn wafer is developed with xylene for 240 seconds to obtain resist patterns. Each resist pattern thus formed is observed by a known method such as a microscope, and it is checked to which pitch the resist pattern has been resolved. Here, the repetitive patterns having the pitches of 80 nm, 90 nm, and 100 nm were resolved.

On the other hand, assuming that the plurality of types of repetitive patterns are formed with a plurality of types of arbitrary beam diameters, their incident exposure dose distributions are obtained. Here the first
To the sixth repetition pattern, each having a beam diameter of 20
The incident exposure dose distributions are respectively obtained on the assumption that the film is formed to have a thickness of nm, 30 nm, 40 nm, 50 nm, and 60 nm. FIG.
FIG. 9 to FIG. 9 collectively show the incident exposure amount distribution in units of the pitch of the repeating pattern. In each of FIGS. 4 to 9, the characteristic I shown by a solid line is an incident light amount distribution assuming a beam diameter of 60 nm, the characteristic II shown by a fine broken line is an incident light amount distribution assuming a beam diameter of 50 nm, The characteristic III indicated by the dashed line is the incident exposure amount distribution assuming that the beam diameter is 40 nm, and the characteristic IV indicated by the double-dashed line is 30 nm.
The incident exposure amount distribution assumed to be nm, and the characteristic V indicated by a rough broken line is the incident exposure amount distribution assumed to be a beam diameter of 20 nm.

Next, among the obtained incident exposure amount distributions,
Attention is paid to the incident exposure distribution in which the resolution is determined in the above experiment and the pitch of which is minimum and the ratio of the maximum value to the minimum value of the beam intensity satisfies a predetermined value. Here, in the above experiment, the pitch was 80 nm, 90 n
Since the repetition patterns of m and 100 nm were resolved respectively, the incident exposure amount distribution for the repetition pattern with the smallest pitch is the incident exposure amount distribution for the repetition pattern with the pitch of 80 nm. That is, the incident exposure amount distribution shown in FIG. 7 is obtained.
Further, attention is focused on the incident exposure amount distribution in which the ratio between the maximum value and the minimum value of the beam intensity satisfies a predetermined value in the incident exposure amount distribution shown in FIG. Here, satisfying the predetermined value means that
This is a condition that can be determined according to the design. Here, it is assumed that a case where the minimum value / maximum value is, for example, 0.4 or less satisfies a predetermined value. Then, in each incident exposure amount distribution shown in FIG. 7, the incident exposure amount distribution assuming that the beam diameter is 20 nm is minimum / maximum {0.03 / 0.52} 0.058, and the beam diameter is 30 nm. The assumed incident light dose distribution is minimum value / maximum value ≒ 0.14 / 0.36 ≒ 0.39, each of which satisfies the criterion of 0.4 or less, but the other incident light dose distributions do not satisfy the criterion. Therefore, the incident light exposure distribution of interest in the case of this example has a pitch of 80 nm.
Beam pattern of 20 nm, 30 n
The incident light exposure distribution is assumed to be formed by m. Next, a distribution having the largest assumed beam diameter is extracted from the focused incident exposure distribution. Then, the incident exposure amount distribution here assumes that the beam diameter in FIG. 7 is 30 nm. Therefore, this beam diameter of 30 nm is estimated as the beam diameter of the beam that has exposed the plurality of types of repetitive patterns.

In the first embodiment, it was necessary to accurately measure the resist pattern dimensions obtained in the experiment. On the other hand, in the case of the second embodiment, the evaluation of the resist pattern obtained in the experiment is performed based on whether or not the repetitive pattern is resolved. This can be performed more easily than in the first embodiment.

2. Description of Reference Examples of Beam Diameter Distribution Estimating Method and Pattern Forming Method 2-1. First Reference Example Next, a description will be given of a reference example for estimating a beam diameter distribution in an area where writing can be performed while a stage of an electron beam writing apparatus is fixed. In this beam diameter distribution estimation method, first,
A pattern is repeatedly drawn on the entire surface of the resist-coated wafer in a range where drawing can be performed only by deflecting the electron beam while the stage of the electron beam drawing apparatus is fixed. Next, the wafer is developed to obtain a resist pattern. Next, this resist pattern is observed with an optical microscope. The resist portion drawn in a state in which the beam diameter is small (specifically, a state in which the beam diameter is equal to or smaller than a certain diameter) is a portion in which a repetitive pattern is resolved, and thus has a different color from other portions (white under the following conditions). appear. From this, it is possible to estimate where a good area where the beam diameter is small (an area where the beam diameter is equal to or less than a certain diameter). Hereinafter, a specific example will be described in detail.

