JPH04100050A - Method for processing exposure data - Google Patents

Abstract

PURPOSE:To enhance the accuracy of generated exposing pattern data and to shorten a processing time by dividing an arranging pattern area on a water to the areas of a specified size, setting the divided area as a 1st area, then dividing the 1st area F to the areas of the specified size and setting the divided area as a 2nd area. CONSTITUTION:A processor is constituted of a data storage means 1, a correction processing means 2 incorporating a data arrangement part 4 and a data correction part 5, and an exposing device 3. The exposed area on the wafer is divided to the areas of the specified size, which is encircled by a thick continuous line and which is called a field F functioning as the 1st area, and the field F is divided to the areas of the specified size, which is encircled by a broken line and which is called a subfield SF functioning as the 2nd area. Arrangement data D is stored in the means 1 every subfield SF, and arrangement information for discriminating the classification of data whether it is a single arrangement part where a datum is singly arranged or a matrix arrangement part where the same data are plurally and continuously arranged, the information on the position of the field, and the information on the number of arrangement, an arranging number and an arranging range, etc., are set in every field SF, and the actual pattern data are allowed to exist in every field SF.

Description

[Detailed description of the invention] [Table of contents] Overview Industrial field of application Conventional technology Problems to be solved by the invention Means for solving the problems (Figures 15 and 16) (Figures 1 and 2) Working examples One embodiment of the present invention (Figures 3 to 14) Effects of the invention [Summary] Regarding the exposure data processing method, all data for one chip is corrected at once, and the boundary portion of the arrangement data is corrected, The purpose of the present invention is to provide an exposure data processing method that improves the accuracy of generated exposure pattern data and shortens processing time. area, and the first area is further divided into areas of a predetermined size to form a second area, which includes data arrangement type information, arrangement information for identifying the data type, the number of data arrangements, and the arrangement. Data is read out for each second area in the data of the first area from a data storage means that stores a data group having positional information including information such as number, arrangement range, etc., and each of the data read out for each second area is An exposure data processing method in which arrangement data is subjected to predetermined processing by a correction processing means to generate exposure pattern data, and the exposure pattern data is provided to an exposure apparatus, the data arrangement section of the correction processing means The arrangement data read out each time and adjacent data adjacent to the arrangement data are registered in a second area where the arrangement data exists, and the arrangement data in the second area is corrected by the data correction unit of the correction means. When doing so, the arrangement data is configured to be corrected based on adjacent data adjacent to the arrangement data to be corrected.

[Industrial application field]

The present invention relates to an exposure data processing method, and more particularly to an exposure data processing method for creating exposure pattern data for printing onto a wafer with an electron beam using an exposure apparatus.

In recent years, for example, in the production of LSI masks, it is common to use an exposure apparatus using an electron beam.

The exposure device prints predetermined exposure pattern data onto the exposure area on the wafer using an electron beam, and the exposure area on the wafer is divided into areas of a predetermined size called field F, as shown in Fig. The field is further divided into regions of a predetermined size called subfields SF. In addition, the LSI design data exists in subfield SF units, and the data type is specified as either an individual placement section that is placed singly, or a matrix placement section that has the same data and is placed multiple times consecutively for each subfield. It has placement information for identification and position information including information such as the number of placements, placement number, placement range, etc. Many exposure data processing methods for correcting design data have been developed.

Correction processing can be broadly divided into ■Dose correction and ■Dimension correction.Dose correction (Dose) is the method to apply the same dose to a pattern of any size when printing exposure pattern data onto a wafer. If the image was printed, it would not be possible to expose the image properly due to forward scattering and backward scattering of electrons.
The design data is corrected according to the pattern size, for example, a large pattern is printed with a lower dose, and a small pattern is printed with a higher dose. When printing onto a wafer, the pattern swells when printed due to the characteristics of the wafer, so the swell of the pattern is taken into consideration in advance and the design data is corrected in size so that the swollen pattern has a predetermined size.

Note that these correction processes such as dose correction and dimension correction are determined in consideration of the surrounding pattern, and are always corrected with reference to information such as the connection partner.

However, as LSIs become larger in capacity and more sophisticated, the calculations required to correct LSI design data also increase significantly.
This increases the calculation processing time, resulting in a longer development period.

Therefore, there is a need to reduce the calculation time required for correction processing of LSI design data.

[Prior Art] As a conventional exposure data processing method of this type, there is a method as shown in FIG.

