JPH08153657A - Bit map developing method and device of graphic data for charged particle beam exposure device - Google Patents

Abstract

PURPOSE: To obtain a bit map developing method and a device, wherein an exposure pattern is finely controlled in a border line without taking its breadth into consideration and processing it along its outline, quickly processed, and turned into a hardware. CONSTITUTION: Graphic data are developed into a bit map twice as high in resolution as a final bit map in both the directions of X and Y, and the bit map is divided into cells C11 which are of 4×4 bits and equal to each other. The cell C11 is composed of beam spot corresponding points P11 , P12 , P22 , and P21 located at the apexes of a square 51 whose side is d in length and bit data acquisition points Q11 , Q12 , Q22 , and Q21 located at the apexes of a rhombus 52. The bit data of the bit data acquisition points are fetched by shifting to the beam spot corresponding points, whereby a final bit map is obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for developing a bitmap of graphic data for a charged particle beam exposure apparatus and a charged particle beam exposure method and apparatus.

[0002]

2. Description of the Related Art FIG. 19 shows a main part of a conventional multi-charged particle beam type exposure apparatus. A semiconductor wafer 10 as an exposure object is mounted on a moving stage 12.
The movement of the moving stage 12 is controlled by the stage control circuit 14, and the position of the moving stage 12 is determined by the laser interferometer 16
Is detected by and is fed back to the stage control circuit 14. A resist film is deposited on the semiconductor wafer 10, and exposure is performed by irradiating the resist film with the electron beam EB2 that has passed through the aperture plate 18.

Electron beam EB on semiconductor wafer 10
The scanning of 2 is performed by the moving stage 12 and the electromagnetic main deflector 20 and the electrostatic sub-deflector 22 arranged above the moving stage 12. The moving stage 12, the main deflector 20, and the sub deflector 22 are in the descending order of the range that can be scanned with the required accuracy, but the reverse order is the descending order of the scanning speed. The scanning as shown in 20 is performed.

That is, the electron beam E is emitted by the main deflector 20.
B2 is continuously deflected in the main scanning direction X, and in parallel with this, the moving stage 12 is continuously moved in the sub scanning direction Y perpendicular to the main scanning direction X. Furthermore, the electron beam EB2
Are continuously deflected in the direction Y by following the movement of the moving stage 12 by the sub deflector 22. For example, the size of one band is 2 mm in length and 10 μm in width, and the scanning time for one band is 100 μs. In this case, the moving stage 1
2 moving speed in Y direction is 10 μm / 100 μs = 100 m
m / s.

Chip area C1 on the semiconductor wafer 10
The same pattern is exposed on C11. The movement of the moving stage 12 is controlled so that the shaded areas are sequentially exposed in the arrow direction. In FIG. 19, the main control circuit 24 supplies the target position to the stage control circuit 14,
A periodic sawtooth wave signal is supplied to the amplifier circuit 26, the stage detection position Y is received from the laser interferometer length measuring device 16, and
(Blanking Aperture Array) The band Y coordinate Yi is received from the control circuit 40, and a signal proportional to the sub deflection distance Y-Yi is supplied to the amplification circuit 28. The drive signals that have been current-amplified and voltage-amplified by the amplifier circuits 26 and 28, respectively, are supplied to the main deflector 20 and the sub-deflector 22.

A blanking aperture array (BAA) plate 30 is arranged above the aperture plate 18. The blanking aperture array plate 30 is shown in FIG.
As shown in (FIG. 21 shows the back surface), a thin substrate 31
The openings 32 are formed in a zigzag pattern. For each opening 33, a pair of electrodes 33 and 34 are erected on the substrate 31 above the paper surface (downward in FIG. 19).

As shown in FIG. 19, a blanking aperture array plate 30 is provided with an electron beam EB having a substantially uniform current density.
0 is projected. Electron beam EB2 passing through the aperture 33
Passes through or is blocked by the lower aperture plate 18 depending on whether the voltage applied between the electrodes 33, 34 is 0V. Therefore, the electrode 3 corresponding to the exposure dot data
By applying a voltage between 3 and 34, the desired fine pattern is exposed on the semiconductor wafer 10.

In FIG. 21, one side of the opening 33 has a length a.
Is a square. Aperture group 30A along the X direction
.About.30D, each opening group is divided into two opening rows, and each opening row is divided into two stages so as to make up for the openings. For example, in the opening row 30A1 of the opening group 30A, the semiconductor wafer 10 is exposed by exposing the first-stage openings and then exposing the second-stage openings after the main scanning time corresponding to the opening pitch 3a in the X direction has elapsed. One line of the band width is exposed. The same exposure point on the semiconductor wafer 10 is exposed with the same exposure dot data in the same column in the same opening group, for example, the hatched openings A1 and A2.