First, as shown in FIGS. 10A and 10B, the repetition pattern used is a pattern in which the line pattern 21 having the line width w is repeated at a predetermined pitch P. However, it is preferable that the line width w be such that the maximum value of the beam intensity changes greatly when the beam diameter is changed. This is because the line pattern is easily resolved or not resolved with respect to the fluctuation of the beam diameter, so that the beam diameter distribution is easily estimated. Also, it is better not to make the pitch P too wide. This is for the same reason as the line width. Specifically, the line width w of the line pattern 21
Should be at most half the beam diameter estimated by the knife-edge method, and the pitch P of the line pattern 21 should be about twice as large as the beam diameter estimated by the knife-edge method. In this example, the line width w is set to 20 nm, and the pitch P
Is considered as an example of a repetition pattern in which is set to 100 nm.

Next, a positive type electron beam resist manufactured by Zeon Corporation called ZEP520 is applied to the silicon wafer for 75 minutes.
Apply to a thickness of nm. Next, the wafer is baked on a hot plate at a temperature of 200 ° C. for 2 minutes. Next, the repetitive pattern described with reference to FIG. 10 was applied to this wafer by the electron beam lithography system HL-700F2 at an exposure amount
/ Cm ² . However, this writing is performed on a resist area of 2 mm × 2 mm, which is wider than the maximum area that can be deflected by the electron beam writing apparatus. In this case, an example in which the electron beam is deflected by the deflection lens closer to the wafer among the deflection lenses provided in the electron beam drawing apparatus is considered. Therefore, it is assumed that a range in which the electron beam can be drawn only by deflecting the electron beam while the stage of the electron beam drawing apparatus is fixed is about 40 μm × 40 μm in this apparatus.

The drawn wafer is cleaned with xylene for 60 hours.
Develop for 2 seconds to obtain resist pattern. The obtained resist pattern is observed with an optical microscope and a photograph thereof is taken. FIG. 11 is a diagram obtained by copying this optical microscope photograph. In FIG. 11, the area defined by a × a is one area in which the electron beam is deflected by the deflecting lens closer to the wafer without moving the stage of the electron beam lithography apparatus, and a repetitive pattern can be drawn. Here, 40 μm
Hit the corner area. In this region defined by a × a, the portion corresponding to the third quadrant appeared white in this case. Moreover,
It was found that the repetitive pattern was resolved as desired in the portion that appeared white, but was not resolved in other portions. From this, a portion 31 (FIG. 1) that looks white when the resist pattern formed by this method is observed.
1) can be presumed to be a portion irradiated with a narrow beam, that is, a portion irradiated with a beam of a certain diameter or less,
Therefore, the beam diameter distribution can be estimated.

According to this method of estimating the beam diameter distribution, it is possible to estimate a portion irradiated with a beam having a diameter equal to or less than a certain diameter only by observing the formed resist pattern with an optical microscope. That is, it is possible to visualize and understand a portion having a good beam diameter. Therefore, it is possible to easily prepare drawing data so that drawing of a desired pattern is performed at a position where the beam diameter is good. Therefore, it is effective in the development of a single electron device and the like.

2-2. Second Reference Example In the above-described first reference example of the method of estimating the beam diameter distribution, an example in which one type of repetition pattern is used has been described. However, multiple types of repetition patterns with different line widths and / or pitches are used, and each is drawn over the entire area that can be drawn only by deflecting the electron beam while the stage of the electron beam drawing apparatus is fixed. Thus, the beam diameter distribution may be estimated. Also in this case, the portion that looks white in the resist pattern formed for each of the repetitive patterns is a portion irradiated with a beam having a certain diameter or less. However, the size and shape of the white-looking part differs for each repetitive pattern. This is because the limit beam diameter at which each repetitive pattern can be resolved varies depending on the line width and pitch of the repetitive pattern. Therefore, if this property is used, for example,
In a range where the electron beam can be drawn only by deflecting the electron beam while the stage of the electron beam lithography apparatus is fixed, it is possible to finely estimate how the beam diameter is distributed depending on the deflection position. Hereinafter, a specific example will be described.

