JPH11111584A - Charged particle beam exposure method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the amount of exposure from being set to an extremely large value by setting the amount of exposure being set for a block mask to that corresponding to the minimum line width candidate when a block exposure pattern data with a line width that is narrower than the preset minimum line width candidate exists. SOLUTION: When a transmission mask 34 with block masks BM1-BM9 is used, the minimum line width candidate in a plurality of block masks BM1-BM9 is set and at the same time the order of priority is provided. Then, pattern data in the transmission mask 34 are checked, and the amount of exposure corresponding to the set minimum line width candidate is set to the amount of exposure of the transmission mask 34 when the pattern data that are narrower than a set line width exist. The set line width candidate is adopted in the order of higher priority, thus preventing the amount of exposure to a block exposure pattern from being set to an extremely large value and setting it to a value being suitable for a design rule.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a charged particle beam exposure method, and more particularly, to a method for creating exposure data for exposing a pattern on a semiconductor wafer by a charged particle beam such as an electron beam.

[0002]

2. Description of the Related Art Exposure to charged particle beams such as electron beams
A submicron pattern can be exposed, and is used in a manufacturing process of an LSI having a high degree of integration. In particular, recently, not only is it used for the manufacture of masks, but also exposure is performed by directly irradiating a resist formed on a semiconductor wafer with a charged particle beam.

In an LSI design process, pattern data having a plurality of layer structures is created in order to form a desired integrated circuit. The resist on the semiconductor wafer or the resist on the mask substrate is exposed according to the pattern data. Exposure is performed by irradiating the resist film with a charged particle beam and causing a chemical reaction in the resist by the energy of the beam.

In this case, a point to be considered is a proximity exposure effect caused by forward scattering and back scattering of the charged particle beam when the resist is irradiated on the resist. The proximity exposure effect is a phenomenon in which when a charged particle beam is irradiated on a certain area, the energy of the beam spreads to an adjacent area due to scattering of the beam. For example, in a region where the exposure pattern density is high, the pattern after development expands due to the influence of the energy of the charged particle beam applied to the adjacent exposure pattern region. Alternatively, in a region where the exposure pattern density is low, there is no influence of energy from an adjacent region, and the pattern after development is reduced or thinned.

Therefore, in consideration of the proximity exposure effect,
It is necessary to correct the designed exposure data. The present applicant has disclosed a method of correcting such exposure data in a patent application filed on January 29, 1996 (Japanese Patent Application No.
3354 (JP-A-8-321462) has been proposed.

In summary, the method proposed in this patent application first sets an exposure amount in consideration of the proximity exposure effect according to the shape of a pattern, generates a plurality of mask areas in a subfield, and The pattern density in the area
By correcting according to the influence of the pattern density of the surrounding area, a substantial pattern density in consideration of the proximity exposure effect is obtained, and the exposure amount (exposure intensity) of the pattern belonging to the area is reduced according to the substantial pattern density. Further, an auxiliary exposure pattern is generated.

[0007]

However, in the case of the charged particle beam exposure, the throughput is deteriorated in the variable rectangular beam exposure system with an increase in the number of patterns.
A region where the same pattern is repeatedly formed is a region wider than the variable rectangle, and a block mask exposure having a plurality of patterns is used. Since exposure can be performed by a single beam irradiation on a relatively large area having a plurality of patterns, throughput can be improved.

When using such a block mask,
There is a case where a block mask has a plurality of patterns, and each pattern has a different line width. However, the exposure of the block mask needs to be performed with the same exposure amount for all the included patterns. Then, according to the above-described method of generating exposure data, the exposure amount is set according to the pattern shape. In particular, when the line width is large, the exposure amount is reduced, and when the line width is small, the exposure amount is increased. This is because, when the line width is large, the actual exposure amount increases due to the proximity exposure effect from the surroundings, and the line width becomes large. When the line width is small, the line width becomes small. Therefore, in the case of a block mask, for example, a minimum line width is detected, and an exposure amount corresponding to the line width is given.

In this case, when a pattern having a small line width exists partially in the block mask, the exposure amount is set to an excessively large value in accordance with the pattern, and the pattern width after development is reduced. May be too thick.

Another problem in using a block mask is that the pattern density in the block mask is high, so that repeated exposure with the block mask causes a so-called Coulomb interaction phenomenon.
The developed pattern tends to be thicker. In particular, the occurrence becomes remarkable in a peripheral portion of a region repeatedly exposed by the block mask.

SUMMARY OF THE INVENTION An object of the present invention is to provide a charged particle beam exposure method capable of giving an optimum exposure amount in a charged particle beam exposure using a block mask.

Another object of the present invention is to provide a charged particle beam exposure method that eliminates the influence of Coulomb interaction and eliminates pattern variations due to the proximity exposure effect in charged particle beam exposure using a block mask. To provide.

[0013]

In order to achieve the above object, the present invention provides a method for setting an exposure amount for a block mask to an exposure amount corresponding to a minimum line width of block exposure pattern data. If there is block exposure pattern data having a line width smaller than the preset minimum line width candidate, the exposure amount is set to the exposure amount corresponding to the minimum line width candidate. as a result,
Even when block exposure pattern data smaller than the design rule exists, setting of an excessively large exposure amount is prevented.

