JP5199786B2 - Method for acquiring drawing region divided data amount and drawing data divided region data amount acquisition device - Google Patents

Description

  The present invention relates to a method for acquiring a divided area data amount of drawing data and a device for acquiring a divided area data amount of drawing data. For example, the present invention relates to a method and an apparatus for acquiring a divided area data amount of drawing data used in a drawing apparatus that draws a predetermined pattern on a sample using an electron beam.

  Lithography technology, which is responsible for the progress of miniaturization of semiconductor devices, is an extremely important process for generating a pattern among semiconductor manufacturing processes. In recent years, with the high integration of LSI, circuit line widths required for semiconductor devices have been reduced year by year. In order to form a desired circuit pattern on these semiconductor devices, a highly accurate original pattern (also referred to as a reticle or a mask) is required. Here, the electron beam (electron beam) drawing technique has an essentially excellent resolution, and is used for producing a high-precision original pattern.

FIG. 14 is a conceptual diagram for explaining the operation of a conventional variable shaping type electron beam drawing apparatus.
The variable shaped electron beam (EB) drawing apparatus operates as follows. In the first aperture 410, a rectangular opening for forming the electron beam 330, for example, a rectangular opening 411 is formed. Further, the second aperture 420 is formed with a variable shaping opening 421 for shaping the electron beam 330 that has passed through the opening 411 into a desired rectangular shape. The electron beam 330 irradiated from the charged particle source 430 and passed through the opening 411 is deflected by a deflector. Then, it passes through a part of the variable shaping opening 421 and irradiates the sample 340 mounted on the stage. The stage continuously moves in a predetermined direction (for example, the X direction) during drawing. That is, a rectangular shape that can pass through both the opening 411 and the variable shaping opening 421 is drawn in the drawing region of the sample 340 mounted on the stage that moves continuously. A method of creating an arbitrary shape by passing both the opening 411 and the variable forming opening 421 is referred to as a variable forming method.

In performing such electron beam drawing, first, a layout of a semiconductor integrated circuit is designed. Then, layout data (design data) in which the pattern layout is defined is generated. Then, the layout data is converted, and drawing data suitable for the electron beam drawing apparatus is generated. The drawing data is input to the drawing apparatus, and is generated as shot data for drawing after a plurality of data processing (see, for example, Patent Document 1). Then, drawing processing is performed according to the shot data. These operations in the drawing apparatus proceed in real time. Here, in the drawing apparatus, when a plurality of data processing is performed, the drawing area is divided into a plurality of block areas, and the data processing is performed by distributing the block areas. The drawing data is expanded for each block area, and intermediate data for each block area before the shot data is generated is generated. The intermediate data for each block area is temporarily stored in the buffer memory. With the recent miniaturization of patterns, the amount of data is increasing. In particular, the amount of data in the critical layer is enormous. If such intermediate data with a large amount of data is generated for a certain block area, the memory capacity of the buffer memory is exceeded, causing an arithmetic processing error such as buffer overflow. If an arithmetic processing error occurs, drawing cannot be performed, and the mask being drawn is discarded.
JP 2007-128933 A

  As described above, when such a large amount of intermediate data is generated for a certain block area, it exceeds the memory capacity of the buffer memory, causing an arithmetic processing error such as buffer overflow. In order to avoid this, it is only necessary to reduce the size of the block area and increase the number of block areas. However, if the block area is set with an unnecessarily small size, each data is processed. Overhead time is increased by an amount corresponding to an increase in the block area, and the time required for arithmetic processing increases. When the time required for the arithmetic processing increases, the actual drawing processing must wait for the result of the arithmetic processing, so that the entire drawing time increases and the throughput deteriorates. Therefore, it is desirable to set the block area size in the drawing apparatus so that the block area does not exceed the memory capacity of the buffer memory but does not become unnecessarily small. However, the method has not been established conventionally. Further, in order to obtain the data amount of the intermediate data for each block area in the drawing apparatus, it is not possible to obtain it unless many arithmetic processes are performed on the inputted drawing data. Therefore, although it may be possible in terms of the function of the device, it may be possible to calculate the amount of intermediate data for each block area by performing blank drawing without irradiating the beam before actual drawing and compare it with the capacity of the buffer memory. Instead of having to perform arithmetic processing, it takes a long time. Furthermore, until the optimum block area is found, it is repeated to obtain the amount of intermediate data for each block area by performing empty drawing, and drawing time becomes longer.

  Accordingly, an object of the present invention is to provide a method and an apparatus for easily acquiring the data amount of each divided region in order to overcome such a problem.

The method for acquiring the divided area data amount of the drawing data according to one aspect of the present invention is as follows.
Inputting drawing data for drawing a pattern on a sample using a charged particle beam;
A step of inputting a predetermined divided area size and dividing a drawing area defined in the drawing data into a plurality of divided areas having a predetermined divided area size;
A step of estimating a data amount of a figure arranged in each divided area using an expected figure number after division derived based on a probability that the figure defined in the drawing data is divided;
Outputting the amount of data for each divided area;
It is provided with.