Between the adjacent aperture groups 30A to 30D is shifted by a / 2 in the Y direction. Also, the aperture groups 30B and 30D
Is 0.5a longer than the interval 3a of the other aperture groups. From these things and the above, the irradiation beam spot on the semiconductor wafer 10 of the electron beam passing through the eight openings A1 to D2 in the substantially same line with diagonal lines is shown in FIG.
It becomes like AD shown in (A). Since FIG. 21 is a back view, the Y direction is opposite to that of FIGS. 22 (A) to 22 (C). The beam spot A corresponds to the openings A1 and A2, and the same applies to the other areas. For example, one side length a =
The area exposed on the semiconductor wafer 10 through the opening 33 of 25 μm is a substantially square having a side length of 2 ka = 0.08 μm. The pitch of the beam spots in the X and Y directions is ka.

To increase the resolution of the exposure pattern, the size of the beam spot may be reduced. However, since the pattern data is expanded into a bitmap to obtain the exposure dot data, the data amount becomes enormous. For example, at least (20000 / 0.08) ² = 62.5 Gbi is required to expose a square semiconductor chip having a side of 20 mm.
t exposure dot data is required. It also causes a decrease in exposure speed.

Therefore, there has been proposed a method of adjusting the boundary line of the exposure pattern as follows. FIG. 22 (B)
Shows a set of beam spots for forming an exposure pattern. 22A to 22D respectively correspond to A to D in FIG. However, the length of one side of the beam spot is shown as half. The boundary line of the exposure pattern is a line that connects the points where the exposure amount becomes equal to the threshold value.
Indicated by

Two beam spots contacting the boundary line BL1
When one is pulled out at a time, it becomes as shown in FIG. In this case, the boundary line BL2 is shifted by ka / 2 from the boundary line BL1. Similarly, the boundary line of the exposure pattern can be finely adjusted by extracting one out of three beam spots near the boundary line or one out of four beam spots.

[0013]

However, when expanding a polygon data pattern into a bitmap for each of a huge number of patterns, it is necessary to perform processing along the contour line according to the pattern width. Since it is a idiot, a huge amount of time is spent on the expansion process. In view of such a problem, an object of the present invention is to finely adjust the boundary line of the exposure pattern without considering the pattern width or performing the process along the contour line, and the hardware of the process and It is an object of the present invention to provide a method and an apparatus for expanding a bitmap of graphic data for a charged particle beam exposure apparatus capable of increasing the speed.

[0014]

According to the first aspect of the present invention, the value of the bit is determined according to whether or not the charged particle beam is irradiated onto the exposure object according to the value of each bit on the first bitmap. In order to form an exposure pattern on the exposure target corresponding to the presence or absence of a beam spot on the exposure target on the exposure target, graphic data corresponding to the exposure pattern In the bitmap expansion method of graphic data to be expanded to a bitmap,
The graphic data is converted into n in the X direction with respect to the first bitmap.
Double (n ≧ 2) and different from the X direction in the Y direction is m times (m ≧ 1)
A first step of developing into a second bitmap having a resolution of, and when divided into (X direction 2n bits) × (Y direction 2m bits) cells on the second bitmap,
The number of rows in the X direction and the number of columns in the Y direction including the selection bits is 3 or more, so that the selection in each cell changes regularly along the X and Y directions. And a second step of creating the first bitmap.

According to the first aspect of the present invention, 4 bits from a cell of (2n bits in X direction) × 2 m bits in Y direction are selected, and the number of rows in the X direction and the number of columns in the Y direction including selection bits is 3 in each case. Since the first bitmap is created by selecting as described above, the boundary line of the exposure pattern can be finely adjusted without considering the pattern width or performing the processing along the contour line. In addition, since it is not necessary to consider the pattern width or perform processing along the contour line, and the selection in each cell changes regularly along the X direction and the Y direction, the processing hardware Speed and speed.

In the first aspect of the first invention, in the second step, the selection in the cells is the same for all cells, as shown in FIG. 1, for example. According to the first aspect, since it is only necessary to repeatedly perform the processing on a cell-by-cell basis, it is easier to implement the processing in hardware. In a second aspect of the first aspect of the present invention, in the second step, the number of rows in the X direction including the selection bit and columns in the Y direction in the cell is 4 and the row in the X direction including the selection bit is added. The interval and the interval of columns in the Y direction including the selected bit are equal to each other.

According to the second aspect, the above "3 or more" has a maximum value of 4 for each of the rows in the X direction and the columns in the Y direction, and the intervals of the rows and columns are equal to each other, so that the exposure pattern You can adjust the boundaries more finely. In the third aspect of the first invention, in the first step, for example, as shown in FIG. 5, the X direction and the Y direction are at right angles to each other and m = n = 2, and in the second step, The selection bits are the first row, first column, second row, third column,
It is 4 bits in the third row, second column and the fourth row, fourth column.