That is, according to the present invention, in a charged particle beam exposure method for irradiating a material to be exposed with a charged particle beam passed through a block mask having a plurality of patterns, a minimum line width candidate for pattern data in the block mask is determined. The step of specifying, and when the minimum line width of the pattern data in the block mask is smaller than the minimum line width candidate, an exposure amount corresponding to the minimum line width candidate is set as an exposure amount of the block mask, and the block Setting the exposure amount corresponding to the minimum line width of the pattern data as the exposure amount of the block mask when the minimum line width of the pattern data in the mask is larger than the minimum line width candidate; The object to be exposed is irradiated by passing a charged particle beam having an amount through the block mask.

In the above invention, in the step of designating the minimum line width candidates, a plurality of the minimum line width candidates are designated in a predetermined priority order, and the block is selected from the plurality of designated minimum line width candidates. The exposure amount corresponding to the minimum line width candidate close to the minimum line width of the pattern data in the mask is set as the exposure amount of the block mask.

Further, the present invention provides a charged particle beam exposure method for shaping a charged particle beam in accordance with variable rectangular pattern data and irradiating a material to be exposed, with a step of designating a minimum line width candidate for the variable rectangular pattern data; When the minimum line width of the variable rectangular pattern data is smaller than the minimum line width candidate, the exposure amount corresponding to the minimum line width candidate is set as the exposure amount of the variable rectangular pattern, and the minimum line width of the variable rectangular pattern data is set. When the width is larger than the minimum line width candidate, a step of setting the exposure amount corresponding to the minimum line width of the variable rectangular pattern as the exposure amount of the variable rectangular pattern, a charged particle beam having the set exposure amount Is shaped into the variable rectangular beam, and the object to be exposed is irradiated.

Also in the case of variable rectangular pattern data, by specifying the minimum line width candidates, it is possible to prevent the exposure amount from becoming excessively high.

Further, the present invention provides a charged particle beam exposure method for repeatedly irradiating a material to be exposed in a matrix with a charged particle beam having passed through a block mask having a plurality of patterns. Exposure of the block exposure pattern to an area located around the matrix, the step of performing a variable rectangular beam,
Exposing a region located at the center of the matrix in the matrix-shaped exposure region to a block exposure pattern using a charged particle beam that has passed through the block mask.

According to the above invention, when exposure is performed using a block mask repeatedly in the form of a matrix, a pattern included in each area around the matrix is exposed using a variable rectangular beam instead of exposure using a block mask. This prevents the Coulomb interaction phenomenon and prevents the peripheral line pattern from becoming thick.

Further, according to the present invention, there is provided a method for obtaining exposure data having exposure pattern data for each subfield from pattern data having pattern data for each of a plurality of subfields in a main field, and exposing a material in accordance with the exposure data. In the particle beam exposure method, a step of generating a plurality of areas in the subfield, a step of obtaining a pattern density in the area, and a review of the pattern density in accordance with the pattern density of the area surrounding the area and the distance between the areas Generating the auxiliary exposure pattern in the area when the reviewed pattern density of the area is lower than a predetermined exposure reference density, and according to the exposure data obtained by adding the auxiliary exposure pattern data to the pattern data. Exposing the document, and exposing the exposure reference density. , To the area having the pattern of the block mask having a lattice-like transmission hole, and wherein the lower than in the other areas.

According to the above invention, when using a block mask in which a beam is inserted in a block exposure pattern in order to prevent the Coulomb interaction phenomenon, an exposure reference ratio which is a reference for generating auxiliary exposure pattern data is set low. In addition, generation of unnecessary auxiliary exposure pattern data can be prevented.

[0022]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, such embodiments do not limit the technical scope of the present invention.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, such embodiments do not limit the technical scope of the present invention. Further, the present invention is applied to a charged particle beam exposure method and an apparatus therefor, and the following embodiments will be described by taking an electron beam exposure method and an apparatus therefor as an example.

FIG. 1 is a schematic configuration diagram of an electron beam exposure apparatus according to an embodiment of the present invention. In this example, the electron beam exposure apparatus uses design data Di having pattern data.
n, and exposure data Dou considering the proximity exposure effect
An exposure data generating device 100 for outputting t, an electron beam control device 200 supplied with the exposure data Dout to control the exposure device, and a lens barrel 300 are provided. Lens barrel 300
Inside, an electron gun 31, a rectangular transmission mask 32, a transmission mask 34 for exposure such as a block mask, and a mask deflector 3
X, Y stage 3 on which a lens 35, a focal lens 36, an electromagnetic deflector 37, an electrostatic deflector 38, and a wafer 40 are placed
9 are provided. The rectangular beam formed by the transmission mask 32 passes through a predetermined mask on the transmission mask selected by the mask deflectors 33 and 35, and is irradiated to a desired position on the wafer 40 by the deflectors 37 and 38.

Control signal S to mask deflectors 33 and 35
1. A control signal S for moving the transmission mask 34 in the horizontal direction
2, control signal S3 to the focal lens, electromagnetic deflector 3
7 and the control signal S to the electrostatic deflector 38.
5. The stage control signal S6 is transmitted to the electron beam controller 2
00 generated.

FIG. 2 shows the above-described exposure data creating apparatus 100.
FIG. 3 is a diagram showing an internal configuration of the device. A design data file 102 storing design data Din, an area generating unit 103 for generating an area in a subfield area, a pattern density generating unit 104 for calculating a pattern density in the area,
A pattern density correction unit 105 that corrects the pattern density in the area in consideration of the influence between the areas, a pattern density file 106 that stores the pattern density data in the area, and the exposure amount of the pattern in the area according to the pattern density An exposure amount correcting unit 107, an auxiliary exposure pattern generating unit 108 for generating an auxiliary exposure pattern in an area according to the pattern density, and an exposure data file 1 for storing exposure data.
09, a block mask pattern generation unit 111, and a block mask pattern file 112. These are connected to the arithmetic unit 111 via the bus 110.