A positive electron beam resist called ZEP520 manufactured by Zeon Corporation and having a thickness of 75 nm is applied to a silicon wafer. Next, the wafer is baked on a hot plate at a temperature of 200 ° C. for 2 minutes. Next, a first repetitive pattern having a design line width of 20 nm and a pitch of 100 nm was exposed on the wafer at an exposure dose of 130 μC / cm ^2.
And a second repetitive pattern having a design line width of 20 nm and a pitch of 120 nm is exposed at an exposure amount of 1
Under the condition of 60 μC / cm ² and the design line width is 20
A third repeating pattern having a pitch of 140 nm and a thickness of 140 nm is drawn in a 2 mm square range under the condition of an exposure amount of 200 μC / cm ² . The drawn wafer is developed with xylene for 60 seconds to obtain a resist pattern. Here, under the above exposure conditions, the first repetitive pattern can be resolved when the beam diameter is b ₁ or less, and the second repetitive pattern can be resolved when the beam diameter is b ₂ or less. The third repetition pattern can be resolved when the beam diameter is b ₃ or less. However, b ₁ <b ₂ <b ₃ . Moreover, when the beam diameter is estimated using the above-described beam diameter estimation method, b ₁径 30 nm, b ₂ ≒
It is estimated that 40 nm and b ₃ ≒ 50 nm. Next, the obtained resist pattern is observed with an optical microscope, and photographs thereof are taken. FIG. 12A is a schematic view of a main part of an observation photograph of a resist pattern for a first repeated pattern, and FIG. 12B is a schematic view of a main part of an observation photograph of a resist pattern for a second repeated pattern.
FIG. 2 (C) is a main part simulated view of an observation photograph of the resist pattern for the third repetition pattern. In these figures, the area defined by a × a is one area where the electron beam is deflected by the deflection lens closer to the wafer without moving the stage of the electron beam lithography apparatus to repeatedly draw a pattern, that is, Here, it corresponds to a 40 μm square area. In FIGS. 12A to 12C, each of the portions 41a to 41c that appear white is a portion that can be estimated to have been irradiated with a beam having a diameter equal to or smaller than that capable of resolving the repetitive pattern. Specifically, the portion indicated by 41a has a diameter of b ₁ n
41b, a portion that can be estimated to have been irradiated with a beam of m or less
The portion indicated by b is a portion that can be estimated to have been irradiated with a beam having a diameter of b ₂ nm or less, and the portion indicated by 41c is b ₃
It is a portion that can be estimated to have been irradiated with a beam of nm or less. The results of FIGS. 12A to 12C are summarized in FIG.
It looks like 3. In FIG. 13, a region a can be estimated as a region irradiated with a beam having a beam diameter of b ₁ or less, a region b can be estimated as a region irradiated with a beam having a beam diameter larger than b ₁ and b ₂ or less, In region c, the beam diameter is b ₂
It can be estimated that the region is irradiated with a beam larger than b ₃ and smaller, and the region d can be estimated as a region irradiated with a beam having a beam diameter larger than b ₃ . From this, it is understood that the accuracy required for the writing pattern can be secured and the beam deflection area as wide as possible can be determined based on the estimated beam diameter distribution. That is, when drawing a pattern that is not very fine, for example, the third repetitive pattern, it is not necessary to make the deflection area narrower, and processing such as controlling the deflection lens so that the deflection area becomes the area c in FIG. 13 can be performed. become. Therefore, it is possible to form a pattern that satisfies both the drawing accuracy and the drawing speed (throughput).

A resist pattern is formed by drawing one type of repetitive pattern used in the first reference example of this method of estimating the beam diameter distribution with different exposure amounts, and observing them with an optical microscope. Also, the beam diameter distribution can be estimated. That way, the pattern used is 1
Therefore, preparation of drawing data is simplified. However, on the other hand, a pattern may be crushed at a level where the exposure amount is increased, so that accuracy is slightly applied.

Although the embodiment of the beam diameter estimating method has been described above, the present invention is not limited to the above embodiment. For example, in the above-described embodiment, an example was described in which the exposure apparatus using a beam was an electron beam lithography apparatus. However, it is considered that the present invention can be applied to estimation of a beam diameter in a focused ion beam exposure apparatus.

[0052]

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

As is clear from the above description, according to the invention of the beam diameter estimating method of the present application, the beam diameter of an exposure apparatus for forming a resist pattern using a beam is reduced.
It is estimated from the result of comparison between the incident light exposure distribution and the resist pattern actually formed. Therefore, a novel method that can estimate the beam diameter more accurately than when the beam diameter is estimated by the knife edge method is realized.

[0061]

In addition to the effects of the invention of the beam diameter estimating method described above, if the beam diameter distribution estimating method of the above-described reference example is also taken into consideration, the pattern design can be performed by accurately considering the beam diameter. Provision of a lithography technology that can be further miniaturized can be expected.

[0063]

[0064]

[Brief description of the drawings]

FIGS. 1A to 1D are explanatory diagrams of an isolated line pattern.