According to the third aspect, the second aspect can be realized. In the fourth aspect of the first aspect of the present invention, in the second step, the selection in the cell is the same in units of clusters composed of a plurality of cells, and the clusters are in the X direction and Y
Arranged regularly along the direction. According to the fourth aspect, by coordinating a plurality of cells in the cluster, the boundary line of the exposure pattern can be adjusted more finely than in the case without the cluster, as shown in FIG. 7, for example.

In a fifth aspect of the first aspect of the invention, in the second step of the fourth aspect, the interval between rows in the X direction including the selection bit in the cluster and Y including the selection bit.
The intervals of the rows in the direction are equal to each other, as shown in FIG. 7, for example. According to the fifth aspect, since the intervals are equal to each other, the boundary line of the exposure pattern can be finely adjusted as compared with the case where a wide pattern and a narrow pattern are mixed.

In the sixth aspect of the first invention, the second step has an n-bit width in the X direction on the second bitmap as shown in FIG. 5 (n = m = 2 in FIG. 5). Within the band along the Y direction, (n bits in the X direction) × (Y
(Direction m bits), selecting one bit regularly along the Y direction, and selecting so that there are a plurality of columns in the Y direction including the selected bit.
Create the first bitmap.

According to the sixth aspect, it is easier to implement the processing hardware. In the seventh aspect of the first invention, the first bitmap data created by any one of the above methods is used, and the charged particle beam is irradiated onto the exposure target object according to the value of each bit on the first bitmap. Depending on whether or not to do so, the value of the bit is made to correspond to the presence or absence of a beam spot on the exposure object, and an exposure pattern corresponding to the pattern on the first bitmap is formed on the exposure object.

According to the seventh aspect, the boundary line of the exposure pattern can be actually finely adjusted. The apparatus and its aspect of the second invention described below correspond to the method and its aspect of the first invention. In the second invention, the value of the bit is made to correspond to the presence or absence of a beam spot on the exposure object depending on whether or not the charged particle beam is irradiated onto the exposure object according to the value of each bit on the first bitmap. , A bitmap data expansion device for expanding graphic data corresponding to the exposure pattern to the first bitmap in order to form an exposure pattern corresponding to the pattern on the first bitmap on the exposure object In, the graphic data is expanded into a second bitmap having a resolution n times (n ≧ 2) in the X direction and m times (m ≧ 1) in the Y direction perpendicular to the X direction with respect to the first bitmap. On the first expansion device and the second bitmap, (2n bits in X direction) × (2m bits in Y direction)
When divided into cells, the four bits from each cell are selected so that the selection in each cell changes regularly along the X and Y directions. And a second expansion device that creates the first bitmap by selecting all the numbers to be three or more.

In the first aspect of the second aspect of the present invention, the second expansion device has (n-bits in the X-direction) × (n-bits in the X-direction) within a band along the Y-direction having an n-bit width in the X-direction on the second bitmap. Select 1 bit from (Y direction m bits),
The selection is performed regularly along the Y direction, and the selection is performed so that there are a plurality of columns in the Y direction including the selection bit, thereby creating the first bitmap.

In the second aspect of the second aspect of the present invention, the graphic data bitmap expanding device and the first bitmap data created by the graphic data bitmap expanding device are used, and Depending on whether or not the charged particle beam is irradiated onto the object to be exposed according to the value of the bit,
An exposure device that forms an exposure pattern on the exposure target corresponding to the presence or absence of a beam spot on the exposure target by forming the exposure pattern corresponding to the pattern on the first bitmap;
Have.

According to a third aspect of the second invention, in the above-described exposure apparatus, a blanking aperture array in which a plurality of openings are formed in a grid in a substrate and a pair of electrodes is formed at the edge of each opening of the substrate is charged. The exposure target is arranged in the optical path of the particle beam, and controls whether the charged particle beam passing through the opening is irradiated onto the exposure target depending on whether or not a voltage is applied between the pair of electrodes. The charged particle beam irradiation position on the object is raster-scanned by the scanning means.

[0026]

Embodiments of the present invention will be described below with reference to the drawings. In the drawings, the same or similar components are designated by the same or similar reference numerals. [First Embodiment] FIG. 1 shows the relationship between bit data acquisition points and beam spot corresponding points on the pattern arrangement surface in the first embodiment.

The intersections of the vertical and horizontal dotted lines are the beam spot corresponding points, which are indicated by a circle and Pij. Where i
And j are numbers attached to horizontal and vertical dotted lines in the figure. The beam spot corresponding point corresponds to the center point of the beam spot obtained by irradiating the beam. When the pitch of the beam spot corresponding points is d, the beam spot is, for example, a square having a side length of 2d.

The bit data is "1" inside the pattern and "0" outside the pattern. For example, vertices S1, S2, S7
When a pattern of polygon data represented by each set of coordinates of S and S8 is expanded into a bitmap, conventionally, grid points P24, P25, P in the rectangles S1, S2, S7, and S8 are used.
The bit data of 34, P35, P44 and P45 are set to "1".