  In such a configuration, the data amount of each divided region is not obtained after the calculation process of a plurality of stages, but the number of figures is estimated by the expected number of figures based on the probability, and the data amount of each divided region is estimated. As a result, the data amount of each divided region can be estimated in a shorter time than when performing a plurality of stages of arithmetic processing performed by the drawing apparatus.

A figure having a size equal to or smaller than the size (2f) of the overlapping portion in a plurality of small regions that divide a cell constituted by at least one figure defined in the drawing data with an overlapping portion is the probability described above for the size. Becomes 0,
When a figure having an intermediate size between the size (2f) of the overlapping portion and the size (S) of the small area is expressed using the figure size (L), the probability described above for the size is (L-2f) / S.
When expressed using a graphic size (L) and an integer (n) of 2 or more, a graphic having a size in the range of (n−1) S ≦ L <nS has a probability of (L−2f) / It is preferable to set to be (nS− (n−1) × 2f).

Then, a figure having a size equal to or smaller than the size (2f) of the overlapping portion has a figure number expected value of 1 for the size,
When a figure having an intermediate size between the size (2f) of the overlapping portion and the size (S) of the small area is expressed using the figure size (L), the figure number expected value for the size is 1+ (L-2f) / S
When the figure size (L) and an integer (n) of 2 or more are used to express the figure, a figure having a size that falls within the range of (n−1) S ≦ L <nS has an expected figure number value of n + (L −2f) / (nS− (n−1) × 2f).

  The above-mentioned data amount is obtained by multiplying the total figure number obtained by calculating the figure number of each figure for each corresponding cell using the figure number expectation value for each figure type by the unit data amount per figure. It is preferable that the obtained value is obtained by accumulatively adding to the corresponding divided area.

An apparatus for acquiring a divided area data amount of drawing data according to an aspect of the present invention includes:
A storage unit for inputting and storing drawing data for drawing a pattern on a sample using a charged particle beam;
A division unit that inputs a predetermined division area size and divides the drawing area defined in the drawing data into a plurality of division areas having a predetermined division area size;
An estimation unit that estimates the data amount of the figure arranged in each divided area using the figure figure expected value after division derived based on the probability that the figure defined in the drawing data is divided;
An output unit for outputting the amount of data for each divided area;
It is provided with.

  According to the present invention, the data amount of each divided area can be easily acquired before drawing. Based on the data amount, the user can determine a divided area size that does not exceed the memory capacity of the buffer memory. Therefore, buffer overflow can be avoided.

  Hereinafter, in the embodiment, a configuration using an electron beam will be described as an example of a charged particle beam. However, the charged particle beam is not limited to an electron beam, and a beam using charged particles such as an ion beam may be used. As an example of the charged particle beam apparatus, a charged particle beam drawing apparatus, particularly, a variable shaping type electron beam drawing apparatus will be described.

Embodiment 1 FIG.
As described above, when performing electron beam drawing, first, the layout of a semiconductor integrated circuit is designed. Then, layout data (design data) in which the pattern layout is defined is generated. Then, the layout data is converted, and drawing data suitable for the drawing apparatus is generated. Then, the drawing data is input to the drawing apparatus. In the drawing apparatus, after a plurality of data processing, it is generated as shot data for drawing. A pattern is drawn on the sample based on the shot data. In the first embodiment, an optimal divided area size (block size) of drawing data is obtained by calculation off-line outside the drawing apparatus before or after drawing data is input to the drawing apparatus.

  FIG. 1 is a flowchart showing main steps of a method for acquiring a divided area data amount of drawing data according to the first embodiment. In FIG. 1, the method for obtaining the divided area data amount of the drawing data in the first embodiment includes an input / storage step (S202), a merge processing step (S204), a block division step (S206), and a cell for each chip. Extraction step (S208), figure number estimation step (S210), graphic data amount estimation step for each cell (S212), determination step (S214), data amount estimation step for each block (S216), data amount A series of steps of a list creation step (S218) and an output step (S224) are performed.

FIG. 2 is a diagram illustrating an example of a hierarchical structure of drawing data according to the first embodiment.
In the drawing data, the drawing area is a layer of the chip 10, a layer of the stripe 20 virtually divided into strips in the y direction, for example, a layer of the block 30 into which the stripe 20 is divided, and at least one figure. A layer of a plurality of internal structural units such as a layer of the cell 32 configured, a layer of a cluster 34 obtained by dividing the cell 32, and a layer of a graphic 36 (pattern) constituting the cell 32 are hierarchized. ing. In general, a plurality of chips are laid out for a drawing region of one sample 101. Therefore, a chip merge process is performed in a merge processing unit described later, and a post-merging hierarchy as shown in FIG. 2 is configured. In this example, the chip area of the stripe 20 is divided into strips, for example, in the y direction. However, this is an example, and the chip area may be divided in the x direction parallel to the drawing surface and perpendicular to the y direction. sell. Alternatively, other directions parallel to the drawing surface may be used.