This method makes use of the fact that the LSI design data in subfield F has placement information and position information, and if it is determined from the placement information that the data to be corrected is a matrix placement section, Recognizes the placement range, sets a matrix development area (shaded area in the figure) in which multiple pieces of identical data are developed within the matrix placement range, and applies dose correction only to the matrix data in one subfield within this area. , and dimension correction, and develops the corrected data in a matrix development area (hereinafter simply referred to as matrix development) to reduce the correction processing time. In other words, in matrix expansion, the correction processing result for one data within a certain matrix arrangement range is the same as the correction processing result for other data within this matrix arrangement range. This is because the correction processing result within the matrix arrangement range can be obtained by performing the correction processing on one piece of data without performing the same correction processing on all data.

However, matrix expansion is effective only for one piece of data, and if there is multiple data to be matrix expanded, only one of the data can be expanded, so as shown in Figure 16, inside the l chip, For example, matrix arrangement parts A, B, and C.

In the case of design data that has multiple pieces of placement data such as D and single placement portion E, the data for the single placement portion E is first corrected and exposed, and then the data for the matrix placement portion A is expanded into a matrix as described above. Similarly,
Matrix arrangement part B.

The C and D data are developed into a matrix and exposed sequentially.

As a result, when exposing the matrix arrangement section, correction processing within the matrix arrangement range can be performed in the correction processing time for only one matrix data.

[Problem to be solved by the invention]

However, in such conventional exposure data processing methods, when there is multiple data to be expanded into a matrix,
Exposure pattern data can only be generated from one piece of data, and the exposure pattern data are overprinted, so the finer the data pattern, the more difficult it becomes to align during exposure. That is, for example, L
As the exposure patterns become finer due to the higher density of SI, the wiring patterns also naturally become thinner, making it more difficult to deal with misalignment of each exposure pattern due to overprinting.

Therefore, as long as the layers are printed, it becomes impossible to expose patterns with a higher density than a certain density.
There was a problem in that the yield deteriorated.

In addition, in the conventional exposure data processing method, correction processing such as dose correction and dimension correction was performed only on the data of the single placement data.
There is a problem that exposure cannot be performed properly near each data boundary in the matrix arrangement section, making it impossible to obtain a highly accurate pattern. This means that correction processes such as dose correction and dimension correction are determined taking into consideration the surrounding patterns, and are always
This is because correction is made by referring to the information of the connection partner, but in the case of data boundaries, the data to be referenced mainly exists outside the subfield SF.

Therefore, the present invention aims to improve the accuracy of the generated exposure pattern data and shorten the processing time by correcting all the data for one chip at once and also correcting the boundary portion of the placement data. The purpose is to provide an exposure data processing method.

[Means to solve the problem]

In order to achieve the above object, the exposure data processing method according to the present invention divides the placement pattern area on the wafer into areas of a predetermined size to form a first area F, as shown in FIG. One area F is further divided into areas of a predetermined size to form a second area SF, which includes data placement type information, placement information for identifying the data type, number of data placements, placement number, placement range, etc. The data is read out for each second area SF in the data of the first area F from the data storage means 1 that stores a data group having position information including position information, and each arrangement read out for each second area SF. The data D is subjected to predetermined processing by the correction processing means 2 to generate exposure pattern data, and the exposure pattern data is transferred to the exposure device 3.
The exposure data processing method provides the arrangement data D read out for each second area SF by the data arrangement unit 4 of the correction processing means 2 and the adjacent data ON adjacent to the arrangement data D to the arrangement data. When registering in the second area SF where D exists and correcting the arrangement data D in the second area SF by the data correction unit 5 of the correction means 2,
The configuration is such that correction processing is performed on the arrangement data D based on the arrangement data D to be corrected and the adjacent adjacent data DN.

In addition, individual arrangement pattern data DS in which a plurality of pieces of arrangement data D read out for each second area SF by the data arrangement unit 4 based on the arrangement information are arranged individually, and matrix arrangement pattern data in which a plurality of pieces of arrangement data D are arranged in a matrix shape. If the classified arrangement data D is matrix arrangement pattern data Kano, the data arrangement unit 4 sets a matrix development area of the matrix arrangement pattern data Kano based on the position information, and Within the matrix development area, each arrangement data D of the second area SF at the area boundary is set to adjacent data ON, and the second area S at the area boundary is set to ON.
F, and the data correction unit 5 extracts the predetermined placement data D from among the placement data D of the second area SF other than the area boundary part in the IJ space development area, and extracts the extracted placement data. It is effective to perform a correction process on the data D to obtain reference pattern data, develop the reference pattern data in an area other than the area boundary within the matrix development area, and then perform a correction process on the arrangement data D.