On the other hand, in the first embodiment, the bit data at the bit data acquisition point Qij represented by ● are shifted and fetched into the beam spot corresponding point Pij. The relationship between the beam spot corresponding point Pij and the bit data acquisition point Qij is repeated with the cell C11 as a unit. The cell C11 has beam spot corresponding points P11, P12, P2 located at the vertices of a square 51 having a side length of d.
2 and P21, and bit data acquisition points Q11, Q12, Q22 and Q21 located at the vertices of the diamond 52. The bit data acquisition point Q12 is the midpoint between the beam spot corresponding points P12 and P22.
21 is the midpoint between the beam spot corresponding points P21 and P22. The bit data acquisition point Q22 is a square P2.
It is the midpoint of 2.P23.P33.P32.

Since d is extremely small, for example, 0.08 μm, the deformation of the pattern due to the displacement of the rhombus 52 from the square 51 is slight. This deformation appears at the boundaries of the pattern. However, since this deformation includes the shift component of the entire pattern that does not cause any problem, it can be regarded as a deformation by removing the shift component, that is, by changing the diamond 52 from the diamond 53. The substantial shift amount is the vertices R11, R12, R on the diamond 53.
22 and R21 and the corresponding vertices on the square 51, which are all equal and √2d / 4 = 0.35.
d = 0.028 μm, which is small. In addition, since the boundary line of the exposure pattern is blurred due to the reflection of electrons in the photoresist, the actual deformation is further reduced.

Regarding the above rectangular pattern, assuming that the data on the boundary line of the pattern is "1", the beam spot corresponding points P24, P25, P34, P35, P44, P4.
5 are the bit data acquisition points Q24, Q25, Q3, respectively.
The data of '1' of 4, Q35, QA4, and Q45 is fetched, and the rectangular beam spot group becomes the same as the conventional one. Rectangular patterns S1 and S3 in which the width of the rectangular pattern is widened by d / 2
-In S6 and S8, bit data acquisition points Q26 and Q4
The data of "0" of 6 are the beam spot corresponding points P, respectively.
Since the beam spot corresponding points P26 and P46 are both taken as "1", the pattern width becomes narrower than that in the conventional case.

In the rectangular patterns S1, S4, S5, and S8 in which the width of the rectangular pattern is further widened by d / 2, the data "1" at the bit data acquisition points Q26 and Q46 are fetched at the beam spot corresponding points P26 and P46, respectively. The exposure pattern width becomes wider. On the other hand, in the conventional case, there is no change in the exposure pattern width. In conclusion, conventionally, the number of beam spots is increased every time the width of the rectangle is shifted by d, but according to the first embodiment, the number of beam spots is increased every time d / 2 is shifted. You can adjust the boundaries more finely. Further, there is no need to consider the pattern width or perform processing along the contour line.

All patterns to be exposed can be decomposed into a rectangular pattern and a right triangle pattern. Figure 2
Is for explaining the increase in the number of beam spots when the triangular pattern on the map of FIG. 1 is gradually expanded. Regarding the triangles T3, T4, T5, the difference from the conventional one is that the beam spot corresponding point P34 becomes "0".

When the triangles T2, T4, and T6 are enlarged, the bit data acquisition point Q34 of "1" is taken into the beam spot corresponding point P34, so that the exposure pattern is slightly enlarged. On the other hand, there is no change in the conventional case. If it is further expanded to form triangles T1, T4, and T7, the number of beam spots increases by 4 points, which is the same as the conventional one.

In conclusion, in the past, the number of beam spots increased every time one side sandwiching a right angle of a triangle was shifted, but according to the first embodiment, the number of beam spots increased every d / 2 shift. Therefore, the boundary line of the exposure pattern can be adjusted more finely. Further, there is no need to consider the pattern width or perform processing along the contour line. The effects described above can be easily understood by referring to FIGS. FIGS. 3A to 3D show the relationship between the parallel movement of the pattern boundary line and the data acquisition point for one cell C11 in FIG.

In FIG. 3A, the right end of the pattern is Y.
When X = 0, 1, 2, and 3 are expanded in parallel with the axis, the number of bit data acquisition points increases by one, and the point at which the bit data corresponding points are taken as '1' is 1 Increase by one. The same applies to the case where this boundary line is the left end of the pattern. Regarding FIGS. 3B to 3D,
The number of dotted lines is equal to the number of lines 3 (not shown) passing through the circles parallel to this dotted line, but as is clear from FIG. 1, the same can be said when considering surrounding cells. In addition,
The rectangular pattern is overwhelmingly more than the triangular pattern,
Moreover, since the beam spot diameter is sufficiently smaller than the pattern size, the effectiveness of the first embodiment is high.