FIG. 3 is a conceptual diagram showing the configuration of the drawing data divided region size acquisition apparatus according to the first embodiment.
In FIG. 3, the drawing data divided area data amount acquisition device 100 includes a control computer 110, a memory 134, an external interface (I / F) 136, and a magnetic disk device 138. The control computer 110, the memory 134, the external I / F 136, and the magnetic disk device 138 are connected to each other by a bus (not shown). In the control computer 110, an input unit 111, merge processing unit 112, cell extraction unit 114, division unit 116, figure number estimation unit 118, cell data amount estimation unit 120, determination unit 122, block data amount estimation unit 124, list creation The unit 126 and the output unit 132 are arranged. The magnetic disk device 138 and the memory 134 are examples of a storage unit or a storage device. The external I / F 136 is also an example of an input unit or an output unit.

  Here, the input unit 111, the merge processing unit 112, the cell extraction unit 114, the division unit 116, the figure number estimation unit 118, the cell data amount estimation unit 120, the determination unit 122, the block data amount estimation unit 124, the list creation unit 126, The output unit 132 may be configured as each processing function by software. Alternatively, the input unit 111, the merge processing unit 112, the cell extraction unit 114, the division unit 116, the figure number estimation unit 118, the cell data amount estimation unit 120, the determination unit 122, the block data amount estimation unit 124, the list creation unit 126, and Each configuration of the output unit 132 may be configured by hardware using an electrical circuit. Or you may make it implement by the combination of the hardware and software by an electrical circuit. Alternatively, a combination of such hardware and firmware may be used. Further, in the case of implementation by software or in combination with software, information input to a computer that executes processing or information during and after arithmetic processing is stored in the memory 134 each time.

  In FIG. 3, components necessary for explaining the first embodiment are described. It goes without saying that the divided region size acquisition apparatus 100 may normally include other necessary configurations.

  In S (step) 202, as an input / storage process, the input unit 111 inputs drawing data for drawing a pattern on the sample using an electron beam from the outside via the external I / F 136, and the magnetic disk device 138. To remember. A pattern of a plurality of chips is usually drawn on one mask substrate. Therefore, the drawing data is usually composed of a plurality of chip data files.

  In S204, as the merge processing step, the merge processing unit 112 reads the drawing data from the magnetic disk device 138, and performs the merge processing on the chips whose distance between the chips is smaller than a predetermined size among the plurality of chips.

In S206, as a block dividing step, the dividing unit 116 inputs the block size B ₁ (predetermined divided area size) from the outside, and the drawing area defined in the drawing data is changed to a plurality of blocks (divided areas) having a block size. To divide. The block size input (set) from the outside via the external I / F 136 or a keyboard (K / B) or mouse (not shown) is stored in the magnetic disk device 138.

  In S208, as a cell extraction process for each chip, the cell extraction unit 114 extracts, for each chip after merging, cells arranged in each chip from cell data defined in the drawing data for each chip. As shown in FIG. 2, the drawing data is defined with a hierarchical structure, and the arrangement position, size, hierarchical relationship, and the like of each layer are defined in the drawing data. Therefore, cells arranged on each chip can be extracted from the drawing data.

  In S210, as a figure number estimation step, the figure number estimation unit 118 estimates the number of figures arranged in the cell for each extracted cell. In the drawing apparatus, each figure is divided so as to be within the cluster 34 (small region) shown in FIG. For this reason, the number of figures increases from the number of figures defined in the original drawing data by the division. Since the amount of data for defining one figure is constant, the amount of data increases as the number of figures increases. Therefore, the figure number estimation unit 118 inputs information such as the size of the cluster and the margin size, and estimates the number of divided figures using the expected figure number after division derived based on the probability. Information such as the size of the cluster and the margin size may be input (set) from the outside via the external I / F 136 or a keyboard (K / B) or mouse (not shown).

  FIG. 4 is a conceptual diagram showing an example of a cluster area in the first embodiment. In FIG. 4, a plurality of clusters 34 having vertical and horizontal widths (S) are defined so as to overlap each other. In each cluster 34, a margin 42 is added to the outside from the cut line 40 of the graphic indicated by the dotted line, and the area of each cluster 34 is surrounded by a cluster boundary 44 to which the margin width (f) of the margin 42 is added for each side. It becomes an area. Each cluster 34 is defined to overlap each other with an overlapping width (2f) (the size of the overlapping portion) twice that of the adjacent cluster 34 and the margin width (f). The width (S) is set to 12.8 μm or 6.4 μm, for example. The margin width (f) is set at 0.8 μm, for example.