[Outline of Exposure Data Generation Method] An outline of a method of generating exposure data from design data will be described below. FIG. 3 is a diagram showing the relationship between the main field MF and the subfield SF in the semiconductor chip 10. Usually, a plurality of semiconductor chips 10 are formed on a semiconductor wafer. FIG. 3 shows a relationship between a main field and a subfield in the semiconductor chip 10. As shown in the exposure apparatus of FIG. 1, the electron beam deflector comprises an electromagnetic deflector 37 having a slow response speed but a large deflection range, and an electrostatic deflector 38 having a fast response speed but a narrow deflection range. The main field MF refers to a region that can be deflected by the electromagnetic deflector 37, and the subfield SF refers to a region that can be deflected by the electrostatic deflector 38.

By driving the X and Y stages 39 of the exposure apparatus, the wafer is moved to the center of a desired main field MF, and the electron beam is deflected by the electromagnetic deflector 37 in the main field MF to further form the desired shape. The deflected electron beam is deflected by the electrostatic deflector 38 and is irradiated to a desired position. In the example of FIG. 3, one main field 1
2 includes 5 rows and 5 columns of subfields SF00 to SF44.
Divided into

In the example of FIG. 3, in the main field 12, a single arrangement subfield SSF having a different pattern and a matrix arrangement subfield MSF repeatedly arranged with the same pattern are arranged. The matrix-arranged subfields MSF repeatedly arranged are subfields often found in, for example, a memory cell region of a memory device. On the other hand, the independently arranged subfield is a subfield found in a peripheral circuit, a logic circuit, or the like. As described above, the chip 10 is divided into a plurality of main fields and a plurality of subfields in each of the main fields, and the design data Din has pattern data existing in each of the subfields.

The subfields do not necessarily have to be laid without overlapping or without an interval as shown in FIG. 3, but may partially overlap or have an interval.

FIG. 4 is a diagram showing an example of the structure of design data relating to subfields in the main field 12 of FIG. In this example, subfields SF00 to SF44
Are the center coordinates (x,
y), the number of patterns n, and the pattern address ad.
In this example, the subfield data includes subfields SSF00, SSF01,.
.. MSF11, MSF12... SSF44.

On the other hand, the pattern data has, for example, lower left coordinates (x, y) of the pattern, and a width w and a height h of the pattern. The address “ad” in the subfield data area indicates an address in the pattern data, and means that the pattern of the address area of the number “n” of the patterns continuously from the address is the pattern data in each subfield. I do.

Therefore, the data of the single arrangement subfield SSF has different pattern data addresses. On the other hand, the data of the matrix arrangement subfield MSF has the same pattern data address. The data structure of this matrix arrangement subfield is referred to as a hierarchical data structure in this specification. With such a hierarchical data configuration, the data amount of the pattern data can be reduced.

FIG. 5 is a flowchart of an exposure process including the creation of exposure data. FIG. 6 is a diagram showing an example of a certain subfield SF. A method of creating exposure data will be described using the example of the subfield in FIG.

In FIG. 5, first, pattern data of a block mask is generated (S10). This step will be described later. Then, in step S11, an exposure amount Q according to the shape of each pattern is set. The exposure amount Q is set to be strong, for example, when the pattern shape is thin, and to be relatively weak when the pattern shape is thick or large. Then, the set exposure amount Q (not shown) is added to each pattern data. Although there are various methods for setting the exposure amount Q according to the pattern shape, they are omitted in the present embodiment because they are not essential.

Next, the hierarchical data is recognized from the design data consisting of all the subfields on the chip and its pattern data (S12). For example, whether the subfield has the attribute of the single arrangement subfield SSF, has the attribute of the matrix arrangement subfield MSF, has the attribute of the first original matrix arrangement subfield of the matrix arrangement subfield, and the like,
Confirmed from design data. Therefore, the attribute data ad (not shown) is added to the data of the subfield.

The design data only includes a pattern to be formed on the resist layer on the wafer. However, when the resist layer on the wafer is irradiated with an electron beam, a region having a high pattern density receives more beam energy due to the proximity exposure effect. On the other hand, in a region where the pattern density is low, there is no proximity exposure effect, and exposure is performed with less beam energy. Therefore, in consideration of the proximity exposure effect, the exposure amount of the pattern of the design data is reviewed, and if necessary, the auxiliary exposure is performed to positively generate the proximity exposure effect, thereby forming the pattern shape after development. It is necessary to increase the accuracy of In the present embodiment, in order to generate exposure data based on design data, a smaller map area is generated in a subfield,
Based on the pattern density in the area, the exposure amount is reviewed and an auxiliary exposure pattern is generated.

The subfield SF shown in FIG.
Patterns P1, P2, and P3 are included. Then, for each of the patterns P1, P2, and P3, the above-described step S1 is performed.
Exposure amounts Q1, Q corresponding to the respective pattern shapes at 1
2 and Q3 are set. Furthermore, areas a11 to a55 of 5 rows and 5 columns are generated in the subfield SF.
13). This area corresponds to the area generating unit 1 shown in FIG.
03. The area is generated by, for example, arranging regions of a predetermined size in a matrix on the basis of the lower left position of the subfield SF. Therefore, the end of the subfield does not always coincide with the end of the area.