[Effect]

In the present invention, when correcting the arrangement data in the second area, the arrangement data is corrected based on the arrangement data to be corrected and the adjacent adjacent data. The surrounding pattern is considered as a correction element, and correction processing such as dose correction and dimension correction is performed.

That is, correction processing is performed on the boundary portion of the arrangement data, and all data for one chip is corrected at once.

Therefore, the accuracy of the generated exposure pattern data is increased, and the processing time is reduced.

〔Example〕

Hereinafter, the present invention will be explained based on the drawings.

Figures 3 to 14 are diagrams showing an embodiment of the exposure data processing method according to the present invention. In these Figures 3 to 14, the same numbers as those in Figures 1 and 2 refer to the same parts. show.

FIG. 3 shows a part of the exposure area on the wafer, and the exposure area on the wafer is divided into areas of a predetermined size called field F, which is the first area surrounded by thick lines, and field F is further divided into It is divided into areas of a predetermined size called subfield SF as a second area surrounded by broken lines, and the data storage means 1 has subfield SF.
Arrangement data D (subfield data) is stored for each subfield SF, and each subfield SF has a data type indicating whether it is a single arrangement section arranged singly or a matrix arrangement section arranged in plural consecutively with the same data. Placement information for identification and position information including information such as the subfield position, the number of placements, placement number, placement range, etc., which is the reference for matrix expansion as placement data, are set, and each subfield SF
There is actual pattern data that serves as input data for each pattern.

Note that ■ to ■ indicate field numbers, and correction processing means 2
This shows the reading order of fields, that is, the exposure order. In the example of FIG. 3, fields
A single placement part E, which is the single placement pattern data DS set in each field ■ to ■, a matrix placement part A, which is the matrix placement pattern data set in the field ■, and a field similar to the matrix placement part A. ■
Matrix arrangement portions B-D set respectively in .

By the way, the deviation of subfield SF is shown in Fig. 4 (a
), and the overlapping of subfields SF refers to the state shown in (b) of the same figure.

In addition, in the reading scanning direction (hereinafter referred to as scanning direction) of the arrangement data D by the correction processing means 2, in FIG.
- As shown by the dotted line, there is an An example of using scanning will be explained.

FIG. 6 is an enlarged view of field ■ in FIG. It shows a state in which placement data D for each SF is read.

It should be noted that the subfields of the matrix arrangement section A and the subfields of the individual arrangement section E are in a different state, and ■~■,
i-X respectively indicate the X coordinate number and Y coordinate number on field (2), and the combination of these numbers indicates the subfield number.

The actual arrangement data D may be a single arrangement part as shown in FIG.
It is not strictly distinguished whether the arrangement data D is a matrix arrangement section, and based on the arrangement information in the arrangement data D read out for each subfield The classified arrangement data D is classified into a plurality of single arrangement pattern data DS arranged singly and a plurality of matrix arrangement pattern data DS arranged in a matrix.

Classified placement data D is independent placement pattern data DS
If so, the classified placement data D is registered together with adjacent data DN adjacent to this placement data D in the subfield SF where the placement data D exists. That is,
Taking subfield U-ii as an example, subfield I[-ii adjacent to subfield I-i, U
-it m-i, l-1i, m -ii, I
-Mountain, ■-Mountain, ■-Mountain each arrangement data D in Fig. 6 ξμ
The data in the range indicated by m becomes adjacent data DN, and this adjacent data ON is registered in subfield 1-1i together with arrangement data D of subfield II-ii.

Also, taking subfield ■-city as an example, this subfield SF protrudes from the field, so the procedure for data in this protruding portion changes depending on whether or not the field data adjacent to subfield ■-mountain is read. In this case, the data of this protruding portion is saved once and referenced after a predetermined time.

To explain this in detail, as shown in FIG. 7, if there is a subfield adjacent to the correction processing subfield to be processed, the pattern data included in the area of ξμm shown by the diagonal line is used as the adjacent data DN. Registered in the correction processing subfield.

In addition, if the correction processing subfield exists in contact with a field boundary line, the position of the correction processing subfield is checked to determine which field boundary line it exists in contact with, and is registered as adjacent data DN in a predetermined file. It is taken out and referenced during correction processing.

On the other hand, if the classified arrangement data D is the matrix arrangement pattern data leaf, the data arrangement unit 4 sets the matrix deployment area of the matrix arrangement pattern data leaf based on the position information in the arrangement data D. Each piece of placement data D in a subfield SF at the area boundary in the matrix development area is registered together with adjacent data DN adjacent to this placement data D in the subfield SF where the placement data D exists. That is, taking the matrix arrangement part A in FIG. 6 as an example, the subfields marked with an X in the figure, which are the subfields at the boundary of the area in the matrix development area, are processed and registered in the same way as the independent arrangement pattern data DS. Ru.