  FIG. 5 is a conceptual diagram showing an example of a figure that can be arranged in the cluster area in the first embodiment. As shown in FIG. 5, in the cluster area, a figure 52 having a size in a certain direction is a size less than or equal to the overlap width (2f), and a figure having an intermediate size between the overlap width (2f) and the cluster size (S). 54 and a graphic 56 larger than the size (S) of the cluster may be arranged. The figure 52 having a size equal to or smaller than the overlap width (2f) is not divided even if it extends over the cut line 40 because there is a possibility that the figure 52 becomes too small to be drawn and becomes a minute figure that cannot be drawn (sliver occurs). When the graphic 52 is at the position shown in FIG. 5, the graphic 52 is defined in the left cluster. Further, the figure 54 having an intermediate size between the overlap width (2f) and the cluster size (S) may be divided or not divided. When the graphic 54 is at the position shown in FIG. 5, the graphic 54 is divided by the cut line 40 and is divided into left and right clusters. Further, a figure 56 that is larger than the size (S) of the cluster is necessarily divided. When the figure 56 is located at the position shown in FIG. 5, the figure 56 is divided by three cut lines 40, the left end is a left cluster (not shown), the right end is a right cluster (not shown), and the two central parts are respectively Will be defined separately in clusters where. Therefore, for each condition, the probability of division for the figure size for which division is estimated is obtained. At this time, the size of the figure for which division is estimated is assumed to be (L). As will be described later, the figure size to be divided is not limited to one direction, but for easy understanding of the contents, the size in one direction will be described here.

(1) In the case of L ≦ 2f The probability that the figure 52 having the size (L) not exceeding the overlap width (2f) is divided is 0 because it is not divided as described above.
(2) When 2f <L <S The probability that the figure 54 having an intermediate size between the overlap width (2f) and the cluster size (S) can be divided can be divided into a maximum of two. -2f) / S. Probability of being divided = 1−Probability of not being divided = 1− (2f // S + (S−L) / S) = (L−2f) / S
(3) (n-1) When S ≦ L <nS (where n is an integer of 2 or more)
Since the figure size (L) is not infinite, the case where it is equal to or larger than the cluster size (S) and (n−1) S ≦ L <nS will be described. In FIG. 5, as an example, a case where it is larger than two clusters and smaller than three clusters (when n = 3) is shown. In such a case, it is surely divided, but this time it becomes a problem how many pieces are divided. Therefore, the probability of being divided into (n + 1) pieces is used. The probability that the figure 56 having a size in the range of (n−1) S ≦ L <nS is divided into (n + 1) pieces is defined by (L−2f) / (nS− (n−1) × 2f). be able to. The probability of being divided into (n + 1) pieces can be derived from 1− (probability of not being divided into (n + 1) pieces).

The figure number expectation value for the figure size for estimating the division, which is derived from the probabilities of the respective ranges described above, is as follows.
(1) In the case of L ≦ 2f Since the probability 52 has a probability (0) and the size (L) of the figure 52 is not divided, the figure remains as it is (one figure remains). The figure count expectation value can be defined as 1.
(2) In the case of 2f <L <S The figure 54 having an intermediate size between the overlap width (2f) and the cluster size (S) has a probability defined by (L-2f) / S, and a maximum of two figures 54 Therefore, the expected number of figures can be defined as 1+ (L−2f) / S. The figure number expectation value can be obtained by (probability of remaining one) × 1 + (probability of being divided into two) × 2.
(3) (n-1) When S ≦ L <nS (where n is an integer of 2 or more)
A figure 56 having a size within the range of (n−1) S ≦ L <nS is defined as (L−2f) / (nS− (n−1) × 2f) with a probability of being divided into (n + 1) pieces. Therefore, the figure number expected value can be defined as n + (L−2f) / (nS− (n−1) × 2f). The figure number expectation value can be obtained by (probability of not being divided into (n + 1) pieces) × n + (probability of being divided into (n + 1) pieces) × (n + 1).

  Here, since the size of a figure is not only in one direction, each size can be divided by the above-mentioned expected number of figures. The expected number of figures that can be divided as one figure can be defined as follows according to the figure type.

FIG. 6 is a conceptual diagram for explaining the expected number of rectangular figures in the first embodiment. 6, rectangular or square, the transverse size (X-direction size) _{L X} and the vertical size (Y-direction size) is defined by _{L Y.} Then, one rectangular or whole figure number expected value N _TOTAL of square, as shown in FIG. 6, figure number expected value of X-direction size N _{D (L X)} a figure number expected value N _D in the Y-direction size It can be defined by the product of (L _Y ).