In the example of FIG. 6, the pattern P1 is the area a2
1, a22, a31, a32, a41, and a42. Further, the pattern P2 is composed of the areas a13, a23, a
33, a43 and a53. Further, the pattern P
3 is an area a24, a25, a34, a35, a44,
a45.

In step S14, the pattern density in each area is determined. That is, it is obtained by the pattern density generation unit 104 in FIG. FIG. 7 is an example in which the pattern density Smn of each area of the subfield in the example of FIG. 6 is entered. That is, the ratio of the pattern area to the area area of the areas a32, a34, etc. is as high as 75%, and the pattern density of the area a11, etc., where no pattern exists, is 0%.

Next, in step S15, the pattern density in each area is reviewed according to the influence of the proximity exposure effect between the areas. The review of the pattern density is performed by the pattern density correction unit in FIG. Due to the proximity exposure effect, the energy of the electron beam applied to the pattern in the area located around the area has an effect on the area according to the distance between the areas. The effect is greater from an area at a closer position, and is less affected from an area at a farther position. Therefore, in the present embodiment, a coefficient β (r) (r is a distance between areas) which is substantially inversely proportional to the distance is set in advance, and the pattern density Smn of the surrounding area is multiplied by the coefficient β (r). Then, it is added to the pattern density of the area of interest.

FIG. 8 is a diagram for explaining the pattern density review in response to the influence of the proximity exposure effect between the areas. This example is an example of reviewing the pattern density of the area a33. Area a 33
The influence of the proximity exposure effect from the areas a11 to a55 around 33 is added. For example the area a11, since pattern density is 0% (= S _11), the effect is, S ₁₁ × beta
(R (a ₁₁ −a ₃₃ )) = 0. The r (a ₁₁ -a _33), indicating the distance between the centers between the areas a11 and a33. The area a12 is also 0. And area a
13 has a pattern density of 25% (= S ₁₃ ),
The degree of influence is S ₁₃ × β (r (a ₁₃ −a ₃₃ )).
Similarly, the processing is performed for the areas a14 to a55. As a result, the pattern density of the area a33 is corrected from 50% to, for example, 60%.

The area a33 has an area with a high pattern density at a position close to the area a33. Therefore, the energy of the electron beam applied to the surrounding area greatly affects the area a33. When considering the impact of this proximity exposure effect, substantial pattern density SR ₃₃ area a33 has a value density [Delta] S ₃₃ is added by the influence from the surroundings on their own pattern density S _33.

FIG. 9 is a diagram showing the pattern density in each area after the result of reviewing the pattern density in the above-mentioned area. For example, an area having a high pattern density around the area has a high pattern density of about 10%, and an area where no such area exists around the area.
For example, the pattern density is increased by about 5%.
Note that the influence from the area in the adjacent subfield is added to the area located around the subfield SF such as the area a11 in the same manner as described above.

Step S16 is a step of detecting a pattern existing area in a high-density subfield. This step will be described later in the section of an auxiliary exposure pattern generation step.

As shown in FIG. 9, when the pattern density of each area is reviewed, the exposure amount of each of the patterns P1, P2, and P3 is corrected based on the revised pattern density SRmn (S17). And an auxiliary exposure pattern is generated (S18).

In step S11, patterns P1, P
While the exposure amount Q is set according to the shape of each of P2 and P3, in the correction of the exposure amount in step S17,
The exposure amount is corrected in consideration of the influence of the proximity exposure effect from the surrounding pattern. For that purpose, the substantial pattern density for each area reviewed in step S15 is used. That is, when the substantial pattern density SRmn is higher than the reference value, the exposure of the pattern belonging to the area is reduced because the proximity exposure effect is greatly affected.

On the other hand, if the substantial pattern density SRmn for each area, which has been reviewed in step S15, is lower than the reference value, the effect of the proximity exposure effect is small. An exposure pattern is generated and added to the exposure data. Here, the auxiliary exposure pattern is an exposure pattern for giving an amount of energy corresponding to the proximity exposure effect to a region having a low pattern density, and a uniform exposure amount of, for example, about several% of the exposed energy is uniform. Is an exposure pattern to be provided. The size of the auxiliary exposure pattern is preferably about the size of an area.
However, the position of the auxiliary exposure pattern does not necessarily need to be the same as the area generated in the pattern density review.

FIG. 10 shows the above-described exposure amount correcting step (S1).
FIG. 7 is a diagram showing an example of a correction table used in 7) and an auxiliary exposure pattern generation step (S18). In this example, the pattern density SRmn of the area is divided into 11 levels, and the exposure correction ratio α of each division and the auxiliary exposure pattern are shown. In this example, the reference value is 45.5.
%, And when the pattern density of the area is larger than the reference value, the exposure amount Q of the pattern of the area is multiplied by the ratio α in FIG. That is, the pattern density SRmn is 9
The exposure amount of the pattern in the area exceeding 0.5% is obtained by multiplying the set exposure amount Q by the ratio α = 0.1. Similarly,
Area pattern density SRmn is 81.5 to 90.5%
In the case of, the set exposure amount Q is multiplied by the ratio α = 0.2.

On the other hand, if the pattern density of the area is lower than the reference value, it means that the density in the vicinity of the area is low, and the developed pattern after exposure becomes thin. Therefore, in such an area, an auxiliary exposure pattern is generated as exposure data in order to perform the auxiliary exposure. In the example of FIG. 10, auxiliary exposure patterns for performing auxiliary exposures 1 to 5 are generated for stages 1 to 5 where the pattern density is low. The auxiliary exposure 1 has a larger exposure amount, and the auxiliary exposure 5 has an auxiliary exposure pattern with a smaller exposure amount. The auxiliary exposure pattern is a pattern having the same size as the area.