FIG. 4 shows an exposure data processing device. The buffer memory 241 is built in the main control circuit 24 of FIG. 19, and the pattern data disk 60, the data expansion device 61, and the bit map memory 6 are provided outside the main control circuit 24.
2. Bit shift circuit 63 and bitmap disk 6
4 are arranged. That is, according to the present invention, the main part of the conventional graphic data bitmap expanding device shown in FIG. 19 can be used as it is. The storage capacities of the pattern data disk 60 and the bitmap disk 64 are the same as those of the conventional one.

The pattern data disk 60 comprises basic pattern data including parameters and data designating the parameters. The basic pattern data includes a shape code, origin coordinates, and size. The data expansion device 61 reads the pattern data from the pattern data disk 60, expands this into a bitmap, and stores it in the bitmap memory 62. The developed 1-bit data is "1" if a square with a side length d / 2 is in the pattern, and is "0" otherwise.

The bit shift circuit 63 removes unnecessary data to reduce the amount of bitmap data to 1/4, and shifts the bit indicated by ● in FIG. 2 to the position indicated by ○. The shifted data is stored on the bitmap disk 64.
Stored in. The data stored in the bitmap disk 64 is read in block units at the time of exposure and stored in the buffer memory 241.

FIG. 5A shows a bit map of two cells. One square is 1-bit data. The data actually used are the bits marked with ●. When the mark ◯ is removed, the mark  is connected by a straight line in the vertical direction in units of two columns, and the two columns on the left side are shifted upward by 1 bit, it becomes symmetrical as shown in FIG. 5B. The bit shift circuit 63 is configured as shown in FIG. 5C by utilizing such a property. FIG. 5C shows a case where one word of the bitmap memory 62 is 4 bits corresponding to the cell width for simplification. Bitmap memory 62 is addressed by the clock count value of counter 65.

The 2-bit register 71 corresponds to the upper 1-bit shift in the left two columns in FIG. 5B. The selectors 72A and 72B are for selecting the data marked with a circle in FIG. 5B, and the selection control signal is given from the rotation shift register 73 having a 2-bit cycle. The bit data selected by the selectors 72A and 72B is held in the 2-bit register 74.

The cycle of the clock supplied to the shift register 73 and the register 74 is twice that of the clock supplied to the counter 65 and the register 71. With the bit shift circuit 63 having such a simple structure, the data at the bit data acquisition points marked with a black circle can be substantially shifted to the beam spot corresponding points at high speed. Further, unnecessary data is removed and the data amount becomes 1/4.

[Second Embodiment] In the first embodiment, the distance between the bit data acquisition points in the X and Y directions is set to d / 2, but it can be made smaller. Figure 6
Shows the relationship between the bit data acquisition points and the beam spot corresponding points in the second embodiment.

In FIG. 1, one cell is repeatedly arranged in a matrix, but in FIG. 6, four different cells C11, C12, C22 and C21 are arranged in a cluster C.
L1 is used as a unit and is repeatedly arranged in a matrix. Cell C11 is the same as that shown in FIG. The bit data acquisition points of each cell are arranged at the vertices of the diamonds 521 to 524 that are congruent to each other. Diamond 522
Is d / 4 shifted downward with respect to the diamond 521, the diamond 524 is d / 4 shifted left with respect to the diamond 521, and the diamond 523 is d / 4 shifted left with respect to the diamond 522.

As in the case of FIG. 1, the data at the bit data acquisition points Q11 to Q44 are fetched at the beam spot corresponding points P11 to P44, respectively. 7A and 7B and FIGS. 8A and 8B show the relationship between the parallel movement of the pattern boundary line and the data acquisition point for one cluster in FIG. As is clear from FIGS. 7A and 7B, the number of beam spots increases each time the pattern boundary line parallel to the Y-axis or the X-axis shifts in units of clusters by d / 4. The line can be adjusted more finely.

In the case of FIGS. 8 (A) and 8 (B), there are also seven straight lines (not shown) which are parallel to the dotted lines and pass through the circles, whereas there are nine dotted lines passing through the circles. The boundary line of the exposure pattern can be adjusted more finely than before. In addition, although there are a wide portion and a narrow portion between the dotted lines passing through the bit data acquisition points, the wide portion has a narrower interval with a line inserted when considering the surrounding clusters.

9 (A), 10 (A) and 11
(A) shows a rectangular pattern, and the beam spot center point and the schematic exposure pattern corresponding to these are shown in FIGS. 9 (B), 10 (B) and 11 (B). The pattern of FIG. 10 (A) is obtained by shifting the pattern of FIG. 9 (A) by d / 4 in the X-axis direction, and the pattern of FIG. 11 (A) is shown in FIG.
The pattern of (A) is shifted by d / 4 in the X-axis direction. From these figures, it can be seen that the exposure pattern is also shifted by approximately d / 4 each time the rectangular pattern is shifted by d / 4.