As described above, the figure number expected value N _D (L _X ) of the size in the X direction can be defined as (1) figure number expected value N _D (L _X ) = 1 in the range of L _X ≦ 2f. it can. (2) In the case of 2f <L _X <S, the figure number expected value N _D (L _X ) = 1 + (L−2f) / S can be defined. (3) In the case of (n−1) S ≦ L _X <nS (where n is an integer of 2 or more), the figure number expected value N _D (L _X ) = n + (L−2f) / (nS− (N-1) × 2f).

As described above, the figure number expected value N _D (L _Y ) of the size in the Y direction can also be defined as (1) L _Y ≦ 2f in the range of the figure number expected value N _D (L _Y ) = 1. it can. (2) In the case of 2f <L _Y <S, the figure number expected value N _D (L _Y ) = 1 + (L−2f) / S can be defined. (3) In the case of (n−1) S ≦ L _Y <nS (where n is an integer of 2 or more), the figure number expected value N _D (L _Y ) = n + (L−2f) / (nS− (N-1) × 2f).

FIG. 7 is a conceptual diagram for explaining the expected figure number of a right isosceles triangle in the first embodiment. In FIG. 7, a right-angled isosceles triangle is defined by a size L of two sides sandwiching a right angle. As shown in FIG. 7, the total figure number expected value N _TOTAL of one right isosceles triangle is equal to the figure number expected value N _D (L) of one size of two sides sandwiching the right angle and the other size. It can be defined by 1/2 of the product of the figure number expected value N _D (L) and the value obtained by adding the correction value 1. Since the number when the right-angled isosceles triangle is not divided is 1, the figure number expectation value N _TOTAL cannot be smaller than 1. Therefore, it can be prevented that the correction value 1 is added to become smaller than 1.

As described above, the figure number expected value N _D (L) of the size L of two sides sandwiching the right angle is defined as (1) L ≦ 2f in the range of the figure number expected value N _D (L) = 1. be able to. (2) In the case of 2f <L <S, it can be defined as figure expected value N _D (L) = 1 + (L−2f) / S. (3) In the case of (n−1) S ≦ L <nS (where n is an integer of 2 or more), the figure number expected value N _D (L) = n + (L−2f) / (nS− (n -1) x 2f).

FIG. 8 is a conceptual diagram for explaining the expected number of trapezoidal figures in the first embodiment. FIG. 8A shows a trapezoid in which the upper and lower sides are parallel, the height is H, the size of the long sides of the two parallel sides is W, and the size of the short sides is WH. FIG. 8B shows a trapezoid in which the left and right sides are parallel, the height is H, the size of the long sides of the two parallel sides is W, and the size of the short sides is WH. Then, the total figure number expected value N _TOTAL of one trapezoid as shown in FIG. 8 is a figure having a figure W expected value N _D (W) and a height H of two parallel long sides of size W. from the product of the number expected value N _{D (H),} the value obtained by subtracting the correction value 1 from the height H of the figure number expected value N _{D (H)} and the height H of the figure number expected value N _{D (H)} The difference can be defined by subtracting 1/2 of the product. Since the number of trapezoids is not divided, the figure number expected value N _TOTAL cannot be smaller than 1. Therefore, it is possible to prevent the correction value 1 from being smaller than 1 by subtracting the correction value 1.

The figure number expectation value N _D (W) of the size W of the two long sides is defined as figure expectation value N _D (W) = 1 in the range of (1) W ≦ 2f as described above. be able to. (2) In the case of 2f <W <S, it can be defined as figure expected value N _D (W) = 1 + (L−2f) / S. (3) In the case of (n−1) S ≦ W <nS (where n is an integer of 2 or more), the figure number expected value N _D (W) = n + (L−2f) / (nS− (n -1) x 2f).

As described above, the figure number expected value N _D (H) at the height H can also be defined as figure figure expected value N _D (H) = 1 in the range of (1) H ≦ 2f. (2) In the case of 2f <H <S, the figure number expectation value N _D (H) = 1 + (L−2f) / S can be defined. (3) In the case of (n−1) S ≦ H <nS (where n is an integer of 2 or more), the figure number expected value N _D (H) = n + (L−2f) / (nS− (n -1) x 2f).

FIG. 9 is a conceptual diagram for explaining the expected number of figures of isosceles trapezoids and isosceles triangles in the first embodiment. FIGS. 9A and 9B show an isosceles triangle in which the size of the remaining side that is not isosceles is H, and the height from the remaining side is H. FIG. 9A shows a case where the vertex is horizontal, and FIG. 9B shows a case where the vertex is upward. FIGS. 9C and 9D show an isosceles trapezoid in which the height is H, the size of the long sides of the two parallel sides is W, and the size of the short sides is W-2H. FIG. 9C shows a case where the left and right sides are parallel. FIG. 9D shows a case where the upper and lower sides are parallel. The total figure expected value N _TOTAL of the entire isosceles trapezoid or isosceles triangle is the figure W expected figure number N _D (W) and the figure H expected figure number of size H as shown in FIG. from the product of the N D _(H), the difference obtained by subtracting the product of the value obtained by subtracting the correction value from 1 to the size H of the figure number expected value N _D graphics number expected value of _(H) and the size H N _{D (H)} Can be defined in Since the number of such isosceles trapezoids or isosceles triangles is not divided, the figure number expectation value N _TOTAL cannot be smaller than 1. Therefore, it is possible to prevent the correction value 1 from being smaller than 1 by subtracting the correction value 1.