FIG. 11 is a diagram showing a subfield in a case where the exposure amount of a pattern in an area having a high pattern density is reduced and an auxiliary exposure pattern is generated in an area having a low pattern density. In this example, areas a22, a32, a23, a33, a43, a
The exposure amount Q of the patterns P1, P2 and P3 belonging to 24, a34 is reduced by multiplying by the ratio α. Further, areas a11 to a21, a25, and
Auxiliary exposure patterns (thick lines) are generated at a31, a35 to a42, and a44 to a55.

In the example of FIG. 11, the area generated for obtaining the pattern density is used as it is as the area of the auxiliary exposure pattern. However, the area of the auxiliary exposure pattern may be different from the area for generating the pattern density.

As shown in FIG. 5, an area is similarly generated for each subfield to determine the pattern density.
The pattern density is reviewed under the influence of the surrounding area, and the pattern exposure amount Q is corrected and the auxiliary exposure pattern is generated using the pattern density as an index. As a result, exposure data is created.

FIG. 12 is a diagram showing a configuration example of the created exposure data. First, in the design data of FIG. 4, the pattern data is the position data (x, y), the width w, and the height h.
Whereas, the pattern data of the exposure data is
Further, it has an exposure amount Q. The exposure amount Q is a value that is initially set according to the shape of the pattern (S11) and is corrected using the pattern density of the area. Second, the design data of FIG. 4 does not have attribute data such as a single arrangement subfield, a matrix arrangement subfield, and its original subfield as subfield data. The attribute data at is added in the hierarchical data recognition step (S12). Third, the design data is stored in the subfield SSF00.
To SSF44 and the pattern data of the subfields. However, the exposure data has subfields TSSF1 to TSSFn for auxiliary exposure having an auxiliary exposure pattern as subfields.

In order to maintain the hierarchical structure of the subfield data and the pattern data, the auxiliary exposure pattern is defined as an exposure pattern in the temporary single arrangement subfields TSSF1 to TSSFn which is a kind of the newly added single arrangement subfield. Registered in exposure data. In FIG.
After the single arrangement subfield SSF44, temporary single arrangement subfields TSSF1 to TSSFn are added.
This temporary single arrangement subfield has an auxiliary exposure pattern of the size of the area. Since the auxiliary exposure pattern in each temporary single placement subfield is not repeated,
The properties of the temporary single arrangement subfield are equivalent to those of the single arrangement subfield. Therefore, as the exposure data, the single placement subfield and the temporary single placement subfield are handled in the same manner. Note that a plurality of auxiliary exposure patterns are generated in the temporary single arrangement subfield.

As shown in FIG. 3, the sub-field is not necessarily an area laid all over the main field, but may be an area partially overlapping each other according to the internal exposure pattern. In the exposure step, the position of each subfield is simply irradiated with an electron beam onto the exposure pattern included in that subfield. Therefore, the subfield of the auxiliary exposure pattern may be registered so as to overlap the subfield of the exposure pattern. This makes it possible to add the auxiliary exposure pattern to the exposure data without breaking the hierarchical data structure.

As described above, the exposure amount of the pattern of the temporary single arrangement subfield added to the subfield of the design data is insufficient to expose the resist by itself.

Then, the exposure data Dout created in this way is supplied to the electron beam controller, and the electron beam exposure is performed according to the exposure data (S20).

A feature of the above-described method of creating exposure data is that an area is generated in a subfield in order to create exposure data in consideration of a proximity exposure effect depending on an exposure density and a distance. Then, based on the substantial pattern density SRmn obtained by adding the proximity exposure effect from the surrounding area to the pattern density Smn of the area, the exposure amount is corrected and the auxiliary exposure pattern is generated. Then, the auxiliary exposure pattern is not the pattern data in the existing subfield, but the pattern data of the newly generated temporary single arrangement subfield.

Next, electron beam exposure using a block mask will be briefly described. FIG. 13 is an example of a transmission mask 34 having a block mask. In this example, nine block masks BM1 to BM9 are provided. The block mask is provided with a plurality of patterns in a wide area of, for example, about 5 μm × 5 μm. However, FIG.
The block mask BM9 of the example has one rectangular pattern, and is used for exposure with a normal variable rectangular beam.

As described with reference to FIG. 1, the electron beam that has passed through the rectangular mask 32 is changed by the mask polarizer 33 to a desired block mask BM area of the transmission mask 34, and the electron beam having the pattern shape is converted into a wafer. Irradiated on 40.

FIG. 14 is a diagram of a subfield having a pattern formed using a block mask. This example is formed by irradiating a stripe pattern formed in a subfield six times using the block mask BM1.

In the case of exposure using a block mask as shown in FIG. 14, the above-mentioned area generation, area pattern density generation, and exposure are performed by treating the pattern existence region of the block mask as a pseudo pattern. The correction of the amount and the generation of the auxiliary exposure pattern can be handled in the same way as a normal pattern.

FIG. 15 is a diagram showing an example in which a pattern formed by a block mask is used as a block mask pattern. In this example, the block mask BM having three stripe patterns 51 has 50 regions. However,
No pattern exists on both sides of the region 50. Therefore, as shown on the left side of FIG. 15, the block mask pattern is reduced to a region 52 where the pattern exists. Then, the pattern density D of the block mask pattern is obtained.