FIG. 2 is a diagram showing a relationship between a line width (actually measured value) obtained by actually measuring a resist pattern actually formed and a line width obtained from an incident light amount distribution;

FIG. 3 shows a beam intensity X _n and a line width W _n from an incident exposure amount distribution.
FIG. 4 is an explanatory diagram of a process for obtaining a.

FIG. 4 is a diagram (part 1) for explaining beam diameter dependence of an incident exposure amount distribution;

FIG. 5 is a diagram (part 2) for explaining the beam diameter dependence of the incident exposure amount distribution.

FIG. 6 is a diagram (part 3) for explaining the beam diameter dependency of the incident exposure amount distribution;

FIG. 7 is a diagram (part 4) for explaining the beam diameter dependency of the incident exposure amount distribution;

FIG. 8 is a diagram (part 5) for explaining the beam diameter dependency of the incident exposure amount distribution;

FIG. 9 is a diagram (part 6) for explaining the beam diameter dependency of the incident exposure amount distribution;

FIG. 10 is an explanatory diagram of a repetition pattern used when estimating a beam diameter distribution.

FIG. 11 is an explanatory diagram of a first reference example of a beam diameter distribution estimation method.

FIG. 12 is an explanatory diagram (part 1) of a second reference example of the method for estimating the beam diameter distribution.

FIG. 13 is an explanatory view (No. 2) of a second reference example of the method for estimating the beam diameter distribution.

[Explanation of symbols]

11a to 11d: isolated line pattern 21a: line pattern w: line width 31, 41a to 41c: part irradiated with a beam of a certain diameter or less

Continuation of the front page (51) Int.Cl. ⁷ Identification code FIG21K 5/04 H01J 37/04 A H01J 37/04 H01L 21/30 541N (56) References JP-A-6-75063 (JP, A) JP-A-63-237525 (JP, A) JP-A-59-135728 (JP, A) JP-A-3-50820 (JP, A) JP-A-2-214860 (JP, A) JP-A-63-132428 ( JP, A) JP-A-3-46220 (JP, A) JP-A-4-58517 (JP, A) JP-A-7-50238 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , (DB name) H01L 21/027 G03F 7/20

Claims (2)

(57) [Claims] In estimating a beam diameter of an exposure apparatus for forming a resist pattern using a beam, a plurality of types of isolated line patterns having different line widths are used as experimental patterns. Is determined as a reference pattern, and an exposure amount (reference exposure amount) for resolving the reference pattern according to the design line width is obtained. When the reference pattern is formed with the reference exposure amount, At the exposure amount where the exposure amount is the same, the resist patterns of the remaining isolated line patterns other than the reference pattern are respectively formed, and the line widths of the formed resist patterns are respectively measured. Are assumed to be arbitrary plural types, and the line width is assumed to be a line width corresponding to the plural types of isolated line patterns, and a combination of these beam diameters and line widths is assumed. Determine the incident exposure distribution of a few minutes each, of these incident exposure distribution, to the respective incident exposure amount distribution obtained assuming the line width line width of the reference pattern,
The beam intensity giving the design line width of the reference isolated line pattern is obtained for each beam diameter, and the obtained beam intensity is applied to the remaining incident exposure amount distribution for each beam diameter to obtain the beam intensity. The line width is calculated from these incident light dose distributions, and the beam diameter assumed for the incident light amount distribution showing the line width most consistent with the line width actually measured from the resist pattern among the obtained line widths, A beam diameter estimating method, comprising estimating a beam diameter of a beam obtained by exposing the plurality of types of isolated line patterns. 2. A method for estimating a beam diameter of an exposure apparatus for forming a resist pattern using a beam, the method comprising the steps of: determining a beam diameter of a plurality of types of repetitive patterns having different pitches, each pitch being estimated by a knife edge method; A plurality of types of repetitive patterns having dimensions of twice or less are used as experimental patterns. A resist pattern based on these repetitive patterns is actually formed, and a resist pattern up to any pitch of the formed resist patterns is used. Observing whether the resolution is achieved by a known method. On the other hand, assuming that the plurality of types of repetitive patterns are respectively formed with a plurality of types of arbitrary beam diameters, their incident exposure dose distributions are obtained. In the exposure amount distribution, it is determined that the Attention is paid to the incident light amount distribution in which the ratio of the maximum value and the minimum value of the beam intensity satisfies a predetermined value even if the pitch is the smallest, and the assumed beam diameter is the largest among the noted incident light amount distributions. And estimating the largest beam diameter as a beam diameter of a beam exposed to the plurality of types of repetitive patterns.