Further, FIGS. 12A, 13A and 14A show triangular patterns, and the beam spot center point and the schematic exposure pattern corresponding to these are shown in FIG.
13B, FIG. 13B, and FIG. 14B. FIG.
The pattern of (A) is obtained by shifting the pattern of FIG. 12 (A) by d / 4 in the X axis direction, and the pattern of FIG. 14 (A) is the pattern of FIG. 13 (A) d / in the X axis direction. It is a four-shifted one. From these figures, it can be seen that the exposure pattern is also shifted by approximately d / 4 each time the rectangular pattern is shifted by d / 4.

FIG. 15A shows the cluster CL1 in FIG.
Shows a half bitmap of. One square is 1-bit data. The data actually used are the bits marked with ●. Remove the ○ mark, connect the ● mark in units of 4 columns in the vertical direction by a straight line, and shift the 4 columns on the left by 2 bits upwards.
It becomes symmetrical as shown in FIG. The bit shift circuit 63A utilizes such a property, and is shown in FIG.
It is configured as shown in. For simplification, FIG. 15C shows the case where one word in the bitmap memory has 8 bits corresponding to the cell width.

4-bit 2-stage registers 71A and 71B
Corresponds to an upward 2-bit shift of the above left four columns. The selectors 72C and 72D are for selecting the data indicated by ●, and the selection control signal is given from the rotary shift register 73A having a 4-bit cycle. The bit data selected by the selectors 72C and 72D is held in the 2-bit register 74.

The cycle of the clock supplied to the shift register 73 and the register 74 is the bit map memory 62A.
Address counter 65 and registers 71A, 71B
4 times that of the clock supplied to. With the bit shift circuit 63A having such a simple configuration, the data at the bit data acquisition points marked with a black circle can be substantially shifted to the beam spot corresponding points at high speed. Also,
Unnecessary data is removed and the data amount becomes 1/16.

The 4-bit 2-stage registers 71A and 71B may be composed of four 2-bit shift registers, and instead of the rotary shift register 73A, 4 registers may be used.
It goes without saying that a configuration using a binary counter and a circuit for detecting that the count value has reached a certain value may be used. [Third Embodiment] There is another method of selecting a cluster that can obtain the same effect as that of the second embodiment.

FIG. 16 shows the relationship between bit data acquisition points and beam spot corresponding points in the third embodiment. The cluster CL2 consists of four different cells C11, D12, D22 and D21. Cell C11 is the same as that shown in FIG. The bit data acquisition points of each cell are arranged at the vertices of the congruent diamonds 521, 525 to 527. The diamond 525 is shifted by d / 4 upwards with respect to the diamond 521, and the diamond 527 is d / 4 by the right side with respect to the diamond 521.
The diamond 526 is shifted to the right by d / 4 with respect to the diamond 525.

FIGS. 17A and 17B and FIG.
16A and 16B show the relationship between the parallel movement of the pattern boundary line and the data acquisition point for one cluster in FIG. As is clear from FIGS. 17A and 17B, the pattern boundary line parallel to the Y axis or the X axis is d in cluster units.
Since the number of beam spots increases every / 4 shift, the boundary line of the exposure pattern can be adjusted more finely.

In the case of FIGS. 18A and 18B, FIG.
The intervals of the dotted lines are more constant than in the cases of (A) and (B). In addition, while there are seven straight lines (not shown) that are parallel to the dotted line and pass through the circles, there are nine dotted lines that pass through the ●, and even for this point, the boundary line of the exposure pattern is adjusted more finely than in the past. it can. The wide portion between the dotted lines passing through the bit data acquisition points is a line that is inserted when the surrounding clusters are taken into consideration and has a narrow interval.

The cells and clusters of the present invention include various modifications, and correspondingly, the bit shift circuit also includes various modifications. Instead of the bit shift circuit, the data of the memory cell at the bit data extraction point may be read from the memory cell array once in a plane. Further, the present invention is applied to an exposure apparatus using a blanking aperture as shown in FIG. 21 in which adjacent aperture groups 30A to 30D are shifted in the Y direction and aperture groups 30B and 30C are shifted in the X direction. However, the present invention is not limited to this, and can be applied to an exposure apparatus using a blanking aperture without this shift.

[0057]

As described above, according to the method and apparatus for developing the bitmap of the figure data for the charged particle beam exposure apparatus according to the present invention, (X direction 2n bits) × (Y direction 2)
4 bits from a cell of (m bits), X including the selection bit
Since the first bitmap is created by selecting so that the number of rows in the direction and the number of columns in the Y direction are both 3 or more, the pattern width is not taken into consideration and the processing along the contour line is not performed. You can finely adjust the border of the exposure pattern,
Since it is not necessary to consider the pattern width or perform processing along the contour line, and the selection in each cell changes regularly along the X direction and the Y direction, the processing hardware and It has an excellent effect that speeding up is possible.