The block mask pattern obtained in this way includes the position (x, y) of the normal pattern data,
It has a pattern density D in addition to the width w, height h, and exposure amount Q. Therefore, this block mask pattern is handled in the same manner as a normal pattern.

FIG. 16 is a diagram showing the block mask patterns BMP1 to BMP6 in the subfield SF and the areas a11 to a34 generated therein. In this example, the block mask pattern BM is included in the area a11 and the area a12.
P1 exists. Therefore, the occurrence of the pattern density in the area a11 is performed using the area S2 (= S1 × D) obtained by multiplying the area S1 of the block mask pattern BMP1 in the area a11 by the density D of the block mask pattern BM1. By doing so, even if the pattern data includes a mixture of block masks, the pattern density in the area is generated to correct the exposure amount and generate the auxiliary exposure pattern.
This can be performed in the same manner as a normal pattern.

FIG. 17 shows an example of the structure of exposure data when block masks are mixed. In this example, the matrix arrangement subfield MSF11 and the like have an ordinary variable rectangular pattern and a block mask. Six block masks are used for the pattern data. The block mask pattern data includes, for example, its position (x, y), block mask number, and exposure amount Q.

However, the data of the block mask pattern BMP is further added for each block mask so that it can be adapted to the algorithm for correcting the exposure amount from the pattern density of the area or generating the auxiliary exposure pattern. The data of the block mask pattern BMP is used only for correcting the exposure amount and the like, and is not used as actual exposure data. The data of the block mask pattern BMP includes a position (x, y), a width w and a height h, a density D, and an exposure amount Q.

By generating the data of the above block mask pattern for each block mask and treating the pattern data in the same manner as normal pattern data, the pattern density of each area can be obtained as described above.

FIG. 18 is a diagram showing an example of area data. When an area is generated in the subfield SF, area data as shown in FIG. 18 is generated. The area data includes, for each of the areas a11 to aij in the subfield SF, the position of the area, the position of the lower left and upper right of the internal pattern existence area, the pattern density, the pattern density after review, and whether or not the auxiliary exposure has occurred. It has a flag, a flag indicating whether or not the area is within the matrix arrangement subfield MSF.

As described above, even when the block mask is used in addition to the variable rectangular pattern, an area is generated in the subfield, the pattern density in the area is obtained, and the exposure is performed based on the pattern density of the surrounding area and the distance between the areas. The pattern density corresponding to the degree of influence is added, the pattern density is reviewed, and the exposure amount can be corrected and the auxiliary exposure pattern can be generated from the pattern density.

The block mask BM has a plurality of internal patterns as shown in the example of FIG. Therefore, in the flowchart of FIG. 5, in the step S11 of determining the exposure amount Q according to the pattern shape, there is a problem how to provide an optimal exposure amount to the block mask. In particular, in the method of calculating the exposure amount Q in accordance with the line width of a pattern existing in a block mask, when an extremely fine pattern data exists, an excessively large exposure amount is set in accordance with the fine pattern data. There is. As a result, when the exposure amount is too large and a pattern wider than that but having a normal line width is developed, the line width becomes larger than necessary.

Such a problem is not limited to the block mask, but also exists in a normal variable rectangular pattern. For example, even if a pattern is designed according to the line width rule of a μm, there is pattern data thinner in the pattern. In this case, if an excessive exposure amount is given in accordance with the fine pattern data, a pattern having a line width of a μm is undesirably affected due to a proximity exposure effect or the like.

Therefore, in this embodiment, in the case of a block mask, a plurality of minimum line width candidates in the block mask are set, and priorities are set. The pattern data in the block mask is checked, and if there is pattern data thinner than the set line width, the exposure amount corresponding to the set minimum line width candidate is set as the exposure amount of the block mask. The set minimum line width candidates are adopted in order of priority. Also, in the variable rectangular pattern,
A minimum line width is set, and an exposure amount corresponding to the set minimum line width is set for a pattern thinner than the minimum line width.

FIG. 19 is a diagram showing an example of a minimum line width setting table included in the design data. FIG. 19 (a)
9 is a table of minimum line width candidates for block mask pattern data. In this example, w1 =
0.18 μm, w2 = 0.20 μm, w3 = 0.22 μm
m, w4 = 0.24 μm and w5 = 0.26 μm are set. FIG. 19B is a table of the minimum line width w _min for normal variable rectangular pattern data. In this example, w _min = 0.18 μm is set.

FIG. 20 is a diagram showing an example of a block mask. FIG. 20A shows that a plurality of patterns exist in the block mask BM, and the line width of the pattern data is
There are ones of 0.18 μm, 0.16 μm and 0.14 μm. However, the line widths of 0.16 μm and 0.14 μm are
This is an example in which pattern data is thinner than the slanted line in the pattern data. FIG. 20 (b)
Has four rectangular patterns in the block mask BM, and their widths are 0.40 μm and 0.80 μm.

FIG. 21 is a flowchart for setting the exposure amount using the minimum line width set in the minimum line width table. Step S11 of the flowchart of FIG.
Corresponds to a partly detailed flowchart of the part. First, the pattern data in the block mask is read (S3
0). Block mask pattern generator 1 shown in FIG.
11 reads out the pattern data corresponding to the block mask number. Then, a line width is detected from the pattern data, and a minimum line width W _min is detected (S3).
1).