As described above, the figure number expected value N _D (W) of the size W can be defined as the figure number expected value N _D (W) = 1 in the range of (1) W ≦ 2f. (2) In the case of 2f <W <S, it can be defined as figure expected value N _D (W) = 1 + (L−2f) / S. (3) In the case of (n−1) S ≦ W <nS (where n is an integer of 2 or more), the figure number expected value N _D (W) = n + (L−2f) / (nS− (n -1) x 2f).

As described above, the figure number expected value N _D (H) of the size H can also be defined as the figure number expected value N _D (H) = 1 in the range of (1) H ≦ 2f. (2) In the case of 2f <H <S, the figure number expectation value N _D (H) = 1 + (L−2f) / S can be defined. (3) In the case of (n−1) S ≦ H <nS (where n is an integer of 2 or more), the figure number expected value N _D (H) = n + (L−2f) / (nS− (n -1) x 2f).

FIG. 10 is a conceptual diagram for explaining the expected number of parallelogram figures in the first embodiment. FIG. 10A shows a parallelogram in which the upper and lower sides are parallel, the height is H, and the size of the two parallel sides is W. FIG. 10B shows a parallelogram in which the left and right sides are parallel, the height is H, and the size of the two parallel sides is W. The total figure expected value N _TOTAL of one parallelogram is the figure figure expected value of the sum of the figure H expected size N _D (H) and the sum of the size W and the size H, as shown in FIG. The product of the size H figure number expected value N _D (H) and the size H figure number expected value N _D (H) minus the correction value 1 is subtracted from the product of the value N _D (W + H). Can be defined by difference. Since the number of such parallelograms that are not divided is 1, the figure number expectation value N _TOTAL cannot be smaller than 1. Therefore, it is possible to prevent the correction value 1 from being smaller than 1 by subtracting the correction value 1.

As described above, the figure number expected value N _D (H) of the size H can also be defined as the figure number expected value N _D (H) = 1 in the range of (1) H ≦ 2f. (2) In the case of 2f <H <S, the figure number expectation value N _D (H) = 1 + (L−2f) / S can be defined. (3) In the case of (n−1) S ≦ H <nS (where n is an integer of 2 or more), the figure number expected value N _D (H) = n + (L−2f) / (nS− (n -1) x 2f).

As described above, the figure figure expected value N _D (W + H) of the sum of the size W and the size H is defined as (1) W + H ≦ 2f in the range of the figure figure expected value N _D (W + H) = 1. Can do. (2) In the case of 2f <W + H <S, the figure number expectation value N _D (W + H) = 1 + (L−2f) / S can be defined. (3) (n−1) In the case of S ≦ W + H <nS (where n is an integer of 2 or more), the figure number expected value N _D (W + H) = n + (L−2f) / (nS− (n -1) x 2f).

As described above, for each extracted cell, the figure number estimation unit 118 calculates the figure number expected value N _TOTAL for each figure arranged in the cell, and estimates the total as the number of divided figures. .

  In step S212, as a graphic data amount estimation process for each cell, the cell data amount estimation unit 120 multiplies the number of divided figures in the corresponding cell obtained in the previous process by the data amount per graphic for each cell. The amount of graphic data for each cell is estimated. For example, when defining one figure, the data amount is 16 bytes.

  In S214, as a determination step, the determination unit 122 determines whether the estimation of the data amount has been completed for all cells. If it has not been completed yet and cells remain, the process returns to S208. Then, S208 to S214 are repeated until the estimation of the data amount for all the cells is completed.

In S216, as a data amount estimation process for each block, the block data amount estimation unit 124 calculates, for each block area divided in the block division process (S206), the data amount of the graphic of the cell arranged in each block area. Cumulative addition is performed to estimate the data amount of the graphic arranged in each block area. Here, for example, when a cell extends over a plurality of block areas, the data amount of the corresponding figure may be added to the block area where the figure in the cell is located. Since the figure number expected value N _TOTAL has already been calculated for each figure, it is easy to distribute the figure data to the corresponding block area even when the cell extends over a plurality of block areas.

As described above, the data amount of the figure arranged in each divided region is estimated using the figure number expected value after division derived based on the probability that the figure defined in the drawing data is divided. At that time, data volume, the unit data per one figure to the total figure number which is the sum of figure number N _TOTAL of the figure obtained by calculation formula for each figure type with figure number expected value N _D to each corresponding cell A value obtained by multiplying the quantity is obtained by cumulative addition to the corresponding block area.