According to the first aspect of the first aspect of the present invention, since it is sufficient to perform the repetitive processing on a cell-by-cell basis, there is an effect that it is easier to implement the processing in hardware. According to the second and third aspects of the first invention, the above “3 or more” has a maximum value of 4 for each of the rows in the X direction and the columns in the Y direction, and the intervals between the rows and the columns are equal to each other. The effect is that the boundary line of the pattern can be adjusted more finely.

According to the fourth aspect of the first aspect of the invention, there is an effect that the boundary line of the exposure pattern can be finely adjusted by cooperation of a plurality of cells in the cluster as compared with the case where there is no cluster. According to the fifth aspect of the first aspect of the invention, since the row interval in the X direction including the selection bit and the column interval in the Y direction including the selection bit are equal to each other, the exposure is performed as compared with the case where a wide area and a narrow area are mixed. The effect is that the boundary line of the pattern can be finely adjusted.

According to the sixth aspect of the first aspect of the present invention, it is possible to make the processing hardware easier. First
According to the seventh aspect of the invention or the second aspect of the second invention, it is possible to actually finely adjust the boundary line of the exposure pattern. The first aspect of the second invention has an effect that the hardware configuration is simple.

[Brief description of drawings]

FIG. 1 is a map showing a relationship between bit data acquisition points and beam spot corresponding points according to the first embodiment of the present invention.

FIG. 2 is an explanatory diagram for increasing the number of beam spots when the triangular pattern on the map of FIG. 1 is gradually expanded.

FIG. 3 is a diagram showing a relationship between parallel movement of a pattern boundary line and a data acquisition point for the cell in FIG.

FIG. 4 is a block diagram showing an exposure data processing apparatus according to the first embodiment of the present invention.

FIG. 5 (A) shows a bitmap of two cells,
(B) shows one half of this bitmap shifted by 1 bit, and (C) shows a bit shift circuit in relation to this bitmap.

FIG. 6 is a map showing a relationship between bit data acquisition points and beam spot corresponding points according to the second embodiment of the present invention.

7 is a diagram showing the relationship between parallel movement of pattern boundary lines and data acquisition points for the clusters in FIG. 6;

8 is a diagram showing the relationship between parallel movement of pattern boundary lines and data acquisition points for the clusters in FIG. 6;

9A is a diagram showing a rectangular pattern, and FIG. 9B is a diagram showing a beam spot center point and a schematic exposure pattern corresponding thereto.

10A is a diagram showing a rectangular pattern, and FIG. 10B is a diagram showing a beam spot center point and a schematic exposure pattern corresponding thereto.

11A is a diagram showing a rectangular pattern, and FIG. 11B is a diagram showing a beam spot center point and a schematic exposure pattern corresponding thereto.

12A is a diagram showing a triangular pattern, and FIG. 12B is a diagram showing a beam spot center point and a schematic exposure pattern corresponding thereto.

13A is a diagram showing a triangular pattern, and FIG. 13B is a diagram showing a beam spot center point and a schematic exposure pattern corresponding thereto.

FIG. 14A is a diagram showing a triangular pattern, and FIG. 14B is a diagram showing a beam spot center point and a schematic exposure pattern corresponding thereto.

FIG. 15A shows a bitmap of half of the cluster, FIG. 15B shows a half-shift of this bitmap by 2 bits, and FIG. 15C shows a bit shift circuit in relation to this bitmap. FIG.

FIG. 16 is a map showing a relationship between bit data acquisition points and beam spot corresponding points according to the third embodiment of the present invention.

FIG. 17 is a diagram showing a relationship between parallel movement of pattern boundary lines and data acquisition points for the clusters in FIG. 16;

FIG. 18 is a diagram showing a relationship between parallel movement of pattern boundary lines and data acquisition points for the cluster in FIG. 16;

FIG. 19 is a block diagram of a main part of a conventional multi-charged particle beam type exposure apparatus.

FIG. 20 is an explanatory diagram of an electron beam scanning method.

21 is a partial rear view of the blanking aperture array in FIG. 19. FIG.

FIG. 22 (A) shows the overlap of beam spots,
(B) And (C) is a figure which shows the boundary line of an exposure pattern.