If the detected minimum line width W _min is smaller than the minimum line width candidate w1 (S32), the exposure amount Q1 corresponding to the minimum line width candidate w1 is set as the exposure amount Q of the block mask. Is performed (S33). When the detected minimum line width W _min is between the minimum line width candidates w1 and w2 (S34), the exposure amount Q corresponding to the minimum line width candidate w2 is set.
2 is set as the exposure amount Q of the block mask (S3
5). Hereinafter, similarly, the exposure amount is set. If the detected minimum line width _Wmin is larger than any of the candidates (S40), an exposure amount corresponding to the detected minimum line width _Wmin is set (S42).

Therefore, when generating the exposure amount of the block mask shown in FIG. 20A, the detected minimum line width W _min =
0.14 μm, the first priority minimum line width w1 =
Since it is thinner than 0.18 μm, the exposure amount Q of the block mask
Is set to the exposure amount Q1 corresponding to w1. In the case of the exposure amount of the block mask shown in FIG. 20B, the detected minimum line width W _min is 0.40 μm, which is larger than any of the minimum line width candidates, and thus corresponds to 0.40 μm. The exposure is set.

The exposure amount of the variable rectangular pattern is set to 0 if the line width of the target pattern data is smaller than the minimum line width candidate w _min = 0.18 μm set as shown in FIG. The exposure amount is set according to 18 μm, and if it is thicker, the exposure amount is set according to the line width of the pattern data.

As described above, in this embodiment, the minimum line width candidate is set in advance, and even if pattern data thinner than the line width exists, the exposure amount corresponding to the minimum line width candidate is set. Avoid setting a larger exposure.
By doing so, it is possible to set the optimum exposure amount for the design rule. Also, even if the data is pattern data that is narrower than the actual pattern width, the exposure value is not set to an excessively large value. Is prevented.

By providing the minimum line width candidate together with the design data, it is possible to set the exposure amount according to the design rule. Moreover, after that, by generating an area and reviewing the exposure amount according to the pattern density of the area,
A more optimal exposure can be generated.

FIG. 22 is a diagram of a subfield having a line and space pattern using a block mask. In this example, the block mask has six line patterns. Therefore, as shown in FIG. 22A, when a block exposure pattern is exposed using a block mask, it is known that the line width of the line pattern becomes large due to Coulomb interaction which is a kind of light interference. Have been. However, when a block exposure pattern is exposed in a matrix using a block mask as shown in FIG. 22, a phenomenon occurs in which the line width of the line pattern becomes large in the central portion 60 of the matrix due to interference between the block exposure patterns. Absent. Therefore, the block exposure patterns in the regions 61 and 62 at both ends of the line pattern tend to have a large line width.

This Coulomb interaction is a phenomenon peculiar to the block mask, and is considered to be caused by the simultaneous presence of the electron beam over a wide area of the block mask. Therefore, in this embodiment, when the line and space are exposed using a block mask, the block masks of the regions 61 and 62 at both ends are changed to a normal variable rectangular pattern. As a result, the Coulomb interaction phenomenon inherent in block exposure can be avoided. In these areas 61 and 62, since the density is high, the exposure amount for each pattern is slightly corrected by the pattern density in the above-mentioned area. Then, the line patterns are exposed one by one. Thereby, the phenomenon that the line width is increased by the block mask can be avoided.

FIG. 23 is a diagram showing an example of a block mask having a large exposure area. In the case of a block mask BM having a large exposure area as shown in FIG.
As shown in (b), a beam is inserted into the exposure area.
That is, a region having a large exposure area is set as a grid-like exposure region, and the above-described phenomenon of Coulomb interaction due to light interference is suppressed, and optimal exposure is performed.

As described above, by inserting a beam, the beam transmission hole area of the block exposure pattern data becomes considerably smaller than the transmission hole area before the beam is inserted. As a result, when an area is generated in the block mask region and the pattern density in the area is obtained, it is expected that the exposure reference ratio will be lower than the exposure reference ratio shown in FIG. As a result, in the above-described exposure data generation method, an area having a low pattern density is determined, and an auxiliary exposure pattern is generated. Moreover,
Since the block exposure pattern by the block mask is repeatedly generated in a matrix, when an auxiliary exposure pattern is generated there, a very large amount of auxiliary exposure pattern data is added.

However, it is not necessary to generate an auxiliary exposure pattern in a high-density block exposure pattern area for preventing Coulomb interaction. Therefore, in the present embodiment, it is expected that a beam will be generated in the block exposure pattern data to reduce the transmission hole area, and the exposure reference ratio which is a reference for generating the auxiliary exposure pattern is set for such a region. Lower. By doing so, the occurrence of the auxiliary exposure pattern can be reduced in the above area.

[0088]

As described above, according to the present invention, even when a block exposure pattern and a variable rectangular pattern are mixed, an area is generated to determine the pattern density, and the exposure amount is corrected or the auxiliary exposure is performed according to the pattern density. A pattern can be generated, and the exposure amount for the block exposure pattern can be set to a value suitable for a design rule.

Further, according to the present invention, even when a pattern smaller than the minimum line width of the design rule exists in the block exposure pattern data, the minimum line width candidate is set, so that the exposure amount becomes excessively large. Setting can be prevented. Also in the case of a variable rectangular pattern, setting the minimum line width according to the design rule prevents an excessively large exposure amount from being set.

Further, according to the present invention, the block exposure pattern in the end region is developed into a variable rectangular pattern and exposed as a variable rectangular pattern, so that when a block mask is used, line-and-line by Coulomb interaction is used. A phenomenon that the accuracy of the space line pattern is reduced can be prevented.