  In S218, as the data amount list creation step, the list creation unit 126 creates a data amount list for each block area.

  FIG. 11 is a diagram illustrating an example of a data amount list according to the first embodiment. As shown in FIG. 11, the list creation unit 126 creates a data amount list 60 in which block coordinates (X, Y) are associated with data amounts (bytes). In the data amount list 60, for each chip after the merge processing, the corresponding data amount is recorded for all the block areas in each chip. The data amount list 60 is stored in the magnetic disk device 138.

  In S224, as an output process, the output unit 132 outputs the data amount d for each block area. Here, it is preferable to output the data amount list 60 described above via the external I / F 136. Or you may display on the monitor etc. which are not illustrated.

  FIG. 12 is a diagram showing an example of the relationship between the estimation error of the number of figures, the data amount, and the memory capacity of the buffer memory in the first embodiment. As shown in FIG. 12, as the amount of data increases, the estimation error of the number of figures sharply decreases. In the estimation method, assuming that figures of a random size are randomly arranged, the larger the number of samples, the more the normal distribution follows the central limit theorem and the width of the distribution becomes narrower. As described above, in obtaining the optimum block size of the drawing data, it is effective to use the expected figure number after division derived based on the probability that the figure defined in the drawing data is divided.

  According to the first embodiment, the user can grasp the data amount d of each block area when divided by a certain block size. Therefore, the user can check the data amount d of each block area and verify whether there is a block exceeding the memory capacity of the buffer memory of the drawing apparatus. As a result, if there is an exceeding block, a new block size with a reduced block size is reset, and the process from block division (S206) to output step (S224) may be executed again. By repeating such processing, the optimum block size can be found outside the drawing apparatus (offline). Then, the found block size may be set in the drawing apparatus.

  FIG. 13 is a flowchart showing a flow of data processing in the drawing apparatus according to the first embodiment. As described above, when performing electron beam drawing, first, the layout of a semiconductor integrated circuit is designed. Then, layout data (design data) in which the pattern layout is defined is generated. Then, the layout data is converted, and drawing data suitable for the drawing apparatus is generated. Then, the drawing data is input to the drawing apparatus. In the drawing apparatus, drawing data is stored in the magnetic disk device 310. Then, as will be described later, after a plurality of data processing, it is generated as shot data for drawing.

  In the data processing step (S102), drawing data of each of the plurality of chips is read from the magnetic disk device 310 and input. Then, the chip merge processing is performed by rearranging in the drawing area. The drawing data after the processing is executed is stored in the magnetic disk device 312.

  As the block division step (S104), the drawing data processed in the upper step is developed, and the chip or each stripe is virtually divided into a plurality of block areas. At this time, as described above, since an optimum block size that does not exceed the buffer capacity has already been obtained, this block size is set in the drawing apparatus. Then, it is virtually divided into a plurality of block areas with the set block size.

  In the cell placement step (S106), the drawing data is further expanded to place the laid out cells in each block.

  In the cluster dividing step (S108), the drawing data is further expanded and each cell is virtually divided into a plurality of clusters.

  In the pattern dividing step (S110), the drawing data is further developed to divide each cluster 34 into a plurality of figures (patterns) laid out in each cluster.

  In the format conversion step (S120), data defined in the drawing data is converted. By converting the format of the drawing data, intermediate data that is a stage before being converted into shot data is generated. The intermediate data is temporarily stored in the buffer memory 314. Here, in the first embodiment, since the block size is specified so as not to exceed the memory capacity of the buffer memory 314, it is possible to prevent an arithmetic processing error such as a buffer overflow. As a result, it is possible to prevent a problem that an arithmetic processing error occurs and drawing cannot be performed.

  As a shot data generation step (S122), intermediate data is read from the buffer memory 314 to generate shot data. The shot data generated as described above is stored in the magnetic disk device 316.

  Then, the drawing apparatus draws a pattern on the sample as follows using the electron beam controlled by the shot data generated as described above.

  As described above, according to the drawing data divided region size acquisition method and apparatus thereof according to Embodiment 1, the block size can be easily acquired so as not to exceed the memory capacity of the buffer memory 314.

  The embodiments have been described above with reference to specific examples. However, the present invention is not limited to these specific examples.

  In addition, although descriptions are omitted for parts and the like that are not directly required for the description of the present invention, such as a device configuration and a control method, a required device configuration and a control method can be appropriately selected and used. For example, the description of the configuration of the control unit that controls the drawing data divided region size acquisition device 100 and the configuration of the drawing device has been omitted, but it goes without saying that the required configuration and the like are appropriately selected and used.

  In addition, any drawing data divided region size acquisition method and drawing data divided region size acquisition apparatus that include elements of the present invention and that can be appropriately modified by those skilled in the art are included in the scope of the present invention.