In this process, as shown in Figure 8, if the subfields are out of alignment or overlap, pattern data is first registered in the subfield to be processed, the region where the pattern exists is checked, and then the pattern data is extracted from that region. Single placement subfields and matrix placement subfields that fall within the range of ξμm are extracted, and only data within the range of ξμm is registered in the processing subfield. Therefore, this process enables dimension correction even when the subfields are shifted or overlap, not only at the boundary between the single placement section and the matrix placement section, but also between the single placement section and the single placement section.
Dimensional corrections can be made between the matrix arrangement section and the matrix arrangement section.

Note that when a processing subfield and an adjacent subfield completely overlap, they are not treated as an overlap and are exceptionally handled as one subfield data.

As a result, each subfield other than the area boundary in the matrix expansion area (the area surrounded by the thick solid line in the matrix placement area A in FIG. 6) can be processed according to a certain matrix layout, similar to the conventional matrix expansion process. Since the correction processing result for one data within this range is the same as the correction processing result for other data within this matrix placement range, the dose correction and Correction processing such as dimension correction is performed, and the corrected data is expanded into a matrix, thereby shortening the correction processing time.

Further, as shown in FIG. 9(a), the shape of the matrix development range that can be processed by the exposure apparatus 3 is generally a rectangle such as a square or a rectangle, but the actual shape of the matrix development range is a shape other than a rectangle ( (hereinafter referred to as special shape)
Therefore, in this embodiment, when the matrix development range has a special shape, dividing lines are generated in the matrix of the special shape to divide multiple rectangular shapes, as shown in FIGS. 9(b) and (c). The special shape matrix development range is divided into shapes that can be processed by the exposure device 3.

After the matrix arrangement information is created, as shown in FIG. 6, patterns that can actually be registered in each subfield are extracted, and the patterns are registered and classified. After pattern registration and classification for each field and each subfield is completed, the data correction unit 5 performs dose and size correction processing, and the matrix placement unit processes and develops only the reference subfield.

After the dimension correction is completed, the data is corrected according to the format of the exposure device 3, compressed, and exposed by the exposure device 3 as final data.

A specific example of the data processing method of the present invention will be explained based on FIGS. 10 to 14 using the above method.

FIG. 10 is an excerpt of a certain field in the input data of one chip. In this example, there is one single placement part and two matrix placement parts in a certain part of the field. are doing.

Further, it is assumed that the subfields of the single arrangement section and the matrix arrangement section are misaligned and overlapped.

First, as shown in FIG. 11, based on the arrangement data D, a single arrangement part and a matrix arrangement part are automatically classified,
Furthermore, if there is a matrix development shape that cannot be processed by the exposure device 3 in the matrix placement section, the division processing is automatically performed again within the matrix development range to create matrix development information.

During processing of the matrix expansion range, if the scan direction moves to the right in the figure for the column marked ■■■, the subfield marked with an x will be set as the matrix expansion reference subfield, but When proceeding toward the center left, a matrix expansion reference subfield is generated in the O-marked subfield, and after the expansion range is determined, the ×-marked subfield is set as the matrix expansion reference subfield. This is because the exposure apparatus 3 processes the left end position of the matrix placement section in the figure as a reference position for matrix expansion, and thus matches the processing of the exposure apparatus 3.

After the matrix expansion information is created, the data is read from the beginning again and the pattern data classification registration process is performed as shown in FIG. At this time, since the subfields in the single arrangement section and the matrix arrangement section do not necessarily match, pattern data is registered and classified in each subfield, taking into consideration the deviation and overlap of the subfields. During this registration/classification, the pattern data of the subfield SF at the boundary of field F may or may not be registered depending on the reading order of the fields, as described above, so the field boundary subfield data may be stored in a separate file at any time. When the data correction unit 5 processes the corresponding subfield SP during dimension correction, it is read from this file as adjacent field pattern data and referred to during correction processing.

FIG. 13 shows data before dimension correction processing, and with this data, the dimensions of the single arrangement and the matrix arrangement are corrected based on the matrix arrangement information. Further, during actual matrix expansion processing, correction processing is performed only on the matrix expansion reference subfield, and data is expanded in the matrix expansion range. Note that the single arrangement section performs the conventional processing.

The data after the correction process is converted into a format compatible with the exposure device 3, and then compressed.
The exposure device 3 exposes the data as final data.