Further, according to the present invention, when coulomb interaction is prevented by inserting a beam into a block exposure pattern, it is possible to prevent an excessive auxiliary exposure pattern from being generated.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of an electron beam exposure apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing an internal configuration of the exposure data creating apparatus 100.

FIG. 3 is a diagram showing a relationship between a main field MF and a subfield SF.

FIG. 4 is a diagram showing a configuration example of design data for a subfield in the main field of FIG. 3;

FIG. 5 is a flowchart of an exposure process including creation of exposure data.

FIG. 6 is a diagram illustrating an example of a certain subfield SF.

7 is an example in which a pattern density Smn for each area of the subfield in the example of FIG. 6 is entered.

FIG. 8 is a diagram for explaining a pattern density review in accordance with the effect of the proximity exposure effect between areas.

FIG. 9 is a diagram showing the pattern density in each area after the result of reviewing the pattern density in the area;

FIG. 10 is a diagram illustrating an example of a correction table used in a correction process of an exposure amount and a generation process of an auxiliary exposure pattern.

FIG. 11 is a diagram showing subfields when the exposure amount of a pattern in an area having a high pattern density is reduced and an auxiliary exposure pattern is generated in an area having a low pattern density.

FIG. 12 is a diagram illustrating a configuration example of created exposure data.

FIG. 13 is a diagram illustrating an example of a transmission mask having a block mask.

FIG. 14 is a diagram of a subfield having a pattern formed using a block mask.

FIG. 15 is a diagram showing an example in which a pattern formed by a block mask is used as a block mask pattern.

FIG. 16 shows block mask patterns BMP1 to BMP6 in a subfield SF and areas a11 to a generated therein.
FIG.

FIG. 17 is a configuration example of exposure data when block masks are mixed.

FIG. 18 is a diagram illustrating an example of area data.

FIG. 19 is a diagram illustrating an example of a minimum line width setting table.

FIG. 20 is a diagram illustrating an example of a block mask.

FIG. 21 is a flowchart for setting an exposure amount.

FIG. 22: Line and AND using a block mask
FIG. 4 is a diagram of a subfield having a pattern of spaces.

FIG. 23 is a diagram showing an example of a block mask having a large exposure area.

[Explanation of symbols]

 MF main field SF subfield MSF matrix arrangement subfield BM block mask

Claims (6)

[Claims] 1. A charged particle beam exposure method for irradiating a material to be exposed with a charged particle beam that has passed through a block mask having a plurality of patterns, a step of designating a minimum line width candidate for pattern data in the block mask. If the minimum line width of the pattern data in the block mask is smaller than the minimum line width candidate, the exposure amount corresponding to the minimum line width candidate is set as the exposure amount of the block mask,
Setting the exposure amount corresponding to the minimum line width of the pattern data as the exposure amount of the block mask, when the minimum line width of the pattern data in the block mask is larger than the minimum line width candidate; Irradiating the material to be exposed by passing a charged particle beam having an exposure amount through the block mask. 2. The method according to claim 1, wherein, in the step of designating the minimum line width candidates, a plurality of the minimum line width candidates are designated in a predetermined priority order. A charged particle beam exposure method, wherein an exposure amount corresponding to a minimum line width candidate close to the minimum line width of the pattern data in the block mask is set as the exposure amount of the block mask. 3. A charged particle beam exposure method for shaping a charged particle beam in accordance with variable rectangular pattern data and irradiating a material to be exposed, a step of designating a minimum line width candidate for the variable rectangular pattern data, If the minimum line width of the data is smaller than the minimum line width candidate, the exposure amount corresponding to the minimum line width candidate is set as the exposure amount of the variable rectangular pattern, and the minimum line width of the variable rectangular pattern data is Setting the exposure amount corresponding to the minimum line width of the variable rectangular pattern as the exposure amount of the variable rectangular pattern if the line width is larger than the line width candidate; And irradiating the material to be exposed with a beam. 4. The method according to claim 1, further comprising the step of: generating a plurality of areas in the subfield; determining a pattern density in the area; Reviewing the pattern density according to the distance of, and, if the reviewed pattern density of the area is higher than a predetermined exposure reference density, correcting the set exposure amount of the block exposure pattern belonging to the area to be low. And a charged particle beam exposure method. 5. A charged particle beam exposure method for repeatedly irradiating a material to be exposed in a matrix with a charged particle beam passed through a block mask having a plurality of patterns, wherein the matrix is exposed around the matrix in the matrix exposure area. The step of performing the exposure of the block exposure pattern on the located area by a variable rectangular beam, and the exposure of the block exposure pattern on the area located at the center of the matrix in the matrix-shaped exposure area is passed through the block mask. A charged particle beam exposure method. 6. A charged particle beam exposure method for obtaining exposure data having exposure pattern data for each subfield from pattern data having pattern data for a plurality of subfields in a main field, and exposing a material according to the exposure data. Generating a plurality of areas in the subfield; determining a pattern density in the area; and reviewing the pattern density according to a pattern density of an area surrounding the area and a distance between the areas; Generating a supplementary exposure pattern in the area when the reviewed pattern density is lower than a predetermined exposure reference density, and exposing the material according to exposure data obtained by adding the supplementary exposure pattern data to the pattern data. Exposure step, wherein the exposure reference density is grid-like. A charged particle beam exposure method for areas having a pattern of the block mask having a transmission hole, and wherein the lower than in the other areas.