[Explanation of symbols]

51 square 52, 53 diamond 60 pattern data disk 61 data expansion device 62, 62A bit map memory 63, 63A bit shift circuit 64 bit map disk 65 counter 241 buffer memory 24 main control circuit 40 BAA control circuit C11, C12, C21, C22 , D12, D21, D
22 cells CL1, CL2 cluster Pij beam spot corresponding point Qij bit data acquisition point

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Jun-ichi Kai 1015 Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa, Fujitsu Limited (72) Inventor Hiroshi Yasuda 1015 Kamedota, Nakahara-ku, Kawasaki, Kanagawa

Claims (12)

[Claims] 1. The value of the bit is made to correspond to the presence or absence of a beam spot on the object to be exposed depending on whether or not the charged particle beam is irradiated onto the object to be exposed according to the value of each bit on the first bit map. , A bitmap data expansion method for expanding graphic data corresponding to the exposure pattern to the first bitmap so as to form an exposure pattern corresponding to the pattern on the first bitmap on the exposure object In the X direction with respect to the first bitmap,
Double (n ≧ 2) and different from the X direction in the Y direction is m times (m ≧ 1)
A second step of expanding to a second bitmap having a resolution of, and (2n bits in the X direction) on the second bitmap
When divided into x (Y direction 2 m bits) cells, 4 bits from each cell are selected so that the selection in each cell changes regularly along the X direction and the Y direction. A second step of selecting the number of rows and columns in the Y direction to be 3 or more and creating the first bitmap, and a bitmap data expansion method for graphic data. 2. The method of claim 1, wherein in the second step, the selection in the cell is the same for all cells. 3. In the second step, the number of rows in the X direction and columns in the Y direction each including the selection bit in the cell is 4, and an interval between rows in the X direction including the selection bit and the selection. 2. A method according to claim 1, characterized in that the columns in the Y direction containing the bits are equally spaced. 4. In the first step, the X direction and the Y direction are at right angles to each other and m = n = 2, and in the second step, the selection bit is in the first row, first column. 4. The method of claim 3, wherein the 4 bits are in the second row, third column, the third row, second column, and the fourth row, fourth column. 5. In the second step, the selection in the cell is the same in a unit of a cluster composed of a plurality of cells, and the clusters are regularly arranged along the X direction and the Y direction. The method according to claim 1 or 4, characterized in that 6. In the second step, an interval between rows in the X direction including the selection bits and an interval between columns including the selection bits in the Y direction in the cluster are equal to each other. The method described. 7. The second step is: (X direction n bits) × (Y direction m bits) within a band along the Y direction having an n bit width in the X direction on the second bitmap.
1 bit is selected from among the 1st bit, the selection is performed regularly along the Y direction, and the selection is performed so that there are a plurality of columns in the Y direction including the selection bit. The method of claim 1, wherein a map is created. 8. A charged particle beam is exposed according to the value of each bit on the first bitmap using the first bitmap data created by the method according to claim 1. Description: The value of the bit is made to correspond to the presence or absence of a beam spot on the exposure object depending on whether or not the object is irradiated, and an exposure pattern corresponding to the pattern on the first bitmap is formed on the exposure object. A charged particle beam exposure method characterized by the above. 9. The value of the bit is made to correspond to the presence or absence of a beam spot on the object to be exposed depending on whether or not the charged particle beam is irradiated onto the object to be exposed according to the value of each bit on the first bit map. , A bitmap data expansion device for expanding graphic data corresponding to the exposure pattern to the first bitmap in order to form an exposure pattern corresponding to the pattern on the first bitmap on the exposure object In the X direction with respect to the first bitmap,
X times (n ≧ 2) and m times in the Y direction perpendicular to the X direction (m ≧ 1)
A first decompressing device for decompressing to a second bit map having a resolution of, and (2n bits in X direction) on the second bit map.
When divided into x (Y direction 2 m bits) cells, 4 bits from each cell are selected so that the selection in each cell changes regularly along the X direction and the Y direction. A second rasterizing device that selects the number of rows and columns in the Y direction to be 3 or more and creates the first bitmap, and a bitmap data rasterizing device for graphic data. 10. The second expansion device is (X direction n bits) × (Y direction m bits) in a band along the Y direction having an n bit width in the X direction on the second bitmap. By selecting 1 bit from the inside, performing the selection regularly along the Y direction, and performing selection so that there are a plurality of columns in the Y direction including the selection bit, the first bitmap can be obtained. The device according to claim 9, wherein the device is created. 11. A bitmap data expansion device for graphic data according to claim 9 or 10, and a first created by the bitmap data expansion device for graphic data.
The bit map data is used to determine whether the charged particle beam irradiates the exposure object according to the value of each bit on the first bit map, and the value of the bit is determined as the presence or absence of a beam spot on the exposure object. A charged particle beam exposure system, comprising: an exposure device which corresponds to and forms an exposure pattern corresponding to the pattern on the first bitmap on the exposure object. 12. The exposure apparatus comprises a blanking aperture array in which a plurality of openings are formed in a grid on a substrate and a pair of electrodes are formed on the edge of each opening of the substrate is arranged in the optical path of the charged particle beam. The charged particle beam on the exposure object is controlled by controlling whether to apply the charged particle beam that has passed through the opening to the exposure object by applying a voltage between the pair of electrodes. The charged particle beam exposure system according to claim 11, wherein the irradiation position is raster-scanned by the scanning means.