Therefore, as shown in FIG. 14, all data for one chip is corrected at once and exposed at once by the exposure device.

In this way, in this embodiment, it is possible to perform correction processing on the boundary portion of the arrangement data, and it is possible to perform correction processing such as dose correction and dimension correction by considering the surrounding pattern of the arrangement data as a correction element. Therefore, all the data for one minute can be corrected at one time, and the accuracy of the generated exposure pattern data can be improved and the processing time can be shortened.

In the above embodiment, the scanning direction is described as an X scan, but the scanning direction is not limited to this, and the same processing can be performed as a Y scan.

[Effects of the Invention] According to the present invention, it is possible to perform correction processing on the boundary portion of the arrangement data, and it is possible to perform correction processing such as dose correction and dimension correction by considering the surrounding pattern of the arrangement data as a correction element. can.

Therefore, all the data for one chip can be corrected at once, the accuracy of the generated exposure pattern data can be increased, and the processing time can be shortened.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an apparatus for implementing the exposure data processing method of the present invention, FIG. 2 is a diagram showing a part of the exposure area on a wafer to which the exposure data processing method of the present invention is applied, and FIG. FIG. 14 shows an embodiment of the exposure data processing method according to the present invention, FIG. 3 shows a part of the exposure area on the wafer, and FIGS. 4(a) and (b) show differences in subfields. 5 is a diagram to explain the readout scanning direction, FIG. 6 is an enlarged view of field ■ in FIG. 3, and FIG. 7 is a sub-field in FIG. An enlarged view of the main part of the field. Figure 8 is a diagram for explaining the processing of subfields that protrude from the field. Figures 9 (a), (b), and (c) are for explaining the division process of the shape of the matrix expansion range. Figures 10 to 14 show specific examples of the exposure data processing method according to the present invention, Figure 10 shows the exposure area of the field part, and Figures 11 and 12 show the matrix development procedure. Figure 13 is a diagram showing a part of the exposure area after the processing is completed; Figure 14 is a diagram explaining the exposure procedure; Figure 15 is a diagram showing the conventional exposure data processing method. FIG. 16 is a diagram illustrating a conventional exposure procedure. 1... Data storage means, 2... Correction processing means, 3... Exposure device, 4... Data arrangement section, 5... Data correction section, F...field (first area), SF...
...Subfield (second area), D...placement data, DN...adjacent data, DS...single placement pattern data, D...
・Matrix arrangement pattern data. Figure ・Subfield deviation (a) ・Subfield overlap Figure 10 is a diagram for explaining the processing of the subfields in the concrete example. Figure 10 is an enlarged view of field 2 in Figure 3. 11
Figure No. 12 for explaining the matrix expansion procedure of the concrete example
Figure Conventional exposure data

Claims (1)

[Claims] (1) The arrangement pattern area on the wafer is divided into areas of a predetermined size to form a first area, and the first area is further divided into areas of a predetermined size to form a second area, and the data arrangement type from the data storage means for storing a data group having arrangement information for identifying the data type and position information including information such as the number of data arrangement, arrangement number, arrangement range, etc. Read data for each second region in the data, perform predetermined processing on each arrangement data read for each second region by a correction processing means to generate exposure pattern data, and provide the exposure pattern data to an exposure device. The exposure data processing method includes registering the arrangement data read out for each second region by the data arrangement section of the correction processing means and adjacent data adjacent to the arrangement data in a second region where the arrangement data exists. and when the data correction unit of the correction means corrects the arrangement data in the second area, the arrangement data is corrected based on the arrangement data to be corrected and adjacent adjacent data. An exposure data processing method characterized by: 2) Classifying the arrangement data read out for each second region by the data arrangement unit based on the arrangement information into a plurality of individual arrangement pattern data in which a plurality of individual arrangement pattern data are arranged, and matrix arrangement pattern data in which a plurality of arrangement data are arranged in a matrix. , If the classified arrangement data is matrix arrangement pattern data, the data arrangement unit sets a matrix development area of the matrix arrangement pattern data based on the position information, and within the matrix development area, the area boundary part is Each arrangement data of the second region is registered as adjacent data in the second region of the region boundary, and the data correction unit selects each arrangement data of the second region other than the region boundary in the matrix expansion region. Extracting predetermined placement data, performing correction processing on the extracted placement data to obtain reference pattern data, expanding the reference pattern data to areas other than area boundaries within the matrix development area, and performing correction processing on the placement data. 2. The exposure data processing method according to claim 1, further comprising performing the following steps.