FIG. 3 is a flowchart showing main steps of a method for acquiring a divided area size of drawing data in the first embodiment. 6 is a diagram illustrating an example of a hierarchical structure of drawing data according to Embodiment 1. FIG. 3 is a conceptual diagram illustrating a configuration of a drawing data divided region size acquisition apparatus according to Embodiment 1. FIG. 3 is a conceptual diagram illustrating an example of a cluster area in Embodiment 1. FIG. 3 is a conceptual diagram illustrating an example of a figure that can be arranged in a cluster area in Embodiment 1. FIG. 5 is a conceptual diagram for explaining an expected value of the number of rectangular figures in the first embodiment. FIG. 6 is a conceptual diagram for explaining an expected figure number of a right isosceles triangle in Embodiment 1. FIG. 6 is a conceptual diagram for explaining an expected value of the number of trapezoidal figures in the first embodiment. FIG. 6 is a conceptual diagram for explaining an expected number of figures of an isosceles trapezoid and an isosceles triangle in Embodiment 1. FIG. 6 is a conceptual diagram for explaining an expected number of figures of a parallelogram in the first embodiment. FIG. 6 is a diagram showing an example of a data amount list in the first embodiment. FIG. 6 is a diagram illustrating an example of a relationship among an estimation error of the number of figures, a data amount, and a memory capacity of a buffer memory in the first embodiment. FIG. 6 is a flowchart showing a flow of data processing in the drawing apparatus according to Embodiment 1. FIG. It is a conceptual diagram for demonstrating operation | movement of the conventional variable shaping type | mold electron beam drawing apparatus.

Explanation of symbols

10 chip 12 intermediate data 20 stripe 30 block 32 cell 34 cluster 36 figure 40 cut line 42 margin 44 cluster boundary 52, 54, 56 figure 60 data amount list 100 divided area size acquisition device 110 control computer 111 input unit 112 merge processing unit 114 Cell extraction unit 116 Dividing unit 118 Figure number estimating unit 120 Cell data amount estimating unit 122 Judging unit 124 Block data amount estimating unit 126 List creating unit 132 Output unit 134 Memory 136 External I / F
138, 310, 312, 316 Magnetic disk device 314 Buffer memory 330 Electron beam 340 Sample 410 First aperture 420 Second aperture 411 Opening 421 Variable shaping opening 430 Charged particle source

Claims (5)

Inputting drawing data for drawing a pattern on a sample using a charged particle beam;
A step of inputting a predetermined divided area size and dividing a drawing area defined in the drawing data into a plurality of divided areas having the predetermined divided area size;
Using the expected figure number after division derived based on the probability that the figure defined in the drawing data is divided, and estimating the data amount of the figure arranged in each divided area;
Outputting the data amount for each divided region;
A method for acquiring a divided area data amount of drawing data. A graphic having a size equal to or smaller than the size (2f) of the overlapping portion in a plurality of small regions that divide a cell constituted by at least one graphic defined in the drawing data with an overlapping portion has the probability of the size. 0,
If a figure having an intermediate size between the size (2f) of the overlapping portion and the size (S) of the small area is expressed using a figure size (L), the probability of the size is (L-2f) / S. ,
When expressed using a figure size (L) and an integer (n) of 2 or more, a figure having a size in the range of (n−1) S ≦ L <nS has the probability (L−2f) / ( 2. The method for acquiring the divided area data amount of drawing data according to claim 1, wherein nS- (n-1) * 2f). The figure having a size equal to or smaller than the size (2f) of the overlapping portion has the figure number expectation value of 1 for the size,
When a figure having an intermediate size between the size (2f) of the overlapping portion and the size (S) of the small area is represented using a figure size (L), the expected figure number for the size is 1+ (L-2f). ) / S,
When expressed using a graphic size (L) and an integer (n) of 2 or more, a graphic having a size that falls within the range of (n−1) S ≦ L <nS has an expected value of the graphic number n + (L -2f) / (nS- (n-1) .times.2f). 3. The method according to claim 2, wherein the divided data amount of drawing data is obtained.   The data amount is a value obtained by multiplying the total figure number obtained by calculating the figure number of each figure by an arithmetic expression for each figure type using the figure number expectation value for each corresponding cell by a unit data amount per figure. 4. The method for obtaining the divided area data amount of the drawing data according to claim 2 or 3, wherein the cumulative area is obtained by cumulatively adding to the corresponding divided area. A storage unit for inputting and storing drawing data for drawing a pattern on a sample using a charged particle beam;
A division unit that inputs a predetermined division area size and divides the drawing area defined in the drawing data into a plurality of division areas having the predetermined division area size;
Using an expected figure number after division derived based on the probability that the figure defined in the drawing data is divided, and an estimation unit for estimating the data amount of the figure arranged in each divided area;
An output unit for outputting the data amount for each divided area;
An apparatus for acquiring a divided area data amount of drawing data.

Images